Arch Microbiol (1984) 137 : 33- 41
Archives of
Microbiology
9 Springer-Veriag 1984
Fermentation of 2,3-butanediol by Pelobacter carbinolicus sp. nov.
and
Pelobacter propionicus sp. nov.,
and evidence for propionate formation from C2 compounds
Bernhard Sehink
Fakultfit f/Jr Biologie, Universit/it Konstanz, Postfach 5560, D-7750 Konstanz, Federal Republic of Germany
Abstract.
From anaerobic enrichments with 2,3-butanediol as
sole substrate pure cultures of new Gram-negative, strictly
anaerobic, non-sporeforming bacteria were isolated. Similar
isolates were obtained with acetoin as substrate. From marine
muds in saltwater medium a short rod (strain Gra Bd 1) was
isolated which fermented butanediol, acetoin and ethylene
glycol to acetate and ethanol. The DNA base ratio of this
strain was 52.3 mol % guanine plus cytosine.
From freshwater sediments and sewage sludge, a different
type of short rod (strain Ott Bd 1) was isolated in freshwater
medium, which fermented butanediol, acetoin, ethanol, lac-
tate and pyrnvate stoichiometrically to acetate and pro-
pionate. Propanol and butanol were oxidized to the respective
fatty acids with concomitant reduction of acetate and bicar-
bonate to propionate. The DNA base ratio of strain Ott Bd 1
was 57.4mo1~ guanine plus cytosine, No other substrates
were used by the isolates, and no other products could be
detected. In cocultures with
Acetobacterium woodii
or
Methanospirillum hungatei,
strain Gra Bd I also grew on
ethanol, propanol, and butanol by fermenting these alcohols
to the respective fatty acids and molecular hydrogen.
Cytochromes could not be detected in any of the new isolates.
Since both types of bacteria can not be affiliated to any of the
existing genera and species, the new species
Pelobacter
carbinolicus
and
Pelobacter propionicus
are proposed. The
mechanism of butanediol degradation and propionate for-
mation from acetate as well as the ecological importance of
both processes are discussed.
Key words: Pelobacter carbinolicus - Pelobacter propionicus
- Species description - 2,3-Butanediol - Acetoin -
Anaerobic degradation - Ethanol metabolism - COa
reduction - Propionate formation
2,3-Butanediol (Butyleneglycol) is an important end product
of fermentations, and is produced mainly by Entero-
bacteriaceae,
Bacillus
species, and Lactobacteriaceae. The
synthesis proceeds either via condensation of two molecules
pyruvate to acetyl lactate, decarboxylation to acetoin and
reduction to the diol as in Enterobacteriaceae and
Bacillus
sp.,
or by condensation of acetyl phosphate and a thiamine
pyrophosphate-bound acetaldehyde residue to diacetyl and
subsequent reduction to acetoin and butanediol as in most
lactic acid bacteria (Speckmann and Collins 1968).
Butanediol dehydrogenase activity is also indffced in strictly
Offprint requests to:
B. Schink
aerobic bacteria under conditions of oxygen limitation as
observed with the knallgas bacterium
Alcaligenes eutrophus
(Schlegel and Vollbrecht 1980). In nearly all of these bacteria,
the butanediol produced is a mixture of all three possible
stereoisomers, the D(-), the L(+), and the meso-isomer (Long
and Patrick 1963). Since butanediol can be chemically
converted to butadiene, an important precursor of synthetic
rubber, production of butanediol from wheat and barley
mashes and waste materials like molasses and whey was
actively studied at the end of World War II (Ledingham et al.
1945; Adams and Stanier 1945). Now, these processes gain
interest again with rising oil prices (Kosikowski 1979; Palsson
et al. 1981; Speckmann and Collins 1982).
For the ecologist the question arose whether butanediol is
formed in anaerobic sediments and, if so, how such a highly
reduced compound can be degraded anaerobically since it
does not accumulate in anaerobic environments. The present
paper reports on new bacteria which anaerobically convert
butanediol and acetoin to methanogenic substrates.
Materials and methods
The following strains were isolated in pure culture from
enrichment cultures inoculated with mud samples:
Strain Gra Bd 1 from anoxic mud of Canale Grande, a
channel in the city of Venice, Italy.
Strain Ott Bd 1 from black anoxic mud of a freshwater
creek near Hannover, FRG.
Strain G6 Bd 1 and Ko Bd 5 from anoxic digester sludge
of municipal sewage plants at G6ttingen and Konstanz, re-
spectively, in Germany.
Methanospirillum hungatei
strain
M 1 h was isolated from digested sludge of the sewage plant at
G6ttingen, FRG.
Acetobacterium woodii
strain NZ Va 16 was
obtained from R. Bache.
A. woodii
strain Gra EG 12,
DSM2396 was isolated with ethylene glycol as substrate
(Schink and Stieb 1983).
Desu~ovibrio vulgaris
strain Marburg was kindly pro-
vided by Prof. Dr. R. K. Thauer, Marburg.
Desulfobulbus
propionicus
strain 1 pr3 was a gift of Dr. Fritz Widdel,
Konstanz.
Media and growth conditions
Carbonate-buffered, sulfide-reduced mineral medium with
low phosphate content was prepared as described earlier
(Widdel and Pfennig 1981 ; Schink and Pfennig 1982a). Trace
element solution SL 7 and vitamin solution (Pfennig 1978)
were added to the complete autoclaved medium. The pH was
First publ. in: Archives of Microbiology 137 (1984), 1, pp. 33-41
Konstanzer Online-Publikations-System (KOPS)
URL: http://www.ub.uni-konstanz.de/kops/volltexte/2007/2513/
URN: http://nbn-resolving.de/urn:nbn:de:bsz:352-opus-25130
34
adjusted to 7.2-7.4. All cultivation procedures were essen-
tially the same as described earlier (Schink and Pfennig
1982a).
Growth was followed in 20 ml tubes in a Bausch and
Lomb Spectronic 70 spectrophotometer. Some tests for
characterization were carried out with commercial media
systems (API20A, BioMerieux, Nfirtingen, FRG). All
growth tests were carried out at least in duplicates at 28~
Isolation
Pure cultures were obtained by repeated application of the
agar shake culture method described by Pfennig (1978).
Tubes were gassed with N2/CO2 mixture (80 ~/20 ~) and
sealed with butyl rubber stoppers. Purity was checked
microscopically and also by growth tests in complex medium
(AC medium, Difco Laboratories, Detroit, MI, USA) with
and without addition of 2 ~ sodium chloride and /0raM
butanediol. Gram staining was carried out according to
Magee et al. (1975) without counterstaining. An unidentified
Bacillus sp. and Escherichia coli were used as controls.
Chemical analyses
Sulfide formation was analyzed by the methylene blue
method (Cline 1969). Formation of nitrite from nitrate was
assayed by azo dye formation with sulfanilic acid and
e-naphthylamine. Butanediol, other alcohols, volatile fatty
acids and methane were assayed by gas chromatographic
procedures as described (Schink and Pfennig 1982a).
The DNA base composition was determined with the
thermal denaturation method according to DeLey (1970)
after extraction according to Marmur (1961).
Cytochromes were assayed in french pressure cell extracts
of butanediol-grown cells. Crude extracts as well as mem-
brane fractions prepared by 120rain centrifugation at
100,000 x g in a Beckman centrifuge were subjected to
difference spectroscopy in a Gilford model 250 spectro-
photometer.
Protein was determined in crude cell extracts after Bei-
senherz et al. (1953).
Enzyme assays
All enzyme assays were carried out with crude French
pressure extracts of butanediol-grown cells after spinning
down unbroken cell debris at 10,000xg for 15rain.
Hydrogenase was assayed spectrophotometrically with
benzyl viologen as electron acceptor after Schink and
Schlegel (1979). Acetate kinase and alcohol dehydrogenase
were measured by standard methods (Bergmeyer 1974).
Butanediol dehydrogenase was assayed as described by
H6hn-Bentz and Radler (1978). Acetoin dehydrogenase was
measured with the same buffer system as NADH-dependent
reduction of diacetyl. Phosphate acetyl transferase was
quantified by a procedure modified after Bergmeyer (1974).
Increase in absorption at 233 ~tm wavelength was followed in
a 1 ml cuvette which contained 0.9 ml 0.1 M Tris-HC1 buffer,
pH 7.6, 30 ~tl 0.1 M acetylphosphate (K-Li-Salt), 30 ~tl 10 mM
coenzyme a, and 40 gl of a suited cell extract dilution. The
reaction was started with acetyl phosphate. A Zeiss PM4
spectrophotometer was used for all enzyme assays.
Chemicals
All chemicals were of reagent grade quality and obtained
from Merck, Darmstadt; Serva, Heidelberg; and Fluka, Neu-
Ulm, FRG.
Results
Enrichment, isolation and enumeration
50 ml enrichments in saltwater and freshwater medium with
10raM 2.3-butanediol as substrate were inoculated with
3 - 5 ml of anoxic mud from creeks, sewage plants and marine
sediments. Gas production started after 4 days and ceased
after about 3 weeks. In subcultures on the same media,
turbidity developed within 2-5 days. After two further
transfers, isolation of butanediol-degrading bacteria was
attempted in agar shake series in the presence and absence of
excess of Desulfovibrio vulgaris cells and sulfate added as a
hydrogen sink. In all cases small lens-shaped, slightly yel-
lowish colonies developed which were again subjected to agar
shake series dilution without D. vulgaris cells. Pure cultures
were isolated from the last positive dilution tubes (strains Gra
Bd 1, Ott Bd 1, G6 Bd 1).
Enumerations of butanediol-degrading bacteria were car-
fled out by the three-tube most probable number technique
(American Public Health Association 1969). In mud of
Canale Grande, Venice, 1100 cells per ml of mud were
counted. The prevailing organisms in the last positive tubes
were short rods similar to strain Gra Bd 1, and acetate was the
main fermentation product formed to about 20 mM con-
centration from 10 mM butanediol. In muds of freshwater
creeks near Hannover and near Konstanz, 1,500 cells per ml
of mud were counted. The prevailing organisms were short
rods similar to strain Ott Bd 1. In digested sludge of the
sewage plant in Konstanz, 2,400 cells per ml were counted.
The prevailing organism was isolated from the last positive
tube as strain Ko Bd 5. The fermentation products in the last
positive tubes inoculated from freshwater sources were
acetate and propionate, and were found in nearly equal molar
amounts.
General properties of the new isolates
All strains grew well in their respective isolation media. Yeast
extract was not required but enhanced cell yields slightly.
Vitamins and the trace elements molybdenum, tungsten and
selenium, although present in the isolation medium, were not
required. Addition of phosphate inhibited growth at con-
centrations higher than 20 retool/1. The marine isolate Gra
Bd I thrived well also in freshwater medium, whereas the
freshwater isolates were inhibited at salt concentrations
higher than 1%. In none of our isolates cytochromes could be
detected, neither in redox difference spectra nor in absorp-
tion spectra of air-oxidized or dithionite-reduced crude cell
extracts or membrane fractions. None of the new isolates
reduced nitrate, sulfate, sulfur, thiosulfate, or sulfite.
Characterization of the marine isolate, stra& Gra Bd 1
In phase contrast microscopy, cells of strain Gra Bdl
appeared as nonmotile, straight short rods, 0.5-0.7 x 1.2-
3.0 ~tm in size, with rounded ends, often in pairs (Fig. 1 a). In
Indian ink preparations, thin layers of extracellular capsular
material were visible (not shown). The Gram reaction was
35
Fig. 1 a-e
Phase contrast photomicrographs
of butanediol-fermenting bacterial
isolates grown on butanediol. Bar
equals 10 gm. a
Pelobacter
carbinolicus
strain Gra Bd 1,
b Pelobacter propionicus
strain
Ott Bd 1, e
Pelobacter propionicus
strain G6 Bd 1
Fig. 2a and b
Electron micrographs of ultrathin
sections of butanediol-fermenting
isolates. Bar equals 0.2 ~tm.
a Pelobacter carbinolicus
strain
Gra Bd 1, b
Pelobacter propionicus
strain Ott Bd i
negative. Ultrathin sections revealed a multilayered cell wall
typical of Gram-negative bacteria (Fig. 2a). Spore formation
was never observed. The guanine-plus-cytosine content of the
DNA was 52.3 _+ 1.0 tool ~. The only substrates degraded in
pure culture were 2.3-butanediol, acetoin, and ethylene glycol
(Table 1). They were fermented stoichiometrically to ethanol
and acetate (Table 2). The molar growth yield on acetoin was
nearly twice as high as on butanediol and ethylene glycol. In
cocultures of strain Gra Bd 1 with either
Acetobacterium
woodii
or
Methanospirillum hungatei
as hydrogen utilizing
partners, no ethanol was formed. The growth yield on
butanediol increased about twice and growth was also
possible on ethanol, propanol and butanol; on the latter two
substrates only in the presence of acetate (Table 2). No other
substrates were used (see Table 1). The growth curve (Fig. 3 a)
shows the dependence of growth and product formation on
butanediol utilization. Optimal growth ~=0.087h -1,
t a = 8.0h) was found between 35 and 40~ no growth
occurred at 10~ and 45~ The pH optimum was at pH
6.5-7.2, the pH limits were pH 6.0 and 8.0.
Characterization of freshwater isolates
Strains Ott Bd 1 and G6 Bd I both appeared in phase contrast
microscopy as nonmotile, straight rods with rounded ends,
0.5-0.7 x 1.2-6.0gm in size. Pairs and chains of cells of
varying length were often observed (Fig. 1 b, c). In Indian ink
preparations, thin layers of capsular material were visible.
The Gram reaction was negative. Ultrathin sections revealed
a multilayered cell wall typical of Gram-negative bacteria
(Fig. 2b). Spores were never found. The guanine-plus-
cytosine content of the DNA was 57.4 _+ 1.0 ~. The only
Table l. Substrates tested for growth of butanediol-fermenting
strains of
Pelobacter carbinolicus
and
Pelobacter propionicus
Substrate
Pelobacter Pelobaeter propionicus
(10 mmol/1)
carbinolicus
strains
strain
Gra Bd I Ott Bd t G6 Bd 1 Ko Bd 5
2,3-Butanediol + + + +
Acetoin + + + +
Diacetyl - - - +
Ethanol _
a + + _
Propanol
_ a + b + b _
Butanol -" + b + b _
Ethylene glycol + - - -
Lactate - + + +_
Pyruvate - + + --
a Growthwas possible in mixed culture
withAcetobacterium woodii
or
Methanospirillurn hungatei
b Growth was only possible in the presence of 10raM acetate
Substrates not utilized:
- Methanol, isopropanol, glycerol, 1,2-propanediol, 1,2-butane-
diol, 1,3-butanediol, 1,4-butanediol, H2/CO 2 + acetate
-
Formate, acetate, glyoxylate, glycolate, oxalate, malonate, suc-
cinate, fumarate, malate, glyeerate, oxaloacetate, tartrate, citrate
-
Glucose, fructose, mannose, lactose, xylose, rhamnose, maltose,
sucrose, cellobiose, mannitol, melezitose, raffinose, sorbose,
salicin
-
Casamino acids, peptone, yeast extract
- Urea, gelatine or esculin not hydrolyzed
No catalase activity
- Substrate concentrations in growth assays were as follows:
Alcohols and acids: t0 mmol/1, sugars and others: 0.1 ~ (w/v)
36
Table 2. Stoichiometry of fermentation and growth yields of marine isolate
Petobacter carbinoticus
strain Gra Bd 1 in pure and mixed culture
Substrate
Amount of Cell dry Substrate Products formed (~tmol) Growth Carbon
substrate weight assimilated yield recovery
degraded formed (gmol) ~ Acetate Ethanol Propio- Methane mg/mmol %
(pmol) (mg)" nate substrate
utilized
Pure culture
Butanediol 200 0.72 7.2 107 303 - - 3.60 104.3
Acetoin 200 1.39 13.9 198 176 - - 6.95 100.4
Ethylene glycol 200 0.72 14.4 104 82 - - 3.60 100.2
Mixed culture with
A. woodii
Butanediol 200 1.75 17.5 542 - - - 8.75 100.1 ~
Ethanol 200 0.86 17.2 308 - - - 4.3 108.1 ~
Propanol 200 0.91 18.2 89 - 204 - 4.6 105.1 ~
Mixed culture with
M. hungatei
Butanediol 200 1.82 18.2 392 - - 146 9.1 102.Y
Ethanol 200 0.82 16.4 186 - - 102 4.1 101.2 ~
Experiments were carried out in 20ml tubes. All figures are means of at least two independent agsays
Cell dry weights were calculated by cell density using the conversion factor 0.1 OD65 o ~ 24.2 mg dry weight per 1, which was obtained by
direct determination in 500 ml cultures
b Substrate assimilated was calculated using the formula < C4H703 > for cell material
Carbon recovery in mixed cultures was calculated after e.g. the following equations:
4 CH3CHOHCHOHCH3 + 6 HCO3 ~ 11 CH3COO- + 5 H + + 4 H20
4 CHsCHOHCHOHCH3 + 3 HCO3 ~ 8 CH3COO- + 5 H + + 3 CH 4 + HaO
0.3 g ,'
0,2 c
0,1
E
o
0 5 10 15 20 "-t23
b
.
5 10 15 20
10 t~
o
LU
TIME (h)
Fig. 3 a and b. Fermentation time course of butanediol-fermenting
isolates. Experiments were performed at 30~ in 20 ml tubes sealed
with Bellco rubber septa. Samples for substrate and product analysis
were removed by a syringe at times indicated and the head-spaces
were flushed with N2/CO z gas mixtures. (O) Optical density,
(O) butanediol, (A) acetate, (A) ethanol, (V) propionate, a
Pelo-
baeter carbinolicus
strain Gra Bd 1, b
Pelobacter propionicus
strain
Ott Bd I
substrates degraded were 2.3-butanediol, acetoin, ethanol,
pyruvate and lactate; in the presence of acetate, also
n-propanol and n-butanol were fermented. The only fermen-
tation products detected were acetate and propionate.
Propanol and butanol were oxidized to propionate and
butyrate with concomitant reduction of acetate and bicar-
bonate to propionate (Table 3). The highest growth yield was
obtained with acetoin as substrate; the yields on lactate and
butanediol were equal, whereas on ethanol, propanol and
butanol the yields were very small. In carbonate-free medium
containing 20 mM potassium phosphate as buffer, strain Ott
Bd I fermented butanediol to ethanol and acetate in a similar
manner as strain Gra Bd 1, however, growth was poor and the
butanediol was not utilized sufficiently for stoichiometric
calculations.
In coculture experiments with
A. woodii, M. hungatei,
or
D. vulgaris,
the fermentation balance of strain Ott Bd 1 was
not significantly shifted to more oxidized products, and
formation of methane or sulfide was negligible. Oxidation of
propionate was not possible in cocultures althouth hy-
drogenase activity in the range of 0.5 U/mg protein was
detected in crude extracts of butanediol-grown cells.
The growth curve (Fig. 3b) shows the dependence of
growth and product formation on butanediol utilization.
Optimal growth (p = 0A44h-1; td = 4.8 h) was observed at
33~ the temperature limits were 4~ and 45~ The pH
optimum for growth was between pH 7.0 and 8:0, the limits
were pH 6.5 and 8.4.
Strain Ko Bd 5 which was isolated as prevalent buta-
nediol-fermenting organism from anaerobic sewage sludge
had similar metabolic properties as the freshwater strains
Ott Bd 1 and G6 Bd I, however, grew much slower (td ~ 24 h).
It was not characterized further.
Enrichments with acetoin as substrate led to similar
isolates as with butanediol in both cases: two ethanol/acetate-
producing strains from marine sediments and three
propionate/acetate-forming strains from limnic muds. Both
types of organisms also exhibited similar substrate utilization
spectra as the strains Gra Bd 1 and Ott Bd I, respectively.
From enrichments on butanediol with pasteurized mud
from a freshwater creek, a homoacetogenic
Clostridium
strain
was isolated which will be described in a separate study
(Schink 1984).
Table3. Stoichiometry of fermentation and growth yields of freshwater isolate
Pelobacter propionicus
strain Ott Bd I
37
Substrate
Amount Cell dry Substrate Products formed (gmol) Growth yield
of substrate weight assimilated mg/mmol
degraded formed (gmol) Acetate Propio- Butyrate substrate
(~tmol) (mg) nate utilized
Carbon
recovery u
%
Butanediol 200 1.25 12.5 200 198 - 6.25
Acetoin 200 1.57 15.7 267 116 - 7.85
Ethanol 200 0.29 5.8 70 122 - 1.4
Propanol" 200 0.29 4.3 - 120 314 - 1.4
Butanol a 200 0.33 3.3 - 131 142 195 1.65
Lactate 200 1.26 t6.6 66 116 - 6.25
105.7
99.6
98.9
99.4
102.8
98.3
Calculation of cell dry weights and yields as described in Table 2
a With 200 gmol acetate added
a Carbon recovery was calculated on the basis of the fermentation equations given in the Discussion
Table4. Catabolic enzymes involved in 2,3-bntanediol degradation
by
Pelobacter carbinolicus
strain Gra Bd I and
Pelobacterpropionicus
strain Ott Bd I
Enzyme
E.C. Specific activity
number (gmol. min- 1
-mg protein- 1)
GraBd] OttBdl
Butanediol dehydrogenase 1.1.1.4 (?) 0.427 0.362
Acetoin dehydrogenase 1.1.1.5 0.189 0.200
Alcohol dehydrogenase 1.1.1.1 0.196 0.014
Phosphate acetyl transferase 2.3.1.8 21.4 5.4
Acetate kinase 2.7.2.1 0.79 2.8
(Propionate kinase) - N.D. 1.13
Assay of enzymes was perfolmed in crude extracts as described in
Materials and methods with 3.2-110 gg protein per assay
N.D. means not determined
Enzymes involved in butanediol degradation
In Table 4, the results of enzyme assays in butanediot-grown
cells of marine and freshwater isolates are compared. Both
contained butanediol dehydrogenase and acetoin dehydro-
genase activity. Alcohol dehydrogenase was found in con-
siderable amounts in strain Gra Bd 1 but was negligible in
strain Ott Bd 1. Phosphate acetyl transferase and acetate
kinase were present in both strains in high activities. In strain
Ott Bd 1, also a propionate kinase activity was detected.
Discussion
Physiology
The new bacteria described in this paper all degraded 2,3-
butanediol by cleavage between the C-Atoms 2 and 3, thus
releasing two C2 units. The first step in degradation was
probably a butanediol dehydrogenase reaction which led to
acetoin. In all our experiments, a butanediol preparation was
used consisting of all three stereoisomers, namely, about 80
of the meso- and about 10 ~ of each the D(-) and the L(+)
enantiomer (communication by Fluka Chemical Co.). Since
we always found complete degradation of the substrate
provided, and butanediol dehydrogenases are reported to be
strictly stereospecific (Taylor and Juni 1960; H6hn-Bentz and
Radler 1978), at least one racemase enzyme for transfor-
mation of all three into one suitable enantiomer has to be
postulated. The product of the dehydrogenase reaction,
acetoin, was also a growth substrate for all our isolates. The
further degradation could either proceed by splitting the
acetoin molecule into two acetaldehyde residues which would
probably be bound to a carrier coenzyme, e.g. thiamine
pyrophosphate. If acetoin is oxidized to diacetyl by acetoin
dehydrogenase, thiolytic cleavage into an acetaldehyde res-
idue and acetyl-CoA would occur. The results of the enzyme
assays do not allow conclusions on which of both pathways is
used since the acetoin dehydrogenase activity detected is
probably a byfunction of bntanediol dehydrogenase. In both
cases, cleavage leads to two C2 residues on the redox state of
acetaldehyde and two reducing equivalents on the redox level
of NAD. The further pathway of C2 metabolism differs
between the marine and the limnic isolates.
The marine strain Gra Bd 1 disproportionated the acetal-
dehyde residues to ethanol and acetate, and more ethanol was
formed by reduction of acetaldehyde with the electrons
derived from the butanediol dehydrogenase reaction. If
acetoin was the growth substrate twice as much acetate was
formed. The growth yields obtained (Table2) are in good
agreement with the assumption that the formation of acetate
by acetate kinase reaction is the only reaction coupled to
substrate-linked phosphorylation. On the basis of our data,
the yield per mol of adenosine triphosphate synthesized is in
the range of 7.0-7.2 g dry cell matter; this value agrees with
observations obtained with other bacteria (Stouthamer 1979).
Ethylene glycol was the only further substrate utilized by this
strain in pure culture. If a degradation pathway similar to that
of
Clostridium glycolicum
(Gaston and Stadtman 1963) and
other anaerobic bacteria (Barker 1972; Toraya et al. 1979) is
assumed, again acetaldehyde is an intermediate which is
disprop0rtionated to acetate and ethanol. The cell yield was
identical with that obtained with butanediol as substrate, and
again suggests that acetate kinase was the only energy
conserving reaction also in ethylene glycol degradation. The
fermentation equations for the substrates mentioned read
as follows (all calculations after Thauer et al. 1977; AGf ~
values for butanediol, -321.8kJ/mol, and for acetoin,