S. Salgueiro Machado • M. A. H. Luttik J. P. van Dijken • J. A. Jongejan • J. T. Pronk
Regulation of alcohol-oxidizing capacity
in chemostat cultures of
Acetobacter pasteurianu$
Received: 20 October 1994/Received revision: 27 January 1995/Accepted: 2 February 1995
Abstract Acetobacter pasteurianus LMG 1635 was studied for its potential application in the enantioselec- tive oxidation of alcohols. Batch cultivation led to accumulation of acetic acid and loss of viability. These problems did not occur in carbon-limited chemostat cultures (dilution rate--0.05 h -~) grown on mineral medium supplemented with ethanol, L-lactate or acet- ate. Nevertheless, biomass yields were extremely low in comparison to values reported for other bacteria. Cells exhibited high oxidation rates with ethanol and ra- cemic glycidol (2,3-epoxy-l-propanol). Ethanol- and glycidol-dependent oxygen-uptake capacities of ethanol-limited cultures were higher than those of cul- tures grown on lactate or acetate. On all three carbon sources, A. pasteurianus expressed NAD-dependent and dye-linked ethanol dehydrogenase activity. Glycidol oxidation was strictly dye-linked. In contrast to the NAD-dependent ethanol dehydrogenase, the ac- tivity of dye-linked alcohol dehydrogenase depended on the carbon source and was highest in ethanol-grown cells. Cell suspensions from chemostat cultures could be stored at 4°C for over 30 days without significant loss of ethanol- and glycidol-oxidizing activity. It is concluded that ethanol-limited cultivation provides an attractive system for production of A. pasteurianus bio- mass with a high and stable alcohol-oxidizing activity.
Introduction
Bacteria of the genus Acetobacter are well known for their application in the production of acetic acid from
S. Salgueiro Machado - M. A. H. Luttik • J. P. van Dijken J. A. Jongejan • J. T. Pronk ([])
Department of Microbiology and Enzymology, Kluyver Laboratory of Biotechnology, Delft University of Technology, Julianalaan 67, 2628 BC Delft, The Netherlands.
Fax: (31) 15 782355
ethanol (Ebner and Follmann 1983). These bacteria also catalyse the oxidation of other aliphatic and aro- matic alcohols to the corresponding aldehydes and organic acids (De Ley and Kersters 1964; Asai 1968). Because of the broad substrate specificity of the dehy- drogenase systems involved in ethanol oxidation, Acetobacter species hold great promise for the produc- tion of a variety of aldehydes and organic acids.
An attractive property of microbial dehydrogenases with regard to the production of fine chemicals is that they often exhibit enantioselectivity (Hummel and Kula 1989). This also holds for the oxidation reactions catalysed by Acetobacter species. For example, when A. pasteurianus cell suspensions are incubated with ra- cemic (R,S)glycidol (2,3-epoxy-l-propanol), the S- isomer is preferentially oxidized (Geerlof et al. 1994). This enantioselective bioconversion can in principle be used for the industrial production of optically pure (R)-glycidol, which is a building block for the synthesis of a number of pharmaceutical products (Klunder et al. 1986; Kloosterman et al. 1988; Avignon-Tropis et al. 1991).
In biotechnological processes, optimum conditions for biocatalyst production (growth) and the bioconver- sion itself may differ strongly. Therefore, industrial production of fine chemicals by whole-cell biotrans- formations generally involves separate process steps for production of biomass and bioconversion processes. Optimization of the first phase of the process, i.e. devel- opment of efficient cultivation techniques yielding highly active, stable biomass, is a prerequisite for indus- trial-scale application.
The broad substrate specificity of dehydrogenase systems from Acetobacter species is in sharp contrast with the narrow range of carbon substrates for growth. For example, the only known carbon substrates for growth of A. pasteurianus are ethanol, acetate and lactate (De Ley 1958). Ethanol and lactate are rapidly oxidized to acetic acid, which accumulates in batch cultures and is only slowly oxidized to carbon dioxide
(De Ley 1958; De Ley and Schell 1959). Dissipation of the trans-membrane pH gradient by acetic acid (Alex- ander et al. 1987) is likely to contribute to the very low biomass yields of Acetobacter species in batch cultures and the sensitivity of such cultures to storage, in par- ticular in the absence of vigorous aeration (Ebner and Follmann 1983).
In theory, it should be feasible to prevent acetic acid formation by growth under carbon-limited conditions
in fed-batch or chemostat cultures. However, although the intermediary carbon metabolism and the molecular biology of electron transport in Acetobacter species have received much attention in the literature (for re- views see Asai 1968 and Matsushita et al. 1994), data on the growth of these organisms in carbon-limited chemostat cultures have not been reported.
The aim of the present study was to investigate the suitability of carbon-limited chemostat cultivation for the production of alcohol-oxidizing biomass of A. pas- teurianus. Research focused on factors relevant for the application of this organism in bioconversions such as the nature, regulation and stability of the alcohol- oxidizing activity.
Materials and methods Microorganism and maintenance
Acetobacter pasteurianus L M G 1635 (formerly called A. peroxydans) was obtained from the culture collection of the Laboratory of Microbiology, University of Gent, Belgium, and stored at - 70°C with 10% (v/v) dimethylsulphoxide in 1-ml aliquots. Samples from these frozen stock cultures were grown on slants of a solidified medium containing, per litre of demineralized water: Difco peptone 10 g, Difco yeast extract t0 g, Difco agar 20 g, ethanol 10 g, and CaCO3 20 g. Inoculum cultures for chemostat cultivation were grown in 500-ml shake flasks containing 100 ml same medium but without agar. These cultures were incubated for 48 h in a rotatory shaker (200 rpm) at 30°C.
Chemostat- cultivation
Aerobic chemostat cultivation was performed at 30°C in 2-1 Ap- plikon laboratory fermenters at a stirring speed of 1000 rpm and at a dilution rate of 0.05 h-1. The cultures were flushed with air (11 rain-1/. The dissolved-oxygen concentration in the cultures was measured with an Ingold polarographic oxygen electrode and re- mained above 25 % of air saturation. The culture pH was controlled at 6.0 by automatic addition of 2 M K O H . The condenser was connected to a cryostat and cooled at 2°C. The working volume of the culture was kept at 1.0 1 by a peristaltic pump coupled to an Applikon level controller. Biomass concentrations in samples taken directly from the culture differed by less than 1% from biomass concentrations in the culture effluent. The mineral medium con- tained, per litre of demineralized water: (NH4)2SO,~ 5 g, KH2PO,~ 3g, M g S O 4 . 7 H 2 0 0.5g, E D T A 15mg, Z n S O 4 ' 7 H 2 0 4.5rag, MnC1 z • 2 H 2 0 1.0 mg, CoC12 • 6 H 2 0 0.3 mg, CuSO 4 • 5H20 0.3 rag, N a z M o O 4 • 2 H 2 0 4 mg, CaC12 . 2 H 2 0 4.5 mg, FeSO 4 • 7H20 3 rag, KI, 0.1 mg, B D H silicone antifoam 50 bd. The mineral medium was autoclaved at 120°C. Pure ethanol, acetic acid or L-lactic acid was added aseptically to the sterile media without prior sterilization.
Control of culture purity
The purity of chemostat cultures was routinely examined by phase- contrast microscopy at 1000 x magnification. Furthermore, absence of catalase activity, a striking characteristic of this aerobic organism (Visser't Hooft 1925) was checked daily by adding 10 gl hydrogen peroxide to a 1-ml culture sample. Contaminations were detected by visible gas formation.
Determination of culture dry weight
Dry weights of culture samples were determined using a microwave oven and 0.45-gin membrane filters as described by Postma et al. (1989). The dry weight of parallel samples varied by less than 1%.
Analysis of substrates and metabolites
Concentrations of ethanol, acetate and lactate in reservoir media and culture supernatants were determined by H P L C on a Rezex organic-acid column (Phenomenex, 300 x 7.8 ram) at 60°C. Detec- tion of acetate and lactate was by means of a Waters 441 UV detector at 214 nm, coupled to a Waters data module. Detection of ethanol was by means of an Erma 7515 A refractive-index detector. Peak areas were linearly proportional to concentrations. The detec- tion limits for acetate, lactate and ethanol were approximately 0.05 raM, 0.05 mM, and 0.5 m M respectively.
Gas analysis
On-line measurement of oxygen-uptake rates of chemostat cultures was performed as described by Van Urk et al. (1988).
Preparation of cell-free extracts
Samples of steady-state chemostat cultures containing 25-50 mg dry weight were harvested by centrifugation, washed twice and resusp- ended in 4 ml ice-cold 0.1 M potassium phosphate buffer (pH 7.5), containing 1 m M dithiothreitol and 2 m M MgC12. Extracts were prepared by sonication with 0.7-mm-diameter glass beads at 0°C for 2 min at 0.5-min intervals with an MSE sonicator (150 W output, 8 I.tm peak-to-peak amplitude). Unbroken cells and debris were removed by centrifugation at 4°C (20 min at 75 000 9). The clear supernatant, typically containing 1-3 mg protein ml-1, was used as the cell-free extract.
Protein determination
Protein concentrations in cell-free extracts were estimated by the Lowry method. The protein content of culture samples was esti- mated by a modified biuret method (Verduyn et al. 1992). In both assays, bovine serum albumin (fatty-acid-free, Sigma) was used as a standard.
Substrate-dependent oxygen' uptake
Substrate-dependent oxygen uptake by culture samples was assayed polarographically with a Clark-type oxygen electrode (Yellow Springs Instruments Inc., Yellow Springs, Ohio, USA) at 30°C after appropriate dilution in air-saturated buffer (100mM potassium phosphate, 10 m M MgSO 4, pH 6.0). Calculations of specific rates
were based on an oxygen concentration in air-saturated buffer of 236 pM at 30°C. Endogenous respiration rates by chemostat-grown cells were negligible.
b s u l N
Growth of A. pasteurianus in chemostat cultures
Enzyme assays
Enzyme assays were performed immediately after preparation of the cell-free extracts. All enzyme assays were carried out at 30°C. In all activity measurements, the reaction rate was linearly proportional to the amount of cell-free extract added.
Alcohol dehydrogenase (pyridine-nucleotide dependent)
The reaction mixture (1 ml) contained: 50 gmol glycine/KOH buffer (pH 9.0), 1 Ixmol N A D or N A D P , 0.05% (v/v) Triton X-100 and cell-free extract. The reaction was started by the addition of either 10 gmol ethanol or 50 gmol racemic glycidol. Formation of NAD(P)H was monitored at 340 nm (e = 6.22 m M -1 cm -~) in a Hitachi spectrophotometer model 100-60.
Acetaldehyde dehydrogenase (pyridine-nucleotide dependent) The reaction mixture (1 ml) contained 100 gmol potassium phos- phate buffer (pH 8.0), 15 gmol pyrazole, 0.4 gmol dithiothreitol, 0.4 gmol N A D or NADP, 10 gmol KC1, 0.05% (v/v) Triton X-100 and cell-free extract. The reaction was started by the addition of 5 p-mol acetaldehyde. Formation of NAD(P)H was monitored at 340 nm in a Hitachi spectrophotometer model 100--60.
Dye-linked alcohol and acetaldehyde dehydrogenases
Activities of dye-linked [probably pyrroloquinoline-quinone (PQQ) dependent; Matsushita et al. 1994] alcohol and aldehyde dehydro- genases were assayed polarographically with a Clark-type oxygen electrode (Yellow Springs Instruments Inc., Yellow Springs, Ohio, USA) at 30°C, using the artificial electron acceptor phenazine methosulphate (PMS) as a mediator. The values presented have been calculated based on an oxygen concentration in air-saturated buffer of 236 gM at 30°C. The reaction mixture (4 ml) contained 400 gmol potassium phosphate buffer (pH 6.0), 8 gmol MgC12, 1.2 gmol (PMS), 26 000 U catalase (from bovine liver, Boehringer) and cell-free extract. The reaction was started by the addition of either 40 gmol ethanol, 200 gmol racemic glycidol or 20 gmol acet- aldehyde.
Stability of alcohol-oxidizing capacity
To determine the stability of the alcohol-oxidizing activity of A.
pasteurianus upon storage; 100-ml samples from steady-state chemostat cultures were transferred tO 100-ml serum flasks. The closed flasks were incubated statically at 4°C and 30°C. At appropri- ate intervals, the contents of the flasks were mixed by inverting them, and samples were withdrawn for determination of substrate-depen- dent oxygen uptake as described above.
Chemicals
Racemic (R, S)-glycidol was obtained from Janssen Chimica and distilled before use to remove polymerized material.
Preliminary experiments with shake-flask cultures demonstrated that A. pasteurianus L M G 1635 is ca- pable of growth on defined mineral salts media without vitamins. However, growth in shake-flask cultures led to rapid loss of viability and alcohol-oxidizing activity. This was probably due to the observed accumulation of acetic acid and concomitant acidification of the growth medium (results not shown). Since the maximum speci- fic growth rate in shake-flask cultures grown in mineral medium with ethanol was about 0.1 h - 1, it was decided to study growth in chemostat cultures at a dilution rate of 0.05 h - 1.
Problems related to the accumulation of acetate were also encountered during the start-up phase of A. pas- teurianus chemostat cultures on ethanol. These prob- lems could be avoided by keeping the initial concentra- tion of ethanol in the growth medium at 5 g 1-1. Once all substrate had been consumed, it was then possible to feed the cultures with higher concentrations of ethanol. Using this strategy, steady-state chemostat cultures could be obtained at reservoir concentrations of ethanol up to 20 g 1-1. Steady-state conditions (i.e. no significant changes in biomass yield and oxygen consumption over a period of 3 days) were generally established after about 10-15 volume exchanges.
In ethanol-grown chemostat cultures, the biomass yield on ethanol was identical at reservoir concentra- tions of 10 g1-1 and 20 g1-1, and no residual ethanol or acetate was detectable in culture supernatants. These observations confirmed that ethanol was the growth- limiting nutrient. Biomass yields on ethanol were very low and growth was accompanied by high oxygen- uptake rates (Table 1). In practice, these high oxy- gen-uptake rates imposed a limit to the biomass concentrations that could be achieved in the chemostat cultures with normal aeration. Nevertheless, biomass
Table I Biomass yield on substrate (Y~x), protein content and speci- fic oxygen-uptake rates (qO2) in chemostat cultures of A. pas- teurianus L M G 1635 with ethanol, acetate or L-Iactate as the growth-limiting nutrient. Growth conditions: dilution, rate = 0.05h -1, T = 30°C, pH 6, PO2 > 25% air saturation. Data are presented as average ± standard deviation of measurements from at least two independent chem0stat cultures grown at different reser- voir concentrations of the growth-limiting nutrient (10g1-1 or 20 gl - 1)
Carbon Y~x Protein qO2
source (g dry weight content (%) (mmol g - 1 h - 1) mol - 2)
Ethanol 6.1 _+ 0.1 68 _+ 1 23 ± 2 Acetic acid 4.1 _+ 0.1 63 _+ 1 23 + 1 L-Lactic acid 11.4 _+ 0.6 64 _+ 1 12 + 1
concentrations exceeding 2 g dry weight 1-1 could be attained in cultures grown on ethanol.
Also with acetic acid or lactic acid as the sole carbon source, steady-state chemostat cultures were obtained at a dilution rate of 0.05 h - 1. Again, no residual sub- strate was detected in these cultures and biomass con- centrations were proportional to the concentration of acetic acid or lactic acid in the reservoir medium (Table 1). As observed with the ethanol-limited cultures, bio- mass yields were very low. The carbon source for growth did not appear to have a substantial effect on the protein content of the biomass, which was about 65% (Table 1).
Regulation of alcohol-oxidizing capacity of intact cells
To investigate regulation of ethanol- and glycidol- oxidizing activity in A. pasteur±anus, substrate-depen- dent oxygen uptake rates were measured in cell suspen- sions harvested from the chemostat cultures.
Although the ability to oxidize ethanol and glycidol appeared to be expressed constitutively, activities were three- to sixfold higher in ethanol-grown cells than in samples from acetate- or lactate-limited chemostat cul- tures (Tables 1, 2). The ethanol-dependent oxygen-up- take capacity of ethanol-grown cells was fourfold high- er than the in situ oxygen uptake rate in ethanol-lim- ited chemostat cultures. The ratio between ethanol- and glycidol-oxidizing activity was not constant in the different cultures (Table 2), suggesting the involvement of at least two enzymes in the oxidation of these com- pounds. To investigate further which enzymes are in- volved in the oxidation of ethanol and glycidol, activ- ities of alcohol-oxidizing enzymes were measured in cell-free extracts.
Regulation of alcohol and acetaldehyde dehydrogenases
Cell-free extracts of A. pasteur±anus exhibited both NAD-dependent and dye-linked (probably PQQ-de- pendent; Matsushita et al. 1994) ethanol dehydrogen- ase activities, but no activity was detected with N A D P as the electron acceptor (Table 3). Oxidation of glycidol was observed with the artificial electron acceptor PMS, but not with N A D or N A D P (Table 3). Acetaldehyde oxidation was observed with NAD, N A D P and the artificial electron acceptor PMS (Table 3).
NAD-dependent ethanol dehydrogenase was ex- pressed at a constant level in ethanol-, lactate- and acetate-grown cultures. In contrast, the activity of dye- linked ethanol dehydrogenase depended on the carbon source on which the cells were grown, the highest activity being measured in ethanol-grown cells (Table 3). A similar pattern was observed for the dye-linked glycidol-oxidizing activity (Table 3).
Table 2 Ethanol- and glycidol-dependent oxygen-uptake rates of Acetobacter pasteur±anus L M G 1635 grown in carbon-limited chemostat cultures (growth conditions as in legend to Table 1). Data are presented as average ± standard deviation of data from at least two independent chemostat culture experiments
Growth-limiting substrate
Substrate-dependent oxygen-uptake capacity (mmol 02 g - 1 h - 1)
10 m M ethanol 50 m M (R, S)-glycidol Ethanol 97 + 4 56 ± 3
Acetic acid 39 _+ 2 10 + 1 L-Lactic acid 29 ± 2 11 ± 1
Table 3 Activities of ethanol-, glycidol- and acetaldehyde-oxidizing enzymes in cell-free extracts of A. pasteur±anus L M G 1635 grown in aerobic, substrate-limited chemostat cultures. Activities are given as average ± standard deviation of data from at least two independent chemostat culture experiments. (ND not determined, PMS phenazine methosulphate)
Dehydrogenase Enzyme activity (U mg protein-1) reaction with growth-limiting substrate:
Ethanol Lactate Acetate Ethanol: N A D 0.21 ± 0.02 0.21 ± 0.02 0.23 ± 0.04 Ethanol: N A D P < 0.01 n.d. < 0.01 Ethanol: PMS 0.74 _+ 0.27 0.24 ± 0.03 0.11 ± 0.01 Glycidol: N A D < 0.01 < 0.01 < 0.01 Glycidol: N A D P < 0.01 N D < 0.01 Glycidol: PMS 0.45 ± 0.17 0.15 ff 0.06 0.04 ± 0.01 Acetaldehyde: N A D 0.11 ± 0.01 0.26 ± 0.08 0.10 ± 0.01 Acetaldehyde: N A D P 0.34 + 0.06 0.89 ± 0.02 0.40 ± 0.01 Acetaldehyde: PMS 0.14 ± 0.01 0.15 ± 0.10 0.11 ± 0.01
Dye-linked acetaldehyde dehydrogenase was ex- pressed constitutively at a constant level. Both NAD- and NADP-dependent acetaldehyde dehydrogenase activities were two- to threefold higher in lactate-grown cells than in extracts from ethanol- or acetate-grown cultures (Table 3).
Stability of alcohol-oxidizing activity of cell suspensions
Stability of the biocatalyst upon storage is a prerequi- site for whole-cell biotransformations in which growth of the biomass and the conversion itself are separated in time. To investigate the stability of the ethanol- and glycidol-oxidizing activity of A. pasteur±anus, cell sus- pensions from ethanol-limited chemostat cultures were stored at 4°C and 30°C without further treatment.
At 4°C, no significant loss of ethanol- or glycidol- oxidizing activity was observed after storage for 30 days (Fig. t). At 30°C, the activity declined with a half- life of approximately 10-15 days (Fig. 1). Since the
1 0 0 T 0 E E so 0 c ® X 0 0
" " • O
-" ¢ B~II • i 0 10 20 30 Time (days) 40Fig. 1 Stability of ethanol- and glycidol-oxidizing activities of cell suspensions of A. pasteurianus pregrown in ethanol-limited chemo- stat cultures. Culture samples were incubated at 4°C (O I ) or 30°C (O []). Ethanol (0 O)- and glycidol ( I [])- dependent oxygen-up- take rates were measured at 30°C with a Clark-type oxygen elec- trode
samples were not stored under aseptic conditions, fun- gal contaminations occurred after prolonged storage at 30°C. This problem did not occur after 30 days of storage at 4°C.
Discussion
Carbon-limited chemostat cultivation of A. pas- teurianus biomass was applied and resulted in high alcohol-oxidizing activities (Tables 2, 3) and an excel- lent storage stability (Fig. 1). Indeed, carbon-limited cultivation appeared to be a prerequisite for the pro- duction of stable alcohol-oxidizing biomass, since at- tempts to grow A. pasteurianus in batch cultures inva- riably resulted in loss of viability as well as alcohol- oxidizing capacity (data not shown). On the basis of the alcohol-oxidizing capacities of cell suspensions (Table 2) and cell-free extracts (Table 3), ethanol was identified as the best carbon source for production of A. pas- teurianus biomass.
Enzyme activity assays in cell-free extracts indicated that a PQQ-dependent (dye-linked) alcohol dehydro- genase (Matsushita et al. 1994) was responsible for the oxidation of glycidol. The periplasmic localization of this enzyme (Matsushita et al. 1994) may be advantage- ous for bioconversions that involve toxic substrates and/or products. The observed ratio of dye-linked and NAD-dependent ethanol dehydrogenase activities, and the similar regulation of ethanol-oxidizing activity of cell suspensions and dye-linked ethanol dehydrogenase activity in cell-free extracts, indicate that PQQ-depen- dent ethanol dehydrogenase is probably predomi- nantly responsible for ethanol oxidation in A. pas- teurianus under conditions of substrate excess. The high activities of PQQ-dependent alcohol dehydrogenase in
extracts of ethanol-grown cells are consistent with the recent report of T a k e m u r a et al. (1993) who showed that the alcohol dehydrogenase gene cluster in a differ- ent A. pasteurianus strain is induced by ethanol.
The biomass yields of A. pasteurianus in carbon- limited chemostat cultures (Table 1) are three- to five- fold lower than those reported for other heterotrophic, aerobic bacteria (Heijnen and Roels 1981). The fact that these low yields were found despite the absence of acetate accumulation indicates that the low growth efficiency in batch cultures is not solely due to uncoup- ling by this metabolite. Instead, the low biomass yields must be due to intrinsic metabolic properties of the organism, which remain to be identified. The biomass yields reported in Table 1 have been measured at a single, low growth rate. It is therefore not clear if the low yields are due to a low maximum ('true') yield, a high maintenance-energy requirement, or both.
The high ethanol- and glycidol-oxidizing activities of A. pasteurianus are consistent with the relation between growth efficiency and m a x i m u m rate of production of catabolic metabolites proposed by Linton (1990). A. pasteurianus clearly represents an extreme case with respect to its apparent low energetic efficiency. There- fore, in addition to the potential benefits for the pro- duction of glycidol-oxidizing biomass, further research into the bioenergetics of this and other acetic acid bacteria is also of fundamental interest.
Acknowledgements Sonia Salgueiro Machado acknowledges the financial support of the Conselho Nacional de Desenvolvimento Cientifico e Technologico (Brazil). We thank our colleague Dr. Arie Geerlof for many stimulating discussions.
References
Alexander B, Leach S, Ingledew WJ (1987) The relationship between chemiosmotic parameters and sensitivity to anions and organic acids in the acidophile Thiobacillusferrooxidans. J Gen Microbi- ol 133:1171-1179
Asai T (1968) Acetic acid bacteria. Classification and biochemical activities. University of Tokyo Press, Japan
Avignon-Tropis M, Treilhou M, Pougny JR, Fr6chard-Ortuno I, Linstrumelle G (1991) Total synthesis of ( + )-leukotriene B4 methyl ester and its 5-epimer from (R)-glycidol. Tetrahedron 49:7279-7286
De Ley J (1958) Studies on the metabolism of Acetobacter per- oxydans. I. General properties and taxonomic position of the species. Antonie van Leeuwenhoek 24:281-297
De Ley J, Kersters K (1964) Oxidation of aliphatic glycols by acetic bacteria. Bacteriol Rev 28:164-180
De Ley J, Schell J (1959). Studies on the metabolism of Acetobacter peroxydans. II. The enzymic mechanism of lactate metabolism. Biochim Biophys Acta 35:154-165
Ebner H, Follmann H (1983) Acetic acid. In: Dellweg H (ed) Bio- technology, vol 3. Biomass, microorganisms for special applica- tions, microbial products, energy from renewable sources. VCH, Weinheim, pp 387-407
Geerlof A, Tol JBA van, Jongejan JA, Duine JA (1994) Enantioselec- rive conversions of the racemic C3-alcohol synthons, glycidol
(2,3-epoxy-l-propanol), and solketal (2,2-dimethyl-4-hy- droxymethyl)-l,3-dioxolane) by quinohaemoprotein alcohol de- hydrogenases and bacteria containing such enzymes. Biosci Bio- technol Biochem 58:1028-1036
Heijnen JJ, Roels JA (1981) A macroscopic model describing yield and maintenance relationships in aerobic fermentation. Biotech- nol Bioeng 23: 739-763
Hummel W, Kula MR (1989) Dehydrogenases for the synthesis of chiral compounds. Eur J Biochem 184:1-13
Kloosterman M, Elferink VHM, van Iersel J, Roskam JH, Meijer EM, Hulshof LA, Sheldon RA (1988) Lipase in the preparation of 13-blockers. Trends Biotechnol 6:251-256
Klunder JM, Ko SY, Sharpless KB (1986) Assymmetric epoxidation of allyl alcohol: efficient routes to homochiral 13-adrenergic blocking agents. J Org Chem 51:3710-3712
Linton JD (1990) The relationship between metabolite production and the growth efficiency of the producing organism. FEMS Microbiol Rev 75:1-18
Matsushita K, Toyama H, Adachi O (1994) Respiratory chains and bioenergetics of acetic acid bacteria. Adv Microb Physiol 36:247-301
Postma E, Verduyn C, Scheffers WA, van Dijken JP (1989) Enzymic analysis of the Crabtree effect in glucose-limited cultures of Saccharomyces cerevisiae. Appl Environ Microbiol 55:468-477
Takemura H, Kondo K, Horinouchi S, Beppu T. (1993) Induction by ethanol of alcohol dehydrogenase activity in Acetobacter pasteurianus. J Bacteriol 175:6857-6866
Van Urk H, Mak PR, Scheffers WA, Dijken JP van. (1988) Meta- bolic responses of Saccharomyces cerevisiae CBS 8066 and Can- dida utilis CBS 621 upon transition from glucose limitation to glucose excess. Yeast 4:283-291
Verduyn C, Postma E, Scheffers WA, Dijken JP van. (1992) Effect of benzoic acid on metabolic fluxes in yeasts: a continuous-culture study on the regulation of respiration and alcoholic fermenta- tion. Yeast 8:501-517
Visser't Hooft F (1925) Biochemische onderzoekingen over het ge- slacht Acetobacter. PhD Thesis, Delft University of Technology, Delft
