Journal of’ General Microbiology (1983), 129, 1651-1 659. Printed in Great Britain
165 1
The Electron Transport Chain of Escherichia coli Grown Anaerobically with
Fumarate as Terminal Electron Acceptor: an Electron Paramagnetic
Resonance Study
By W . J O H N I N G L E D E W
Department of Biochemistry and Microbiology, University of St Andrews,
St Andrews KY16 9AL. U.K.
(Received 10 September 1982; revised 13 November 1982)
The electron transport chain of Escherichia coli grown anaerobically on glycerol with fumarate as
terminal electron acceptor, has been studied using electron paramagnetic resonance (EPR)
spectroscopy. The analysis did not include cytochromes, but was confined to the lower potential
(dehydrogenase) section of the electron transport chain. Several ferredoxin-type centres were
detected and partially characterized, and the possible presence of iron-sulphur centres paramagnetic in their oxidized form (‘HiPIP-type’) was also noted. An EPR-detectable signal that
may have been due to the presence of molybdenum in the electron transport chain is described
and assessed. Most of the centres detected were reducible by substrate in the absence of oxidant
and some were found to be wholly or partly reduced during steady state oxidation of substrates in
the presence of oxygen.
INTRODUCTION
The composition of the electron transport chain of Escherichia coli, in terms of its oxidationreduction enzymes, has been the subject of a number of investigations (e.g. Shipp, 1972; Hendler
et al., 1975; Pudek & Bragg, 1976; Reid & Ingledew, 1979; Ingledew et al., 1980; Poole et al.,
1980). There are four principle modes of oxidative growth that have been widely studied:
aerobic (high aeration), aerobic (low aeration), anaerobic (NO; as terminal electron acceptor),
and anaerobic (fumarate as terminal electron acceptor). The cytochrome composition of the
electron transport chain alters greatly between these different growth conditions, and a rationale
for these changes is beginning to emerge. However, there has been only a small number of
investigations into the iron-sulphur centres and the effects of alteration of growth conditions on
the number and type of these centres. The studies on E. coli iron-sulphur centres that have been
published are confined to aerobically grown cells (Nicholas et al., 1962; Hamilton et al., 1970;
Hendler & Burgess, 1974; Poole & Haddock, 1975; Ingledew et al., 1980) or to a haem-deficient
mutant grown fermentatively with glucose (Blum et a/., 1980). Ingledew et a / . (1980), using
electron paramagnetic resonance (EPR) spectroscopy, have reported that membrane fragments
derived from aerobically grown E. coli contain a minimum of four iron-sulphur centres, three of
which appear to be very similar to the iron-sulphur centres of succinate dehydrogenase of
eukaryotic mitochondria (Ohnishi et al., 1976a, b) and of other prokaryotes (e.g. Ingledew &
Prince, 1977) and it was suggested that these centres did belong to the E. coli succinate dehydrogenase. This paper describes an EPR study of the iron-sulphur centres and a possible
molybdenum signal found in membranes of E. coli grown anaerobically with glycerol as carbon
and energy source and fumarate as terminal electron acceptor.
Abbreviation: ETP, electron transport particles.
Downloaded from www.microbiologyresearch.org by
0022-1287/83/0001-0788 $02.00 0 1983 SGM
IP: 88.99.165.207
On: Sun, 18 Jun 2017 23:14:48
1652
W . J . INGLEDEW
METHODS
Organism, growth conditions and preparation oj'cells.Escherichia coli strain EMG-2 (prototroph) was grown at
37 "C in 20 1 batches on a mineral salts medium (Cohen & Rickenberg, 1956) but with manganese omitted; the
medium was supplemented with 0.1 % (w/v) vitamin-free Casamino acids. The major reductant was glycerol (1 %,
w/v), and the electron acceptor was potassium fumarate (100 mM).
Cells were harvested in late exponential phase using an MSE continuous flow rotor running in an MSE 18
centrifuge at 18000 r.p.m. and a flow rate of approximately 250 ml min-'. The cells were washed twice by resuspension in 50 mM-potassium phosphate buffer (pH 7.5), and centrifugation at IOOOOg for 15 min at 4 "C. The
cell paste was frozen in small samples in liquid N, and stored at - 30 "C until required.
Cell breakage and the preparation ojelectron transport particles (ETP). Cells were resuspended in a buffer that
contained 20 mM-TES plus 2 mM-EDTA (pH 7-O), to a concentration of approximately 10 mg protein ml-I. The
cell suspension was passed twice through a French pressure cell at approximately 120 MPa. Unbroken cells and
debris were removed by centrifugation at lOOOOg for 15 min and the decanted supernatant centrifuged at
lOOOOOg for 1 h to pellet the ETP fraction. The particles were resuspended in the TESiEDTA medium,
centrifuged again at 1OOOOOg for 1 h, frozen in liquid N2 and stored at - 30 "C until used.
Sample preparation, redox titrations and assays. Samples for EPR were prepared in 3 mm internal diameter
quartz tubes and frozen quickly in an iso-pentane/cyclohexane (5 : 1, v/v) freezing mixture cooled with a liquid N,
cold-finger. The samples were stored under liquid N, until the EPR spectra could be taken. Oxidation-reduction
titrations, and the determination of mid-point potentials and equivalencies (n-values) were performed as described
by Dutton (1978); computer analysis of the data was as described by Reid & Ingledew (1979). Titrations were
performed in 50 mM-TES buffer (pH 7.0) with protein, concentrations in excess of 10 mg ml-'. The following
oxidation-reduction mediators were added at concentrations of between 25 and 150 p ~ phenazine
:
etho2-hydroxysulphonate; vitamin K, ; indigotetrasulphonate; indigodisulphonate; 5-hydroxy-l,4-naphthoquinone;
1,4-naphthoquinone; anthraquinone- 1,5-disulphonate ; anthraquinone-2-sulphonate; anthraquinone-2,6-disulphonate; resorufin; safranine; methylviologen; benzylviologen; duroquinone; pyocyanine and cysteine. The
redox titration vessel was purchased from the University of Pennsylvania glass workshop; a combination
platinum-calomel reference electrode was used to measure the ambient E , (Russell pH, Auchtermuchty, Fife,
U.K.). The vessel was continuously flushed with White Spot N, (British Oxygen Company) which had been
passed through a Nil-Ox apparatus (Jencons Scientific, Mark Rd, Heme1 Hempstead, U.K.) to remove residual
0 2 E.p.r.
.
spectra were obtained using a Bruker E R 200 D EPR spectrometer (Bruker Analytische Messtechnik,
Silberstreiten, D-75 12 Rheinstetten, F.R.G.) fitted with a variable temperature cryostat and liquid helium transfer
line (Oxford Instruments, Osney Mead, Oxford, U.K.). Rates of substrate oxidation were determined using a
Clark O2 electrode at 3 0 ° C and protein concentrations measured by the Lowry method using BSA as the
standard.
Chemicals. Chemicals for bacterial growth and general media were from BDH, except for Casamino acids
which was from Oxoid. Oxidation-reduction mediators were obtained from Aldrich, Koch-Light and Eastman
Kodak. BSA (V) and NADH were obtained through Sigma.
RESULTS A N D DISCUSSION
EPR spectra and oxidase activities
Formate is the substrate most rapidly oxidized by freshly prepared ETP [570 ng-atom 0
min-* (mg protein)-' 1, followed by NADH, succinate, lactate and a-glycerophosphate at 260,
140, 62 and 5 ng-atom 0 min-' (mg protein)-', respectively (30 OC, pH 7.0 with 20 mMpotassium phosphate buffer). The a-glycerophosphate oxidation rate in these particles was low
because the dehydrogenase in anaerobically grown cells is soluble and is lost during ETP
preparation (Kistler & Lin, 1972). Glycollate, glutamate, malate, and P-hydroxybutyrate
showed negligible oxidase activities.
EPR spectra of the reduced ETP are shown in Fig. 1 (a, b). The spectra in Fig. 1 (a) are of
dithionite-reduced particles and taken at different temperatures, whilst the spectra in Fig. I (b)
are of ETP allowed to go anaerobic in the presence of substrate (a-glycerophosphate in the case
shown, but other substrates gave similar results); the gain used in the latter case was twice that
used in Fig. 1 (a). In both cases, at relatively high temperatures (150 K), a ferredoxin is observed
with apparently rhombic symmetry and g , at 2.037, g, centred at 1.937 and g , at 1.923; this
centre is as fully reduced by substrate as by dithionite (note different gains). For convenience,
this centre will be termed centre I, pending a more accurate description of its role. As the
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 23:14:48
1653
E. coli electron transport chain
100 K
,
--150 K
2.045
I
2.07
2.037
F
tq----1.915
1.937
1.
923
I
1.875
2.07
I
1.875
Fig. 1. EPR spectra of reduced ETP at different temperatures. The numbers by each spectrum refer to
the temperature (K). The numbers at the bottom of the figure are g-values and, where appropriate, are
linked to indicate resonances belonging to a single centre. Membranes were reduced with (a) dithionite
or (b)by adding 10 mM-a-glycerophosphate for 5 min. The receiver gain was 2.5 x 10" in (a) and 5 x
lo4 in (b). Protein concentration was 52 mg ml-I. In both ( a ) and (h), the microwave frequency was
9.47 GHz, microwave power 24 mW, modulation frequency 100 kHz, and modulation intensity 1 mT.
temperature was lowered, the g, and g , absorbance of centre I were increasingly obscured by
other signals. Theg, absorbance decreased as the temperature was lowered to 30 K, below which
temperature it was also obscured by other signals. This decrease in signal height on lowering the
temperature is due to a phenomenon called saturation. On lowering the temperature, the rate at
which the photon-excited state can relax back to the ground state decreases and there comes a
point at which the relaxation rate cannot keep up with the excitation rate (the rate of photon
absorption), so as a consequence the latter has to decrease. This saturation phenomenon is thus
dependent on the intrinsic relaxation rate of the species under investigation, the temperature,
and the photon flux density (microwave power). As a result of this process, each centre will have
its own temperature profile, with the signal height increasing with decreasing temperature to a
temperature below which the signal height decreases.
On lowering the temperature from 150 K, a second species was increasingly observed at 100 K
and 50 K, overlapping with centre I. This centre appears to have slight rhombic distortion with
g, at 1.915, g , centred at 1.927 and g, at 2.022. These resonances increased in amplitude as the
temperature was lowered further: in the case of the dithionite-reduced sample (Fig. 1 u),
saturation did not occur until temperatures below 10 K were obtained, whereas in the substratereduced sample, saturation was observed below 30 K. The component that is not reduced by
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 23:14:48
1654
W . J. I N G L E D E W
J
0 -100 -200 -300 -400
Etl (mv)
Fig. 2. Analysis of redox titration data. Redox titrations were carried out as described in the text. The
signal heights of absorbances at designated g-values are plotted as a function of ambient redox potential
(E,,).The plots are made from spectra obtained at (a) 150 K, (b)25 K , (c) 12.5 K, and (d)6 K ; g-values
are given on the right. EPR conditions are as for Fig. 1 except for receiver gain which was varied as
appropriate for each measurement. Protein concentration was 29.5 mg ml-l.
100
substrate but is reduced by dithionite has a relatively rapid relaxation rate (i.e. is most
prominent at very low temperatures) which also enables the component to be distinguished in
the redox titrations (see later). The centres described thus far will be referred to as IIa, b, c on the
basis of resolution into three components by redox titration (IIa having the highest mid-point
potential and IIc the lowest).
Below 30 K, two further sets of signals were observed. The resonances are paired on the basis
of their similar temperature profiles and redox behaviour (see later) and will be referred to as
ferrodoxin centres 3 (g = 2.045, 1.89) and 4 (g = 2.07, 1.875).
In addition to these ferrodoxin-type iron-sulphur centres, signals in the g = 1-98 to 1-96
region were observed. These signals were relatively larger in the substrate-reduced sample. The
signals in this region are due to at least two species with different relaxation times and redox
behaviour (Figs 1 and 2). These signals and their possible implications are discussed later.
Oxidation-reduction titrations of EPR signals
The results of a redox titration are shown in Fig. 2. Samples were examined at different
temperatures to aid the resolution of components. At 150 K, a single ferrodoxin-like component
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 23:14:48
E. coli electron transport chain
1655
is seen, centre I (Fig 2a). The curve drawn for this component is a theoretical n = 1 curve and
this gave the best statistical fit (Table 1). Also seen at 150 K is a g = 1.978 signal, titrating in a
bell-shaped manner; the mid-points of the two half-reactions are - 20 mV and - 270 mV, each
with an n-value of 1. This signal, observed at relatively high temperatures, has a different
temperature profile and different mid-point potentials to a g = 1.978 signal which was observed
at low temperatures (see below). The signal seen at 150 K is referred to as the ‘slow’g = 1.978
signal as its temperature dependence shows that it has the slower relaxation rate of the two 1.978
absorbing centres.
When the redox titration samples were observed at 25 K a different profile was observed (Fig.
2 b): the g = 1,978 signal was less prominent (in Fig. 2 the signal units are arbitrary) and titrated
differently than at 150K. Centre I could not be distinguished. At 25 K, ferredoxins with
g, = 1.915 (not shown in Fig. 2), g, = 1.927 and g, = 2-02 (Fig. 1) were observed titrating as
multiple centres (centres IIa, b, c). These ferredoxins were resolved better at 12.5 K (Fig. 2c).
Three centre I1 components were resolved by redox titration, having Ern,,values of IIa - 50 mV,
IIb - 244 mV and IIc - 340 mV; best fits to the points were obtained with theoretical curves
having n-values of 1. A comparison of the plots at 25 K and 12.5 K revealed that the lower
potential components had more rapid relaxation rates than the higher potential component.
These observations correlate with the spectra in Fig. 1 to indicate that it is the lowest potential
component which is not substrate-reducible. The g = 1.89 (centre 111) and g = 1-875(centre IV)
components (Fig. 2c) titrated with Ern,,values of - 150 mV and - 250 mV, respectively (both
these centres being reducible by substrate). The ‘fast’ (low temperature) g = 1.978 signal also
gave a ‘bell-shaped’ titration curve, but with mid-points at approximately
50 mV and
- 100 mV.
When the samples were observed at 6 K (Fig. 2d), the g = 1-89 and 1.875 signals gave a
slightly different redox titration profile with apparent Ern,,values of approximately - 290 mV.
It is, however, impossible to say whether these differences are due to spectral distortion caused
by overlapping (or interacting) components or are new species (centres V and VI?).
+
The EPR signals o j oxidized ETP
Shown in Fig. 3 are EPR spectra, taken at different temperatures, of oxidized ETP. A
prominent peak occurs at g = 2.023 to 2.017 and troughs from g = 2.01 1 to 1.976. The peak at
g = 2.023 appears to pair with the trough at 2-011, giving a high potential iron-sulphur protein
(HiP1P)-like signal similar to that of mitochondria1 succinate dehydrogenase (Ohnishi et d.,
1976~).The designation ‘HiPIP-like’ is based on (1) the centre being paramagnetic in the
oxidized state and (2) its spectral similarity to the HiPIP signal of succinate dehydrogenase. This
centre is most clearly discerned in spectra taken above 15 K (Fig. 3) where the broad trough at
g = 1.976 and the overlapping peak at g = 2-017are no longer so prominent. The temperaturedependencies of these peaks and troughs confirm the presence of at least two distinct centres,
one of which has a peak at g = 2.017 and a trough at g = 2-011, and the other with a peak at
g = 2.017 and a trough at about g = 1.976; the exact position of this latter trough changes with
temperature. The temperature-dependencies in Fig. 3 are complicated by the fact that the
g = 2.017 and g = 2-073 signals extensively overlap. These signals titrate potentiometrically
with mid-point potentials of approximately 120 mV (results not shown).
+
Redox poise of EPR-detectable components in the steady state
The redox poise of the centres during steady state oxidation of NADH, succinate or lactate by
oxygen was studied. In all cases, the ‘slow’ g = 1.978 signal was detected, indicating some
partial reduction of this component as the centre is only observed at an intermediate redox state.
The amplitude of the g = 1.978 signal was similar in the samples oxidizing NADH or lactate,
but in the presence of succinate the signal approximately doubled in amplitude, reaching almost
maximum amplitude for this centre. As the g = 1-978 signal is due to an intermediate redox
state, it cannot be said whether the increased amplitude during succinate oxidation is due to
greater oxidation or reduction than with other substrates.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 23:14:48
1656
W . J . INGLEDEW
A
2.023
yo
2.017
*,
1
1.976
Fig. 3 . Temperature dependence of spectra of oxidized ETP. EPR conditions were as for Fig. 1 except
for receiver gain which was 1 x loJ. Protein concentration was 60 mg ml-’. ETP were oxidized by a
small amount of hydrogen peroxide in the absence of substrate. The g-values for prominent signals are
shown at the bottom of spectra.
During NADH oxidation, only centre I of the ferredoxins was extensively reduced (Fig. 4a,
6). The temperature profile of the ‘ g = 1.94’signal (Fig. 4a) is typical of a single homogeneous
component. This is confirmed in Fig. 46 where a microwave power profile over a saturating
range shows that a single component contributes to the ‘g = 1.94’ signal. Analysis of the redox
poise of the ferredoxins during succinate and lactate oxidation is more complex, as more than
one centre was reduced (Fig. 4a). In the case of membranes oxidizing lactate, the profile
indicates partial reduction of centre I (approximately 50%) and some reduction of a second
centre with a more rapid relaxation state (probably centre IIa, see Figs 1 and 2). During the
steady state oxidation of succinate, at least three different ‘g = 1-94,ferredoxins are partially
reduced; by comparisons of line-shapes and temperature dependencies these are centre I and
probably centres IIa and b. Although these samples were freeze-trapped and then stored in
liquid N 2 until used, it is possible that intracomplex electron transfer could continue to occur
under these conditions.
Conclusions
The characteristics of the centres resolved are compiled in Table 1. The picture is complex,
perhaps not surprisingly so, as these membranes contain hydrogenase, formate dehydrogenase,
fumarate reductase and succinate dehydrogenase. In other electron transport chains examples
are known of these dehydrogenases which contain EPR-detectable iron-sulphur centres (Yoch
& Carithers, 1979). There is no indication that the almost ubiquitous Rieske iron-sulphur centre
is present in these membranes; this may correlate with the absence of a cytochrome bc (or 6s)
complex in E . coli with which the Rieske centre is associated in other systems. Three of the ironDownloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 23:14:48
1657
E . coli electron transport chain
v
0.02
c'
.-8n
0
c
7
I
0-08 0.10
l/Temperature (K-')
0.04 0.06
0.12
(b)
0.0002 0.002 0.02
0.2
2
Microwave power (mW)
20
Fig. 4. An analysis by temperature and microwave power profiles of the ferredoxins reduced under
aerobic steady state conditions. (a) Temperature dependence of the 'g = 1.94' signal of the samples
freeze-trapped in the aerobic steady state during oxidation of NADH (a),succinate (A)
or lactate (0).
(b) Power saturation profile at 12 K of the single ' g = 1.94' signal observed during steady state
oxidation of NADH. EPR conditions were as for Fig. 1. Protein concentration was 52 mg ml-l.
Substrate concentrations were: lactate, 10 mM; succinate, 10 mM; and NADH, 6 mM. A small quantity
of hydrogen peroxide was added to prolong aerobiosis long enough to freeze-trap the samples.
Table 1. Characteristics of resolved iron-sulphur centres
Centre*
I
IIa
IIb
IIC
111
IV
V?
VI?
HiPIPa
HiPIPb
'Slow' 1a978
'Fast' 1.978
Em.7
g-value
(mV)
n-value
2.037 1-937 1.923
2.022 1.927
2.022 1.927
2422 1.927
.89
2.045
.875
2.07
-89
a875
2.023 2.01 1
2-017 1.976
1.978
-20
- 50
- 240
- 340
-150
-250
-290
-290
+120
+120
- 20
- 270
+50
-100
1
1
1
1
1
1
1
1
1 ~978
-
1
1
-
<1
Optimum
temperature
Possible source
(K)
150
30
10
12
12
6
6
20
Succinate dehydrogenase
Succinate dehydrogenaset
Succinate dehydrogenase
Formate dehydrogenase Mo (V)
Formate dehydrogenase Mo (V)
-, n-value not adequately resolved.
* Centres I-VI are paramagnetic in the reduced form and are referred to as ferredoxin-like. The name HiPIP is
used only to indicate that the centre is paramagnetic and therefore EPR-detectable in its oxidized form.
7 Not substrate-reducible.
sulphur centres detected can be tentatively assigned to the succinate dehydrogenase on the basis
of similarities between their properties and those described for succinate dehydrogenase centres
from a variety of sources (e.g. Ohnishi et al., 1976b, Ingledew & Prince, 1977); these centres are
IIa, IIc and the HiPIPa centre (see Table 1).
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 23:14:48
1658
W . J. INGLEDEW
The source of the g = 1.978 signals is most probably molybdenum in its pentavalent state,
because of similarities between the redox behaviour of the centre giving rise to the signal, and its
g-value, to the behaviour reported for other molybdenum centres (Barber et al., 1977; Bray,
1975). Furthermore, growth experiments show the dependence of the signal on molybdenum in
the growth medium (W. J . Ingledew, unpublished results). Molybdenum is known to be a
constituent of two E. coli oxido-reductases : the nitrate reductase (absent in membranes of
glycerol fumarate-grown cells) and formate dehydrogenase (present in membranes of glycerol
fumarate-grown cells) (Haddock & Mandrand-Berthelot, 1982; Boxer et al., 1982). Consequently, it is possible that the signals observed come specifically from the molybdenum centre
of formate dehydrogenase. This requires confirmation.
The complex picture represented by Table 1 should clarify as more becomes known about the
individual dehydrogenases. Some of these dehydrogenases have been cloned and amplified
(Cole & Guest, 1982); this will assist in future studies, both in the resolution of the redox centres
and in specifically determining their sidedness and orientation in the membrane.
I am indebted to Mr A. Houston and Mrs I. Heeramaneck for technical assistance. This investigation was
supported by Science and Engineering Research Council of the U.K.
REFERENCES
BARBER,M. J., BRAY, R. C., CAMMACK,R. &
COUGHLAN,M. P. (1977). Oxidation-reduction
potentials of turkey liver xanthine dehydrogenase
and the origins of oxidase and dehydrogenase
behaviour in molybdenum containing hydroxylases.
Biochemical Journal 163, 279-289.
BLUM,H., POOLE,R. K . & OHNISHI,T. (1980). The
orientation of iron-sulphur clusters in membrane
mu1tilayers prepared from aerobically-grown Escherichia coli K12 and a cytochrome deficient mutant.
Biochemical Journal 190, 385-393.
BOXER,D. H., MALCOLM,
A. & GRAHAM,A. (1982).
Escherichia coli formate to nitrate respiratory pathway : structural analysis. Biochemical Society Transactions 10, 480-481.
BRAY,R. C. (1975). Molybdenum iron-sulphur flavin
hydroxylases and related enzymes. In The Enzymes,
vol. 12, pp. 299-419. Edited by P. D. Boyer. New
York : Academic Press.
COHEN,G. N. & RICKENBERG,
H. W. (1956). Concentration specifique reversible des amino acides chez
Escherichia coli. Annales de l'fnstitut Pasteur 91,693720.
COLE,S. T. & GUEST,J. R. (1982). Molecular genetic
aspects of the succinate :fumarate oxido-reductases
of Escherichia coli. Biochemical Society Transactions
10, 473475.
DUTTON,P. L. (1978). Redox potentiometry : determination of midpoint potentials of oxidation-reduction
components of biological electron transfer systems.
Methods in Enzymology 54, 41 1-435.
H A D ~ KB., A. & MANDRAND-BERTHELOT,
M-A.
(1982). Escherichia coli formate to nitrate respiratory
chain : genetic analysis. Biochemical Society Transactions 10, 478480.
HAMILTON,
J. A., Cox, G . B., LOONEY,
F. D. & GIBSON,
F. (1970). Ubisemiquinone in membranes from
Escherichia coli. Biochemical Journal 116, 3 19-320.
A. H. (1974). FractionHENDLER,R. W. & BURGESS,
ation of the electron-transport chain of Escherichia
coli. Biochimica et biophysica acta 357, 2 15-230.
HENDLER,
R. W., TOWNE,D. W. & SHRAGER,R. I.
(1975). Redox properties of 3-type cytochromes in
Escherichia coli and rat liver mitochondria and
techniques for their analysis. Biochimica et biophysica
acta 376, 42-62.
INGLEDEW,
W. J. & PRINCE,R. C. (1977). Thermodynamic resolution of the iron-sulphur centers of the
succinic dehydrogenase of Rhodopseudomonas
sphaeroides. Archices of Biochemistry and Biophysics
178, 303-307.
INGLEDEW,
W. J., REID,G . A., POOLE,R. K., BLUM,H.
& OHNISHI,T. (1980). The iron-sulphur centres of
aerobically grown Escherichia coli K12. An electron
paramagnetic resonance study. FEBS Letters 111,
223-227.
KISTLER,
W. S. & LIN,E. C. C. (1972). Purification and
properties of the flavine stimulated aerobic L-aGlycerophosphate dehydrogenase of Escherichia coli.
Journal of Bacteriology 112, 539-547.
NICHOLAS,D. J. D., WILSON,P. W., HEINEN,W.,
PALMER,G. & BEINERT,H. (1962). Use of electron
paramagnetic resonance spectroscopy in investigations of functional metal components in microorganisms. Nature, London 196, 433-436.
OHNISHI,
T., SHIRASHI,
S. & INGLEDEW,
W. J. (19764.
Resolution and functional characterization of two
mitochondria1 iron-sulphur centres of the high
potential iron-sulphur protein type. Biochemical
Journal 153, 3948.
OHNISHI,T., SALERNO,
J. C., WINTER,D. B., LIM, J.,
Yu, C. A., Yu, L. &KING,T. E. (19763). Thermodynamic and epr characteristics of two ferredoxintype iron-sulphur centers in the succinate-ubiquinone reductase segment of the respiratory
chain. Journal of Biological Chemistry 251, 20942 104.
POOLE,R. K. & HADDOCK,B. A. (1975). Effects of
sulphate-limited growth in continuous culture on the
electron transport chain and energy conservation in
Escherichia coli K 12. Biochemical Journal 152, 537546.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 23:14:48
E. coii electron transport chain
POOLE,R. K., SCOTT,R. I . & CHANCE,
B. (1980). Low
temperature spectral and kinetic properties of cytochromes of Escherichiu coli K 1 2 grown at lowered
oxygen tension. Biochimica et biophysica acta 591,
47 1 4 8 2 .
PUDEK,M. R. & BRAGG,P. D. (1976). Redox potentials
of the cytochromes in the respiratory chain of
aerobically grown E. coli. Archices of'Biochemistry
and Biophysics 174, 546-552.
1659
REID,G. A . & INGLEDEW, W . J . (1979). Characterisation and phenotypic control of the cytochr~me
content of Escherichia coli. Biochemical Journal 182,
465-472.
SHIPP,W . S. (1972). Cytochromes of Escherichia coli.
Archives oj'Biochemistry and Biophysics 150,459-472.
YOCH,D. C. & CARITHERS,
R. P. (1979). Bacterial ironsulphur centres. MicrobiologicalReciews 43,384-42 1 .
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 23:14:48