Identification and optimization of quinolone-based inhibitors against cytochrome bd oxidase using an electrochemical assay

Article abstract of DOI:10.1016/j.electacta.2021.138293

A target-focused library of small molecules, containing a set of quinones, quinolones, coumarins, flavonoids and other diverse drugs was screened to identify potential inhibitors of the cytochrome bd-I oxidase from Escherichia coli by means of an electroc

Full text of DOI:10.1016/j.electacta.2021.138293

Electrochimica Acta 381 (2021) 138293
Contents lists available at ScienceDirect
Electrochimica Acta
journal homepage: www.elsevier.com/locate/electacta
Identification and optimization of quinolone-based inhibitors against
cytochrome bd oxidase using an electrochemical assay
I. Makarchuka, A. Nikolaeva, A. Thesselingb, L. Dejonc, D. Lambertyc, L. Stiefc, A. Speicherc,
T. Friedrichb, P. Hellwiga, H.R. Nasirid,∗, F. Melina,∗
a Laboratoire de Bioelectrochimie et Spectroscopie, UMR 7140, Chimie de la Matière Complexe, Université de Strasbourg, CNRS, Strasbourg 67081, France
b Albert-Ludwigs-Universität Freiburg, Institut für Biochemie, Freiburg 79104, Germany
c Department of Organic Chemistry, Saarland University, Campus C4.2, Saarbrücken 66123, Germany
d Department of Cellular Microbiology, University Hohenheim, Stuttgart 70599, Germany
a r t i c l e i n f o
a b s t r a c t
Article history:
A
target-focused library of small molecules, containing a set of quinones, quinolones, coumarins,
Received 28 January 2021
Revised 1 March 2021
Accepted 27 March 2021
Available online 31 March 2021
flavonoids and other diverse drugs was screened to identify potential inhibitors of the cytochrome bd-
I oxidase from Escherichia coli by means of an electrochemical assay based on the immobilization of the
proteins on gold nanoparticles. As primary hits, three quinolone fragments were identified with inhibition
activity in this study. As a hit-to-lead optimization approach, a series of more complex natural product-
based quinolones substituted in position 3 by an isoprenyl chain of various lengths were synthetized in
order to increase the inhibitory efficiency of the initial hit fragments. We show that the efficiency of
these molecules depends on the length of their isoprenyl side chain.
Keywords:
Cytochrome bd oxidase
Membrane proteins
Bioelectrochemistry
Inhibition
© 2021 Elsevier Ltd. All rights reserved.
Quinolones
1. Introduction
based on production of reactive oxygen species (ROS), NO and re-
active nitrogen species (RNS), with these molecules being impor-
Cytochrome bd oxidases are membrane proteins encountered
in the respiratory chains of only prokaryotes [1], including sev-
eral pathogens such as Escherichia coli, Mycobacterium tuberculosis,
Salmonella and Klebsiella pneumoniae. They couple the reduction of
O2 to the oxidation of quinol and contribute to the proton motive
force required for adenosine triphosphate (ATP) synthesis by taking
the four protons required for O2 reduction from the cytoplasmic
side and releasing protons from quinol oxidation to the periplas-
mic side of the membrane. Beyond their role in cellular bioener-
getics, bd oxidases are apparently involved in the bacterial viru-
lence [2-4] and antibiotics resistance [5-8]. There is a large body of
evidence showing that these enzymes enhance bacterial tolerance
to oxidative and nitrosative stress conditions [9-13], although the
molecular mechanisms are not fully understood. They efficiently
metabolize hydrogen peroxide and peroxynitrite and they show
lower sensitivity to cyanide, nitric oxide and hydrogen sulfide inhi-
bition as compared to the heme-copper oxidases [13-16]. It is cur-
rently suggested that the expression of bd oxidases in pathogenic
bacteria represents a strategy to evade the host immune attack
tant signaling molecules in human physiology [11]. Cytochrome bd
oxidases also allow bacteria to colonize environments poor in O2
and rich in hydrogen sulfide, such as the intestine of the host or-
ganism. E. coli, in particular, is a facultative anaerobe which ex-
presses three terminal oxidases depending on the conditions: the
heme-copper bo3 oxidase at high concentration of O2 and two cy-
tochrome bd oxidases under O2-limited conditions [17,18]. The bo3
oxidase is structurally related to the aa3 oxidase found in mam-
mals and it exhibits a low affinity for O2 [19], whereas the bd-
I oxidase has a very high affinity for O2 [20]. At low micromo-
lar concentration of sulfide, the bo3 oxidase enzyme is fully inhib-
ited [15,16], so the bacterium strictly relies on the cytochrome bd
oxidases for continued respiration. In vivo studies on mice have
shown that the colonization of the intestine by cytochrome bd oxi-
dase assembly mutants of E. coli is severely impaired [21]. In addi-
tion, some strict anaerobes are also able to survive under nanomo-
lar concentrations of O2 by expressing a cytochrome bd oxidase
that is essential for O2 consumption [22]. Specific inhibitors of cy-
tochrome bd oxidases thus may find immediate applications as an-
tibacterial drugs with new mode of action.
The X-ray structure of the enzyme from Geobacillus thermoden-
itrificans [23] and the very recent cryo-electron microscopy struc-
tures of the bd-I from E. coli [24,25] have shown that bd oxidases
∗
Corresponding authors.
E-mail addresses: Hamid.Nasiri@uni-hohenheim.de (H.R. Nasiri), fmelin@unistra.
fr (F. Melin).
https://doi.org/10.1016/j.electacta.2021.138293
0013-4686/© 2021 Elsevier Ltd. All rights reserved.

I. Makarchuk, A. Nikolaev, A. Thesseling et al.
Electrochimica Acta 381 (2021) 138293
immobilization of the proteins on electrodes modified with gold
nanoparticles. We have now used this assay to identify inhibitors
of the protein in a focused library of 34 compounds which belong
to the families of quinones, naphthoquinones, phenols, quinolones,
coumarins and flavonoids. We report here on the quinolones hits
that we have identified from this study and on the testing of a
series of more complex natural product-based molecules that we
have synthesized in order to increase the inhibiting efficiency of
the initial hits.
2. Experimental
2.1. Chemicals
Sodium citrate, hydrogen tetrachloroaurate trihydrate (HAuCl4,
3H2O), potassium dihydrogen phosphate (KH2PO4), sodium hy-
dride (NaH), phosphorus tribromide (PBr3), sodium chloride (NaCl),
magnesium sulfate (MgSO4), hexanethiol, 6-mercaptohexan-
1-ol, ubiquinone-1 (UQ1), N-oxo-2-heptyl-4-hydroxyquinoline
(HQNO), Lauryl maltose neopentyl glycol (MNG), 3-(N-morpholino)
propanesulfonic acid (MOPS), phospatidylethanolamine (PE),
dimethyl sulfoxide (DMSO), tetrahydrofuran (THF), n-hexane,
ethyl acetate (EtOAc) and ethanol (EtOH) were purchased from
Sigma-Aldrich and used without further purifications.
Fig. 1. schematic view on the cofactor localization in the E. coli cytochrome bd-I
oxidase (pdb file 6rko).
are highly hydrophobic and nearly completely embedded in the
membrane (See Fig. 1). They contain three heme cofactors, namely
heme d, heme b558 and heme b595, which are deeply buried in
the interior of the protein. The high spin heme d is believed to
be the active site where O2 is reduced. Unlike the terminal oxi-
dases found in mitochondria, the bd oxidases do not have any cop-
per center. A hydrophilic loop (known as the “Q-loop”) connecting
transmembrane helices 6 and 7 in the CydA subunit is believed to
be involved in quinone binding [26-29]. It has also been recently
demonstrated that the Q-loop has a role in the stabilization of the
entire enzyme complex [30].
2.2. Protein preparation
Cytochrome bd-I oxidase samples were prepared from E. coli
ꢀcyo-CydAhBX strain as previously described [25]. The protein at
final concentration of 10 mg/ml was equilibrated in buffer (20 mM
MOPS, 20 mM NaCl, 0.003% MNG, pH 7.0). For protein immobiliza-
tion on electrodes, the concentration of detergent was decreased
by washing 14 μL of the protein stock solution with phosphate
ꢀ
buffer (pH 7) for 45 min at 75 000 rpm and 4 °C on a 50 kDa
AMICON membrane filter. The final 10 μM protein solution was in-
cubated overnight at 4 °C with PE (2.5% of molar concentration of
cyt bd) and various amounts of UQ1 from 0 to 120 μM.
Only a few selective inhibitors of the quinone binding site of
cytochrome bd oxidases have been reported so far, such as the
isoprenoid quinolone alkaloids aurachins C and D [31-33], some
prenylphenols isolated from fungi [34] and very recently the indole
derivative MQL-H2 [35]. Aurachin D shows a high inhibitory activ-
ity against cytochrome bd-I oxidase from E. coli, but not against the
cytochrome bo3 oxidase. To study the redox activity of cytochrome
bd oxidases, we have set up an electrochemical assay based on the
2.3. Inhibitor synthesis
Geranylgeranyl aurachin D (AD-4) has been synthesized in the
following steps according to scheme 1.
Scheme 1. Multistep synthesis procedure of the compound AD-4.
2

I. Makarchuk, A. Nikolaev, A. Thesseling et al.
Electrochimica Acta 381 (2021) 138293
Compound 2: 1-Bromo-3,7,11,15-tetramethylhexadeca-2,6,10,14-
tetraene (geranylgeranyl bromide). To a solution of geranylgeraniol
(1) (5.80 g, 20.0 mmol, 1.0 eq) in Et2O (40 ml) at 0 °C was added
dropwise within 10 min PBr3 (1.80 g, 0.65 ml, 6.60 mmol, 0.33
eq) and the solution was stirred at 0 °C for additional 1 h. After
dilution with H2O, it was extracted with n-hexane (3 × 75 ml)
and the combined organic layers were washed with sat. NaCl solu-
tion, dried (MgSO4), filtered and concentrated. The geranylgeranyl
bromide (6.57 g, 18.6 mmol, 93%) was obtained as a pale-yellow
liquid. 1H NMR (CDCl3): δ (ppm) = 5.54 (t, J = 8.0 Hz, 1 H, =CH),
5.15–5.08 (m, 3 H, 3 × =CH), 4.03 (d, J = 8.0 Hz, 2 H, CH2–Br),
2.15–1.95 (combined m, 12 H, 3 × CH2CH2), 1.74/1.69/1.62/1.61
(4 × s, 15 H, 5 × CH3). 13C NMR (CDCl3): δ (ppm) = 143.62,
135.64, 134.97, 131.26, 124.36, 124.16, 123.38, 120.53, 39.71,
39.65, 39.52, 34.27, 29.68, 26.76, 26.59, 26.10, 25.68, 17.68, 16.04,
16.00.
2.4. Gold colloid solution preparation
The gold nanoparticles were prepared according to the conven-
tional method proposed by Turkevich and Frens [36-38]. Briefly, to
obtain nanoparticles of 15 nm average diameter, 125 mL of a 1 mM
solution of HAuCl4 were boiled before adding 13 mL of a 39 mM
sodium citrate solution. The solution was kept under boiling for
15 min, and was then allowed to cool down to room temperature.
The average size of the nanoparticles was estimated to be 15 nm
from the ratio of the absorbance at 521 and 450 nm [39]. Before
use, the gold colloid was concentrated by centrifugation for 30 min
ꢀ
at 10 000 rpm, and removal of 99% of the supernatant.
2.5. Electrode preparation and electrochemical setup
The surface of the gold rotating disk electrode (RDE) was me-
chanically polished with 0.05 μm alumina paste and activated by
subsequent oxidation at +2.0 V for 5 s, reduction at −0.35 V for
10 s and then cycling 100 times between −0.35 V and 1.5 V at
Compound
4:
Ethyl
2-acetyl-5,9,13,17-tetramethyloctadeca-
4,8,12,16-tetraenoate (Ethyl 2-(geranylgeranyl)acetoacetate). NaH
(14.8 mmol, 1.1 eq) was suspended under argon in anhydrous
THF (10 ml), cooled to 0 °C and ethyl acetoacetate (3) (1.70 ml,
1.75 g, 13.4 mmol, 1.0 eq) was added slowly and stirring was
continued for additional 20 min at 0 °C. Geranylgeranyl bromide
(2) (5.23 g, 14.8 mmol, 1.1 eq) in anhydrous THF (5 ml) was added
and the mixture was stirred at rt for 12 h. To the suspension was
added H2O (20 ml) followed by extraction with Et2O (3 × 20 ml).
The combined organic layers were washed with sat. NaCl so-
lution (2 × 30 ml), dried (MgSO4), filtered and concentrated.
The crude product was purified by flash chromatography (SiO2,
EtOAc/n-hexane 5:95, Rf = 0.18), yield 3.29 g (8.17 mmol, 61%)
of a pale-yellow liquid. 1H NMR (CDCl3): δ (ppm) = 5.12–5.01
(combined m, 4 H, 4 × =CH), 4.19 (q, J = 8.0 Hz, 2 H, O–CH2),
3.44 (t, J = 8.0 Hz, 1 H, α–H), 2.57 (t, J = 8.0 Hz, 2 H, CH2), 2.22
−
1
4 V.s
in 0.1 M H2SO4 solution. A single deposition of 7 μL of
a concentrated gold colloid solution was done on the gold disk,
and the electrode was allowed to dry under air. The modified el−ec-
1
trode was cycled again between −0.35 V and 1.5 V at 0.1 V.s
until a stable voltammogram was obtained. The real surface area
of the electrode was estimated from the integration of the gold-
oxide reduction peak at 1.1 V. The charge associated with the re-
duction of the gold oxide layer was assumed to be 390 μC.cm−2
[40]. The electrode was then immersed for 30 min in a (1:1) mix-
ture of 1 mM 6-mercaptohexan-1-ol and 1-hexanethiol in EtOH
and rinsed with fresh ethanol and dried. Finally, 5 μL of a 10 μM
protein solution in phosphate buffer were deposited on the surface
and left at 4 °C for two hours. Before the electrochemical stud-
ies, the modified electrode was rinsed with fresh buffer to remove
excess protein. All voltammograms were measured in a standard
three-electrode cell using a Princeton Applied Research VERSAS-
TAT 4 potentiostat. An AgCl/Ag 3 M NaCl electrode and a platinum
wire were used as the reference and counter electrode, respec-
tively. However, all the potentials mentioned in the manuscript are
referenced to the standard hydrogen electrode (SHE). All the mea-
surements were carried out at room temperature in a 20 mL pH 7
phosphate buffer solution. Three voltammograms were recorded at
0, 10 and 20 min after the preparation of the electrode to evaluate
the stability of the protein film.
–
(s, 3 H, H3C C = O), 2.10–1.90 (combined m, 12 H, 6 × CH2),
1.69 (broad s, 6 H, 2 × CH3), 1.64/1.63/1.61/1.60/1.50 (5 × s, 9
H, 3 × CH3), 1.27 (t, J = 7.1 Hz, 3 H, CH3). 13C NMR (CDCl3):
δ (ppm) = 203.10, 169.58, 138.46, 135.21, 135.16, 134.92, 131.24,
125.01, 124.98, 124.66, 124.60, 124.37, 124.33, 124.19, 124.10,
123.90, 123.88, 119.71, 119.60, 61.27, 59.82, 59.81, 39.98, 39.95,
39.68, 31.95, 29.10, 29.08, 26.89, 26.75, 26.70, 26.64, 26.52, 26.27,
25.70, 25.67, 23.37, 17.66, 17.61, 16.12, 15.98, 14.08.
Compound AD-4: Geranylgeranyl aurachin D. Ethyl 2-
(geranylgeranyl)acetoacetate (4) (300 mg, 0.74 mmol, 1.0 eq),
aniline (5) (0.08 ml, 0.82 mmol, 1.1 eq) and molecular sieves
˚
3 A (1 g) were heated to reflux in toluene (7 ml) for 16 h. After
cooling, the reaction mixture was filtrated through Celite and
the solvent was removed in vacuum. The crude enamine 6 was
added quickly to heated diphenyl ether (6 ml) and heating to
at least 250 °C was continued for 1 h. After cooling, the prod-
uct was separated from the solvent by column chromatography
(SiO2, EtOAc/n-hexane 5:95 to 1:1), yield 170 mg (4.00 mmol
53%) of a pale-yellow solid. 1H NMR (DMSO–d6): δ (ppm) = 7.98
(d, J = 8.0 Hz, 1 H, Ar–H), 7.78 (d, J = 8.0, 1 H, Ar–H), 7.58
(dd, J = 8.0/8.0 Hz, 1 H, Ar–H), 7.40 (dd, J = 8.0/8.0 Hz, 1 H,
Ar–H), 5.10–4.95 (combined m, 4 H, 4 × =CH), 2.72 (mc, 2 H,
quinolone-CH2), 2.52 (s, 3 H, CH3), 2.20–1.80 (combined m, 12
H, 6 × CH2), 1.60/1.53/1.51/1.50/1.34 (s, 15 H, 5 × CH3). 13C NMR
(DMSO–d6): δ (ppm) = 158.74, 158.73, 154.47, 146.61, 134.60,
134.57, 134.54, 134.47, 134.46, 134.34, 134.27, 130.79, 130.53,
128.49, 127.85, 124.79, 124.76, 124.37, 124.06, 124.02, 123.98,
123.81, 123.67, 120.73, 119.73, 110.37, 110.34, 77.29, 77.21, 39.42,
39.17, 38.83, 31.64, 31.45, 31.39, 31.23, 30.22, 30.14, 26.15, 26.12,
26.02, 25.97, 25.82, 25.78, 25.45, 25.42, 23.35, 23.26, 23.14, 23.10,
23.04, 22.73, 21.66, 21.53, 18.77, 17.46, 17.39, 15.72, 15.64, 15.62,
15.59. HR–MS (Cl) for [M] C30H41NO: 431.3191 (found); 431.3188
(calculated).
2.6. Inhibition studies
20 mM stock solutions of all inhibitors were prepared in DMSO.
A freshly prepared electrode was used for each inhibitor, and the
measurements were repeated with 3 different electrodes in order
to report the mean values and standard deviations. For the sin-
gle measurement at 10 μM, 10 μL of the solution of inhibitor
was added to the 20 mL buffer solution. After an equilibration
time of 10 min, the voltammogram of the immobilized proteins
in presence of the inhibitor was recorded. For the concentration-
response analysis, several aliquots of the inhibitor stock solution
were added to the 20 mL buffer solution. A voltammogram was
recorded 10 min after each addition of inhibitor.
3. Results and discussion
3.1. Preparation of the electrodes
Highly hydrophobic membrane proteins such as cytochrome bd
oxidase are more difficult to handle and study than soluble pro-
teins. Our strategy consists in their immobilization on 3D gold
3

I. Makarchuk, A. Nikolaev, A. Thesseling et al.
Electrochimica Acta 381 (2021) 138293
Fig. 2. Cyclic voltammograms of the bare gold disk electrode (black curve) and a gold disk electrode modified with gold NPs (red curve) in 0.1 M H2SO4 solution at scan
−
1
rate of 0.1 V.s
and 100 nm (C).
(A); Scanning electron microscopy images of the electrode surface after deposition of the gold nanoparticles at different magnification scales: 1 μM (B)
Table 1
nanoparticle electrodes modified with self-assembled monolayers
of thiols, which can be customized to the nature of the protein
surface [41-43]. The morphology of the modified gold disk elec-
trode which is obtained after a single deposition of a concentrated
solution of gold nanoparticles of 15 nm average diameter is shown
in Figs. 2B and C. It consists of interconnected ligaments of 30 to
100 nm in width, which form uniformly distributed pores that are
40–80 nm in diameter. The modified electrode was also studied by
cyclic voltammetry in 0.1 M H2SO4 (see Fig. 2A). A eight-fold in-
crease in the intensity of the gold oxide reduction peak at + 1.1 V
was observed after addition of the gold NPs, which confirms that
the gold NPs lead to a significant increase in the effective sur-
Influence of the substituents on the inhibitory efficiency of the quinolone deriva-
tives at 10 μM.
Substituent on C-2
Substituent on C-3
Inhibition at 10 μM (%)
CH3
CH3
CH3
CH3
CH3
COOH
COOEt
H
H
1
0.7
3.1
2.3
CH3
Br
I
5
6
11
16
0
0.4
1.1
1.7
–
=
CH2 CH C(CH3)2
H
H
3
1.4
1.4
COOEt
1
face area of the electrode (from 0.13 to 1.1 cm2
5%). This highly
porous network of gold nanoparticles is suggested to both allow
high enzyme loading on the electrode and to facilitate the electron
transfer from the electrode to the heme cofactors located deep in-
side the enzyme [44].
red curve in Fig. 3B) has only little effect on the voltammograms.
This antibiotic is known to inhibit the transcription and replica-
tion of DNA in gram-negative bacteria [47] but it shows no activity
against cytochrome bd oxidase. The 2–bromo-1,4-naphthoquinone
shows a rather unexpected activating effect on the immobilized
proteins (solid red curve in Fig. 3C). It leads to both an increase in
the catalytic current and in the catalytic potential. Thus, the reduc-
tion of O2 at the electrode requires less overpotential and occurs
with a higher turnover. As observed in the control voltammogram
in Fig. 3C (dashed red curve), the 2–bromo-1,4-naphthoquinone
shows a reduction wave at 0.03 V, which is in the same range of
potentials where the reduction of O2 by the protein occurs in the
presence of this molecule. So, the naphthoquinone probably acts as
a redox mediator, making the electron transfer between the elec-
trode and the immobilized proteins more efficient.
In previous reports, we have optimized the immobilization pro-
tocol to obtain stable E. coli cytochrome bd-I oxidase films [43].
Concerning the choice of the thiols, a (1/1) mixture of neutral 6-
mercaptohexan-1-ol and 1-hexanethiol seems to be suitable for
this highly hydrophobic protein. It is also crucial to control the
protein density on the gold surface. The highest catalytic activity
and reproducibility was observed for a surface coverage of active
protein of approximately 0.1 pmol.cm−2. The enzymes also require
the presence of phospholipids, such as phosphatidylethanolamine
(PE) or phosphatidylglycerol (PG) for an efficient oxidase activity
[45,46]. These lipids are both present in high amounts in the E.
coli membranes.
This label-free electrochemical assay thus allows to readily
identify inhibitors of cytochrome bd oxidases, which presumably
compete with the ubiquinone substrate at the “Q-loop” and block
the transfer of electrons to the active site. On this basis, the in-
hibitory potency of a library of 34 compounds, which belong to
the families of quinones, quinolones, coumarins, flavonoids and
other diverse drugs was evaluated at a single 10 μM concentra-
tion (Fig. 4). The inhibition efficiency was determined from the de-
crease of the maximum current value at −0.4 V.
3.2. Catalytic activity and inhibition of the immobilized proteins
Fig. 3 shows the typical stationary voltammograms under air of
the E. coli cytochrome bd-I oxidase immobilized on gold nanopar-
ticles modified with a (1:1) mixture of 6-mercaptohexan-1-ol and
1-hexanethiol (black curves). These sigmoidal-shaped voltammo-
grams are not observed in absence of either O2 or the protein, so
they are due to the reduction of molecular oxygen at the electrode
catalyzed by the immobilized proteins. The maximum current at
−0.4 V depends on both, the flux of substrate towards the elec-
trode and the turnover rate of the enzymes, which we previously
Three hits with an inhibitory effect better than 10% were iden-
tified in this primary screening campaign. These compounds are
quinolones substituted by alkyl or iodine substituents in position
C-2 and C-3. The quinolones with no substituent or polar sub-
stituents such as hydroxy, carboxylic acids or esters showed no
or little inhibition towards cytochrome bd oxidase (see Table 1),
which strongly suggests that the quinone binding pocket contains
hydrophobic residues. In the halogen series, the quinolone deriva-
tive with bromine was less active than its counterpart contain-
ing iodine. The tested coumarins, flavonoids and diverse drugs,
which include the vasodilator minoxidil, the anticancer drug tira-
−
1
determined to be 100 s
25% on these nanostructured gold
electrodes [43]. The exposure of the protein to molecules that are
structurally related to their quinol substrate may result in three
different behaviors. After addition of 50 μM HQNO (solid red curve
in Fig. 3A), the voltammogram turns back to background levels, so
most of the immobilized proteins have lost their activity. In con-
trast, the addition of the same concentration of nalidixic acid (solid
4

I. Makarchuk, A. Nikolaev, A. Thesseling et al.
Electrochimica Acta 381 (2021) 138293
Fig. 3. Effect of the addition of 50 μM HQNO (A), nalidixic acid (B) and 2–bromo-1,4-naphtoquinone (C) on the voltammograms of E. coli cytochrome bd-I oxidase immo-
bilized on gold nanoparticles. The solid black curves show the voltammograms before addition and the solid red curves after addition of the quinol analogues. The con−trol
1
voltammograms recorded for the quinol analogues in the absence of enzyme are shown as dashed red curves. Conditions: air-saturated pH 7 buffer, scan rate: 0.02 V.s
,
electrode rotation speed: 1000 rpm. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
Fig. 4. Inhibition efficiency of various compounds at 10 μM against E. coli cytochrome bd-I oxidase and structures of the three identified molecules showing 10% or more
inhibition efficiency.
pazamine and the anti-obesity agent cetilistat were also barely
active against the protein. Interestingly, 1,4-naphtoquinones sub-
the quinone is clearly dependent on its redox potential. To be-
have as a redox mediator, the quinone should have a reduction
potential which is higher than the unmediated catalytic poten-
tial (observed at −0.15 V here), but not too high to be able to
transfer its electrons to the protein cofactors which have mid-
point potentials of +0.17, +0.18 (hemes b) and +0.26 V (heme
stituted by bromine in position
2 or hydroxy in positions 5
and 8 showed an activation of the protein. In contrast, the 1,4-
dihydroxy-9,10-anthraquinone was inactive and the 2,5-dibromo-
1,4-benzoquinone was moderately desactivating. The behavior of
5

I. Makarchuk, A. Nikolaev, A. Thesseling et al.
Electrochimica Acta 381 (2021) 138293
Fig. 5. Constitution of the synthetic aurachin D derivatives used in this study.
Table 2
d) at pH 7 [48]. The cytochrome bd-I oxidase from E. coli is a
very versatile quinoprotein, which can accommodate both a 1,4-
benzoquinone derivative (ubiquinone) and a 1,4-naphthoquinone
derivative (menaquinone) as electron donors [45,49]. These phys-
Inhibition activity at
a
single concentration of 10 μM with or without added
quinone cytochrome bd oxidase of HQNO and synthetic aurachin D analogues.
Compound
Inhibition at 10 μM (%)
iological substrates have respective potentials of +0.11
V and
without added UQ1
with 3 eq UQ1
with 12 eq UQ1
−0.08 V at pH 7. The 2–bromo-1,4-naphthoquinone also exhibits
a reduction potential in this range (+ 0.03 V). In contrast, the
1,4-dihydroxy-9,10-anthraquinone, which is inactive here, shows
a very low reduction potential (−0.58 V at pH 5) [50] and the
2,5-dibromo-1,4-benzoquinone has a too high redox potential (+
0.33 V at pH 7) to act as an efficient redox mediator.
HQNO
AD-1
AD-2
AD-3
AD-4
21
16
37
27
6
3.1
1.1
1.8
1.2
24
17
23
19
7
3.4
2.4
3.8
3.9
27
19
15
10
ND
4.4
1.1
0.7
0.1
2.3
1.5
3.3. Synthesis of aurachin D analogues
the growing length of the isoprenyl side chain. An increase of in-
hibitory activity is observed here from one to two isoprene repeti-
tive units followed by a rapid decrease upon further increasing the
chain length. The most active inhibitor in this series thus seems to
be the one with two isoprene repetitive units, and not the nat-
ural product AD-3 identified before. Interestingly, no systematic
variation of inhibitor potency was reported before for quinolone
derivatives with saturated n-alkyl chains with 8 to 12 carbons [31].
Considering that the enzyme uses ubiquinone with eight isoprene
repetitive units as physiological substrate, steric effects alone can-
not explain the influence of chain length observed here. Thus, we
believe that the interaction of the tested compounds with the flex-
ible N-terminal domain of the Q-loop is based on specific inter-
action rather than the stickiness of the molecules. It is also pos-
sible that the lower solubility of the AD-3 and AD-4 molecules
in DMSO/buffer mixtures limits their efficiency. However, solubility
tests based on UV measurements suggests that the standard con-
centration of 10 μM used in this assay should still be acceptable
for these molecules.
We then synthetized a series of quinolone molecules with iso-
prenyl chains of various length in position C-3 (see Fig. 5) in or-
der to increase the lipophilicity and hence improve their affin-
ity against the target. As expected, an increase in lipophilicity is
observed when the isoprenic tail is increased (see the calculated
clogP values). These more complex molecules are analogues of the
already known aurachin D selective inhibitor of the enzyme (AD-
3). Here, we determined the influence of the isoprenyl chain length
on the inhibition efficiency of substituted quinolones.
These compounds can be efficiently synthetized in a few steps,
using the Conrad-Limpach cyclization [51]. We describe here the
multistep synthesis of the compound AD-4 that has not been re-
ported before (See Scheme 1). Geranylgeraniol (1), readily available
from farnesylacetone [52], was transformed into the correspond-
ing geranylgeranyl bromide (2) by reaction with phosphorus tribro-
mide and used for alkylation of ethyl acetoacetate (3). The result-
ing 2-substituted ethyl acetoacetate (4) was converted with aniline
(5) to the enamine 6 (not isolated due to lability) and cyclized di-
rectly to AD-4 under high temperature Conrad-Limpach conditions.
Upon adding quinone to the protein sample, the inhibition ef-
ficiency of the AD-2 and AD-3 molecules significantly decreases.
This confirms that aurachin D acts as a competitive inhibitor. For
the less active molecules AD-1 and AD-4, no strong effect of the
quinone is observed.
3.4. Inhibition activity of the aurachin D analogues
The inhibitory effect against the E. coli cytochrome bd-I of the
synthetic aurachin D analogues at a single concentration of 10 μM
with or without added ubiquinone substrate is shown in Table 2.
Ubiquinone-1 was added in 3 or 12 eq excess directly to the pro-
tein sample before immobilization on the surface of the nanostruc-
tured electrode to test whether the inhibition is competitive or not.
It can be observed that the length of the isoprenyl chain signif-
icantly affects the inhibition efficiency in the present assay. But it
does not simply follow the increasing order of hydrophobicity with
When successive aliquots of the inhibitors are added, the max-
imum catalytic current (i) at −0.4 V gradually decreases to back-
ground levels (Fig. 6A). The IC50 can be determined by plotting
the ratio i/i0 of the current at −0.4 V, where i0 is the value of
the current in absence of inhibitor, as a function of the concen-
tration of the inhibitor (Fig. 6B). The IC50 values in absence of
added ubiquinone substrate were in the low micromolar range
for the AD-1 (3.8
0.1 μM), AD-2 (1.1
0.7 μM) and AD-3
(2.2 0.5 μM) compounds, and a higher value was obtained for
6

I. Makarchuk, A. Nikolaev, A. Thesseling et al.
Electrochimica Acta 381 (2021) 138293
Fig. 6. Effect of the addition of aliquots of AD-2 on the voltammograms of the protein (A); plot of the normalised current (i/i0) at −0.4 V versus the concentration ([I]) of
the synthetic aurachin D derivatives (AD-1 to 4) and HQNO (B). Henderson plots obtained for AD-2 at three different amounts of ubiquinone-1 (C). Conditions: air-saturated
−
pH 7 buffer, scan rate: 0.02 V.s 1, electrode rotation speed: 1000 rpm.
AD-4 (14.4
0.5 μM). These measurements confirm that AD-2 is
increases from one to two isoprene repetitive units and decreases
for longer chains.
the most active compound in this family, but all of them behave
as stronger inhibitors in the concentration-response analysis than
in the initial screening at 10 μM where they all showed less than
50% inhibition (see Table 2). It is well documented in the litera-
ture that the traditional screening method at a single concentra-
tion can more easily generate false positives and false negatives
and is thus less reliable than titration-based approaches [53]. In
particular, highly hydrophobic molecules can show underestimated
activity at high concentration in bioassays due to poor solubility
[54].
Funding sources
PH acknowledges support by the University of Strasbourg Insti-
tute for Advances Studies (USIAS); IM was supported by the IDEX
PDI program from the University of Strasbourg; TF is supported
by the Deutsche Forschungsgemeinschaft by grant 235777276/RTG
1976; HN receives “INPA” funding from the Else Kröner Fresenius
Foundation.
Interestingly, aurachin D and its decyl derivatives were previ-
ously reported to have an even lower IC50 using standard quinol
oxidase activity assays based either on the measurement of oxy-
gen [31] or quinol consumption rates. These measurements were
carried out with proteins in dispersed membranes which can pro-
mote the interaction between lipophilic molecules and the mem-
brane proteins. To improve the sensitivity of the present assay, it
would thus be interesting to immobilize the protein within a lipid
bilayer membrane on the surface of the electrode.
Notes
The authors declare no competing financial interest.
Declaration of Competing Interest
None.
We have further analyzed our data on the most active inhibitor
(AD-2) using the Henderson model for tight-binding inhibitors
with one binding site per protein [55]. For this purpose, we have
plotted the quantity [I]/(1 − i/i0) as a function of i0/i at various
amounts of added ubiquinone substrate (see Fig. 6C). In this ex-
pression, [I] is the concentration of inhibitor, i the residual catalytic
current at – 0.4 V and i0 the catalytic current in absence of in-
hibitor. Unlike the results reported previously with the A. vinelandii
enzyme [56] the slope of the Henderson plot (see Fig. 6C) clearly
increases with the concentration of the substrate (UQ1). This is
again suggesting that AD-2 is a competitive inhibitor of the E. coli
bd-I oxidase.
Credit authorship contribution statement
I. Makarchuk: Investigation, Visualization. A. Nikolaev: Inves-
tigation, Visualization. A. Thesseling: Resources. L. Dejon: Re-
sources. D. Lamberty: Resources. L. Stief: Resources. A. Speicher:
Supervision, Resources. T. Friedrich: Funding acquisition, Supervi-
sion, Resources. P. Hellwig: Funding acquisition, Formal analysis,
Supervision, Writing – review & editing. H.R. Nasiri: Conceptual-
ization, Formal analysis, Resources, Writing – review & editing. F.
Melin: Conceptualization, Methodology, Formal analysis, Supervi-
sion, Writing – original draft.
References
[1] V.B. Borisov, S.A. Siletsky, A. Paiardini, D. Hoogewijs, E. Forte, A. Giuffrè,
R.K. Poole, Bacterial oxidases of the cytochrome bd family: redox enzymes of
unique structure, function, and utility as drug targets, Antioxid. Redox Signal.
(2021), doi:10.1089/ars.2020.8039.
Conclusion
Protein film voltammetry was successfully used to detect the
inhibition of key redox proteins by small quinol analogs. Here,
we focused on cytochrome bd oxidases, which are involved in
the virulence, adaptability and antibiotics resistance of bacteria.
Quinolones with alkyl or iodine substituents in position C-2 and
C-3 were identified as potential inhibitors of the enzyme from a
library of 34 small molecules related to the quinol substrate of the
enzyme. More potent inhibitors were obtained by chemical mod-
ification of the quinolone core and introduction of an isoprenyl
chain in position C-3. The activity of these competitive inhibitors
[2] L. Shi, C.D. Sohaskey, B.D. Kana, S. Dawes, R.J. North, V. Mizrahi, M.L. Gennaro,
Changes in energy metabolism of Mycobacterium tuberculosis in mouse lung
and under in vitro conditions affecting aerobic respiration, Proc. Natl. Acad.
Sci. U.S.A. 102 (2005) 15629–15634.
[3] J. Jones-Carson, M. Husain, L. Liu, D.J. Orlicky, A. Vázquez-Torres, Cytochrome
bd-dependent bioenergetics and antinitrosative defenses in salmonella patho-
genesis, MBio 7 (2016) e02052 16.
[4] D. Corbett, M. Goldrick, V.E. Fernandes, K. Davidge, R.K. Poole, P.W. Andrew,
J. Cavet, I.S. Roberts, Listeria monocytogenes has both cytochrome bd-type
and cytochrome aa3-type terminal oxidases, which allow growth at different
oxygen levels, and both are important in infection, Infect. Immun. 85 (2017)
e00354 17.
7

I. Makarchuk, A. Nikolaev, A. Thesseling et al.
Electrochimica Acta 381 (2021) 138293
[5] A. Koul, L. Vranckx, N. Dhar, H.W.H. Göhlmann, E. Özdemir, J.-.M. Neefs,
M. Schulz, P. Lu, E. Mørtz, J.D. McKinney, K. Andries, D. Bald, Delayed bacterici-
dal response of Mycobacterium tuberculosis to bedaquiline involves remodelling
of bacterial metabolism, Nat. Commun. 5 (2014) 3369.
[29] T. Mogi, S. Akimoto, S. Endou, T. Watanabe-Nakayama, E. Mizuochi-Asai,
H. Miyoshi, Probing the ubiquinol-binding site in cytochrome bd by site-di-
rected mutagenesis, Biochemistry 45 (2006) 7924–7930.
[30] A. Theßeling, S. Burschel, D. Wohlwend, T. Friedrich, The long Q-loop of Es-
cherichia coli cytochrome bd oxidase is required for assembly and structural
integrity, FEBS Lett 594 (2020) 1577–1585.
[6] P. Lu, M.H. Heineke, A. Koul, K. Andries, G.M. Cook, H. Lill, R. van Spanning,
D. Bald, The cytochrome bd-type quinol oxidase is important for survival of
Mycobacterium smegmatis under peroxide and antibiotic-induced stress, Sci.
Rep. 5 (2015) 10333.
[31] B. Meunier, S.A. Madgwick, E. Reil, W. Oettmeier, P.R. Rich, New inhibitors of
the quinol oxidation sites of bacterial cytochromes bo and bd, Biochemistry 34
(1995) 1076–1083.
[7] N.P. Kalia, E.J. Hasenoehrl, N.B. Ab Rahman, V.H. Koh, M.L.T. Ang, D.R. Sajorda,
K. Hards, G. Grüber, S. Alonso, G.M. Cook, M. Berney, K. Pethe, Exploiting the
synthetic lethality between terminal respiratory oxidases to kill Mycobacterium
tuberculosis and clear host infection, Proc. Natl. Acad. Sci. U.S.A. 114 (2017)
7426–7431.
[32] H. Miyoshi, K. Takegami, K. Sakamoto, T. Mogi, H. Iwamura, Characterization
of the ubiquinol oxidation sites in cytochromes bo and bd from Escherichia coli
using Aurachin C analogues1, J. Biochem. 125 (1999) 138–142.
[33] T. Mogi, H. Miyoshi, Properties of cytochrome bd plastoquinol oxidase from the
cyanobacterium Synechocystis sp. PCC 6803, J. Biochem. 145 (2009) 395–401.
[8] P. Lu, A.H. Asseri, M. Kremer, J. Maaskant, R. Ummels, H. Lill, D. Bald, The
anti-mycobacterial activity of the cytochrome bcc inhibitor Q203 can be en-
hanced by small-molecule inhibition of cytochrome bd, Sci. Rep. 8 (2018) 2625.
[9] A. Giuffrè, V.B. Borisov, D. Mastronicola, P. Sarti, E. Forte, Cytochrome bd oxi-
dase and nitric oxide: from reaction mechanisms to bacterial physiology, FEBS
Lett 586 (2012) 622–629.
[10] V.B. Borisov, E. Forte, A. Davletshin, D. Mastronicola, P. Sarti, A. Giuffrè, Cy-
tochrome bd oxidase from Escherichia coli displays high catalase activity: an
additional defense against oxidative stress, FEBS Lett 587 (2013) 2214–2218.
[11] A. Giuffrè, V.B. Borisov, M. Arese, P. Sarti, E. Forte, Cytochrome bd oxidase and
bacterial tolerance to oxidative and nitrosative stress, Biochim. Biophys. Acta,
Bioenerg. 1837 (2014) 1178–1187.
¯
[34] T. Mogi, H. Ui, K. Shiomi, S. Omura, H. Miyoshi, K. Kita, Antibiotics LL-Z1272
identified as novel inhibitors discriminating bacterial and mitochondrial quinol
oxidases, Biochim. Biophys. Acta, Bioenerg 1787 (2009) 129–133.
[35] A. Harikishore, S.S.M. Chong, P. Ragunathan, R.W. Bates, G. Grüber, Targeting
the menaquinol binding loop of mycobacterial cytochrome bd oxidase, Mol.
Divers. 25 (2021) 517–524.
[36] J. Kimling, M. Maier, B. Okenve, V. Kotaidis, H. Ballot, A. Plech, Turkevich
method for gold nanoparticle synthesis revisited, J. Phys. Chem. B 110 (2006)
15700–15707.
[37] P. Zhao, N. Li, D. Astruc, State of the art in gold nanoparticle synthesis, Coord.
Chem. Rev. 257 (2013) 638–665.
[12] M. Shepherd, M.E.S. Achard, A. Idris, M. Totsika, M.-.D. Phan, K.M. Peters,
S. Sarkar, C.A. Ribeiro, L.V. Holyoake, D. Ladakis, G.C. Ulett, M.J. Sweet,
R.K. Poole, A.G. McEwan, M.A. Schembri, The cytochrome bd-I respiratory oxi-
dase augments survival of multidrug-resistant Escherichia coli during infection,
Sci. Rep. 6 (2016) 35285.
[38] M. Sengani, A.M. Grumezescu, V.D. Rajeswari, Recent trends and methodolo-
gies in gold nanoparticle synthesis – a prospective review on drug delivery
aspect, OpenNano 2 (2017) 37–46.
[39] W. Haiss, N.T.K. Thanh, J. Aveyard, D.G. Fernig, Determination of size and con-
centration of gold nanoparticles from UV−vis spectra, Anal. Chem. 79 (2007)
4215–4221.
[13] E. Forte, V.B. Borisov, J.B. Vicente, A. Giuffrè, Chapter four - cytochrome bd
and gaseous ligands in bacterial physiology, in: R.K. Poole (Ed.), Advances in
Microbial Physiology, Academic Press, 2017, pp. 171–234.
[40] S. Trasatti, O.A. Petrii, Real surface area measurements in electrochemistry,
Pure Appl. Chem. 63 (1991) 711–734.
[14] A. Quesada, M.I. Guijo, F. Merchán, B. Blázquez, M.I. Igeño, R. Blasco, Essential
role of cytochrome bd-related oxidase in cyanide resistance of Pseudomonas
pseudoalcaligenes CECT5344, Appl. Environ. Microbiol. 73 (2007) 5118–5124.
[15] E. Forte, V.B. Borisov, M. Falabella, H.G. Colaço, M. Tinajero-Trejo, R.K. Poole,
J.B. Vicente, P. Sarti, A. Giuffrè, The terminal oxidase cytochrome bd promotes
sulfide-resistant bacterial respiration and growth, Sci. Rep. 6 (2016) 23788.
[16] S. Korshunov, K.R.C. Imlay, J.A. Imlay, The cytochrome bd oxidase of Escherichia
coli prevents respiratory inhibition by endogenous and exogenous hydrogen
sulfide, Mol. Microbiol. 101 (2016) 62–77.
[17] M.D. Rolfe, A.T. Beek, A.I. Graham, E.W. Trotter, H.M.S. Asif, G. Sanguinetti,
J.T. de Mattos, R.K. Poole, J. Green, Transcript profiling and inference of Es-
cherichia coli K-12 ArcA activity across the range of physiologically relevant
oxygen concentrations, J. Biol. Chem. 286 (2011) 10147–10154.
[41] F. Melin, T. Meyer, S. Lankiang, S.K. Choi, R.B. Gennis, C. Blanck, M. Schmutz,
P. Hellwig, Direct electrochemistry of cytochrome bo3 oxidase at
a series
of gold nanoparticles-modified electrodes, Electrochem. Commun. 26 (2013)
105–108.
[42] T. Meyer, F. Melin, H. Xie, I. von der Hocht, S.K. Choi, M.R. Noor, H. Michel,
R.B. Gennis, T. Soulimane, P. Hellwig, Evidence for distinct electron transfer
processes in terminal oxidases from different origin by means of protein film
voltammetry, J. Am. Chem. Soc. 136 (2014) 10854–10857.
[43] E. Fournier, A. Nikolaev, H.R. Nasiri, J. Hoeser, T. Friedrich, P. Hellwig, F. Melin,
Creation of a gold nanoparticle based electrochemical assay for the detec-
tion of inhibitors of bacterial cytochrome bd oxidases, Bioelectrochemistry 111
(2016) 109–114.
[44] Y. Xiao, F. Patolsky, E. Katz, J.F. Hainfeld, I. Willner, Plugging into enzymes":
[18] K. Bettenbrock, H. Bai, M. Ederer, J. Green, K.J. Hellingwerf, M. Holcombe,
S. Kunz, M.D. Rolfe, G. Sanguinetti, O. Sawodny, P. Sharma, S. Steinsiek,
R.K. Poole, Chapter two - towards a systems level understanding of the oxy-
gen response of Escherichia coli, in: R.K. Poole (Ed.), Adv. Microb. Physiol.,
(Ed.)Academic Press, 2014, pp. 65–114.
[19] R. D’Mello, S. Hill, R.K. Poole, The oxygen affinity of cytochrome bo’ in Es-
cherichia coli determined by the deoxygenation of oxyleghemoglobin and
oxymyoglobin: Km values for oxygen are in the submicromolar range, J. Bac-
teriol. 177 (1995) 867–870.
nanowiring of redox enzymes by
1877–1881.
a gold nanoparticle, Science 299 (2003)
[45] K. Kita, K. Konishi, Y. Anraku, Terminal oxidases of Escherichia coli aerobic res-
piratory chain. II. Purification and properties of cytochrome b558-d complex
from cells grown with limited oxygen and evidence of branched electron-car-
rying systems, J. Biol. Chem. 259 (1984) 3375–3381.
[46] A. Nikolaev, I. Makarchuk, A. Thesseling, J. Hoeser, T. Friedrich, F. Melin, P. Hell-
wig, Stabilization of the highly hydrophobic membrane protein, cytochrome bd
oxidase, on metallic surfaces for direct electrochemical studies, Molecules 25
(2020) 3240.
[47] R.M. Burger, D.A. Glaser, Effect of nalidixic acid on DNA replication by
toluene-treated Escherichia coli, Proc. Natl. Acad. Sci. U.S.A. 70 (1973)
1955–1958.
[20] R. D’Mello, S. Hill, R.K. Poole, The cytochrome bd quinol oxidase in Escherichia
coli has an extremely high oxygen affinity and two oxygen-binding haems: im-
plications for regulation of activity in vivo by oxygen inhibition, Microbiology
142 (1996) 755–763.
[21] S.A. Jones, F.Z. Chowdhury, A.J. Fabich, A. Anderson, D.M. Schreiner, A.L. House,
S.M. Autieri, M.P. Leatham, J.J. Lins, M. Jorgensen, P.S. Cohen, T. Conway, Res-
piration of Escherichia coli in the Mouse Intestine, Infect. Immun. 75 (2007)
4891–4899.
[22] A.D. Baughn, M.H. Malamy, The strict anaerobe Bacteroides fragilis grows in
and benefits from nanomolar concentrations of oxygen, Nature 427 (2004)
441–444.
[48] I. Belevich, V.B. Borisov, D.A. Bloch, A.A. Konstantinov, M.I. Verkhovsky, Cy-
tochrome bd from Azotobacter vinelandii: evidence for high-affinity oxygen
binding, Biochemistry 46 (2007) 11177–11184.
[49] J. Zhang, W. Oettmeier, R.B. Gennis, P. Hellwig, FTIR spectroscopic evidence
for the involvement of an acidic residue in quinone binding in cytochrome bd
from Escherichia coli, Biochemistry 41 (2002) 4612–4617.
[50] D. Nematollahi, A. Sayadi, F. Varmaghani, Electrochemical study of quinizarin
in the presence of arylsulfinic acids: synthesis of new sulfone derivatives of
quinizarin, J. Electroanal. Chem. 671 (2012) 44–50.
[23] S. Safarian, C. Rajendran, H. Müller, J. Preu, J.D. Langer, S. Ovchinnikov, T. Hi-
rose, T. Kusumoto, J. Sakamoto, H. Michel, Structure of a bd oxidase indicates
similar mechanisms for membrane-integrated oxygen reductases, Science 352
(2016) 583–586.
[24] S. Safarian, A. Hahn, D.J. Mills, M. Radloff, M.L. Eisinger, A. Nikolaev, J. Meier–
Credo, F. Melin, H. Miyoshi, R.B. Gennis, J. Sakamoto, J.D. Langer, P. Hellwig,
W. Kühlbrandt, H. Michel, Active site rearrangement and structural divergence
in prokaryotic respiratory oxidases, Science 366 (2019) 100–104.
[25] A. Theßeling, T. Rasmussen, S. Burschel, D. Wohlwend, J. Kägi, R. Müller,
B. Böttcher, T. Friedrich, Homologous bd oxidases share the same architecture
but differ in mechanism, Nat. Commun. 10 (2019) 5138.
[51] L. Dejon, A. Speicher, Synthesis of aurachin
D and isoprenoid analogues
from the myxobacterium Stigmatella aurantiaca, Tetrahedron Lett. 54 (2013)
6700–6702.
[52] H. Serizawa, B. Argade Ankush, A. Datwani, N. Spencer, Geranylgeranylacetone
derivatives, US, 2013, WO2013/052148Al.
[53] J. Inglese, D.S. Auld, A. Jadhav, R.L. Johnson, A. Simeonov, A. Yasgar, W. Zheng,
C.P. Austin, Quantitative high-throughput screening: a titration-based approach
that efficiently identifies biological activities in large chemical libraries, Proc.
Natl. Acad. Sci. U.S.A. 103 (2006) 11473–11478.
[26] T.J. Dueweke, R.B. Gennis, Epitopes of monoclonal antibodies which inhibit
ubiquinol oxidase activity of Escherichia coli cytochrome d complex localize
functional domain, J. Biol. Chem. 265 (1990) 4273–4277.
[54] L. Di, E.H. Kerns, Biological assay challenges from compound solubility: strate-
gies for bioassay optimization, Drug Discov. Today 11 (2006) 446–451.
[55] P.J. Henderson, A linear equation that describes the steady-state kinetics of
enzymes and subcellular particles interacting with tightly bound inhibitors,
Biochem. J. 127 (1972) 321–333.
[27] T.J. Dueweke, R.B. Gennis, Proteolysis of the cytochrome
d complex with
trypsin and chymotrypsin localizes a quinol oxidase domain, Biochemistry 30
(1991) 3401–3406.
[56] S. Jünemann, J.M. Wrigglesworth, P.R. Rich, Effects of decyl-aurachin D and re-
versed electron transfer in cytochrome bd, Biochemistry 36 (1997) 9323–9331.
[28] Y. Matsumoto, M. Murai, D. Fujita, K. Sakamoto, H. Miyoshi, M. Yoshida,
T. Mogi, Mass spectrometric analysis of the ubiquinol-binding site in cy-
tochrome bd from Escherichia coli, J. Biol. Chem. 281 (2006) 1905–1912.
8