Combining in silico and biophysical methods for the development of pseudomonas aeruginosa quorum sensing inhibitors: An alternative approach for structure-based drug design

Article abstract of DOI:10.1021/jm401102e

The present work deals with the optimization of an inhibitor of PqsD, an enzyme essential for Pseudomonas aeruginosa quorum sensing apparatus. Molecular docking studies, supported by biophysical methods (surface plasmon resonance, isothermal titration cal

Full text of DOI:10.1021/jm401102e

Subscriber access provided by UNIV OF PITTSBURGH
Article
Combining In Silico and Biophysical Methods for the Development
of Pseudomonas aeruginosa Quorum Sensing Inhibitors #
An Alternative Approach for Structure-Based Drug Design.
Jan Henning Sahner, Christian Brengel, Michael Philipp Storz,
Matthias Groh, Alberto Plaza, Rolf Müller, and Rolf W. Hartmann
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/jm401102e • Publication Date (Web): 01 Oct 2013
Downloaded from http://pubs.acs.org on October 8, 2013
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Journal of Medicinal Chemistry is published by the American Chemical Society. 1155
Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 28
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
Combining In Silico and Biophysical Methods for the
Development of Pseudomonas aeruginosa Quorum
Sensing Inhibitors – an Alternative Approach for Struc-
ture-Based Drug Design
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
‡,ǂ
‡,ǂ
ǂ
ǂ
¥
J. Henning Sahner , Christian Brengel , Michael P. Storz , Matthias Groh , Alberto Plaza , Rolf Mül-
¥
ǂ,
ler and Rolf W. Hartmann *.
ǂ
Helmholtz Institute for Pharmaceutical Research Saarland (HIPS), Department of Drug Design and
Optimization & Pharmaceutical and Medicinal Chemistry, Saarland University Campus C2 3, 66123
Saarbrücken, Germany
¥
Helmholtz Institute for Pharmaceutical Research Saarland (HIPS), Department of Microbial Natural
Products, Saarland University Campus C2 3, 66123 Saarbrücken, Germany.
ABSTRACT
The present work deals with the optimization of an inhibitor of PqsD, an enzyme essential for Pseudo-
monas aeruginosa quorum sensing apparatus. Molecular docking studies, supported by biophysical
methods (surface plasmon resonance, isothermal titration calorimetry, saturation transfer difference
NMR), were used to illuminate the binding mode of the 5-aryl-ureidothiophene-2-carboxylic acids. En-
1
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 28
abled to make profound predictions, structure-based optimization led to increased inhibitory potency.
Finally a covalent inhibitor was obtained. Binding to the active site was confirmed by LC-ESI-MS and
MALDI-TOF-MS experiments. Following this rational approach, potent PqsD inhibitors were efficient-
ly developed within a short period of time. This example shows that a combination and careful applica-
tion of in silico and biophysical methods represents a powerful complement to co-crystallography.
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
INTRODUCTION
Structure-based drug design is a rational strategy to develop bioactive molecules without the necessity
of performing many rounds of modifications to derive structure activity relationships (SAR). Frequently
data from X-ray co-crystallography are used to accelerate the lead optimization process. Although this
method is undeniably efficient, there are some drawbacks: a co-crystal structure does not necessarily
represent the biological state, as it results in a single “frozen” conformation that is affected by the crys-
1,2
tallization conditions. Furthermore, especially the ligand may not be well defined even at high resolu-
3
tion. In some cases, attempts of co-crystallizing ligands and their targets were not successful forcing
the research groups to make use of alternative strategies. The described problems can be overcome by
the use of biophysical methods such as surface plasmon resonance (SPR), nuclear magnetic resonance
(NMR) and isothermal titration calorimetry (ITC). An advantage of SPR, NMR and ITC is that they can
be performed in aqueous solution, which can be considered almost “physiological” conditions.
In the forefront of this work, compound 1 (Figure 1a) was identified as an inhibitor of PqsD. The target
protein PqsD, is a key player in the quorum sensing system of P. aeruginosa. It mediates the formation
of heptyl-4-hydroxy-quinoline (HHQ) which is the precursor of Pseudomonas quinolone signal (PQS)
(Scheme 1). Both molecules are potent virulence factors and function as signal molecules of P. aeru-
4,5
ginosa, coordinating group behavior like the formation of biofilms. The substrate channel of PqsD is
6
about 15 Å long and rather narrow. The channel can be divided into three parts. A positively charged
entrance followed by a mainly hydrophobic middle segment, ending in a polar bottom part. The latter is
2
ACS Paragon Plus Environment

Page 3 of 28
Journal of Medicinal Chemistry
delimited by the catalytic residues Cys112, His257 and Asp287. Suppression of PqsD activity has been
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
7
shown recently by our group to inhibit biofilm formation. This makes PqsD an attractive target for the
8–10
therapy of chronic P. aeruginosa infections, especially in immunosuppressed individuals.
Herein, we report on the optimization of 1 supported by biophysical methods and molecular docking.
The approach described in the following, represents a hit to lead optimization process that does not in-
volve co-crystallographic structure determination.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
a
Scheme 1. Biosynthesis of signaling molecules HHQ and PQS in Pseudomonas aeruginosa.
a
Anthranilate is transferred to PqsD via thio-esterification of ACoA with Cys112. The next step is a condensation with acetyl carrier pro-
tein (ACP) bound -ketodecanoic acid to form HHQ, which is finally transformed to PQS by PqsH.
RESULTS AND DISCUSSION
11
The IC value of compound 1 (Figure 1a) against PqsD was determined in a functional enzyme assay
5
0
to be 6 µM. The same compound has recently been reported as a weak inhibitor of bacterial RNA poly-
1
2
merase (IC : 241 µM) , a classical target for antibacterial therapy. While targeting quorum sensing an
5
0
antibacterial effect is not desired to avoid selection pressure inevitably leading to the development of
1
3
resistant strains. Thus, other potential hit candidates from the 5-aryl-ureidothiophene-2-carboxylic acid
class, displaying poor selectivity, were not considered for further optimization (see Table S1 in Support-
ing Information).
3
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 28
Figure 1. a) Compound 1. Biophysical characterization of 1: b) Thermogram (red) and titration curve
black) obtained from ITC of PqsD with 1; c) SPR experiment: ACoA binds to PqsD (black). ACoA
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
(
binding is inhibited by preincubation with 1 (red). The reference signal is shown in grey.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Attempts to obtain a co-crystal-structure of 1 and PqsD, using soaking techniques, were not successful.
Therefore, a combination of biophysical methods and molecular docking was used to elucidate the bind-
ing mode of 1. To exclude a non-specific binding behavior (e.g. via aggregation), 1 was examined in an
isothermal titration calorimetry (ITC) experiment to ensure stoichiometric binding. Thereby an equimo-
lar binding ratio (0.82 ± 0.05) with a K of 6.3 ± 2.6 µM was determined (Figure 1b), confirming that
D
the discovered class of compounds represents a promising starting point for further optimization.
In order to clarify the binding site on PqsD, SPR experiments were conducted. It was shown, that the
natural substrate ACoA can no longer bind to the enzyme when PqsD was preincubated with 1 (Figure
1c). This result indicates that 1 binds inside the ACoA-channel, thus blocking the first step of the cata-
lytic reaction.
From docking simulations with 1, two putative binding modes with similar scoring values but reverse
orientations were obtained. In the first pose (Figure 2; orange), the ureido N-ethyl-benzyl moiety points
to the bottom of the pocket (delimited by Cys112, His257 and Asn287), whereas in the second pose
(Figure 2; green), the methoxy-group is placed in this position.
4
ACS Paragon Plus Environment

Page 5 of 28
Journal of Medicinal Chemistry
−
1
Figure 2. Docking poses of 1 within the ACoA channel of PqsD: (orange; Vina-Score: 8.3 kcal mol )
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
−1
(
green; Vina-Score: 8.2 kcal mol ).
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
5
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 28
Figure 3. a) Schematic illustration of the SPR competition experiments: PqsD was preincubated with
competitor compounds A and B binding to the ACoA-channel, delimited by Cys112. The model com-
pounds 1−3 were subsequently added and the influence of A or B on their binding behaviors was inves-
tigated. The results shed light on the binding mode of the aryl-ureidothiophene-2-carboxylic acids.
b) Structures of the model compounds 1−3 c) Results from SPR-competition experiments suggest the
aryloxy-moiety to point into the pocket.
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
To get a reliable starting point for a structure-based optimization, discrimination between the two poses
was necessary. Therefore, we designed an SPR competition experiment with model compounds 1−3
from the 5-aryl-ureidothiophene-2-carboxylic acid class, differently substituted in the ureido and ar-
yloxy parts (Figure 3b). As competitor compounds the 2-nitrophenyl-methanols A and B were used
7
(
Figure 3a). They were recently reported as PqsD inhibitors and were shown to interact with the catalyt-
1
4
ic residues, especially Cys112, at the bottom of the ACoA-channel.
As illustrated in Figure 3a, the presence of A and B should influence the binding of the ureidothiophene
model compounds depending on their substitution pattern. Compared with 2, 1 bears a large substituent
6
ACS Paragon Plus Environment

Page 7 of 28
Journal of Medicinal Chemistry
at the ureido part, whereas 3 is elongated in the aryloxy part, at the opposite site. If the ureido part
points to the bottom of the pocket, one would expect that 1 is more sterically hindered than 2 and 3,
when the competitors are present. In contrast, the longer aryloxy part of 3 should lead to a reduced bind-
ing if the ureidothiophenes bind in the reversed orientation. In the absence of a competitor compound,
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
1
−3 displayed comparable responses (Figure 3c). When the sterically less demanding 2-nitrophenyl-
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
methanol derivative A was present, a reduction of affinity was observed exclusively for 3 bearing a
bulky phenyl substituent in the aryloxy part. In contrast, the ethyl-benzyl moiety at the ureido motif of 1
did not prevent a binding. These results fit to the docking prediction with the aryloxy part pointing into
the pocket (Figure 2, green). In competition with B, the binding affinity of all three compounds was
reduced as expected from the docking results. (see Figure S1 and S2) Thus a deeper insight into the po-
sition of the new inhibitors in the binding channel was provided.
As the starting point for structure activity exploration compound 2 was chosen, a truncated derivative of
the initial hit 1. It displays comparable ligand efficiency (LE, Table 1) whilst being more suitable for
further modification at the ureido motif. The docking pose of 2 (Figure 4a) suggests an anchoring of the
ureido moiety by interaction of the carbonyl oxygen with Arg223. To validate this assumption, binding
of 2 to the R223A mutant, that should display weaker affinity, was investigated by SPR. The response of
compound 2 was reduced to 6 response units (RU) in comparison to 25 RU for the wild type enzyme
(see Figure S4), confirming this hypothesis. Arg262, placed in the upper part of the binding pocket,
should be a potential interaction partner for a negatively charged moiety. The introduction of a carbox-
ylate resulting in compounds 4 and 5 (Table 1) indeed led to an increased activity, confirming this pre-
diction. In the docking pose, the introduced carboxylic acid moiety of 4 is located closely to Arg262
enabling the formation of a new interaction (Figure 4b). Based on the finding that 4 is more active than
5
, we concluded that a methylene linker is more appropriate than a longer spacer unit. This allowed
making use of α amino acids for simple variations in the next steps, taking advantage of their hydro-
philicity to increase the polarity of the inhibitors, as low lipophilicity is an important requirement for
drugs targeting Gram negative bacteria. A study of O’Shea et al. shows that almost all antibiotics, effec-
7
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 28
1
5
tive against P. aeruginosa, possess a clogD7.4 value below 0. While our hit compound 1 (logD : 2.58)
7.4
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
did not meet this criterion, amino acid derivative 4 (logD : −1.62) did.
7
.4
Figure 4. a) Docking pose of 2, the starting point for structure-based optimization. b) Docking pose of
glycine derivative 4. c) Docking pose of optimized compound 7. d) Illustration of the limited space in
the narrow cleft formed by Arg223 and Phe226. e) Close-up of the benzyl moiety of 7, perfectly fitting
into the narrow cleft.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
8
ACS Paragon Plus Environment

Page 9 of 28
Journal of Medicinal Chemistry
As illustrated in Figure 4b, docking simulations predict that the glycine derivative 4 is anchored by an
interaction of the carboxylate group with Arg262. The narrow cleft, subjacent to the α-position of 4 and
delimited by Arg223 and Phe226, is supposed to be addressable by expansion of 4. A targeted approach
should replace the pro-S hydrogen since the space adjacent to the pro-R hydrogen is blocked by Arg223,
which could lead to a sterical clash (Figure 4d). Following this strategy, the introduction of S-amino acid
residues improved the activity resulting in up to submicromolar IC values in case of the S-alanine and
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
5
0
S-phenylalanine derivative 6 and 7 (Figure 4b−e). Compound 8, the counterpart of 7 that lacks the
COOH group next to the ureido moiety displays a significantly weaker inhibition. This again highlights
the importance of the carboxylic acid group, which is in accordance with our binding hypothesis.
In order to complete the picture, the R-enantiomers 9 and 10 were investigated. In both cases the activity
decreased dramatically. These data corroborate the highly selective three-dimensional interaction
formed by the S-enantiomeric moieties with the target enzyme. To evaluate whether the decreased activ-
ity of the R-enantiomer 9 is due to a sterical clash, as expected, or associated with the absence of the S-
methyl group of 6, the dimethyl compound 11 was investigated. The fact that 11 was less active than 4
and 6 confirms the prediction of a sterical collision.
To test for selectivity towards bacterial RNA polymerase, the most potent derivatives 6 and 7 were test-
ed for inhibition of this enzyme. Both displayed only slight inhibitory potency (~10−20%) at a concen-
tration of 200 µM.
9
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 28
Table 1. Inhibition of PqsD in vitro, ligand efficiencies and clogD values of compounds 1−11
7
.4
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
[
a]
[b]
Cpd
R
IC50
PqsD
LE
clogD7.4
2.58
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
6 µM
0.25
2
4
170 µM 0.26
7 µM 0.30
37 µM 0.25
2 µM 0.32
0.15
−1.62
5
6
−1.47
−1.27
7
8
0.5 µM 0.28
58 µM 0.21
54 µM 0.24
26 µM 0.19
56 µM 0.23
0.59
2.61
9
−1.27
0.59
10
11
−0.93
[
a]
Ligand efficiency calculated as LE = 1.4 x (pIC /HAC)
50
[b]
clogD7.4 calculated with ACDLabs Percepta 2012
1
6
Saturation transfer difference (STD) NMR was used to identify the binding epitopes of 7 (Figure S5).
Strong STD enhancements were observed for all protons of 7 with the methylene linker protons exhibit-
ing the strongest enhancements. This is consistent with the docking simulations, which predict these
methylene protons to be positioned at the entrance of the narrow subpocket delimited by Arg223 and
Phe226. Moreover, the fact that the other protons of 7 also displayed strong STD enhancements suggests
that 7 is nearly completely surrounded by PqsD.
1
0
ACS Paragon Plus Environment

Page 11 of 28
Journal of Medicinal Chemistry
To confirm that the optimized inhibitors bind in a similar way as the model compounds 1−3, the SPR-
competition experiment with A and B was conducted for compound 7. As expected, the binding re-
sponse was only significantly reduced in the presence of B (Figure S3). In addition, we performed an
STD-NMR competition experiment to validate the SPR results. Increasing amounts of compound B
were added to a 100:1 complex of 7/PqsD (Figure 5). Difference spectra were monitored for a change in
intensity of signals belonging to either 7 or B during titration. As seen in the spectral expansions show-
ing the aromatic region, stepwise addition of up to 3.0 equivalents of B with respect to 7 resulted in a
uniform decrease in intensity for signals corresponding to 7 while signals ascribable to B appeared and
steadily increased. This confirms that 7 binds deep inside the ACoA-channel of PqsD.
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
Figure 5. Competition of 7 and B binding to PqsD by STD NMR. Expanded H STD NMR spectrum of
7 (1 mM) in the presence of PqsD (10 µM) (black curve). STD NMR spectra recorded on the same
sample after addition of 0.5 (purple), 1.0 (green), 2.0 (red) and 3.0 (blue) equivalents of compound B.
Signals of 7 are labeled in black, from B in blue.
1
1
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 28
Finally, we focused on the substituted 5-aryl ring that is predicted to be located near the active site with-
in the ACoA channel (Figure 4). As already shown with phenoxy compound 3 (IC = 20 µM), larger
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
5
0
substituents are tolerated in this part of the binding pocket. As Cys112 is pivotal for the catalytic activity
of PqsD, a covalent trapping of this amino acid should lead to an effective blockade of the enzymatic
1
7
machinery. For that purpose a β-chloroacetyl moiety was introduced, as an “electrophilic warhead”, at
the appropriate position, linked to the 5-aryl ring via an amide function (Figure 6a).
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 6. a) Schematic illustration of the proposed trapping of Cys112 by the introduced electrophilic
warhead of 12. MALDI-TOF-MS after tryptic digestion. The mass of the fragment containing Cys112 is
highlighted in red b) PqsD c) PqsD after incubation with 12. Signals below 3% intensity were removed
for the sake of clarity (raw data is shown in Figure S6 and S7).
The resulting compound 12 (clogD : −0.59) displays an IC value of 2 µM. The covalent binding of
7
.4
50
+
12 to PqsD was demonstrated by LC-ESI-MS. The PqsD peak (m/z = 36687.5 [M+H] ) shifted after
1
2
ACS Paragon Plus Environment

Page 13 of 28
Journal of Medicinal Chemistry
+
preincubation with 12 to the calculated adduct mass (m/z = 37153.1 [M+H] ) (Figure S8). To evaluate
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
whether indeed Cys112 is targeted by 12, the tryptically digested protein was analyzed by MALDI-
+
TOF-MS. The signal of the fragment containing Cys112, with a calculated m/z = 1510.8 [M+H] , was
found in the reference sample without inhibitor. In the sample treated with 12, this signal was absent
+
while two new peaks at m/z = 1975.9 and 1931.9 [M+H] appeared (Figure 6b,c). The former fits to the
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Cys112 fragment with covalently bound 12. The latter refers to its fragmentation product missing a car-
boxylic acid moiety. Taking into account that a long residence time at the active site is considered im-
1
8
portant for in vivo potency of the inhibitor, covalently binding compound 12 carries the potential of
strong biological effects.
CONCLUSIONS
In conclusion, we optimized the PqsD inhibitor 1, following a novel approach. First, an equimolar bind-
ing behavior to the target protein PqsD was determined by ITC. The binding mode was elucidated by
examination of three model compounds (1−3) in SPR competition experiments. To the best of our
knowledge, this is the first example of such an approach. These findings allowed profound docking cal-
culations guiding our subsequent structure-based optimization process. This approach culminated in an
irreversible inhibitor, which covalently binds to the active site of PqsD as predicted. The optimization of
activity was accompanied by reduced lipophilicity (Table 1), caused by the introduction of amino acids
at the ureido motif. According to the literature, this should facilitate permeation into Gram negative
1
5
cells.
For several drug targets, inhibitors with known binding sites and the protein crystal structures are avail-
able. In such cases, the presented approach of determining the binding mode can be helpful to guide a
structure-based design of new inhibitors, especially if co-crystallization fails.
While we mainly focused on the inhibitory potency at this early point of development, structural varia-
tions leading to more drug-like properties (e.g. bio-isosteric exchange of the carboxylic groups) remains
1
3
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 28
a challenging task and will be an element of future work. Additional efforts will also include studies on
intracellular activity and inhibition of biofilm formation in P. aeruginosa. In the present study, short-
comings of in silico docking simulations (prediction of multiple binding modes) were compensated by
the use of biophysical methods. This rational approach represents a powerful complement to X-ray co-
crystallography and should be applicable to many drug targets with known ligands.
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
EXPERIMENTAL SECTION
Materials and Methods. Starting materials were purchased from commercial suppliers and used with-
out further purification. Column flash chromatography was performed on silica gel (40–63 µM), and
reaction progress was monitored by TLC on TLC Silica Gel 60 F254 (Merck). All moisture-sensitive
reactions were performed under nitrogen atmosphere using oven-dried glassware and anhydrous sol-
1
13
vents. H and C NMR spectra were recorded on Bruker Fourier spectrometers (500/300 or 176/126/75
MHz) at ambient temperature with the chemical shifts recorded as δ values in ppm units by reference to
the hydrogenated residues of deuteriated solvent as internal standard. Coupling constants (J) are given
in Hz and signal patterns are indicated as follows: s, singlet; d, doublet; dd, doublet of doublets; t, tri-
plet; m, multiplet, br., broad signal. The purity of the final compounds was >95% except for 12 (90%)
measured by HPLC. The Surveyor LC system consisted of a pump, an autosampler, and a PDA detector.
Mass spectrometry was performed on a MSQ electrospray mass spectrometer (ThermoFisher, Dreieich,
Germany). The system was operated by the standard software Xcalibur. A RP C18 NUCLEODUR 100-5
(125 mm x 3 mm) column (Macherey-Nagel GmbH, Dühren, Germany) was used as the stationary
phase. All solvents were HPLC grade. In a gradient run the percentage of acetonitrile (containing 0.1%
trifluoroacetic acid) was increased from an initial concentration of 0% at 0 min to 100% at 15 min and
kept at 100% for 5 min. The injection volume was 10 µL, and flow rate was set to 800 µL/min. MS
analysis was carried out at a spray voltage of 3800 V and a capillary temperature of 350 °C and a source
CID of 10 V. Spectra were acquired in positive mode from 100 to 1000 m/z at 254 nm for the UV trace.
1
4
ACS Paragon Plus Environment

Page 15 of 28
Journal of Medicinal Chemistry
Chemistry
General synthesis procedures.
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Experimental details on synthesis of 12 can be found in the Supporting Information.
Scheme S1: Synthesis of 5-aryl-3-ureidothiophene-2-carboxylic acids.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
9
Method A, general procedure for the synthesis of 5-aryl-3-amino-2-carboxylic acid methylester (II):
POCl (26.1 g, 0.17 mol) was added dropwise to DMF (24.9 g, 0.34 mol) maintaining the temperature
3
below 25 °C (cooling in ice bath) and stirred for additional 15 min. The acteophenone I (85.0 mmol)
was added slowly and the temperature was kept between 40 and 60 °C. After complete addition, the
mixture was stirred for 30 minutes at room temperature. Hydroxylamine hydrochloride (23.6 g, 0.34
mol) was carefully added portionwise (exothermic reaction!) and the reaction was stirred for additional
3
0 min without heating. After cooling to room temperature, the mixture was poured into ice water (300
mL). The precipitated β-chloro-cinnamonitrile was collected by filtration, washed with H O (2 x 50 mL)
2
and dried under reduced pressure over CaCl . In the next step sodium (1.93 g, 84.0 mmol.) was dis-
2
solved in MeOH (85 mL) and methylthioglycolate (6.97 g, 65.6 mmol) was added to the stirred solution.
The β-chloro-cinnamonitrile (61.1 mmol) was added and the mixture was heated to reflux for 30 min.
After cooling to room temperature, the mixture was poured in ice water (300 mL). The precipitated solid
was collected by filtration, washed with H O (2 x 50 mL) and dried under reduced pressure over CaCl .
2
2
If necessary, recrystallisation was performed from EtOH.
1
5
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 28
Method B, general procedure for the synthesis of 5-aryl-3-amino-2-carboxylic acid (III):
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
The 5-aryl-3-amino-2-carboxylic acid methyl ester II (16.6 mmol) was added to a solution of KOH (60
mL, 0.6M in H O) and MeOH (60 mL). The mixture was heated to reflux for 3 h, concentrated, and
2
washed with EtOAc (2 x 50 mL). The aqueous layer was cooled with ice and acidified with a saturated
aqueous solution of KHSO . The precipitated solid was collected by filtration, washed with H O (2 x 30
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
4
2
mL) and dried under reduced pressure over CaCl2.
2
0,21
Method C, general procedure for the synthesis of 5-aryl-2-thiaisatoic-anhydrid (IV):
To a solution of the 5-aryl-3-amino-2-carboxylic acid (III) (5.28 mmol) in THF (50 mL) a solution of
phosgene (6.10 mL, 20 wt% in toluene, 11.6 mmol) was added dropwise over a period of 30 min. The
reaction mixture was stirred for 2 h at room temperature, followed by the addition of saturated aqueous
solution of NaHCO (30 mL) and H O (50 mL). The resulting mixture was extracted with EtOAc/THF
3
2
(1:1, 3 x 100 mL). The organic layer was washed with saturated aqueous NaCl (100 mL), dried
(
MgSO ) and concentrated. The crude material was suspended in a mixture of n-hexane/EtOAc (2:1, 50
4
mL) heated to 50 °C and after cooling to room temperature separated via filtration.
2
2
Method D, general procedure for the synthesis of of 5-aryl-3-ureidothiophene-2-carboxylic acid (V):
The 5-aryl-2-thiaisatoic-anhydrid (IV) (0.46 mmol) was suspended in water (7.5 mL) and the appropri-
ate amine (4.60 mmol) was added. The reaction mixture was stirred, heated to 100 °C and then cooled to
room temperature. The reaction mixture was poured into a mixture of concentrated HCl and ice (1:1)
and extracted with EtOAc/THF (1:1, 60 mL). The organic layer was washed with aqueous HCl (2M),
followed by saturated aqueous NaCl (2 x 50 mL), dried (MgSO ) and concentrated. The crude material
4
was suspended in a mixture of n-hexane/EtOAc (2:1, 20 mL) heated to 50 °C and after cooling to room
temperature separated via filtration.
1
6
ACS Paragon Plus Environment

Page 17 of 28
Journal of Medicinal Chemistry
Spectroscopic data of final compounds and intermediates can be found in the Supporting Information.
Compound 7 is presented as example.
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
(S)-3-(3’’-(1-Carboxy-2-phenylethyl)ureido)-5-(4’-methoxyphenyl)thiophene-2-carboxylic acid (7).
1
H NMR (DMSO-d , 300 MHz): δ = 12.84 (br. s., 2 H, COOH), 9.42 (s, 1 H), 8.15 (d, J = 7.9 Hz, 1 H),
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
8
.09 (s, 1 H), 7.69−7.51 (m, J = 8.8 Hz, 2 H), 7.37−7.25 (m, 4 H), 7.25−7.17 (m, 1 H), 7.03−6.97 (m, J
8.8 Hz, 2 H), 4.56−4.28 (m, 1 H), 3.79 (s, 3 H, OCH ), 3.11 (dd, J = 13.9, 4.6 Hz, 1 H), 2.88 (dd, J =
=
3
1
3.9, 9.8 Hz, 1 H) ppm.
1
3
C NMR (DMSO-d , 75 MHz): δ = 173.6, 164.6, 160.1, 153.7, 147.1, 145.9, 137.7, 129.0, 128.2, 127.1,
6
1
26.4, 125.4, 116.7, 114.7, 105.8, 55.3, 54.4, 36.9 ppm.
Biology and Biophysics
General procedure for expression and purification of recombinant PqsD WT and R223A mutant
in E. coli. His6-tagged PqsD (H6-PqsD) and mutants were expressed in E. coli and purified using a sin-
gle affinity chromatography step. Briefly, E. coli BL21 (λDE3) cells containing the pET28b(+)/pqsD
(
kindly provided by Prof. Rolf Müller) were grown in LB medium containing 50 µg/mL kanamycin at
7 °C to an OD600 of approximately 0.8 units and induced with 0.2 mM IPTG for 16 h at 16 °C. The
cells were harvested by centrifugation (5,000 rpm, 10 min, 4 °C) and the cell pellet was resuspended in
00 mL binding buffer (10 mM Na HPO , 2 mM KH PO pH 7.4, 3 mM KCl, 137 mM NaCl, 20 mM
3
1
2
4
2
4
imidazole, 10% glycerol (v/v)) and lysed by sonication for a total process time of 2.5 min. Cell debris
were removed by centrifugation (18500 rpm, 40 min, 4 °C) and the supernatant was filtered through a
syringe filter (0.20 μm). The clarified lysate was immediately applied to a Ni-NTA column, washed with
binding buffer and eluted with 500 mM imidazole. The protein containing fractions were buffer-
exchanged into 125 mM Na HPO , 50 mM KH PO pH 7.6, 50 mM NaCl, 10% glycerol (v/v), using a
2
4
2
4
PD10 column and judged pure by SDS-PAGE analysis. Then protein was stored in aliquots at –80 °C.
1
7
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 28
11
Screening assay procedure for in vitro PqsD inhibition. The assay was performed monitoring en-
zyme activity by measuring HHQ formed by condensation of anthraniloyl-CoA and β-ketodecanoic ac-
id. The reaction mixture contained MOPS buffer (0.05 M, pH 7.0) with 0.005% (w/v) Triton X-100, 0.1
µM of the purified enzyme and inhibitor. The test compounds were dissolved in DMSO and diluted with
buffer. The final DMSO concentration was 0.5%. After 10 min preincubation at 37 °C, the reaction was
started by the addition anthraniloyl-CoA to a final concentration of 5 µM and -ketodecanoic acid to a
final concentration of 70 µM. Reactions were stopped by addition of MeOH containing 1 µM amitripty-
line as internal standard for LC/MS-MS analysis. HHQ was quantified using a HPLC-MS/MS mass
spectrometer (ThermoFisher, Dreieich, Germany) in ESI mode. Ionization of HHQ and the internal
standard amitriptyline was optimized in each case. The solvent system consisted of 10 mM ammonium
acetate (A) and acetonitrile (B), both containing 0.1% trifluoroacetic acid. The initial concentration of B
in A was 45%, increasing the percentage of B to 100% in 2.8 min and keeping it at 98% for 0.7 min with
a flow of 500 µL/min. The column used was a NUCLEODUR-C18, 100-3/125-3 (Macherey & Nagel,
Duehren, Germany). Control reactions without the inhibitor, but including identical amounts of DMSO,
were performed in parallel and the amount of HHQ produced was set to 100%. All reactions were per-
formed in triplicates.
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
RNAP-transcription assay. E. coli RNA polymerase holo enzyme was purchased from Epicentre Bio-
technologies (Madison, WI). Final concentrations in a total volume of 30 µL were one unit of RNA pol-
3
ymerase (0.5 µg) which were used along with 60 nCi of [5,6- H]-UTP, 400 µM of ATP, CTP and GTP as
well as 100 µM of UTP, 20 units of RNAse inhibitor (RiboLock, Fermentas), 10 mM DTT, 40 mM Tris-
HCl (pH 7.5), 150 mM KCl, 10 mM MgCl and 0.1% CHAPS. As a DNA template 3500 ng of religated
2
2
3
pcDNA3.1/V5-His-TOPO were used per reaction. Prior to starting the experiment, the compounds
were dissolved in DMSO (final concentration during experiments: 2%). Dilution series of compounds
were prepared using a liquid handling system (Janus, Perkin Elmer, Waltham, MA). The components
1
8
ACS Paragon Plus Environment

Page 19 of 28
Journal of Medicinal Chemistry
described above (including the inhibitors) were preincubated in absence of NTPs and DNA for 10 min at
5 °C. Transcription reactions were started by the addition of a mixture containing DNA template and
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
2
NTPs and incubated for 10 min at 37 °C. The reaction was stopped by the addition of 10% TCA, fol-
lowed by a transfer of this mixture to a 96 well Multiscreen GFB plate (Millipore, Billerica, MA) and
incubation for 45 min at 4 °C. The plate underwent several centrifugation and washing steps with 10%
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
3
TCA and 95% ethanol to remove residual unincorporated H-UTP. After that the plate was dried (30
min, 50 °C) and 30 µL of scintillation fluid (Optiphase Supermix, Perkin Elmer) was added to each
3
well. After 10 min the wells were assayed for presence of H-RNA by counting using a Wallac Mi-
croBeta TriLux system (Perkin Elmer). To obtain inhibition values for each sample, their counts were
related to DMSO controls.
Surface plasmon resonance. SPR binding studies were performed using a Reichert SR7500DC instru-
ment optical biosensor (Reichert Technologies, Depew, NY, USA) and CMD500M (carboxymethyl dex-
trane-coated) sensor chips obtained from XanTec (XanTec Bioanalytics, Düsseldorf, Germany).
Immobilization of wild type PqsD and R223A mutant. PqsD was immobilized on a CMD500M sensor
chip at 12 °C using standard amine coupling analogous to the manufacturers’ instructions. PqsD was
diluted into sodium acetate buffer (10 mM, pH 4.5) to a final concentration of 100 µg/mL. PqsD was
immobilized at densities between 5000-6000 RU for binding studies and for mutagenesis studies. The
PqsD mutant was immobilized analogously to the wild type.
Competition studies. The competition studies were performed at a constant flow rate of 50 µL/min in
instrument running buffer (50 mM MOPS, pH 8.0, 150 mM NaCl, 5% DMSO (v/v), 0.1% Triton X
100 (v/v)). 10 mM stock solutions of compounds 1, 2 and 3 in DMSO were directly diluted to a concen-
tration of 500 µM (50 mM MOPS, pH 8.0, 150 mM NaCl, 0.1% Triton X-100 (v/v)) and then diluted to
a final concentration of 100 µM in running buffer (50 mM MOPS, pH 8.0, 150 mM NaCl, 5% DMSO
(v/v), 0.1% Triton X 100 (v/v)). Final concentration of DMSO was retained at 5% (v/v). Before the
1
9
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 28
compounds were injected, 6 warm-up blank injections were performed. Buffer blank injections and
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
DMSO calibration were included for double referencing.
ACoA competition. PqsD was preincubated with compound 1, therefore 1 was added to the running
buffer (100 µM). The sensor chip surface was flushed for several hours at a constant flow rate of
50 µL/min until the baseline was stable. Afterwards the flow rate was decreased to 10 µL/min and
ACoA was injected (100 µM) twice for 120 s association and 15 min dissociation times.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Competitor competition. PqsD was preincubated with compounds A or B, therefore the competitor
compound was added to the running buffer (100 µM). The sensor chip surface was flushed for several
hours at a constant flow rate of 50 µL/min until the baseline was stable. Afterwards the flow rate was
decreased to 10 µL/min and compound 1, 2 or 3 were injected (100 µM) twice for 120 s association and
3
00 s dissociation times. Due to a slow k rate of the competitor compounds, k values referring to the
off D
concentrations do not play a decisive role during the competition experiment.
Studies with R223A mutant. The experiments were performed at a constant flow rate of 50 µL/min in
instrument running buffer (50 mM MOPS, pH 8.0, 150 mM NaCl, 5% DMSO (v/v), 0.1% Triton X
1
00 (v/v)) at 12 °C. 10 mM stock solutions of compounds 1 or 2 in DMSO were diluted to a concentra-
tion of 100 µM analogous to the binding studies. Before the compound was injected, twelve warm-up
blank injections were performed. The obtained data were referenced against blank injections and DMSO
calibration. The compounds were injected twice for 120 s association and 300 s dissociation time.
Scrubber software was used for processing and analysing data.
Isothermal titration calorimetry. ITC experiments were carried out using an ITC200 instrument (Mi-
crocal Inc., GE Healthcare). Concentration of the ligand in DMSO stock solutions was 20 mM. Final
ligand concentrations were achieved by diluting 1:20 (v/v) in the experimental buffer resulting in a final
DMSO concentration of 5% (v/v). Protein concentration was determined by measuring the absorbance
−1
−1
at 280 nm using a theoretical molarity extinction coefficient of 17,780 M cm . DMSO concentration
2
0
ACS Paragon Plus Environment

Page 21 of 28
Journal of Medicinal Chemistry
in the protein solution was adjusted to 5% (v/v). ITC measurements were routinely performed at 25 °C
in PBS-buffer, pH 7.4, 10% glycerol (v/v), 5% DMSO (v/v). The titrations were performed on 100 μM
PqsD-His6 in the 200 µL sample cell using 2 µL injections of 1.0 mM ligand solution every 180 s. Raw
data were collected and the area under each peak was integrated. To correct for heats of dilution and
mixing the final baseline consisting of small peaks of the same size at the end of the experiment was
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
subtracted. The experimental data were fitted to a theoretical titration curve (one site binding model)
−1
using MicroCal Origin 7 software, with ΔH (enthalpy change in kcal mol ), K (association constant in
A
−
1
M ), and N (number of binding sites) as adjustable parameters. Thermodynamic parameters were calcu-
lated from equation:
ΔG = ΔH−TΔS = RT ln K = −RT ln KD
A
where ΔG, ΔH, and ΔS are the changes in free energy, enthalpy, and entropy of binding, respectively, T
−
1
−1
is the absolute temperature, and R = 1.98 cal mol K . For compound 1 four independent experiments
were performed.
Saturation transfer difference NMR. All NMR data were recorded at 298 K on a Bruker Avance 500
NMR instrument equipped with a cryogenically cooled z-shielded gradient probe. Experiments were
recorded with the carrier set at -2 ppm for on-resonance irradiation and 40 ppm for off-resonance irradi-
ation. Control spectra were recorded under identical conditions on samples containing free compound 7
to test for artifacts. Selective protein saturation (3 s) was accomplished using a train of 50 ms Gauss-
shaped pulses, each separated by a 1 ms delay, at an experimental determined optimal power (50 dB on
our probe); a T1ρ filter (15 ms) was incorporated to suppress protein resonances. Experiments were rec-
orded using a minimum of 512 scans and 32K points. On- and off-resonance spectra were processed
independently, and subtracted to provide a difference spectrum. Samples containing 7 and PqsD at final
concentrations of 1 mM and 10 µM, respectively, were prepared in 20 mM sodium phosphate, 50 mM
NaCl, and 5 mM MgCl , pH 7.0.
2
2
1
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 28
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
LC-ESI-MS. PqsD (25 µM) was preincubated with compound 12 (25 µM) for 30 min at 37°C in Tris-
HCl buffer (0.05 M, pH 8.0) with 0.5% DMSO. Dithiothreitol was added and the samples were ana-
lyzed by HPLC-ESI. All ESI-MS-measurements were performed on a Dionex Ultimate 3000 RSLC sys-
tem using an Aeris Widepore XB-C8, 150 x 2.1 mm, 3.6 µm dp column (Phenomenex, USA). Separa-
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
tion of 2 µl sample was achieved by a linear gradient from (A) H O + 0.05% TFA to (B) ACN + 0.05%
2
TFA at a flow rate of 250 µl/min and 45 °C. The gradient was initiated by a 1.0 min isocratic step at
1
5% B, followed by an increase to 80% B in 4.5 min to end up with a 6 min step at 80% B before re-
equilibration with initial conditions. UV spectra were recorded by a DAD in the range from 200 to 600
nm. The LC flow was split to 75 µl/min before entering the maXis 4G hr-ToF mass spectrometer
(Bruker Daltonics, Bremen, Germany) using the standard Bruker ESI source. In the source region, the
temperature was set to 180°C, the capillary voltage was 4000 V, the dry-gas flow was 6.0 l/min and the
nebulizer was set to 1.1 bar. Mass spectra were acquired in positive ionization mode ranging from 600 –
1
800 m/z at 2.5 Hz scan rate. Protein masses were deconvoluted by using the Maximum Entropy algo-
rithm (Copyright 1991-2004 Spectrum Square Associates, Inc.).
MALDI-TOF-MS. PqsD (10 µM) was preincubated with compound 12 (10 µM) for 60 min at 37°C
using identical buffer composition as in the screening assay procedure (0.05 M MOPS, pH 7.0 with
0
.005 % (w/v) Triton X-100 and 0.5 % DMSO). The buffer was exchanged by an NH HCO buffer
4 3
(0.05 M, pH 8.1) in three diafiltration steps. Diafiltration was performed at 1200 g for 6 min at 4°C in
Nanosep® Centrifugal Devices (MWCO = 10K) of Pall Corporation (Port Washington, NY, USA). The
protein was digested with trypsin overnight and dithiothreitol was added. α-Cyano-4-hydroxycinnamic
acid was used as matrix component. Analyzes of the peptides were performed on a 4800 TOF/TOF Ana-
lyzer mass spectrometer (Applied Biosystems, Darmstadt, Germany) in positive reflector mode using a
pulsed 200 Hz solid state Nd:YAG laser with a wavelength of 355 nm. Laser energy was set to 2000–
2
2
ACS Paragon Plus Environment

Page 23 of 28
Journal of Medicinal Chemistry
2
300 units for standards and to 2700–3200 units for real samples. Source 1 voltage was set to 20 kV
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
with a grid voltage of 16 kV. Reflector detector voltage was 2.19 kV. Spectra of standard peptides used
for wide range calibration ranging from 0.8 to 4 kDa (des-arg1-bradykinin, angiotensin I, glu1-
fibrinopeptide B, ACTH 1–17 clip, ACTH 18–39 clip and ACTH 7–38 clip) were measured with a delay
time of 600 ns. One single mass spectrum was formed from 20 subspectra per spot using 25 accepted
laser impulses each. From the standard peptides exclusively monoisotopic ions were used with a mini-
mum signal-to-noise (S/N) ratio of 20 and a resolution >10000. Mass tolerance was set to 0.3 Da with
maximum outlier of 5 ppm. Accepted calibration settings were used to measure real sample spectra in
the range of 1 to 3.5 kDa with a minimum S/N range of 10 and a resolution >8000. An internal algo-
rithm defined the isotope cluster area subsequently named intensity (I), based on the peptides’ molecular
weight and their general elemental composition. MALDI-TOF-MS resulted in pmfs consisting of mass-
intensity spectra (m/z–Iabs, ai).
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Molecular Modeling
2
4
Docking. Inhibitors were built in Molecular Operating Environment (MOE). The receptor was derived
6
from the co-crystal structure of anthraniloyl coenzyme A with PqsD (PDB Code: 3H77) ACoA, the
covalently bound anthranilate and H O were removed and Cys112 was restored considering its confor-
2
6
mation in 3H76. AutoDockTools V.1.5.6 was used to add polar hydrogens and to save the protein in the
2
5
appropriate file formate for docking with Vina. For docking AutoDockVina was used. The docking
parameters were kept at their default values. The docking grid was sized 18 Å x 24 Å x 24 Å, covering
the entire ACoA channel.
ASSOCIATED CONTENT
1
3
Supporting Information. Further experimental procedures, analytical data and C NMR spectra, selec-
tivity data (PqsD vs. RNAP) and more details concerning mutagenesis, SPR-competition, MALDI-TOF-
2
3
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 28
MS and LC-ESI-MS investigations. This material is available free of charge via the Internet at
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
http://pubs.acs.org
AUTHOR INFORMATION
Corresponding Author
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
*
Phone: +(49) 681 302 70300. E-mail: rolf.hartmann@helmholtz-hzi.de.
Author Contributions
The manuscript was written through contributions of all authors. / All authors have given approval to the
final version of the manuscript. / ‡These authors contributed equally.
ACKNOWLEDGMENT
The authors thank Carina Scheid for technical assistance, Patrick Fischer for his help performing the
synthesis, Dr. K. Hollemeyer for MALDI-TOF investigations and Michael Hoffmann for LC-ESI-MS
measurements. J.H. Sahner gratefully acknowledges a scholarship from the Stiftung der Deutschen
Wirtschaft (SDW).
ABBREVIATIONS
ACN,
acetonitrile;
ACoA,
anthraniloyl
coenzyme
A;
CHAPS,
3-[(3-
Cholamidopropyl)dimethylammonio]-1-propanesulfonate hydrate; CMD, carboxymethyl dextrane, cyti-
dine triphosphate; EtOAc, ethylacetate; EtOH, ethanol; GTP, guanosine triphosphate; HAC, heavy atom
count; HHQ, heptyl-4-hydroxy-quinoline; ITC, isothermal titration calorimetry; MeOH, methanol;
MOE, Molecular Operating Environment; MOPS, 4-Morpholinepropanesulfonic acid; NTA, nitrilotri-
acetic acid; NTP, nucleotide triphosphate; PDA, photodiode array; PQS, Pseudomonas quinolone signal;
RNAP, RNA polymerase; RP, reversed phase; RU, response unit; SPR, surface plasmon resonance,
STD, saturation transfer difference; UTP, uridine triphosphate
2
4
ACS Paragon Plus Environment

Page 25 of 28
Journal of Medicinal Chemistry
REFERENCES
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
(1) Wagner, G.; Hyberts, S. G.; Havel, T. F. NMR structure determination in solution: a critique and
comparison with X-ray crystallography. Annu. Rev. Biophys. Biomol. Struct. 1992, 21, 167–198.
(2) Davis, A. M.; Teague, S. J.; Kleywegt, G. J. Application and limitations of X-ray crystallograph-
ic Data in structure-based ligand and drug design. Angew. Chem. Int. Ed. 2003, 42, 2718–2736.
(3) Boström, J. Reproducing the conformations of protein-bound ligands: a critical evaluation of
several popular conformational searching tools. J. Comp-Aided Mol. Des. 2001, 15, 1137–1152.
(4) Xiao, G.; Dézièl, E.; He, J.; Lépine, F.; Lesic, B.; Castonguay, M.-H.; Milot, S.; Tampakaki, A.
P.; Stachel, S. E.; Rahme, L. G. MvfR, a key Pseudomonas aeruginosa pathogenicity LTTR-
class regulatory protein, has dual ligands. Mol. Mirob. 2006, 62, 1689–1699.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(5) For a recent review see: Dubern, J.-F.; Diggle, S. P. Quorum sensing by 2-alkyl-4-quinolones in
Pseudomonas aeruginosa and other bacterial species. Mol. Biosyst. 2008, 4, 882–888.
(6) Bera, A.K.; Atanasova, V.; Robinson, H.; Eisenstein, E.; Coleman, J. P.; Pesci, E. C.; Parsons, J.
F. Structure of PqsD, a Pseudomonas quinolone signal biosynthetic enzyme, in complex with an-
thranilate. Biochemistry 2009, 48, 8644–8655.
(7) Storz, M. P.; Maurer, C. K.; Zimmer, C.; Wagner, N.; Brengel, C.; de Jong, J. C.; Lucas, S.;
Müsken, M.; Häussler, S.; Steinbach, A.; Hartmann, R. W. Validation of PqsD as an anti-biofilm
target in Pseudomonas aeruginosa by development of small-molecule inhibitors. J. Am. Chem.
Soc. 2012, 134, 16143–16146.
(
8) Gomez, M. I.; Prince, A. Opportunistic infections in lung disease: Pseudomonas infections in
cystic fibrosis. Curr. Opin. Pharmacol. 2007, 7:244–251.
(
9) Hakki, M.; Limaye, A. P.; Kim, H. W.; Kirby, K. A.; Corey, L.; Boeckh, M. Invasive Pseudomo-
nas aeruginosa infections: high rate of recurrence and mortality after hematopoietic cell trans-
plantation. Bone Marrow Transpl. 2007, 39, 687–693.
2
5
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 28
(
(
(
10) Asboe, D.; Gant, V.; Aucken, H. M.; Moore, D. A.; Umasankar, S.; Bingham, J. S.; Kaufmann,
M. E.; Pitt, T. L. Persistence of Pseudomonas aeruginosa strains in respiratory infection in AIDS
patients. AIDS 1998, 12,1771–1775.
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
11) Pistorius, D.; Ullrich, A.; Lucas, S.; Hartmann, R. W.; Kazmaier, U.; Müller, R. Biosynthesis of
2
-alkyl-4(1H)-quinolones in Pseudomonas aeruginosa: potential for therapeutic interference
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
with pathogenicity. ChemBioChem 2011, 12, 850–853.
12) Sahner, J. H.; Groh, M.; Negri, M.; Haupenthal, J.; Hartmann, R. W. Novel small moleucle in-
hibitors targeting the “switch region” of bacterial RNAP: structure-based optimization of a virtu-
al screening hit. Eur. J. Med. Chem. 2013, 65, 223–231.
(
13) Hentzer, M.; Givskov, M. Pharmacological inhibition of quorum sensing for treatment of chronic
bacterial infections. J. Clin. Invest. 2003, 112, 1300–1307.
(
14) Storz, M. P.; Brengel, C.; Weidel, E.; Hoffmann, M.; Empting, M.; Steinbach, A.; Müller, R.;
Hartmann, R. W. unpublished results.
(15) O‘Shea, R.; Moser, H.E. Physicochemical properties of antibacterial compounds: implications
for drug discovery. J. Med. Chem. 2008, 51, 2871–2878.
(
(
(
16) Mayer, M.; Meyer, B. Characterization of ligand binding by saturation transfer difference NMR
spectroscopy. Angew. Chem. Int. Ed. 1999, 38, 1784–1788.
17) Henise, J. C.; Taunton, J. Irreversible Nek2 kinase inhibitors with cellular activity. J. Med.
Chem. 2011, 54, 4133–4146.
18) Copeland, R. A.; Pompliano, D. L.; Meek, T. D. Drug-target residence time and its implications
for lead optimization. Nat. Rev. Drug Discov. 2006, 5, 730–739.
(19) Hartmann, H.; Liebscher, J. A simple method for the synthesis of 5-aryl-3-amino-2-
alkoxycarbonylthiophenes. Synthesis 1984, 275–276.
(20) Fabis, F.; Jolivet-Fouchet, S.; Robba, M.; Landelle, H.; Rault, S. Thiaisatoic anhydrides: effi-
cient synthesis under microwave heating conditions and study of their reactivity. Tetrahedron
1
998, 54, 10789–10800.
2
6
ACS Paragon Plus Environment

Page 27 of 28
Journal of Medicinal Chemistry
(
21) Foulon, L.; Braud, E.; Fabis, F.; Lancelot, J.C.; Rault, S. Synthesis and combinatorial approach
of the reactivity of 6-and 7-arylthieno [3, 2-d][1, 3] oxazine-2, 4-diones. Tetrahedron 2003, 59,
10051–10057.
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
(
22) Le Foulon, F.X.; Braud, E.; Fabis, F.; Lancelot, J.C.; Rault, S. Solution-phase parallel synthesis
of a 1140-member ureidothiophene carboxylic acid library. J. Comb. Chem. 2005, 7, 253–257.
23) Haupenthal, J.; Hüsecken, K.; Negri, M.; Maurer, C.K.; Hartmann, R.W. Influence of DNA tem-
plate choice on transcription and inhibition of E. coli RNA polymerase. Antimicrob. Agents Ch.
2012, 56, 4536–4539.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(
(24) Molecular Operating Environment (MOE), 2010.10; Chemical Computing Group Inc., 1010
Sherbooke St. West, Suite #910, Montreal, QC, Canada, H3A 2R7, 2010.
(25) Trott, O.; Olson, A. J. AutoDock Vina: improving the speed and accuracy of docking with a new
scoring function, efficient optimization, and multithreading. J. Comput. Chem. 2009, 31, 455–
4
61.
2
7
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 28
Table of Contents Graphic
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
8
ACS Paragon Plus Environment