Structure optimization of 2-benzamidobenzoic acids as PqsD inhibitors for Pseudomonas aeruginosa infections and elucidation of binding mode by SPR, STD NMR, and molecular docking

Article abstract of DOI:10.1021/jm4006302

Pseudomonas aeruginosa employs a characteristic pqs quorum sensing (QS) system that functions via the signal molecules PQS and its precursor HHQ. They control the production of a number of virulence factors and biofilm formation. Recently, we have shown that sulfonamide substituted 2-benzamidobenzoic acids, which are known FabH inhibitors, are also able to inhibit PqsD, the enzyme catalyzing the last and key step in the biosynthesis of HHQ. Here, we describe the further optimization and characterization of this class of compounds as PqsD inhibitors. Structural modifications showed that both the carboxylic acid ortho to the amide and 3′-sulfonamide are essential for binding. Introduction of substituents in the anthranilic part of the molecule resulted in compounds with IC₅₀ values in the low micromolar range. Binding mode investigations by SPR with wild-type and mutated PqsD revealed that this compound class does not bind into the active center of PqsD but in the ACoA channel, preventing the substrate from accessing the active site. This binding mode was further confirmed by docking studies and STD NMR.

Full text of DOI:10.1021/jm4006302

Subscriber access provided by University of Massachusetts Amherst Libraries
Article
Structure Optimization of 2-Benzamidobenzoic Acids as PqsD
Inhibitors for Pseudomonas aeruginosa Infections and Elucidation
of Binding Mode by SPR, STD NMR and Molecular Docking
Elisabeth Weidel, Johannes C. DeJong, Christian Brengel, Michael P. Storz, Andrea
Braunshausen, Matthias Negri, Alberto Plaza, Anke Steinbach, Rolf Mueller, and Rolf W. Hartmann
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/jm4006302 • Publication Date (Web): 08 Jul 2013
Downloaded from http://pubs.acs.org on July 13, 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 36
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Structure Optimization of 2ꢀBenzamidobenzoic
Acids as PqsD Inhibitors for Pseudomonas
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
aeruginosa Infections and Elucidation of Binding
Mode by SPR, STD NMR and Molecular Docking
Elisabeth Weidel,¶ ‡ Johannes C. de Jong,†‡ Christian Brengel,† Michael P. Storz,† Andrea
⊥
Braunshausen†, Matthias Negri,† Alberto Plaza,§ Anke Steinbach,† Rolf Müller,§ Rolf W.
Hartmann†¶*
† HelmholtzꢀInstitute for Pharmaceutical Research Saarland (HIPS), Department of Drug Design
and Optimization, Campus C2.3, 66123 Saarbrücken, Germany
¶
Pharmaceutical and Medicinal Chemistry, Saarland University, Campus C2.3, 66123
Saarbrücken, Germany
§
HelmholtzꢀInstitute for Pharmaceutical Research Saarland (HIPS), Department of Microbial
Natural Products, Campus C2.3, 66123 Saarbrücken, Germany
⊥ ElexoPharm GmbH, Im Stadtwald A1.2, 66123 Saarbrücken, Germany
1
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 36
1
2
3
4
KEYWORDS
5
6
7
Pseudomonas aeruginosa, PqsD inhibitors, 2ꢀbenzamidobenzoic acids, binding mode, SPR, STD
8
9
NMR, Molecular Docking
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ABSTRACT
Pseudomonas aeruginosa employs a characteristic pqs quorum sensing (QS) system that
functions via the signal molecules PQS and its precursor HHQ. They control the production of a
number of virulence factors and biofilm formation. Recently, we have shown that sulfonamide
substituted 2ꢀbenzamidobenzoic acids, which are known FabH inhibitors, are also able to inhibit
PqsD, the enzyme catalyzing the last and key step in the biosynthesis of HHQ. Here, we describe
the further optimization and characterization of this class of compounds as PqsD inhibitors.
Structural modifications showed that both the carboxylic acid ortho to the amide and 3'ꢀ
sulfonamide are essential for binding. Introduction of substituents in the anthranilic part of the
molecule resulted in compounds with IC50 values in the low micromolar range. Binding mode
investigations by SPR with wildꢀtype and mutated PqsD revealed that this compound class does
not bind into the active center of PqsD, but in the ACoA channel preventing the substrate from
accessing the active site. This binding mode was further confirmed by docking studies and STD
NMR.
INTRODUCTION
The Gramꢀnegative bacterium Pseudomonas aeruginosa is an opportunistic pathogen that is
commonly associated with hospitalꢀacquired lifeꢀthreatening infections. Especially patients with
2
ACS Paragon Plus Environment

Page 3 of 36
Journal of Medicinal Chemistry
1
2
severe burns and immunocompromised individuals,1 like those undergoing chemotherapy or
having HIV/AIDS2 are highly susceptible to infections with P. aeruginosa. A wide variety of
infection sites can be observed, such as ocular, ear, urinary tract, bloodstream, skin and soft
tissue infections, and hospitalꢀacquired pneumonia. Furthermore it is the dominant cause of
chronic lung infections in the majority of people with cystic fibrosis (CF).3ꢀ5 Due to the high
intrinsic resistance of P. aeruginosa to many antimicrobial agents, antimicrobial therapy is
challenging. The resistance arises from the combination of low outer membrane permeability and
resistance mechanisms such as efficient multidrug efflux pumps and βꢀlactamases.6 Treatment is
further hampered when the bacterium is growing in biofilms.7, 8 As a consequence infections
caused by bacterial biofilms are generally chronic and very difficult to eradicate.9 Due to the
rising levels of resistance to conventional antibiotics, there is an urgent need for new therapeutic
options.
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
The production of several virulence factors as well as biofilm formation10, 11 is coordinated by a
cell density dependent mechanism termed quorum sensing (QS) using a number of QS signal
molecules.12, 13 This mechanism of communication enables bacteria to coordinate the regulation
of gene expression with subsequent effects on metabolism, protein synthesis and virulence.14
Three different QS systems are known for P. aeruginosa. The las15, 16 and rhl17, 18 systems use
3ꢀoxoꢀC12ꢀHSL (Nꢀ(3ꢀoxododecanoyl) homoserine lactone) and C4ꢀHSL (Nꢀbutyryl homoserine
lactone), respectively, as signal molecules, whereas the pqs system utilizes the 2ꢀheptylꢀ3ꢀ
hydroxyꢀ4(1H)ꢀquinolone, called Pseudomonas Quinolone Signal (PQS) and its immediate
precursor 2ꢀheptylꢀ4ꢀquinolone (HHQ).19, 20 Both PQS and HHQ bind to the transcriptional
regulator PqsR, the receptor of the pqs system. Among other factors, this system is involved in
the regulation of multiple virulence determinants21, 22 such as elastase, rhamnolipids and
3
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 36
1
2
pyocyanin and also influences biofilm formation.23, 24 Therefore PqsD is considered as an
attractive therapeutic target for drug development and treatment of P. aeruginosa infections. It is
believed that disrupting the cellꢀtoꢀcell communication instead of killing the bacteria will result
in less selective pressure than with conventional antibacterial agents such as antibiotics25 and
thus will circumvent the problem of resistance.
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Using a ligandꢀbased approach we developed the first highly potent competitive PqsR
antagonists, which were able to reduce the production of the virulence factor pyocyanin.26, 27 As
HHQ and PQS activate PqsR, inhibition of their synthesis should lead to a decrease of virulence
factor production without interfering with bacterial growth.28, 29 PqsD is an essential enzyme in
the biosynthesis of HHQ and PQS as it catalyzes the last step in HHQ formation, the
condensation of anthraniloylꢀCoA (ACoA) with ßꢀketodecanoic acid (Scheme 1).30
Scheme 1. Biosynthesis of signal molecules HHQ and PQS
Recently we have shown for the first time that inhibition of PqsD with a small molecule
(compound I, Scheme 2) leads to a strong reduction of extracellular HHQ and PQS levels and a
significant reduction of biofilm volume.31 Structurally, PqsD most closely resembles βꢀketoacylꢀ
4
ACS Paragon Plus Environment

Page 5 of 36
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
ACP synthase (FabH) having the same catalytic triad consisting of Cys112, His257 and Asn287.
Therefore, we postulated that inhibitors of FabH might also act as PqsD inhibitors. For three
FabH inhibitors (compound II, III and IV, Scheme 2)30, 32, 33, two of them were anthranilic acid
derived, we could demonstrate that they indeed inhibit PqsD.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Scheme 2. Structures of the PqsD inhibitor I and the FabH inhibitors II, III and IV as well as the
according PqsD IC50 values. V represents the general structure of synthesized compounds.
In the current work, we report about the optimization of compound II for which a series of 3ꢀ
sulfonamide substituted benzamidobenzoic acids (general structure V) were synthesized. The
binding mode of the potent inhibitors was investigated using surface plasmon resonance
spectroscopy (SPR), molecular docking studies and saturation transfer difference nuclear
magnetic resonance (STD NMR).
RESULTS AND DISSCUSSION
Chemistry Reaction of amines 1, 3, 7ꢀ8 and 11ꢀ12 with the appropriate acid chloride afforded
the amides 2a, 4a, 9aꢀ10a and 13aꢀ14a (see SI). Subsequent basic hydrolysis of the carboxylic
5
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 36
1
2
3
4
5
6
ester groups yielded the corresponding acids 2, 4, 9ꢀ10 and 13ꢀ14. Carboxamide 6 was prepared
from 2ꢀaminobenzamide (5) and 3ꢀ(N,Nꢀdiethylsulfamoyl)benzoyl chloride (Scheme 3).
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Scheme 3. Reaction of benzoyl chlorides with methyl aminobenzoates and 2ꢀaminobenzamidea
aReagents and conditions: (a) 3ꢀ(N,Nꢀdiethylsulfamoyl)benzoyl chloride 1, 3: K2CO3, Nꢀ
methylimidazole, TMEDA, CH3CN; 5: toluene; (b) 4ꢀ(N,Nꢀdiethylsulfamoyl)benzoyl chloride;
K2CO3, Nꢀmethylimidazole, TMEDA, CH3CN; (c) 3ꢀ(diethylcarbamoyl)benzoyl chloride,
K2CO3, Nꢀmethylimidazole, TMEDA, CH3CN; (d) 1N NaOH (aq.), MeOH, THF (not 5).
As outlined in Scheme 4 the synthesis of 3ꢀsulfamoylbenzoic acids 22ꢀ33 started from paraꢀ
substituted benzoic acids 15ꢀ17, which were converted into 3ꢀ(chlorosulfonyl)benzoic acids 18ꢀ
20 by reaction with chlorosulfonic acid at 100 °C in 82ꢀ88% yield.
6
ACS Paragon Plus Environment

Page 7 of 36
Journal of Medicinal Chemistry
1
2
3
4
Scheme 4. Synthesis of 3'ꢀsulfonamide substituted benzamidobenzoic acids and their
5
6
7
intermediatesa
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
aReagents and conditions: (a) ClSO3H, 100 °C, overnight; (b) HNR2R3, CH2Cl2; (c) (i) SOCl2,
DMF (cat.), CH2Cl2, overnight or SOCl2, DMF (cat.), toluene, 80 °C, 4h; (ii) methyl anthranilate;
(iii) 1N NaOH (aq.), MeOH, THF, rt; (d) phenylboronic acid, cesium carbonate, DME/water 1/1,
100 °C, 3.5 h.
Compound 21 was commercially available. The sulfonyl chlorides 18ꢀ21 were then reacted with
ammonia or a secondary amine to afford 3ꢀsulfamoylbenzoic acids 22ꢀ33. Reaction of the
carboxylic acids with thionyl chloride, either in dichloromethane at room temperature or in
toluene at 80 °C gave the corresponding acid chlorides, which were reacted under standard
conditions with the methyl ester of anthranilic acids. The resulting compounds 34aꢀ52a and 57aꢀ
64a (see SI) were hydrolyzed with a mixture of 1N NaOH (aq.), MeOH and THF, yielding the
corresponding 2ꢀ(3'ꢀ(sulfamoyl)benzamido) benzoic acids 34ꢀ52 and 57ꢀ64. Phenyl substituted
compounds 53ꢀ56 were obtained in 60ꢀ76% yield by means of a Suzuki reaction of methyl 5ꢀ
7
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 36
1
2
3
4
5
6
7
8
9
bromoꢀ2ꢀ(3ꢀ(N,Nꢀdiethylsulfamoyl) benzamido) benzoate (43a) with phenylboronic acids. Under
the reaction conditions used, the intermediate methyl esters were already hydrolyzed to the
desired carboxylic acids.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
PqsD inhibition In order to investigate the importance of the carboxylic acid and the
sulfonamide moiety of the 3'ꢀsulfonamide substituted benzamidobenzoic acids (general structure
V) 2ꢀ(3'ꢀ(N,Nꢀdiethylsulfamoyl)benzamido)benzoic acid (2) and two regioisomers (4 and 9) were
synthesized. Furthermore, the carboxylic acid and sulfonamide were replaced by a carboxamide
(6 and 13, Scheme 5).
Scheme 5. Influence of sulfonamide and benzoic acid modifications on PqsD inhibition.
Compound 2 lacking the Br substituent showed reduced activity compared to 34 (IC50 = 19.8 vs
6.9 ꢁM). The carboxylic acid ortho to the NH turned out to be essential for activity. Changing
the position of the carboxylic acid to the meta position (4) as well as replacement with a
carboxamide (6) led to a dramatic loss in activity. Shifting the N,Nꢀdiethylsulfamoyl group from
3'ꢀposition to 4'ꢀposition34 (9) also resulted in reduced activity, whereas replacement with an
N,Nꢀdiethylcarboxamide (13) gave an inactive compound. As the structure of compound 2,
8
ACS Paragon Plus Environment

Page 9 of 36
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
having a carboxylic group next to the amide and with the sulfonamide in 3'ꢀposition, seemed to
be optimal, we turned our attention to the introduction of additional substituents on both
aromatic rings (Table 1).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Table 1. PqsD inhibition by compounds 10, 14 and 2ꢀ(3'ꢀsulfamoyl)benzamido)benzoic acids 2
and 34-64.
Compd
2
R1
H
R2
R3
IC50 (ꢁM)
19.8 ± 4.5
6.9 ± 0.8
27.3 ± 1.9
39%*
3'ꢀSO2NEt2
H
4'ꢀBr, 3'ꢀSO2NEt2
4'ꢀMe, 3'ꢀSO2NEt2
4'ꢀEt, 3'ꢀSO2NEt2
3'ꢀSO2NEt2
34
35
36
37
38
39
40
41
42
43
44
45
10
14
H
H
4ꢀCl
4ꢀF
9.4 ± 1.3
8.0 ± 1.2
6.3 ± 1.1
18.4 ± 3.2
12.4 ± 3.0
11.4 ± 1.9
9.9 ± 2.1
26.2 ± 10.4
8.9 ± 1.8
13.0 ± 1.8
25.5 ± 5.5
3'ꢀSO2NEt2
4ꢀNO2
5ꢀMe
5ꢀCF3
5ꢀF
3'ꢀSO2NEt2
3'ꢀSO2NEt2
3'ꢀSO2NEt2
3'ꢀSO2NEt2
5ꢀBr
5ꢀCN
5ꢀNO2
5ꢀBr
5ꢀBr
3'ꢀSO2NEt2
3'ꢀSO2NEt2
3'ꢀSO2NEt2
4'ꢀSO2NEt2
3'ꢀC(O)NEt2
9
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 36
1
2
3
4
5
6
7
8
9
5ꢀF
5ꢀBr
4'ꢀBr, 3'ꢀSO2NEt2
4'ꢀBr, 3'ꢀSO2NEt2
3'ꢀSO2NEt2
6.6 ± 0.1
3.8 ± 1.0
3.0 ± 0.7
44 %*
46
47
48
49
50
51
52
53
54
55
56
57
58
59
5ꢀPh
6ꢀOMe
6ꢀCl
3'ꢀSO2NEt2
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
3'ꢀSO2NEt2
39.0 ± 1.9
24.9 ± 4.0
1.2 ± 0.1
3.8 ± 0.4
1.9 ± 0.1
1.5 ± 0.3
2.7 ± 0.3
43.6 ± 1.3
35%*
6ꢀF
3'ꢀSO2NEt2
6ꢀOH
3'ꢀSO2NEt2
3'ꢀSO2NEt2
3'ꢀSO2NEt2
3'ꢀSO2NEt2
3'ꢀSO2NEt2
3'ꢀSO2NH2
3''ꢀC(O)NH2ꢀ
4''ꢀC(O)NH2
3''ꢀCO2H
4''ꢀCO2H
H
H
H
3'ꢀSO2NMe2
3'ꢀSO2N(nꢀPr)2
5.4 ± 1.0
H
H
H
H
H
47%*
60
61
62
63
64
14.4 ± 2.4
14.8 ± 2.4
16%*
16.5 ± 3.8
Data shown are mean ± SD, n = 3.
* Inhibition at 50 ꢁM
Binding mode characterization
SPR measurements In order to gain some insight into the binding mode of the potent inhibitors
SPR experiments were performed. At first compounds 43 and 48 were tested in absence of
ACoA and strong interactions with PqsD were observed (Figure 1A). In a second experiment the
10
ACS Paragon Plus Environment

Page 11 of 36
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
PqsD loaded sensor chip was first treated with ACoA. Nucleophilic attack of Cys112 to the
thioester of ACoA and subsequent elimination of CoA is known to result in an anthranilateꢀPqsD
complex,36 in which the catalytic center is occupied by the covalent bound anthranilate.
Subsequent injection of compounds 43 and 48 resulted in Figure 1B showing comparable signals
for the ACoAꢀtreated and the ꢀuntreated PqsD. This indicates that the compounds do not occupy
the catalytic site of PqsD. These results are in contrast to what was observed for the sulfone (IV)
(Figure 1C and 1D). Performing the experiment in the same way no signal for IV was observed
when injected after preincubation with ACoA, indicating that this compound is competing with
anthranilate for the same binding site.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 1. SPR binding studies of compounds 43, 48 and IV. A) Binding of 43 (blue) and 48 (red)
to PqsD. B) Binding of 43 (blue) and 48 (red) after pretreatment of the chip with ACoA, which
results in a covalent anthranilateꢀCys112ꢀPqsD complex with part of the active site being
blocked. CꢀD) Same experiments performed with IV.
This also means that the two inhibitor classes have different modes of action: compound IV is
competing with the catalytic triad, whereas the anthranilic acid derivatives are “channel
blockers”. Another SPR experiment was performed inverting the compound order. PqsD was
11
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 36
1
2
3
4
5
6
7
8
9
preincubated with 48 and ACoA was injected subsequently. The finding that in contrast to the
untreated surface no binding of ACoA to PqsD was observed indicates that 48 blocks the ACoA
channel and prevents ACoA from reaching its binding site.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Further support for this mode of action comes from our screening assay. In the standard
procedure purified PqsD is first incubated with test compound. After 10 minutes ACoA and ßꢀ
ketodecanoic acid are added and the enzyme activity is monitored by measuring the amount of
HHQ being formed during 40 minutes reaction time. In a modified assay procedure, a 30 min
preincubated enzyme / ACoA mixture is first added to the test compound. After 10 minutes, the
reaction is started by addition of ßꢀketodecanoic acid and the enzyme activity is monitored as
mentioned before. In contrast to experiments with compound IV, very similar IC50 values were
observed in both settings, indicating that the benzamidobenzoic acids do not compete in the
active site of PqsD with the covalently bound anthranilate (Table 2).
12
ACS Paragon Plus Environment

Page 13 of 36
Journal of Medicinal Chemistry
1
2
3
4
Table 2. Comparison of IC50 values obtained with standard procedure (SP) and modified
5
6
procedure (MP).
7
8
9
Compd
IC50 [ꢁM] SP IC50 [ꢁM] MP
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
0.15 ± 0.1
19.8 ± 4.5
9.9 ± 2.1
3.0 ± 0.7
1.2 ± 0.1
1.5 ± 0.3
2.7 ± 0.3
16.5 ± 3.8
1.1 ± 0.1
19.4 ± 0.8
7.0 ± 0.4
4.2 ± 0.3
1.1 ± 0. 2
1.6 ± 0.1
2.0 ± 0.2
12.9 ± 0.1
IV
2
43
48
52
55
56
64
Data shown are mean ± SD, n = 3.
For further binding mode investigation, additional SPR experiments were performed with
compounds 34, 43 and 48 differing in the size of the substituents in position 5 of the anthranilate.
To investigate how deep they reach into the channel the compounds were tested on wildꢀtype and
on three PqsD mutants in which active site residues were exchanged (Ser317Phe, Cys112Ala and
His257Phe). Notably, different binding behaviors were observed. For compound 34 no
differences in binding signals were seen, indicating that the binding is not influenced by the
amino acid exchanges. Increasing the size of the substituent (H to Br, 43) resulted in slightly
decreased binding signals compared to wildꢀtype PqsD. This observation is more pronounced for
compound 48, where the additional phenyl ring leads to strongly reduced binding signals (Table
3).
13
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 36
1
2
3
4
5
6
Table 3. SPR based binding studies with wildꢀtype and mutated PqsD (Ser317Phe, Cys112Ala,
and His257Phe), for compounds 34, 43 and 48 (n=2).
7
8
9
Compound
Wild-type
Ser317Phe
Cys112Ala
His257Phe
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
34
43
48
These results suggest that bulky substituents on the anthranilate are likely to reach the active
center potentially interacting with Ser317, Cys112 and His257. The decreased binding signals
observed in the mutants might either be due to missing interactions caused by amino acid
replacement or due to conformational changes in the active site. The fact that the three
investigated compounds are differently influenced by the amino acid exchanges is very likely to
be caused by the size of the substituents the more bulky ones reaching deeper into the catalytic
center.
Docking studies In order to rationalize the SPR experiments and to establish a basis to explain
the SAR of the PqsD inhibitors molecular docking studies were performed with selected
compounds. For compound 2 three main binding clusters were found all localized within the
ACoA channel (Figure 2 and S1 ). In binding mode I (Figure 2I) 2 is placed deep into the
14
ACS Paragon Plus Environment

Page 15 of 36
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
channel: it is stabilized by hydrogen bonds to Cys112 (carboxylic acid) and Arg262
(sulfonamide oxygens), CHꢀπ interactions to Leu193 (anthranilic ring) and Asn260 (N,Nꢀ
diethylsulfonamidebenzene moiety), and hydrophobic interactions with further amino acids. In
binding mode II (Figure 2II) the inhibitor is placed more central in the channel: hydrogen bonds
are formed to Asn287 (carboxylic acid) and Arg36/Arg223 (sulfonamide oxygens), πꢀstackings
to Pro259/Met225 (anthranilic ring) and Met220 and Arg262 (sulfonamidebenzene), and
additional hydrophobic interactions. In mode III (Figure 2III) 2 is rotated 180 degrees placing the
anthranilic ring at the channel entrance: the carboxylic acid binds to Arg262 and Arg36, while
CHꢀπ interactions with Thr37, Pro259 and Ile263 stabilize the aromatic scaffold. The N,Nꢀ
diethyl group is buried into to the catalytic center forming hydrophobic interactions
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
15
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 36
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 2. Three main binding modes I, II and III of compound 2 in the ACoA channel of PqsD
and (IV) superimposition of the three main binding modes of compound 2 shown as orange
16
ACS Paragon Plus Environment

Page 17 of 36
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
(binding mode I), green (binding mode II) and cyan (binding mode III) sticks. The main
interacting amino acids are shown as lines and key interactions are highlighted as dotted lines
(polar in yellow, hydrophobic in dark cyan).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
In the most common pose of compound 48 the phenyl ring in position 5 is sandwiched between
Met225 and Pro259, as seen for the anthranilic ring of compound 2 in binding mode II. The rest
of the molecule is slightly pushed outwards (Figure 3 and S1) with the anthranilate forming
hydrogen bonds to Arg262, Asn154 or Asn260. In contrast to compound 2 no hydrogen bonds
are found to the sulfonamide oxygens. πꢀStacking with Phe226 (anthranilic ring) and Phe32
(diethylꢀsulfonamideꢀbenzene) and hydrophobic interactions (Figure S1) further stabilize the
inhibitor.
Figure 3. Docking pose of compound 2 (binding mode II; green) and 48 (magenta).
For compound 48 a dissociation constant (KD) of 5.2 ꢁM was determined by ITC (Figure 4). An
explanation for these results could be that the main interactions for this compound are
hydrophobic. The low value of enthalpy could be due to a desolvation penalty (ꢂG = ꢀ7.2
17
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 36
1
2
3
4
5
6
7
8
9
kcal/mol, ꢂH = ꢀ2.3 kcal/mol, ꢀTꢂS = ꢀ5.0 kcal/mol). In two of the PqsD structures (3H76 and
3H78) a water molecule is present bridging to the catalytic residues Cys112, His257 and Asn287.
When this water molecule was included in the docking process as part of the protein compound
52 was found deeper in the channel. The OH group forms an intramolecular hydrogen bond to
the carboxylic acid (maintaining its planarity), which binds to Asn287, but also interacts via the
structural water molecule with Cys112 and His257 (Figure S2). This would explain the strong
increase in activity and is further supported by the fact that for the methoxy derivative less than
50% inhibition at 50 ꢁM is observed.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 4. ITC experiments with compound 48 and wildꢀtype PqsD. A: Integrated heats (black
squares) plotted against the molar ratio of the reactants. B: Thermodynamic profile of 48. Data
shown are mean ± SD, n = 3.
18
ACS Paragon Plus Environment

Page 19 of 36
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
STD NMR studies To identify the orientation and mode of binding to PqsD, we recorded
Saturation Transfer Difference (STD) NMR37 spectra of 2, 48, 52, and 64 in the presence of
PqsD (Figure 5 and S4ꢀS7). STD NMR is a powerful technique that allows the precise definition
of binding epitopes on small molecule ligands. Samples contained an 80ꢀfold excess of the
compounds relative to PqsD with respective concentrations of 0.8 mM and 10 ꢁM were recorded
at 298 K. As seen in Figure 5 for compound 2, normalization of the signal of greatest intensity
(Hꢀ6) in the difference spectrum (red) to the one of the reference spectrum (black) showed
signals for the anthranilate protons Hꢀ5 (95%), Hꢀ4 (86%), and Hꢀ3 (81%) to display the
strongest enhancements. Signals for the benzene protons Hꢀ2' and Hꢀ4', Hꢀ5', and Hꢀ6' showed
enhancements of ~70% while significantly reduced enhancements were observed for the
methylene (42%) and methyl protons (39%, Figure S4). These results indicate that the protons of
the anthranilic acid ring of 2 are in closer contact to the protein.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 5. Reference (black) and STD NMR difference (red) spectra of 2 in complex with PqsD.
Samples contained an 80ꢀfold excess of the compounds relative to PqsD with respective
19
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 36
1
2
3
4
concentrations of 0.8 mM and 10 ꢁM were recorded at 298 K. Overlaid spectra were normalized
5
6
to the signal for Hꢀ6 (δ 7.90), which gave the strongest enhancement.
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Similar epitope binding profiles were observed for 52 where once again the anthranilate protons
showed the strongest enhancements (Figure S5). The STD NMR spectrum of 64 in presence of
PqsD (Figure S6) displayed the strongest enhancements for anthranilate protons Hꢀ6 (100%), Hꢀ
4 (99%), and Hꢀ3 (90%), for benzene proton Hꢀ5' (93%), and for the Nꢀbenzylic proton Hꢀ
4'''(90%). It is worthwhile to note that enhancements of 87 and 80% were observed for
methylene protons Hꢀ1'' and methyl protons Hꢀ2'', respectively. The spectra of 2, 52 and 64
revealed a similar binding mode for all compounds, which is independent of the substitution on
the anthranilic acid ring or on the sulfonamide moiety. Low signal to noise ratio and spectral
overlap did not allow to integrate individual peaks corresponding to the benzene protons of 48 in
the STD NMR spectrum (Figure S7). However, it was clear that the benzene protons together
with the methylene protons showed the highest intensities.
The results of SPR, STD NMR and docking as presented are consistent with binding mode II,
where the methylene protons are within van der Waals contact distances (2,5ꢀ 3,5 Å) to Pro259
and Met225. In particular, the SPR experiment with ACoA injection after pretreatment with
inhibitor clearly speaks in favor of a channelꢀblocking mechanism for these inhibitors. Further,
NMR results exclude binding mode III, where the anthranilic ring is almost exposed at the
entrance of the ACoA channel (no correspondence between the intense signals and close enough
20
ACS Paragon Plus Environment

Page 21 of 36
Journal of Medicinal Chemistry
1
2
3
4
5
6
residues), and disfavor binding mode I, where the methyl groups are predicted to be in closer
contact to the protein.
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
SAR Discussion
From binding mode II of 2 it becomes apparent that an important interaction partner is the
carboxylic acid group in ortho position to the NH, which forms hydrogen bonds with Asn 287.
Shifting to meta position (4) as well derivatization to carboxamide (6) strongly reduces activity
(Scheme 5), as this interaction can no longer be sustained. Also of high importance is the 3ꢀ
position of the N,Nꢀdiethylsulfamoyl group to perfectly fit into the channel, and which must not
be exchanged by N,Nꢀdiethylcarboxamide as only the sulfur oxygens can establish hydrogen
bonds with Arg 36 and Arg 223 (compounds 9 and 13, respectively, Scheme 5).
The finding that the sulfonamides lacking the diethyl residues (57ꢀ 64) exhibit weak or no
inhibition can be explained by the strong interactions between the methylene and methyl protons
with the protons of the protein as shown in the NMR experiments. Introduction of substituents in
2 resulted in different effects which are in accordance with the proposed binding mode II. An
example is the Br substitution ortho to the sulfonamide group, which showed improved binding
(34, 46 ꢀ 47) compared to the unsubstituted parent compounds. In binding mode II the benzene
ring is oriented in a way in which the Br can build orthogonal halogen bonds with amino acids of
the protein backbone for example with Arg36 and Thr37 (Figure S8). A reduced activity
depending on the size of the substituents is observed when they were introduced into the 6ꢀ
position of the anthranilic acid ring (49 ꢀ 51). This is possibly due to a steric interference with the
carboxylic acid group, which in turn can no longer optimally interact with Asn287. 52 is an
exception, where introduction of OH in 6ꢀposition leads to a high inhibitory activity. As shown
21
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 36
1
2
3
4
5
6
7
8
9
above, a plausible reason for this could be the ability of the OH group to build a hydrogen bond
network with the carboxylic acid group and the amino acid residues of the active site (ITC and
docking results). The introduction of small substituents in 4 and 5 position of the anthranilic acid
ring leads to increased PqsD inhibition (37 ꢀ 47). As shown in the SPR experiments with mutated
PqsD, the docking results and NMR experiments the anthranilic acid ring is oriented towards the
active center, whereby substituents at positions 4 and 5 reach deeper into the channel and thus
can form additional interactions. This effect is more pronounced for 48, where a phenyl ring is
attached in 5ꢀposition, leading to a slight shift in the binding mode compared to compound 2 (see
above). Obviously, the gain in binding activity by the phenyl ring sandwiched between Met225
and Pro259 is higher than the loss of binding activity due to the missing interactions of the
sulfonamide ethyl groups. The introduction of an amide or carboxylic acid group in meta or
paraꢀposition (53ꢀ56) to address Asn 287 and His257 however, led only to slightly improved
inhibition compared to the parent compound 48.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
CONCLUSION
In summary, we synthesized a series of sulfonamide substituted 2ꢀbenzamidobenzoic acids
which inhibit PqsD in the low ꢁM range. The carboxylic acid was essential for activity.
Systematic variation of substituents in 4ꢀ and 5ꢀposition of the 2ꢀaminobenzoic acid part of the
scaffold resulted in the most potent PqsD inhibitors. Activity was reduced significantly when
substituents were introduced in 6ꢀposition. However, introduction of a hydroxyl group in this
position resulted in a compound with increased potency (IC50 = 1.2 ꢁM). SPR, ITC, modeling,
and NMR experiments suggest these inhibitors to bind within the ACoA channel of PqsD,
22
ACS Paragon Plus Environment

Page 23 of 36
Journal of Medicinal Chemistry
1
2
3
4
5
6
thereby acting as entropyꢀdriven channelꢀblockers that prevent ACoA from reaching the catalytic
site. Further biological evaluation of our compounds is under investigation.
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
EXPERIMENTAL SECTION
Chemical Synthesis. General Methods. Chemicals were purchased from commercial sources
1
and used without further purification. H NMR spectra were recorded on a Bruker DRXꢀ500
(500 MHz) and a Bruker Fourier 300 (300 MHz) spectrometer. Chemical shifts are given in parts
per million (ppm) with the solvent resonance as internal standard. Data are reported as follows;
chemical shift, multiplicity (s = singlet, d = doublet, t = triplet, q = quartet, m = multiplet, bs =
broad signal or as a combination of these, e.g. dd, dt, etc.), coupling constants (Hz) and
13
integration. C NMR spectra were recorded on a Bruker DRXꢀ500 (125 MHz) with complete
proton decoupling. Chemical shifts are given in parts per million (ppm) with the solvent
resonance as internal standard. Mass spectrometry (HPLC/MS) was performed on a MSQ®
electro spray mass spectrometer (Thermo Fisher). The system was operated by the standard
software Xcalibur®. A RP C18 NUCLEODUR® 100ꢀ5 (125 × 3 mm) column (Macherey ꢀ Nagel
GmbH) was used as stationary phase. The mobile phase was a mixture of CH3CN (0.1% TFA)
and H2O (0.1% TFA). All solvents were HPLC grade. Flash chromatography was performed on
silica gel 60, 70ꢀ230 mesh (Fluka) and the reaction progress was determined by thinꢀlayer
chromatography (TLC) analyses on silica gel 60, F254 (Merck). Visualization was accomplished
with UV light and staining with basic potassium permanganate (KMnO4). Melting points were
determined in open capillaries using a SMP3 melting point apparatus of Bibby Sterilin Ltd with a
gradient of 2 °C/min. The reported yields are the actual isolated yields of purified material,
unless stated otherwise, and are not optimized. Purities of all test compounds used in the
23
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 36
1
2
3
4
5
6
7
8
9
biological assays were determined by HPLC/MS using the area percentage method on the UV
trace recorded at a wave length of 254 nm. All compounds were found to have ≥95% purity.
Yields and characterization of all compounds are provided in the Supporting Information.
General Procedure A for the Preparation of 3ꢀSulfamoylbenzoic Acids. At 0 °C the amine (30.0
mmol) was added to a stirred solution of 3ꢀ(chlorosulfonyl)benzoic acid (2.21 g, 10.0 mmol) in
dichloromethane (25 mL). After stirring overnight at room temperature 1N HCl (aq., 25 mL) was
added. The organic solvent was removed by evaporation under reduced pressure. The solid was
isolated by suction filtration, washed thoroughly with water and dried under reduced pressure at
50 °C.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
General Procedure B for the Preparation of 3ꢀSulfamoylbenzoyl Chlorides. Thionyl chloride
(2.9 mL, 40.0 mmol) was added slowly to a stirred solution of 3ꢀsulfamoylbenzoic acid (4.00
mmol) in dichloromethane (25 mL) followed by 1ꢀ2 drops of DMF. After stirring overnight at
room temperature the solution was evaporated under reduced pressure. The crude 3ꢀ
sulfamoylbenzoyl chloride was used without further purification.
General Procedures C for the Preparation of 2ꢀBenzamidobenzoic Acid Methyl Esters.
Procedure C1. 3ꢀ(N,NꢀDiethylsulfamoyl)benzoyl chloride (1.00 mmol) was added to a stirred
solution of anthranilic acid methyl ester (1.00 mmol) in pyridine (4 mL). After stirring overnight
the solvent was removed by evaporation under reduced pressure. The residue was stirred with 1N
HCl (aq., 4 mL). The solid was isolated by suction filtration, washed with water and heated
briefly with a small amount of methanol. After cooling to room temperature the product was
isolated by suction filtration, washed with some methanol and dried under reduced pressure at
50 °C.
24
ACS Paragon Plus Environment

Page 25 of 36
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Procedure C2. 3ꢀ(N,NꢀDiethylsulfamoyl)benzoyl chloride (1.50 mmol) was added to a stirred
suspension of anthranilic acid methyl ester (1.00 mmol), potassium carbonate (207 mg, 1.50
mmol) Nꢀmethylimidazole (8 mg, 0.10 mmol) and N,N,N’,N’ꢀtetramethylethylenediamine (12
mg, 0.10 mmol) in acetonitrile (1 mL). After stirring for 2.5 h at room temperature 1N HCl (aq.)
was added. The solid was isolated by suction filtration, washed thoroughly with water and heated
briefly with a small amount of methanol (2 mL). After cooling to room temperature the product
was isolated by suction filtration, washed with some methanol and dried under reduced pressure
at 50 °C.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Procedure C3. A solution of 3ꢀ(N,Nꢀdiethylsulfamoyl)benzoyl chloride (3.00 mmol) and
anthranilic acid methyl ester (3.00 mmol) in toluene (10 mL) was heated to reflux for 1.5 h. The
solvent was evaporated under reduced pressure and the residue was heated briefly with a small
amount of methanol. After cooling to room temperature the solid was isolated by suction
filtration, washed with some methanol and dried under reduced pressure at 50 °C.
General Procedure D for the Hydrolysis of 2ꢀBenzamidobenzoic Acid Methyl Esters. A mixture
of 2ꢀbenzamidobenzoic acid methyl ester (1.30 mmol), THF/MeOH 2/1 (9 mL) and 1N NaOH
(aq., 2.6 mL, 2.6 mmol) was stirred overnight at room temperature. 1N HCl (aq., 2.6 mL, 2.6
mmol) was added and the solid was isolated by suction filtration, washed with water and dried
under reduced pressure.
Procedure for the Synthesis of compound IV. 1,4ꢀNaphtoquinone (31.6 mmol) was dissolved in
DCM (45 mL). Than TFA (2.45 mL) and pꢀtoluensulfinate (33.18 mmol) dissolved in water (30
mL) were added and the mixture was stirred for 1 hour. Solid precipitates were obtained. The
solution was purified over a column and stirred for additional 5 hours. The solid was filtered off
and washed several times with water and DCM.32, 38
25
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 36
1
2
3
4
5
6
7
8
9
Screening assay procedure for in vitro PqsD inhibition. The assay was performed monitoring
the enzyme activity by measuring the HHQ formation as described by Storz and coworkers31.
Quantification of HHQ was performed analogous to Storz et al.,31 but with some modifications:
the flow rate amounted 750 ꢁL/min and an Accucore RPꢀMS column, 150 x 2.1 mm, 2.6 ꢁm
(Thermo Scientific, Waltham, Massachusetts, USA), was used. All test compound reactions were
performed in triplicate in three independent runs. In the modified protocol, PqsD and the first
substrate were preincubated for 30 min prior to adding them to the test compounds. Than ßꢀ
ketodecanoic acid was added and the further steps were performed analogous to the standard
protocol.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Surface Plasmon Resonance. SPR binding studies were performed using a Reichert SR7500DC
instrument optical biosensor (Reichert Technologies, Depew, NY, USA) and CMD500M sensor
chips obtained from XanTec (XanTec Bioanalytics, Düsseldorf, Germany). Scrubber software
was used for processing and analyzing data.
Preparation of PqsD mutants. Ser317Phe, Cys112Ala and His257Phe PqsD mutants were
generated using the QuikChange SiteꢀDirected Mutagenesis Kit (Stratagene, La Jolla, CA)
according to the manufacturer’s instructions using the pET28b(+)/pqsD plasmid as a template.
Briefly, pqsD gene was amplified through 16 cycles of PCR. After treatment with DpnI, the PCR
product was transformed into E. coli strain XL1ꢀBlue. Plasmid DNA was purified using the
GenEluteTM HP Plasmid Miniprep Kit (SigmaꢀAldrich, St. Louis, MO) and sequenced to confirm
the siteꢀdirected mutations. For the primer sequence of the mutations see Supporting Information.
Expression and purification of recombinant PqsD wildꢀtype and mutants in E. coli. E. coli
BL21 (λDE3) cells were transformed with plasmid harboring PqsD (PET28b(+)/pqsD).30
Overexpression, purification and storage of the His6ꢀtagged PqsD was performed as described by
26
ACS Paragon Plus Environment

Page 27 of 36
Journal of Medicinal Chemistry
1
2
Storz et al..31 The expression and purification of recombinant PqsD used for the NMRꢀstudies is
described in SI
3
4
5
6
7
8
9
Immobilization of His6ꢀPqsD. PqsD (38 kDA, >90% pure based on SDSꢀPAGE) was
immobilized on a CMD500M (carboxymethyldextranꢀcoated) sensor chip at 12 °C analogous to
the method described by Henn and coworkers.39 The PqsD immobilization levels were 2217 RU
for the binding studies with ACoA and 3085 RU in the mutagenesis studies to serve as reference.
The PqsD mutants were immobilized analogous to the wildꢀtype at densities of 6789 RU
(Ser317Phe), 3903 RU (Cys112Ala) and 4272 RU (His257Phe).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Binding studies. The binding studies were performed at a constant flow rate of 50 ꢁL/min in
instrument running buffer (50 mM MOPS, pH 8.0, 150 mM NaCl, 5% DMSO (v/v), 0.1%
TritonꢀX 100 (v/v)). IV (100 ꢁM), 43 and 48 (20 ꢁM) were injected consecutively for 180 s
association and 300 s dissociation times. Experiments were performed twice. In the second
experiment ACoA (100 ꢁM) was injected for 40 minutes with a constant flow of 5 ꢁL/min to
reach saturation of the ACoA binding site. Afterwards the flow rate was increased to 50 ꢁL/min
(30 min) in order to flush all CoA away. Once the baseline signal is stable additional ACoA (10
ꢁM) was injected to assure that the anthraniloyl binding site is completely saturated (no
additional signal was observed). Afterwards IV (100 ꢁM), 43 and 48 (20 ꢁM) were
consecutively injected.
In the experiment with inverted compound order PqsD was preincubated with compound 48,
therefore the compound was added to the running buffer (20 ꢁM). The sensor chip surface was
flushed for several hours at a constant flow rate of 50 ꢁL/min until the baseline was stable.
Afterwards the flow rate was decreased to 10 ꢁL/min and ACoA was injected (100 ꢁM) twice
for 120 s association and 15 min dissociation. The binding studies with PqsD mutants were
27
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 36
1
2
3
4
5
6
performed analogously to the wildꢀtype experiments. The compounds were injected twice for
120 s association and 300 s dissociation time.
7
8
9
Molecular Docking. Molecular docking studies were performed by means of GOLDv5.040, 41
using the CHEMPLP scoring function. 100 runs per molecule were performed. Default GOLD
parameters were used with carbons, halogens, and nonꢀpolar sulfur atoms matching hydrophobic
regions. All three 3Dꢀstructures of PqsD (PDBꢀid 3H76, 3H77, 3H78) were used. For 3H77
(anthranilateꢀCys112) and 3H78 (Ala112) the residue 112 of both chain A and B was mutated
back into wildꢀtype Cys112. Before docking geometry optimization of the enzymes has been
performed using the LigX module of Molecular Operating Environment (MOE).42 The active site
was determined including all residues within 8 Å of the coꢀcrystallized ACoA molecule of 3H77.
Saturation-Transfer Difference NMR. Experiments were recorded with the carrier set at ꢀ2
ppm for onꢀresonance irradiation and 40 ppm for offꢀresonance irradiation. Control spectra were
recorded under identical conditions on samples containing free compound 2 to test for artifacts.
Selective protein saturation (2 s) was accomplished using a train of 50 ms Gaussꢀshaped pulses,
each separated by a 1 ms delay, at an experimentally determined optimal power (60 dB on our
probe); a T1ρ filter (25 ms) was incorporated to suppress protein resonances. Experiments were
recorded using a minimum of 2048 scans and 32K points. Onꢀ and offꢀresonance spectra were
processed independently, and subtracted to provide a difference spectrum. The expression and
purification of recombinant PqsD from E. coli for STD NMR analysis is described in Supporting
Information.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Isothermal Titration Calorimetry. ITC experiments were carried out using an ITC200
instrument (Microcal Inc., GE Healthcare). ITC measurements were routinely performed at
25 °C in PBSꢀbuffer, pH 7.4, 10% glycerol (v/v), 5% DMSO (v/v). The titrations were
28
ACS Paragon Plus Environment

Page 29 of 36
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
performed on 83ꢀ102 ꢃM His6ꢀPqsD in the 200 ꢁL sample cell using 2 ꢁL 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 subtracted. The experimental data were fitted
to a theoretical titration curve (one site binding model) using MicroCal Origin 7 software, with
ꢂH (enthalpy change in kcal mol−1), KA (association constant in M−1), and N (number of binding
sites) as adjustable parameters. Thermodynamic parameters were calculated from equation;
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ꢂG = ꢂH − TꢂS = RT ln KA = −RT ln KD
where ꢂG, ꢂH, and ꢂS are the changes in free energy, enthalpy, and entropy of binding,
respectively, T is the absolute temperature, and R = 1.98 cal mol−1 K−1. For compound 48 four
independent experiments were performed.
ASSOCIATED CONTENT
Supporting Information
Additional figures docking study and STD NMR, ITC measurement, experimental details,
biological methods and compound characterizations. This material is available free of charge via
the Internet at http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
*Phone: +49 681 302 70300; fax: +49 681 302 70308; eꢀmail: rolf.hartmann@helmholtzꢀhzi.de.
29
ACS Paragon Plus Environment