Carboxyl residues in the iron-sulfur protein are involved in the proton pumping activity of P. denitrificans bc(1) complex.

Abstract

A study is presented on chemical modification of the three subunit Paracoccus denitrificans bc(1) complex. N-(Ethoxycarbonyl)-2-ethoxy-1,2-dihydroquinoline (EEDQ) treatment caused a loss of the proton pumping activity of liposome-reconstituted bc(1) complex. A similar effect, which is referred to as the decoupling effect, resulted upon reaction of N,N'-dicyclohexylcarbodiimide (DCCD) with the complex. Direct measurement of the binding of EEDQ to the complex subunits, performed in the presence of the fluorescent hydrophobic nucleophile 4'-[(aminoacetamido)methyl]fluorescein (AMF), showed that the iron-sulfur protein (ISP) and cytochrome c(1) were labeled by EEDQ, whereas cytochrome b was not. Tryptic digestion and sequencing analysis of the fluorescent fragment of the ISP revealed this to consist of a segment with six acidic residues, among which the highly conserved aspartate 160 is present. Analogous experiments on DCCD binding showed that all the three subunits of the complex were labeled. However, DCCD concentration dependence of carboxyl residue modification in the individual subunits and of proton pumping activity showed that the decrease of the H(+)/e(-) ratio correlated only with the modification of the ISP. Tryptic digestion of labeled ISP and sequencing analysis of the fluorescent fragment gave results superimposable upon those obtained with EEDQ. Chymotryptic digestion and sequencing analysis of the single fluorescent fragment of cytochrome b showed that this fragment contained glutamate 174 and aspartate 187. We conclude that, in the P. denitrificans bc(1) complex, carboxyl residues in cytochrome b do not appear to be critically involved in the proton pump mechanism of the complex.

Full text

Page 1

Carboxyl Residues in the Iron-Sulfur Protein Are Involved in the Proton Pumping
Activity of P. denitrificans bc1Complex†
Tiziana Cocco,*,‡Giuseppe Cutecchia,‡Bernd Ludwig,§Marcus Korn,§Sergio Papa,‡and Michele Lorusso*,‡,|
Department of Medical Biochemistry and Biology, UniVersity of Bari, Bari, Italy, Centre for the Study of Mitochondria and
Energy Metabolism, Consiglio Nazionale delle Ricerche, Italy, and Molekulare Genetik, Institut fur Biochemie, Biozentrum,
Johann Wolfgang Goethe-UniVersitat, Frankfurt, Germany
ReceiVed July 9, 2001; ReVised Manuscript ReceiVed September 20, 2001
ABSTRACT: A study is presented on chemical modification of the three subunit Paracoccus denitrificans
bc1complex. N-(Ethoxycarbonyl)-2-ethoxy-1,2-dihydroquinoline (EEDQ) treatment caused a loss of the
proton pumping activity of liposome-reconstituted bc1complex. A similar effect, which is referred to as
the decoupling effect, resulted upon reaction of N,N′-dicyclohexylcarbodiimide (DCCD) with the complex.
Direct measurement of the binding of EEDQ to the complex subunits, performed in the presence of the
fluorescent hydrophobic nucleophile 4′-[(aminoacetamido)methyl]fluorescein (AMF), showed that the iron-
sulfur protein (ISP) and cytochrome c1were labeled by EEDQ, whereas cytochrome b was not. Tryptic
digestion and sequencing analysis of the fluorescent fragment of the ISP revealed this to consist of a
segment with six acidic residues, among which the highly conserved aspartate 160 is present. Analogous
experiments on DCCD binding showed that all the three subunits of the complex were labeled. However,
DCCD concentration dependence of carboxyl residue modification in the individual subunits and of proton
pumping activity showed that the decrease of the H+/e-ratio correlated only with the modification of the
ISP. Tryptic digestion of labeled ISP and sequencing analysis of the fluorescent fragment gave results
superimposable upon those obtained with EEDQ. Chymotryptic digestion and sequencing analysis of the
single fluorescent fragment of cytochrome b showed that this fragment contained glutamate 174 and
aspartate 187. We conclude that, in the P. denitrificans bc1complex, carboxyl residues in cytochrome b
do not appear to be critically involved in the proton pump mechanism of the complex.
The cytochrome bc1 complex (ubiquinol-cytochrome c
oxidoreductase) is localized in the inner membrane of
mitochondria, where it catalyzes electron transfer from
ubiquinol to cytochrome c and coupled proton translocation
from the matrix to the intermembrane space. Analogous bc1
complexes are integral part in the electron transport chains
of chloroplasts (the bf complex) and of aerobic and photo-
synthetic bacteria. The various complexes differ from one
another in the number of constituent subunits, the mammalian
enzyme being made up of eight supernumerary subunits in
addition to those holding the redox centers, namely, the b
and c1cytochromes and the Rieske Fe/S protein (ISP).1The
bacterial bc1complexes, on the other hand, contain only three
to four polypeptides.
For each electron transferred to cytochrome c by the
mitochondrial bc1 complex, one proton is taken up at the
negative phase (N) of the inner membrane and vectorially
released at the positive phase (P). A second proton, which
derives formally from the quinol oxidation, is directly
released at the P site, thus giving an overall H+/e-ratio of
2. This ratio has been shown to decrease under conditions
not affecting the electron-transfer activity. This effect, which
is referred to as the decoupling effect, has been demonstrated
in mammalian (1-5) and yeast (6-8) mitochondria, in
chloroplast (9-10) and membrane preparations of Rhodo-
bacter sphaeroides (11).
The decoupling effect exerted by the amino acid modifier
DCCD led to suggest the involvement of acidic residues in
the pump mechanism of the complex. Evidence was actually
provided indicating that the decoupling effect by DCCD was
correlated with the modification of the bovine bc1complex
8 kDa subunit IX (2, 12), thereafter referred to as the DCCD
binding protein (13). Subsequently Cocco et al. (14) showed
that glutamate 53 is the target for DCCD modification of
this subunit. Beattie and co-workers (15) have reported that
aspartate 160 in the non-membrane-spanning helix cd of
yeast cytochrome b and its counterparts, aspartate 155 (or
glutamate 166) in the cytochrome b6 of the thylakoid
membrane cytochrome bf (16-17) and aspartate 187 of
†This work was financially supported by Grants from the National
Research Project (PRIN) on Bioenergetics and Membrane Transport
of MURST, the Finalized Project for Biotechnology (99.3622.PF49)
of CNR, Italy, and Deutsche Forschungsgemeinschaft (SFB 472).
* Corresponding authors: Department of Medical Biochemistry and
Biology, Medical Faculty, University of Bari, Piazza G. Cesare,
Policlinico,70125 Bari,Italy.
+39-080-5478429; e-mail m.lorusso@biochem.uniba.it or cocco.t@
biochem.uniba.it.
‡University of Bari.
§Johann Wolfgang Goethe-Universitat.
|Consiglio Nazionale delle Ricerche.
1Abbreviations: DCCD, N,N′-dicyclohexylcarbodiimide; EEDQ,
N-ethoxycarbonyl-2-ethoxy-1,2-dihydroquinoline; AMF, 4′-[(amino-
acetamido)methyl]fluorescein; ISP, iron-sulfur protein; SDS-PAGE,
sodium dodecyl sulfate-polyacrylamide gel electrophoresis; DBH, 2,3-
dimethoxy-5-methyl-6-undecyl-1,4-benzoquinol; CCCP, carbonyl cya-
nide (m-chlorophenyl)hydrazone; PVDF, poly(vinylidene difluoride).
Tel
+39-080-5478432; fax
15396
Biochemistry 2001, 40, 15396-15402
10.1021/bi011421b CCC: $20.00 © 2001 American Chemical Society
Published on Web 11/16/2001

Page 2

cytochrome b in Rb. sphaeroides (11) are modified by
DCCD. These observations led this group to ascribe the
decoupling effect of DCCD to modification of these cyto-
chrome b residues. However, it should be noted that these
are residues not conserved in cytochrome b of all species.
Furthermore, Shinkarev et al. (18) recently reported that the
change of aspartate 187 of Rb. sphaeroides cytochrome b to
asparagine did not alter the effect of DCCD on the functions
of the bc1complex.
Cocco et al. (19, 14) have introduced EEDQ, which
specifically modifies buried acidic residues, into the study
of the proton pump mechanism of the bc1 complex from
bovine heart. The results obtained suggested that acidic
residues in the core protein II and in ISP may be involved
in the pump function of the complex. In particular, ISP
aspartate 166, which is a conserved residue, was indicated
as a residue critically involved in this function.
Because of the apparent involvement of acidic residues
residing in supernumerary subunits of mammalian bc1
complex (the core protein II and the 8 kDa subunit IX) in
addition to those belonging to the redox subunits (the
cytochrome b and ISP), it appeared of interest to study the
effect of DCCD and EEDQ modification on the proton pump
function of a three-subunit bacterial bc1complex.
Here we report a detailed study on acidic residue modi-
fication in Paracoccus denitrificans bc1complex, which was
previously reported to be not susceptible to undergoing
decoupling after DCCD treatment (20). The results we
present show that (i) the P. denitrificans bc1complex was
indeed decoupled by treatment with either DCCD or EEDQ
and (ii) acidic residues in the ISP and cytochrome b were
modified. Among these, residues in the ISP do, while
glutamate 174 (or aspartate 187) in the cytochrome b do not,
appear to be critically involved in the proton pump mech-
anism of the complex.
EXPERIMENTAL PROCEDURES
Cell Growth and Purification of P. denitrificans bc1
Complex. MK10, a P. denitrificans strain overexpressing the
bc1complex (21), was grown at 32 °C on succinate as the
major carbon source (22).
Disruption of cells, isolation of membranes, protein
determination, solubilization of membranes, and one-step
protein purification by ion-exchange chromatography were
performed as described earlier (21, 23, 24).
Preparation of bc1Vesicles. Reconstitution of P. denitri-
ficans bc1complex into phospholipid vesicles was performed
essentially as described by Yang and Trumpower (20), with
some modifications. Acetone-washed soybean phospholipids
were suspended in 50 mM KCl and 50 mM K-HEPES, pH
7.2, to yield a final concentration of 30 mg/mL phospholipids
with 15% (w/v) cardiolipin. The suspension was first
vortexed and then sonicated under N2 with a microtip
sonicator for about 10 min until the mixture turned clear.
Purified bc1complex (0.04 mL of 30 mg of protein/mL) was
gently mixed with lauryl maltoside (0.06 mL of 200 mg/
mL), after which 1.90 mL of 30 mg/mL sonicated phospho-
lipids was added. This mixture was incubated on ice for 1 h
and then dialyzed against 1 L of 50 mM KCl and 50 mM
K-HEPES, pH 7.2, for 3 h and against 1 L of 99.6 mM KCl
and 1 mM K-HEPES, pH 7.2, overnight.
Measurement of Cytochrome c Reductase Activity and
Protonmotive Activity in bc1 Vesicles. Reductase activity
of liposome-reconstituted bc1complex was measured with
a dual-wavelength spectrophotometer (Johnson Research
Foundation, Philadelphia, PA) following the cytochrome c
reduction at the wavelength couple 550-540 nm (? ) 19.1
mM-1·cm-1) in a basic reaction mixture containing 1 mM
K-HEPES, pH 7.2, 100 mM KCl, 1 mM KCN, and 7.5 µM
ferricytochrome c. DBH (2,3-dimethoxy-5-methyl-6-undecyl-
1,4-benzoquinol, 30 µM) was used as substrate. Respiratory
control ratio, measured as described (20, 25), gave values
not lower than 4.
In the oxidant pulse experiments, bc1vesicles, suspended
in the last dialysis medium also containing 1 mM KCN, 1
µM ferricytochrome c, and 1 µg/mL valinomycin, were
reduced with 50 µM DBH. The reaction was started by
adding 3 µM ferricyanide. Acidification of the external
medium was followed electrometrically (20, 25).
In the reductant pulse experiments (2, 25), bc1 vesicles
suspension was supplemented with 1 mM KCN, 7.5 µM
ferricytochrome c, and 1 µg/mL valinomycin. The reaction
was started by adding 10 µM DBH. Spectrophotometric
determination of cytochrome c reduction at 550-540 nm
and electrometric determination of proton release were
carried out simultaneously on the same sample of vesicle
suspension.
Modification of bc1 Complex with EEDQ and DCCD.
Cytochrome bc1complex (20 mg of protein/mL) in 10 mM
K-HEPES, pH 7.0, and 0.02% Tween 80 was incubated at 0
°C with methanolic solutions of EEDQ and DCCD at the
concentrations specified in the figure captions. For labeling
studies the incubation mixture also contained 16 mM 4′-
[(aminoacetamido)methyl]fluorescein (AMF) (14, 19). After
1 h of incubation, aliquots of the bc1complex suspension
were directly added to a sonicated phospholipid suspension
or precipitated in 90% cold acetone. The pellets were
solubilized in 5% SDS, 15% glycerol, 50 mM Tris,, pH 6.8,
and 2% ?-mercaptoethanol and subjected to SDS-PAGE.
Electrophoretic Analysis of the bc1Complex. SDS-PAGE
analysis was performed on slab gels (10 cm long, 0.1 cm
thick) according to Schagger and von Jagow (26) with a 12%
separating gel. Slab gels were run at 30 V for 2 h and at 90
V for a further 5 h and either stained with Coomassie Brilliant
Blue G and then destained or soaked in 50% methanol for 6
h and then photographed on a dark surface under an
ultraviolet light source.
Proteolytic Digestion of EEDQ- and DCCD-Labeled bc1
Complex Subunits and Sequence Analysis. EEDQ- and
DCCD-labeled ISP and DCCD-labeled cytochrome b bands
were cut from the gel and electroeluted. The proteins, at 0.5
mg/mL, were digested with the indicated proteolytic enzymes
at a ratio of 1:20 (proteolytic enzyme:protein w/w) at 4 °C
and then analyzed by a modified Laemmli low molecular
weight polypeptide SDS-PAGE system, in a 20.1% T-0.5%
C separating gel, pH 9.3, and a 9.4% T-4.8% C stacking
gel, pH 6.8 (14). Gels were either stained with Coomassie
Brilliant Blue G and then destained or soaked in 50%
methanol for 6 h and then photographed on a dark surface
under an ultraviolet light source. For sequence analysis, gels
were blotted electrophoretically onto a poly(vinylidene
difluoride) (PVDF) protein sequencing membrane, in a
medium containing 10 mM CAPS [3-(cyclohexylamino)-1-
Proton Pump of P. denitrificans bc1Complex
Biochemistry, Vol. 40, No. 50, 2001 15397

Page 3

propanesulfonic acid] and 10% methanol, pH 11.0. Peptides
were sequenced with a fully automatic peptide sequencer
(model 473A of Applied Biosystems).
MATERIALS
DBH, EEDQ, and DCCD were obtained from Sigma
Chemical Co.; AMF was obtained from Molecular Probes
Inc. Trypsin, trypsin inhibitor, and chymotrypsin were from
Boehringer Mannheim. All other reagents were of the highest
purity grade commercially available.
RESULTS
Protonmotive Activity Measurement. The effect of EEDQ
on the redox and protonmotive activity of P. denitrificans
bc1 complex was tested by treating the enzyme with the
reagent before reconstitution into phospholipid vesicles.
Measurement of proton translocation coupled to electron-
transfer activity by the liposome containing bc1complex is
shown in Figure 1. Original traces of a representative
experiment are reported in Figure 1a showing that addition
of ferricyanide to bc1vesicles incubated in the presence of
DBH, valinomycin, and cytochrome c gave rise to a rapid
ejection of protons from the vesicles, whose extent was
almost halved in the presence of the uncoupler CCCP and
drastically reduced in EEDQ-reacted bc1complex-containing
vesicles. The H+/e-ratio, obtained from the extent of H+
release and the amount of added ferricyanide, dropped from
around 2.0 in the control to 1.20 for the treated enzyme.
Subtracting from these values 1 H+/e-which is uncoupler-
insensitive, around 80% inhibition of the proton pump
activity occurred after EEDQ modification. It has to be noted
that in these experiments the rate of proton leak is definitely
negligible. This is quite expected for the liposome reconsti-
tuted system, in which the passive proton permeability is
much lower than in the native membrane (27).
In Figure 1b the results of reductant pulse experiments
are reported, in which bc1vesicles, incubated in the presence
of valinomycin and 7.5 µM cytochrome c, were pulsed with
DBH. EEDQ treatment of the enzyme caused a slight
decrease of the rate of electron-transfer activity and a more
marked decrease of the rate of proton release. A statistical
evaluation of the effect of EEDQ treatment on the H+/e-
ratio for the uncoupler-sensitive proton translocation is
reported in Figure 1c. EEDQ caused a concentration-
dependent decrease of the H+/e-ratio measured with either
method, under conditions slightly affecting the reductase
activity.
Complex Subunits Modified by EEDQ. Direct measure-
ment of the binding of EEDQ to the complex was performed
in the presence of the fluorescent hydrophobic nucleophile
AMF, which specifically requires an EEDQ-modified car-
boxyl residue to bind to, by formation of an amide bond
(28, 29). Of the three constituent subunits of the complex,
cytochrome c1 and ISP were labeled by AMF whereas
cytochrome b was not (Figure 2a). An analogous experiment
carried out on the bovine enzyme (19), did not shown any
labeling of the cytochrome c1subunit. However, it should
be noted that P. denitrificans cytochrome c1presents in its
structure, when compared with the counterpart in the bovine
enzyme, an additional glutamate-rich sequence (30); thus its
labeling by AMF was quite expected. The concentration
dependence of carboxyl residue modification (Figure 2b) was
similar for the two subunits being labeled.
To identify the carboxyl residues in the ISP that were
labeled by EEDQ/AMF, the fluorescent band was excised
from the gel and electroeluted, and the resulting peptide was
digested with trypsin. The digested sample was then sub-
FIGURE 1: Effect of EEDQ modification on proton translocation
activity of the bc1complex. Cytochrome bc1complex at 20 mg of
protein/mL (80 nmol of cytochrome c1/mL) was treated, prior to
the reconstitution into phospholipid vesicles with EEDQ concentra-
tions ranging from 2 to 6 mM, this resulting in a EEDQ/c1ratio of
25-75 mol/mol. (a) Oxidant pulse experiment: bc1vesicles were
suspended at a concentration of 0.6 µM cytochrome c1 in the
reaction mixture described under Materials and Methods. The
reaction was started by the addition of 3 µM ferricyanide. (s)
Control; (---) 4 mM EEDQ-treated bc1complex; (‚‚‚) plus 3 µM
CCCP. (b) Reductant pulse experiment: bc1 vesicles were sus-
pended at a concentration of 0.6 µM cytochrome c1in the reaction
mixture described under Materials and Methods. The reaction was
started by adding 10 µM DBH. (b) Electron transfer; (9) proton
release. (c) H+/e-ratio for uncoupler-sensitive proton translocation.
The values reported are the means ( SD of measurements from
nine different experiments.
15398 Biochemistry, Vol. 40, No. 50, 2001
Cocco et al.

Page 4

jected to SDS-PAGE (Figure 3, lane b). Only one fragment
out of seven produced displayed AMF fluorescence (Figure
3, lane c). Sequencing analysis of this fragment, of estimated
molecular mass 4.5 kDa, gave the sequence TMDEAG. This
suggests that the cleavage product contains the residues from
Thr-118 to Arg-164 (see Figure 6). Six carboxyl residues
are contained in this fragment. Glu-124, Asp-146, and Asp-
160 are also present in the bovine ISP (see Figure 6). It
remains to be established which of these residues is actually
modified by EEDQ. However, determination of solvent-
accessible area of the bovine counterparts (31, 32) has
revealed that Asp-160 (Asp-166 in the bovine numbering)
is the only buried residue, as requested by the chemistry of
EEDQ modification.
Studies with DCCD. Yang and Trumpower have reported
that DCCD treatment does not inhibit the proton translocation
activity of bc1complex purified from P. denitrificans (20).
In light of the results obtained with EEDQ modification, we
have reexamined the effect of DCCD on the P. denitrificans
complex. As compared to EEDQ, we had to use 1 order of
magnitude higher concentrations of DCCD in order for the
effect to be elicited. With regard to the experimental
conditions used in ref 20, the incubation mixture of the
present experiments contained 0.02% Tween 80, with the
DCCD to cytochrome c1ratios being quite similar. Figure
4a shows the results of an experiment in which binding of
DCCD to the subunits of the complex was measured in the
presence of AMF. All the three subunits could indeed be
labeled. However, the binding of DCCD to the individual
subunits showed a markedly different concentration depen-
dence. Cytochrome c1, whose intrinsic fluorescence is evident
in the control, was strongly labeled by AMF, with the signal
appearing largely saturated at 10-20 mM DCCD (see Figure
4b). Cytochrome b was also labeled by AMF, with the
relative intensity band being independent of the DCCD
concentration, in the range 10-40 mM. The ISP, on the other
hand, appeared to react with DCCD only when the concen-
tration of the reagent was raised to 40 mM (Figure 4a,b).
The Coomassie blue-stained gel (Figure 4a, right panel) of
the control and DCCD-treated enzyme shows that DCCD
did not cause any cross-linking side reactions between the
subunits.
The effect of the reaction of DCCD on the protonmotive
activity of the complex was also followed, and the results
are reported in Figure 4c. The H+/e-ratio for vectorial proton
translocation measured in the liposome-reconstituted bc1
complex, with both oxidant and reductant pulses, dropped
only when the concentration of DCCD was raised to 40 mM,
that is, at concentrations causing modification of ISP subunit.
Under these conditions, DCCD treatment caused only a
small, if any, decrease of the rate of electron transfer (not
shown); the redox activity was, however, completely sensi-
tive to antimycin. It thus appears from these experiments
that modification by DCCD of either b and c1cytochromes
does not correlate with the observed decoupling effect. On
the other hand, a role of carboxyl residue(s) in the ISP,
already suggested by the experiment with EEDQ, is further
evidenced.
Tryptic digestion of DCCD-labeled ISP and subsequent
SDS-PAGE of the digestion sample revealed the presence
of a single fluorescent band. After sequencing analysis, this
fragment resulted to be exactly the same as that obtained by
tryptic digestion of the EEDQ-treated enzyme (not shown).
The electroeluted fluorescent cytochrome b was digested
with chymotrypsin and the resulting peptides were subjected
FIGURE 2: Binding of EEDQ to bc1complex subunits. Cytochrome
bc1complex was treated with the indicated concentrations of EEDQ
in the presence of 16 mM AMF. After 1 h of incubation, aliquots
of the bc1 suspension were precipitated in acetone and the
solubilized pellets were subjected to SDS-PAGE. (a) Gel exposed
to ultraviolet light. Lane Co shows the Coomassie-stained gel. (b)
Fluorescence of the bands was determined by scanning densitometry
of the photograph and plotted, as arbitrary units. The c1fluorescent
bands were corrected for the intrinsic fluorescence of the cyto-
chrome, in the control.
FIGURE 3: Trypsin digestion of EEDQ-labeled ISP. EEDQ-labeled
ISP fluorescent band (lane a) was cut from the gel and electroeluted.
The sample (0.5 mg of protein/mL) was digested with trypsin for
24 h and the reaction was stopped by adding a 5-fold excess of
trypsin inhibitor. The gel was either stained with Coomassie Brilliant
Blue G and destained (lane b) or soaked in 50% methanol for 6 h
and then photographed on a dark surface under an ultraviolet light
source (lane c). For sequence analysis, gels were blotted electro-
phoretically to a PVDF protein sequencing membrane and the
fluorescent peptide was sequenced with a fully automatic peptide
sequencer. The N-terminal sequence of the tryptic fluorescent
peptide is reported.
Proton Pump of P. denitrificans bc1Complex
Biochemistry, Vol. 40, No. 50, 2001 15399

Page 5

to SDS-PAGE (Figure 5). One out of six fragments
displayed fluorescence. Sequencing analysis of this fragment,
whose apparent molecular mass was about 5 kDa, gave the
sequence GQMSFW, thus suggesting it contains the residues
from Gly-152 to His-198. Two carboxyl residues are present
in this sequence, Glu-174 and Asp-187. Their counterparts
in the bovine enzyme reside in the non-membrane-spanning
cd2helix and in the cd loop, respectively (see Figure 6).
DISCUSSION
When experiments on DCCD modification of the bc1
complex are performed, DCCD treatment may result in
inhibition of both proton translocation and electron transfer
to almost the same extent (18, 33, 34), with no change in
the H+/e-ratio. We too observed such an effect in bovine
as well as in P. denitrificans bc1complex. To our experience,
multiple DCCD modifications, involving extended segments
of most of the complex subunits, do occur under these
conditions. On the contrary, if experimental conditions are
chosen resulting in a pure decoupling effect, then the results
obtained allow us to get insights into the involvement of the
individual subunits, and of residues in these, in the proton
pumping activity of the enzyme.
Here we have shown that P. denitrificans bc1 complex
shares with mammalian, yeast, chloroplast, and other aerobic
or photosynthetic bacteria the property to undergo decou-
pling. The use of the fluorescent probe AMF ensures that
carboxyl residues are indeed the target of the modification.
Parallel experiments on DCCD concentration dependence of
carboxyl residue modification, in the individual subunits, and
of proton pumping activity by the liposome-reconstituted
complex (Figure 4), show an involvement of the ISP in the
proton pump mechanism. We consider the binding of EEDQ
and DCCD to the P. denitrificans cytochrome c1subunit as
a consequence of the presence of a glutamate-rich N-terminal
sequence, absent in the eukaryotic counterpart. It should be
noted that binding of DCCD and EEDQ to the eukaryotic
cytochrome c1 has never been reported; neither has an
involvement of this subunit in the proton pump function been
suggested.
Ten times higher concentrations of DCCD than those
usually used for modification of the bovine complex had to
be used to modify the P. denitrificans complex. This is likely
the reason why DCCD was not found to modify the bovine
enzyme ISP (19). However, the bovine enzyme was still
decoupled after modification of Glu-53 in the 8 kDa subunit
IX (14). This observation would indicate the presence in
eukaryotic complex of different critical residues located in
different subunits.
A role of carboxyl residues in the non-membrane-spanning
helix cd and in the cd loop (see Figure 6) of cytochrome b
in the proton pumping activity has been repeatedly suggested
FIGURE 4: Binding of DCCD to bc1complex subunits and effect
on the protonmotive activity. Cytochrome bc1complex was treated
with the indicated concentrations of DCCD as described under
Materials and Methods. After 1 h of incubation, aliquots of the bc1
suspension were precipitated in acetone and subjected to SDS-
PAGE or added to a sonicated phospholipid suspension. (a) Gel
exposed to ultraviolet light. Lanes on the right refers to the
Coomassie-stained gel. (b) Fluorescence of the bands was deter-
mined by scanning densitometry of the photograph and plotted as
a percentage of the fluorescence intensity measured at 40 mM
DCCD. (c) Effect of DCCD on bc1vesicle protonmotive activity.
Proton translocation and electron-transfer activities were measured
with both oxidant and reductant pulse experiments, as described
under Materials and Methods and in the caption to Figure 1.
FIGURE 5: Chymotrypsin digestion of DCCD-labeled cytochrome
b. DCCD-labeled cytochrome b fluorescent band (lane a) was cut
from the gel and electroeluted. The sample (0.5 mg of protein/mL)
was digested with chymotrypsin for 7 days; afterward the digestion
was stopped by the addition of 2 mM phenylmethanesulfonyl
fluoride (PMSF). Gels were either stained with Coomassie Brilliant
Blue G and destained (lane b) or soaked in 50% methanol for 6 h
and then photographed on a dark surface under an ultraviolet light
source (lane c). For sequence analysis, gels were blotted electro-
phoretically to a PVDF protein sequencing membrane and the
fluorescent peptide was sequenced with a fully automatic peptide
sequencer. The N-terminal sequence of the tryptic fluorescent
peptide is reported.
15400 Biochemistry, Vol. 40, No. 50, 2001
Cocco et al.

Page 6

(11, 15-17). From the present data, it appears that modifica-
tion by DCCD of this residue(s) does not cause decoupling
of the P. denitrificans enzyme (Figure 4). Furthermore, as
noted by Crofts et al. (35), Asp-187 in the cd loop of Rb.
sphaeroides cytochrome b does not appear close enough to
the proton extrusion pathway site to have an obvious
functional role. Finally, substitution of this residue with
asparagine did not hamper the bc1complex functions, nor
did it modify the effect caused by DCCD (18).
The experiments reported here on the use of EEDQ as a
carboxyl residue modifier produced results superimposable
upon those obtained with DCCD. The same carboxyl residue-
containing region in the ISP was modified under conditions
in which the two reagents caused decoupling of the complex.
Asp-160, present in the fluorescent fragments, is homologous
with Asp-166 in the ISP of the bovine enzyme, which was
previously suggested to be involved in the proton pump
function of the complex (14, 19). In the bovine enzyme it
was found that, under decoupling conditions, carboxyl
residues in the core protein II were also modified by EEDQ
(14, 19). Thus, in addition to the ISP subunit, which appears
to contribute with its carboxyl residue(s) to the proton pump
mechanism of both bacterial and eukaryotic enzymes, the
supernumerary subunits core protein II and the 8 kDa subunit
IX seem to play a role in this process, in the bovine complex.
In the description of the crystal structure of the enzyme from
bovine heart (36, 37), core proteins enclose a large cavity
also housing subunit IX, in an arrangement that is suggested
to contribute a possible proton access pathway at the N side
of the membrane and, at the same time, to prevent proton
leakage back to the negative phase of the membrane. It is
tempting to speculate that these supernumerary subunits may
play a role in the optimization of the proton uptake process
at the N site of the membrane and of the coupling efficiency
of the process.
On the basis of the present and related observations (38,
10), it is proposed that modification by DCCD or EEDQ of
carboxyl residue(s) in the ISP blocks the proton-conducting
channel delivering at the quinol oxidation center the protons
generated upon quinol oxidation. It has to be noted that
aspartate 166 of the bovine bc1 complex ISP (and its
counterpart Asp-160 in the P. denitrificans enzyme) is a
buried residue involved in an internal bridge/hydrogen-bond
network essential for Fe/S cluster stability in the protein (39).
Furthermore, its engagement, through hydrophobic-hydro-
philic interactions, with other residues in the cluster domain
suggests a role in the proton outlet at the quinol oxidation
side (40).
It is worthwhile emphasizing that the H+/e-ratio for
vectorial proton translocation in the bovine bc1complex is
decreased by the steady-state transmembrane pH difference.
The ratio value was indeed found to be in linear inverse
correlation with the extent of the transmembrane pH gradient
(41, 42). It is conceivable that this decoupling between
electron flow and proton translocation, which is observed
also in the bovine cytochrome oxidase (43), may have a role
in lowering, at high value of the protonmotive force, the
oxygen concentration and the reduction level of electron
transport carriers, thus preventing superoxide anion genera-
tion (44, 45).
FIGURE 6: Sequence alignment of the carboxyl region of the ISP protein and of the cd loop region of cytochrome b. The arrows indicate
the trypsin (Tr) and chymotrypsin (Ch) cleavage sites. The conserved residue Asp-160 in the ISP fragment is highlighted. The nonconserved
acidic residues present in the P. denitrificans fluorescent fragment are in boldface type. The cd1and cd2non-membrane-spanning helices
(bovine sequence) of cytochrome b are in light boxes; the cd loop is in a bold box. The nonconserved acidic residues present in the P.
denitrificans cytochrome b fluorescent fragment are in boldface type. CAB88757, cytochrome b6of spinach thylakoids bf complex. (/)-
identical or conserved residues in all sequences in the alignment; (:) conserved substitutions; (.) semiconserved substitutions. Numbering
refers to the P. denitrificans sequence. Sequences were obtained from SWISSPROT database (Rel. 39), and multiple sequence alignment
was performed with CLUSTALW software (www.ch.embnet.org).
Proton Pump of P. denitrificans bc1Complex
Biochemistry, Vol. 40, No. 50, 2001 15401

Page 7

ACKNOWLEDGMENT
We thank F. Di Lernia for his helping in preparing the
illustrations.
REFERENCES
1. Degli Esposti, M., Saus, J., Timoneda, J., Bertoli, E., and
Lenaz, G. (1982) FEBS Lett. 147, 101-105.
2. Lorusso, M., Gatti, D., Boffoli, D., Bellomo, E., and Papa, S.
(1983) Eur. J. Biochem. 137, 413-420.
3. Clejan L., Bosch C. G., and Beattie D. S. (1984) J. Biol. Chem.
259, 13017-13020.
4. Lorusso, M., Cocco, T., Boffoli, D., Gatti, D., Meinhardt, S.
W., Ohnishi, T., and Papa, S. (1989) Eur. J. Biochem. 179,
535-540.
5. Cocco, T., Lorusso, M., Sardanelli, A. M., Minuto, M., Ronchi,
S., Tedeschi, G., and Papa, S. (1991) Eur. J. Biochem. 195,
731-734.
6. Beattie, D. S., and Villalobo, A. (1982) J. Biol. Chem. 257,
14745-14752.
7. Clejan, L., and Beattie, D. S. (1983) J. Biol. Chem. 258,
14271-14275.
8. Bruel, C., Manon, S., Gue ´rin, M., and Lemesle-Meunier, D.
(1995) J. Bioenerg. Biomembr. 27, 527-539.
9. Wang, Y. D., and Beattie, D. S. (1991) Arch. Biochem.
Biophys. 291, 363-705.
10. Joliot, P., and Joliot, A. (1998) Biochemistry 37, 10404-10410.
11. Wang, Y., Obungu, V., and Beattie, D. S. (1998) Arch.
Biochem. Biophys. 352, 193-198.
12. Degli Esposti, M., Meier, E. M. M., Timoneda, J., and Lenaz,
G. (1983) Biochim. Biophys. Acta 725, 349-360.
13. Borchart, U., Machleidt, W., Schagger, H., Link, T. A., and
von Jagow, G. (1985) FEBS Lett. 191, 125-130.
14. Cocco, T., Di Paola, M., Papa, S., and Lorusso, M. (1999)
FEBS Lett. 456, 37-40.
15. Wang, Y., Howton, M. M., and Beattie, D. S. (1995)
Biochemistry 34, 7476-7482.
16. Wang, Y., and Beattie, D. S. (1992) Biochemistry 31, 8455-
8459.
17. Wang, Y., and Beattie, D. S. (1993) Biochemistry 32, 9586-
9591.
18. Shinkarev, V. P., Ugulava, N. B., Takahashi, E., Crofts, A.
R., and Wraight, C. A. (2000) Biochemistry 39, 14232-14237.
19. Cocco, T., Di Paola, M., Papa, S., and Lorusso, M. (1998)
Biochemistry 37, 2037-2043.
20. Yang, X. H., and Trumpower, B. L. (1988) J. Biol. Chem.
263, 11962-11970.
21. Schro ¨ter, T., Hatzfeld, O. M., Gemeinhard, S., Korn, M.,
Friedrich, T., Ludwig, B., and Link, T. A. (1998) Eur. J.
Biochem. 255, 100-106.
22. Ludwig, B. (1986) Methods Enzymol. 126, 153-159.
23. Gerhus, E., Steinru ¨cke, P. and Ludwig, B. (1990) J. Bacteriol.
172, 2392-2400.
24. Witt, H., Zickermann, V., and Ludwig, B. (1995) Biochim.
Biophys. Acta 1230, 74-76.
25. Papa, S., Lorusso, M., Boffoli, D., and Bellomo, E. (1983)
Eur. J. Biochem. 137, 405-412.
26. Schagger, H., and von Jagow, G. (1987) Analytical Biochem.
166, 368-379.
27. Krishnamoorthy, G., and Hinkle, P. C. (1984) Biochemistry
23, 1640-1645.
28. Carraway, K. L., and Koshland, D. E. (1972) Methods
Enzymol. 25, 616-623.
29. Paprica, P. A., Margaritis, A., and Petersen, N. O. (1992)
Bioconjugate Chem. 3, 32-36.
30. Kurowski, B., and Ludwig, B. (1987) J. Biol. Chem. 262,
13805-13811.
31. http://bioinfo.weizmann.ac.il:8500/oca-bin/csu?PDB_ID)
1be3.
32. Sobolev, V., Sorokine, A., Prilusky, J., Abola, E. E., and
Edelman, M. (1999) Bioinformatics 15, 327-332.
33. Nalecz, M. J., Casey, R. P., and Azzi, A. (1983) Biochim.
Biophys. Acta 724, 75-82.
34. Shinkarev, V. P., Ugulava, N. B., Crofts, A. R., and Wraight,
C. A. (2000) Biochemistry 39, 16206-16212.
35. Crofts, A. R., Guergova-Kuras, M., Huang, L., Kuras, R.,
Zhang, Z., and Berry, E. A. (1999) Biochemistry 38, 15791-
15806.
36. Xia, D., Yu, C. A., Kim, H., Xia, J. Z., Kachurin, A. M.,
Zhang, L., Yu, L., and Deisenhofer, J. (1997) Science 277,
60-66.
37. Iwata, S., Lee, J. W., Okada, K., Lee, J. K., Iwata, M.,
Rasmussen, B., Link, T. A., Ramaswamy, S., and Jap, B. K.
(1998) Science 281, 64-71.
38. Brandt, U., and Trumpower, B. L. (1994) Crit. ReV. Biochem.
Mol. Biol. 29, 165-197.
39. Iwata, S., Saynovits, M., Link, T. A., and Michel, H. (1996)
Structure 5, 567-579.
40. Yu, C. A., Xia, D., Deisenhofer, J., Zhang, L., Kachurin, A.
M., and Yu, L. (1998) Biochim. Biophys. Acta 1365, 151-
158.
41. Cocco, T., Lorusso, M., Di Paola, M., Minuto, M., and Papa,
S. (1992) Eur. J. Biochem. 209, 475-481.
42. Lorusso, M., Cocco, T., Minuto, M., Capitanio, N., and Papa,
S. (1995) J. Bioenerg. Biomembr. 27, 101-108.
43. Capitanio, N., Capitanio, G., De Marinis, D. A., De Nitto, E.,
Massari, S., and Papa, S. (1994) Biochemistry 35, 10800-
10806.
44. Skulachev, V. P. (1996) Q. ReV. Biophys. 29, 169-202.
45. Papa, S., and Skulachev, V. P. (1997) Mol. Cell. Biochem.
174, 305-319.
BI011421B
15402 Biochemistry, Vol. 40, No. 50, 2001
Cocco et al.