N-Phenyl-4,5-dibromopyrrolamides and N-Phenylindolamides as ATP Competitive DNA Gyrase B Inhibitors: Design, Synthesis, and Evaluation

Article abstract of DOI:10.1021/acs.jmedchem.5b00775

Bacterial DNA gyrase is a well-known and validated target in the design of antibacterial drugs. However, inhibitors of its ATP binding subunit, DNA gyrase B (GyrB), have so far not reached clinical use. In the present study, three different series of N-ph

Full text of DOI:10.1021/acs.jmedchem.5b00775

Subscriber access provided by NATIONAL TAIWAN UNIV
Article
N-Phenyl-4,5-dibromopyrrolamides and N-Phenylindolamides as ATP
competitive DNA Gyrase B Inhibitors: Design, Synthesis, and Evaluation
Nace Zidar, Helena Macut, Tihomir Tomasic, Matjaz Brvar, Sofia Montalvão, Päivi
Tammela, Tom J Solmajer, Lucija Peretlin Masic, Janez Ilaš, and Danijel Kikelj
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.5b00775 • Publication Date (Web): 30 Jun 2015
Downloaded from http://pubs.acs.org on July 1, 2015
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 51
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
NꢀPhenylꢀ4,5ꢀdibromopyrrolamides and Nꢀ
Phenylindolamides as ATP Competitive DNA
Gyrase B Inhibitors: Design, Synthesis, and
Evaluation
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
†
†
†
‡
§
Nace Zidar, Helena Macut, Tihomir Tomašič, Matjaž Brvar, Sofia Montalvão, Päivi
§
‡
†
†
,†
Tammela, Tom Solmajer, Lucija Peterlin Mašič, Janez Ilaš and Danijel Kikelj*
†
‡
Faculty of Pharmacy, University of Ljubljana, Aškerčeva 7, 1000 Ljubljana, Slovenia;
National Institute of Chemistry, Laboratory for Biocomputing and Bioinformatics, 1001
§
Ljubljana, Slovenia; Centre for Drug Research, Division of Pharmaceutical Biosciences,
Faculty of Pharmacy, University of Helsinki, P.O. Box 56 (Viikinkaari 5 E), Helsinki FIꢀ00014,
Finland
KEYWORDS: antibacterial; inhibitor; ATP competitive; DNA gyrase B; GyrB; structureꢀ
based design; topoisomerase; 4,5ꢀdibromopyrrole.
ABSTRACT
Bacterial DNA gyrase is a wellꢀknown and validated target in the design of antibacterial
drugs. However, inhibitors of its ATP binding subunit, DNA gyrase B (GyrB), have so far not
reached clinical use. In the present study, three different series of Nꢀphenylꢀ4,5ꢀ
dibromopyrrolamides and Nꢀphenylindolamides were designed and prepared as potential
1
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 51
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
DNA gyrase B inhibitors. The IC values of compounds on DNA gyrase from Escherichia
50
coli were in the low micromolar range, with the best compound, (4ꢀ(4,5ꢀdibromoꢀ1Hꢀpyrroleꢀ
ꢀcarboxamido)benzoyl)glycine (18a), displaying an IC50 of 450 nM. For this compound a
2
highꢀresolution crystal structure in complex with E. coli DNA gyrase B was obtained,
revealing details of its binding mode within the active site. The binding affinities of three
compounds with GyrB were additionally evaluated by surface plasmon resonance and the
results were in good agreement with the determined enzymatic activities. For the most
promising compounds the inhibitory activities against DNA gyrase from Staphylococcus
aureus and topoisomerases IV from E. coli and S. aureus were determined. Antibacterial
activities of the most potent compounds of each series were evaluated against two Gramꢀ
positive and two Gramꢀnegative bacterial strains. The results obtained in this study provide
valuable information on the binding mode and structureꢀactivity relationship of Nꢀphenylꢀ4,5ꢀ
dibromopyrrolamides and Nꢀphenylindolamides as promising classes of ATP competitive
GyrB inhibitors.
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
By the midꢀ1960s, the major advances of the “antibiotic era” had led to the belief among
some scientists that infectious diseases are largely dealt with. These views were reflected in
1
priorities of research funding by the governments and the pharmaceutical industry. The overꢀ
optimism of those years was soon followed by the realization that, for every antibacterial drug
introduced into the clinic, resistance was quickly developed by bacteria. As a consequence,
infectious diseases remain one of the major global health concerns, so new targets and
appropriate inhibitory compounds are urgently needed in the fight against resistant
2
microorganisms.
One of the best established and validated targets in antibacterial drug discovery are the
bacterial topoisomerases, enzymes essential for control of the topological state of DNA during
2
ACS Paragon Plus Environment

Page 3 of 51
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
3
cell division and DNA replication. There are two major classes of topoisomerases: type I,
that catalyze single strand breaks and type II, that are responsible for the transient break of
4
two strands of DNA. DNA gyrase is a type IIA topoisomerase whose main function is to
introduce negative supercoils into the DNA molecule. It is a heterotetrameric protein
composed of two A subunits (DNA gyrase A or GyrA) and two B subunits (DNA gyrase B or
GyrB). The main role of subunit A is cleavage and reunion of the DNA molecule, while
subunit B binds ATP and, through its hydrolysis, provides energy for the cleavage/ligation
reactions. A further important bacterial type IIA topoisomerase is topoisomerase IV (topo IV).
This plays a critical role in the control of DNA supercoiling and is responsible for unlinking,
or decatenating, daughter chromosomal DNA following DNA replication. Topo IV has a
heterotetrameric structure similar to that of DNA gyrase, with two ParC subunits homologous
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
5
to GyrA, and two ParE subunits homologous to GyrB. DNA gyrase and topo IV exhibit 40%
6
3b, 7
overall sequence identity and possess significantly similar active sites.
Currently, the only
clinically relevant DNA gyrase/topo IV inhibitors are fluoroquinolones. These bind to the
GyrA/ParC subunit, stabilizing the covalent gyrase/topo IVꢀDNA complexes, thereby
preventing the reunion of the two DNA strands, stopping the replication cycle and, ultimately,
8
leading to cell death. Because of some serious side effects and the emergence of
fluoroquinoloneꢀresistant bacteria, the clinical use of fluoroquinolones is limited.
An attractive, but so far less successfully exploited target for the design of antibacterials is
the ATP binding site of DNA gyrase that is located on the GyrB subunit. It has been shown
that the likelihood of cross resistance between fluoroquinolones and ATP competitive
inhibitors is low, thus making GyrB an interesting target in the fight against fluoroquinolone
resistance. Among the earliest known ATPase inhibitors of DNA gyrase were aminocoumarin
natural products such as novobiocin (1) and clorobiocin (2) (Figure 1a). Novobiocin was
introduced into the clinic in the 1960s, but was soon withdrawn because of its acute toxicity
3
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 51
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
and resistance problems. The binding site of aminocoumarin antibiotics overlaps partially
with the ATP binding site and has been characterized from several high resolution crystal
9
structures (Figure 1b). The substituted deoxysugar moiety, novobiose, binds into the adenine
part of the ATP binding pocket and interacts with Asp73 (E. coli numbering), while 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
coumarin ring of aminocoumarins is directed away from the ATP binding pocket and makes
3c
contact with Arg136. Though the binding modes of novobiocin and clorobiocin are similar,
clorobiocin has a somewhat stronger affinity for GyrB, attributed to the additional interactions
formed by the 5ꢀmethylpyrroloꢀ2ꢀcarboxylate group of clorobiocin, than those formed by the
carbamate group of novobiocin. The methyl group at position 5 of the pyrrole ring forms
additional hydrophobic contacts with the residues Val43, Ala47, Val71 and Val 167 in the
3c
adenineꢀbinding pocket of the enzyme.
Figure 1. (a) Structures of novobiocin (1) and clorobiocin (2), natural aminocoumarin
inhibitors of DNA gyrase B; (b) Schematic representation of interactions between clorobiocin
9
a
(
2) and E. coli GyrB (PDB code: 1KZN). Hꢀbonds are indicated by dotted lines.
Hydrophobic interactions are shown as dashed curves.
4
ACS Paragon Plus Environment

Page 5 of 51
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
In recent years, there has been extensive research directed at discovering new inhibitors of
DNA gyrase B with improved enzyme affinities, more effective antibacterial activities and
favorable pharmacokinetic properties. The majority of recently discovered compounds are
lowꢀmolecular weight ligands, identified through highꢀthroughput screening (HTS) or
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
a, b
structureꢀbased virtual screening (VS) campaigns and optimized by structureꢀbased design.
Recently discovered GyrB inhibitors belong to a number of chemical classes, such as 4,5’ꢀ
1
0
11
12
13
bithiazoles (3),
indolinꢀ2ꢀones (4),
indazoles (5),
imidazo[1,2ꢀa]pyridines (6),
14
15
16
17
benzimidazole ureas (7), pyrazolthiazoles (8), azaindoles (9), pyrrolopyrimidines (10),
1
8
19
pyrrolamides (11), and tetrahydrobenzothiazoles (12) (Figure 2). Their binding modes in
the GyrB active sites have been well characterized from several Xꢀray crystal structures. A
common feature of the majority of GyrB inhibitors is a hydrogen bond donor – hydrogen
bond acceptor pattern that is essential for the formation of a hydrogen bond network with
Asp73 and a conserved water molecule in the adenineꢀbinding pocket. Although several
synthetic compounds possess excellent inhibitory activities against DNA gyrase B, none have
so far reached the clinic. Further research is needed in order to produce drug candidates with
stronger in vitro and in vivo antibacterial activities, less toxicity and good ADME properties.
The aim of the present study was to specifically address these problems with the focus on
improving the enzymatic activities through a detailed structureꢀbased design.
5
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 51
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
Figure 2. Representative synthetic inhibitors of DNA gyrase B from the chemical classes of
10
11
12
13
4
,5’ꢀbithiazoles (3), indolinꢀ2ꢀones (4), indazoles (5), imidazo[1,2ꢀa]pyridines (6),
1
4
15
16
17
benzimidazole ureas (7), pyrazolthiazoles (8), azaindoles (9), pyrrolopyrimidines (10),
18
19
pyrrolamides (11), and tetrahydrobenzothiazoles (12)
RESULTS AND DISCUSSION
Design. Three series of Nꢀphenylꢀ4,5ꢀdibromopyrrolamides and Nꢀphenylindolamides were
designed and prepared as potential DNA gyrase B inhibitors in the course of modification of
1
9
our 5,6,7,8ꢀtetrahydrobenzothiazole inhibitor series (12, Figure 2): (i) compounds with a 4ꢀ
aminobenzamide central part (A, Figure 3), (ii) compounds containing a 4ꢀaminophenol
central core (B, Figure 3), and (iii) compounds comprising a 4ꢀaminobenzene ring with
heterocyclic substituents or basic side chain functionalities (C, Figure 3). In all three series of
compounds, 4,5ꢀdibromopyrrolamide or indolamide groups were selected to constitute the
western part of the inhibitor molecules. The 4,5ꢀdibromopyrrolamide and indolamide moieties
6
ACS Paragon Plus Environment

Page 7 of 51
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
both comprise an essential hydrogen bond donorꢀacceptor motif, a common feature of the
majority of known GyrB inhibitors. The strategy of incorporating substituted pyrrolamides as
adenine bioisosteres has been employed successfully in the series of GyrB inhibitors (e.g. 11)
18
produced at AstraZeneca. 4,5ꢀDibromopyrroleꢀ2ꢀcarboxamideꢀcontaining compounds are
present in nature, for example in alkaloids oroidin and dibromosceptrin. They are secondary
metabolites produced by various marine sponges and exhibit a variety of biological
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
2
0
activities. The role of the 4,5ꢀdibromo functionality in binding these compounds is, on the
one hand to occupy the hydrophobic pocket of the enzyme and, on the other hand, to increase
the acidity of the pyrrole amino group and thus improve its interaction with the Asp73
residue. By replacing the 4,5ꢀdibromopyrrole group with a sterically more bulky indole group,
we aimed to explore the size of the hydrophobic pocket in the adenineꢀbinding site. Aniline
moiety was introduced as the central core of the molecules. It does not form additional
interactions with the enzyme, but is suitable for functionalization with various substituents to
produce compounds with the spatial arrangements of functional groups that interact with
specific amino acid residues in the active site. On the aniline paraꢀposition we introduced
groups that could make favorable contacts with Arg76 or Arg136. To enable ionic interactions
with Arg136 we attached substituents that comprise terminal carboxylic acid functionalities
and, in order to explore the optimal lengths of the inhibitors, in type A series, analogs with 2,
3 and 4 atoms linkers between the phenyl ring and the COOH group were prepared. To assess
the importance of the free carboxylic acid group for biological activity, derivatives with
methyl ester functionalities and derivatives in which the carboxylic group was transformed
into the primary carboxamide group (Scheme 1, compounds 19a-c) were prepared and
evaluated. To further elucidate the structureꢀactivity relationship (SAR), type B derivatives
with an additional carboxylic acid substituent on the phenyl ring (Scheme 2, compounds 24b,
24d, 25b and 25d), and type C derivatives with basic side chain functionalities, i.e.,
7
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 51
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
dimethylamino group (Scheme 3, compound 29a) and a morpholinꢀ4ꢀyl group (Scheme 3,
compound 29b), were prepared. Finally, to determine whether conformational freedom of the
side chain is important for biological activity, compound 35 (Scheme 4) with an (S)ꢀ4,5ꢀ
dihydrooxazoleꢀ4ꢀcarboxylic acid substituent on the paraꢀposition of the phenyl ring was
prepared.
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 3. Design of Nꢀphenylꢀ4,5ꢀdibromopyrrolamides and Nꢀphenylindolamides as ATP
competitive DNA gyrase B inhibitors.
Chemistry. The synthesis of 4,5ꢀdibromopyrroleꢀ and indoleꢀ2ꢀcarboxamidoꢀbenzamides
18a-d and 19a-c is outlined in Scheme 1. In the first step, 4ꢀnitrobenzoyl chloride (1) was
reacted with glycine or βꢀalanine methyl ester hydrochloride (14a or 14b) in the presence of
potassium carbonate, to afford 4ꢀnitrobenzamides 15a-b. After reduction of the nitro groups
of 15a-b, the obtained 4ꢀaminobenzamides 16a-b were coupled with 4,5ꢀdibromoꢀpyrroleꢀ2ꢀ
carboxylic acid or indoleꢀ2ꢀcarboxylic acid in a TBTUꢀpromoted reaction, to give compounds
17a-d. The target compounds 18a-d were obtained after hydrolysis of the methyl ester groups
with aqueous lithium hydroxide. In derivatives 17a-c the methyl ester functionalities were
3
further converted to primary carboxamide groups with gaseous NH , to give 19a-c.
8
ACS Paragon Plus Environment

Page 9 of 51
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Scheme 1. Synthesis of 4-(4,5-dibromo-pyrrole-2-carboxamido)benzamides 18a-b, 19a-b,
a
and 4-(indole-2-carboxamido)benzamides 18c-d, 19c
O
O
O
n
n
OCH3
OCH3
n
a
b
Cl
OCH3
N
N
-
+
+
Cl H3N
H
H
O
O
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
O2N
O
O N
H N
2
2
13
14a (n = 1)
15a (n = 1)
15b (n = 2)
16a (n = 1)
16b (n = 2)
1
4b (n = 2)
O
O
n
n
OCH3
R
O
N
H
O
N
H
c
H
N
e or f
H
1
6a-b
O
N
O
N
H
N
Br
Br
H
Br
Br
17a (n = 1)
7b (n = 2)
18a (n = 1, R = OH) 19a (n = 1, R = NH2)
18b (n = 2, R = OH) 19b (n = 2, R = NH2)
1
O
O
n
n
OCH3
R
O
N
H
O
N
d
e or f
H
H
N
H
O
N
O
N
H
N
H
17c (n = 1)
7d (n = 2)
18c (n = 1, R = OH) 19c (n = 1, R = NH2)
18d (n = 2, R = OH)
1
a
Reagents and conditions: (a) K
2
CO
3
, CH
3
CN, 0 °C → rt, 15 h; (b) H
Cl , 50 °C, 15 h; (d) indoleꢀ2ꢀ
, 50 °C, 15 h; (e) LiOH, THF/H O, 5 h, rt; (f) NH3(g)
2
, PdꢀC, MeOH, rt, 5 h;
(c) 4,5ꢀdibromoꢀpyrroleꢀ2ꢀcarboxylic acid, TBTU, NMM, CH
2
2
carboxylic acid, TBTU, NMM, CH
MeOH/THF, 0 °C → rt, 15 h.
2
Cl
2
2
,
Substituted 2ꢀ(4ꢀ(4,5ꢀdibromoꢀpyrroleꢀ2ꢀcarboxamido)phenoxy)acetic acids 25a-b and 2ꢀ
(4ꢀ(indoleꢀ2ꢀcarboxamido)phenoxy)acetic acids 25c-d were synthesized according to Scheme
2
. After alkylation of the phenolic hydroxyl groups of 4ꢀnitrophenol (20a) or methyl 5ꢀ
hydroxyꢀ2ꢀnitrobenzoate (20b) with methyl 2ꢀbromoacetate (21) and potassium carbonate as
reagents, the products 22a-b were submitted to catalytic hydrogenation to reduce the aromatic
nitro groups to amino groups. TBTUꢀpromoted coupling of 23a-b with 4,5ꢀdibromoꢀpyrroleꢀ
2
ꢀcarboxylic acid or indoleꢀ2ꢀcarboxylic acid, and subsequent hydrolysis of methyl esters
gave target compounds 25a-d.
9
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 51
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
Scheme 2. Synthesis of 2-(4-(4,5-dibromo-pyrrole-2-carboxamido)phenoxy)acetic acids
a
13a-b and 2-(4-(indole-2-carboxamido)phenoxy)acetic acids 25c-d
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
2 3 3 2
Reagents and conditions: (a) K CO , CH CN, rt, 15 h; (b) H , PdꢀC, MeOH, rt, 5 h; (c) 4,5ꢀ
dibromoꢀpyrroleꢀ2ꢀcarboxylic acid, TBTU, NMM, CH Cl , 50 °C, 15 h; (d) indoleꢀ2ꢀ
2
2
2 2 2
carboxylic acid, TBTU, NMM, CH Cl , 50 °C, 15 h; (e) LiOH, THF/H O, 15 h, rt.
The synthetic route to 4,5ꢀdibromoꢀNꢀ(4ꢀethoxyphenyl)ꢀ1Hꢀpyrroleꢀ2ꢀcarboxamides 29a-b
is presented in Scheme 3. 4ꢀNitrophenol (20a) was reacted with 2ꢀchloroꢀN,Nꢀ
dimethylethylamine hydrochloride (26a) or 2ꢀchloroethylmorpholine hydrochloride (26b) to
afford 27a-b. After reduction of the nitro groups of 27a-b, the products 28a-b were coupled
with 4,5ꢀdibromoꢀpyrroleꢀ2ꢀcarboxylic acid to give the target compounds 29a-b.
Scheme 3. Synthesis of 4,5-dibromo-N-(4-ethoxyphenyl)-1H-pyrrole-2-carboxamides
a
29a-b
1
0
ACS Paragon Plus Environment

Page 11 of 51
Journal of Medicinal Chemistry
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
a
Reagents and conditions: (a) K CO , CH CN, 50 °C, 15 h; (b) H , PdꢀC, EtOH, rt, 5 h; (c)
2
3
3
2
4
,5ꢀdibromoꢀpyrroleꢀ2ꢀcarboxylic acid, TBTU, NMM, CH
2
2
Cl , 50 °C, 15 h.
(S)ꢀ2ꢀ(4ꢀ(4,5ꢀDibromoꢀpyrroleꢀ2ꢀcarboxamido)phenyl)ꢀ4,5ꢀdihydrooxazoleꢀ4ꢀcarboxylic
acid (35) was prepared according to Scheme 4. 4ꢀNitrobenzonitrile (30) was first converted to
the imino ester hydrochloride 31 in a Pinner reaction with methanol and hydrochloric acid as
reagents. The cyclisation of an imino ester functionality to a 4,5ꢀdihydrooxazole ring
(derivative 32) was performed with Lꢀserine methyl ester hydrochloride in the presence of
triethylamine. The nitro group of 32 was then reduced to an amino group and the product 33
coupled with 4,5ꢀdibromoꢀpyrroleꢀ2ꢀcarboxylic acid to give derivative 34. Hydrolysis of the
methyl ester functionality led to the target compound 35.
Scheme 4. Synthesis of (S)-2-(4-(4,5-dibromo-pyrrole-2-carboxamido)phenyl)-4,5-
a
dihydrooxazole-4-carboxylic acid (35)
1
1
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 51
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
a
Reagents and conditions: (a) HCl(g), MeOAc/MeOH (10:1), 0 °C → 4 °C, 48 h; (b) Lꢀserine
methyl ester hydrochloride, Et N, CH Cl , rt, 48 h; (c) H , PdꢀC, EtOH, rt, 15 h; (d) 4,5ꢀ
3
2
2
2
dibromoꢀpyrroleꢀ2ꢀcarboxylic acid, TBTU, NMM, CH
MeOH/H O, 15 h, rt.
2 2
Cl , 50 °C, 15 h; (e) LiOH,
2
Inhibitory activities against DNA gyrase and topoisomerase IV. Twentyꢀone Nꢀphenylꢀ
,5ꢀdibromopyrrolamides and Nꢀphenylindolamides of all three structural classes (AꢀC,
4
Figure 3) were prepared and evaluated for inhibitory activity against E. coli DNA gyrase. The
results are presented in Tables 1ꢀ3 as IC50 values or, for less active inhibitors, as residual
activities (RAs) of the enzyme in the presence of 100 ꢁM tested compound. From the results
presented in Tables 1 and 2, differences in inhibitory potencies of Nꢀphenylꢀ4,5ꢀ
dibromopyrrolamide and Nꢀphenylindolamide compounds can be observed. The 4,5ꢀ
dibromopyrrolamide derivatives were always more potent than their indolamide counterparts,
indicating that the sterically bulkier indole moiety is unable to occupy optimally the
hydrophobic pocket in the adenine binding site, thus making the interactions of the indole NH
group with Asp73 less optimal and weaker. Compounds with free carboxylic acid
functionalities were generally more potent than their methyl ester precursors, with the only
exceptions of compound 18b (E. coli DNA gyrase IC50 = 1.4 ꢁM) and its methyl ester
derivative 17b (E. coli DNA gyrase IC50 = 0.85 ꢁM), where the methyl ester displayed an
1
2
ACS Paragon Plus Environment

Page 13 of 51
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
approximately 2ꢀfold greater activity. The optimal length of the linker between the phenyl
ring and the COOH group could not be established with certainty since, in some cases, the
compounds with a three atoms linker were more active than their counterparts with a four
atoms linker (comparison of compounds 18a with 0.45 ꢁM and 18b with 1.4 ꢁM IC50 values).
In other cases, longer molecules displayed higher enzymatic activity (comparison of
compounds 17a with 23 ꢁM and 17b with 0.85 ꢁM IC50 values). Compounds in which
carboxylic acids were transformed to primary carboxamides (19a-c) displayed slightly lower
inhibitory activities, as seen from the comparison of compounds 19a (IC50 = 2.2 ꢁM) and 18a
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
(IC50 = 0.45 ꢁM). On the other hand, even though the DNA gyrase activity of 19a was
approximately 5ꢀfold lower than that of compound 18a the less acidic CONH group proved
2
favorable for increasing the compounds’ antibacterial activity (Table 2S), probably due to its
greater ability to permeate the bacterial cell wall. Compounds 29a and 29b, with basic
dimethylamino and morpholinꢀ4ꢀyl side chains, did not possess any inhibitory activity.
Comparison of compounds 25a (IC50 = 1.4 ꢁM) and 25b (IC50 = 15 ꢁM) shows that
introduction of an additional carboxylic substituent to phenyl ring leads to a decrease in
inhibitory activity. Apparently, the 2ꢀcarboxylic substituent does not form favorable
interactions with the enzyme. Compound 35, with a (S)ꢀ4,5ꢀdihydrooxazoleꢀ4ꢀcarboxylic acid
substituent on the paraꢀposition of the phenyl ring and, thus, restrained conformational
freedom of the side chain, displayed a promising 5.0 ꢁM IC50 value on E. coli DNA gyrase.
For the most potent inhibitors of E. coli DNA gyrase, also the inhibitory activities against
E. coli topoisomerase IV, activities against S. aureus DNA gyrase and activities against S.
aureus topoisomerase IV, were determined (Tables 1ꢀ3). In general, inhibitory activities
against topo IV and DNA gyrase from S. aureus were weaker than those against E. coli DNA
gyrase. The most active inhibitor was compound 18a, with 58% residual activity on S. aureus
DNA gyrase and 66% residual activity on E. coli topo IV at 100 ꢁM concentrations. The
1
3
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 51
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
weaker activity on DNA gyrase from S. aureus can be explained by the fact that the adenineꢀ
binding pocket of S. aureus GyrB is smaller than that of the E. coli enzyme. In E. coli, Ile51
3b
and Ile175 are replaced by the smaller valines and Leu103 by methionine. Similarly, the
different inhibitory activities of compounds on DNA gyrase and topo IV from the same
bacterial species could be explained by a small conformational or dynamic difference in the
GyrB and ParE active sites. Although the active sites of GyrB and ParE are very similar, it has
been postulated that the replacement of Ile78 in GyrB by Met74 in ParE causes a partial
narrowing of the binding pocket and a change in the conformation of the active site, which
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
4a, 21
can result in the altered interaction with inhibitors.
Surface Plasmon Resonance (SPR) Experiments. Compounds 17b, 18a and 35, that
showed promising activities in the enzymatic assays, were subjected to surface plasmon
22
resonance (SPR) measurement. K values were calculated from the concentrationꢀresponse
d
curves, using Origin software with a steady state affinity binding model (one site binding).
The 24 amino acid long part of the GyrB subunit, G24 protein, which comprises the ATP
d
binding site, was used, with novobiocin as a standard. The measured K of 28 nM for
2
3
novobiocin was in good agreement with 19 nM reported in the literature. All three inhibitors
were tested in at least eight concentrations in three parallels, depending on the response of the
immobilized protein. The sensorgrams of compounds 17b, 18a and 35 at different
concentrations are presented in Supporting information Figure 3S. The K
compounds 17b, 18a and 35 of 0.97 ꢁM (IC50 = 0.85 ꢁM), 0.47 ꢁM (IC50 = 0.45 ꢁM) and 1.7
M (IC50 = 5.0 ꢁM) are in good agreement with the determined IC50 values. These results
d
values for
ꢁ
further validate the Nꢀphenylꢀ4,5ꢀdibromopyrrolamides as an interesting class of DNA gyrase
inhibitors and, since only a 24 amino acids long part of the ATP binding GyrB subunit was
used for the experiments, they confirm that the tested compounds bind to the enzyme’s ATP
binding site.
1
4
ACS Paragon Plus Environment

Page 15 of 51
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Table 1. Inhibitory activities of 4,5ꢀdibromoꢀNꢀphenylꢀ1Hꢀpyrroleꢀ2ꢀcarboxamides and Nꢀ
phenylꢀ1Hꢀindoleꢀ2ꢀcarboxamides containing a 4ꢀaminobenzamide central part against DNA
gyrase and topoisomerase IV.
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
a
b
IC50 (ꢁM) or RA (%)
1
2
Compd.
R
R
n
E. coli
gyrase
S. aureus
E. coli
topo IV
S. aureus
topo IV
gyrase
4
,5ꢀDibromoꢀ
pyrrolꢀ2ꢀyl
,5ꢀDibromoꢀ
17a
17b
17c
OCH
OCH
OCH
OCH
OH
3
3
3
3
1
2
1
2
1
2
1
2
1
2
1
/
23 ± 6 ꢁM
n.d.
138%
n.d.
n.d.
n.d.
107%
n.d.
4
0.85 ± 0.10 ꢁM
K = 0.97 ± 0.12 ꢁM
c
100%
n.d.
pyrrolꢀ2ꢀyl
d
Indolꢀ2ꢀyl
90%
17d
Indolꢀ2ꢀyl
91%
n.d.
n.d.
n.d.
4
,5ꢀDibromoꢀ
0.45 ± 0.08 ꢁM
= 0.47 ± 0.15 ꢁM
1
8a
8b
c
58%
78%
n.d.
66%
61%
n.d.
100%
99%
pyrrolꢀ2ꢀyl
4,5ꢀDibromoꢀ
pyrrolꢀ2ꢀyl
K
d
1
OH
1.4 ± 0.4 ꢁM
18c
Indolꢀ2ꢀyl
Indolꢀ2ꢀyl
OH
20 ± 6 ꢁM
23 ± 5 ꢁM
2.2 ± 0.3 ꢁM
43%
n.d.
1
8d
9a
OH
n.d.
n.d.
n.d.
4,5ꢀDibromoꢀ
pyrrolꢀ2ꢀyl
,5ꢀDibromoꢀ
1
NH
NH
NH
/
2
2
2
90%
n.d.
102%
n.d.
99%
4
19b
n.d.
pyrrolꢀ2ꢀyl
19c
Indolꢀ2ꢀyl
109%
n.d.
n.d.
n.d.
d
1
/
0.17 ± 0.02 ꢁM
0.041 ± 0.07 ꢁM
11 ± 2 ꢁM
27 ± 7 ꢁM
a
Concentration of compound (mean ± SD) that inhibits the enzyme activity by 50%.
Residual activity of the enzyme at 100 µM of the compound.
Dissociation constant (mean ± SD) determined from surface plasmon resonance experiments.
Novobiocin.
b
c
d
n.d. = not determined
Table 2. Inhibitory activities of 4,5ꢀdibromoꢀNꢀphenylꢀ1Hꢀpyrroleꢀ2ꢀcarboxamides and Nꢀ
phenylꢀ1Hꢀindoleꢀ2ꢀcarboxamides containing a 4ꢀaminophenol central part against DNA
gyrase and topoisomerase IV.
1
5
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 51
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
IC50 (ꢁM) or RA (%)
1
2
3
Compd
R
R
R
E. coli
gyrase
S. aureus
E. coli
topo IV
S. aureus
topo IV
gyrase
4,5ꢀDibromoꢀ
pyrrolꢀ2ꢀyl
24a
H
CH
CH
H
3
3
88 ± 19 ꢁM
71%
n.d.
n.d.
n.d.
n.d.
n.d.
24c
Indolꢀ2ꢀyl
H
n.d.
4
,5ꢀDibromoꢀ
2
5a
5b
H
1.4 ± 0.2 ꢁM
15 ± 3 ꢁM
47 ± 5 ꢁM
78%
100%
88%
100%
n.d.
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
pyrrolꢀ2ꢀyl
4,5ꢀDibromoꢀ
pyrrolꢀ2ꢀyl
2
COOH
H
n.d.
n.d.
n.d.
25c
Indolꢀ2ꢀyl
Indolꢀ2ꢀyl
/
H
H
n.d.
n.d.
25d
COOH
/
H
n.d.
n.d.
n.d.
c
1
/
0.17 ± 0.02 ꢁM
0.041 ± 0.07 ꢁM
11 ± 2 ꢁM
27 ± 7 ꢁM
a
Concentration of compound (mean ± SD) that inhibits the enzyme activity by 50%.
Residual activity of the enzyme at 100 µM of the compound.
Novobiocin.
b
c
n.d. = not determined
Table 3. Inhibitory activities of 4,5ꢀdibromoꢀNꢀphenylꢀ1Hꢀpyrroleꢀ2ꢀcarboxamides and Nꢀ
phenylꢀ1Hꢀindoleꢀ2ꢀcarboxamides containing a central aniline moiety with heterocyclic
substituents or basic side chain functionalities against DNA gyrase and topoisomerase IV.
a
b
IC50 (ꢁM) or RA (%)
Compd
R
E. coli
gyrase
S. aureus
E. coli
topo IV
S. aureus
gyrase
topo IV
29a
94%
88%
n.d.
n.d.
n.d.
n.d.
n.d.
2
3
3
9b
n.d.
n.d.
4
50%
n.d.
n.d.
5.0 ± 1.1 ꢁM
= 1.7 ± 0.3 ꢁM
5
c
83%
105%
98%
K
d
d
1
/
0.17 ± 0.02 ꢁM
0.041 ± 0.07 ꢁM
11 ± 2 ꢁM
27 ± 7 ꢁM
a
Concentration of compound (mean ± SD) that inhibits the enzyme activity by 50%.
Residual activity of the enzyme at 100 µM of the compound.
Dissociation constant (mean ± SD) determined from surface plasmon resonance experiments.
Novobiocin.
b
c
d
n.d. = not determined
1
6
ACS Paragon Plus Environment

Page 17 of 51
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Antibacterial Activity. The antibacterial activities of the most potent compounds of each
series were evaluated against two Gramꢀpositive (S. aureus ATCC 25923, E. faecalis ATCC
29212) and two Gramꢀnegative (E. coli ATCC 25922, P. aeruginosa ATCC 27853) bacterial
strains (Supporting information, Table 2S). Except for compound 19a, that showed 23%
inhibition of growth of Gramꢀnegative P. aeruginosa at 50 µM concentration after 24 h
incubation, none of the tested compounds displayed significant antibacterial activity. The
higher antibacterial activity of compound 19a can be attributed to the more favorable physicoꢀ
chemical properties for bacterial cell entry that are due to the presence of the less acidic
primary carboxamido functionality than that of the carboxylic acid group. These results are in
accordance with the recently reported linear correlation between activity (pMIC) of
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
24
compounds on S. pneumoniae and their pKa.
X-Ray Crystallography. The structure of the 43 kDa Nꢀterminal fragment of E. coli DNA
gyrase B (GyrB43) in complex with ligand 18a solved at a resolution of 2.20 Å (PDB code:
4ZVI) revealed the detailed binding mode of the ligand (Figure 4). The structure of E. coli
GyrB adopts a bilobal architecture, with the ligand bound to the ATP binding site of the Nꢀ
terminal lobe. The crystals contain one monomer of GyrB in the asymmetric unit and the
model comprises residues Leu16 to Thr392, except for the ones indicated in Supporting
information Table 1S. The amino acid residues forming the ligand binding site and the ligand
are well defined in the electron density map and show clearly the binding mode and the
orientation and conformation of the ligand bound to its binding site. Based on a distance of
less than 3.5 Å between the donor and acceptor atoms, two specific hydrogen bonds of the
ligand 18a to the side chain atoms of Arg76 and Asp73 were identified. By the same criteria,
the presence of additional hydrophilic interactions between the main chain atoms of Thr165
and ligand 18a, as well as the interactions between side chain atoms of Asp73, Thr165, and
Asn46 and ligand 18a bridged by water molecules, are suggested. The water molecules and
1
7
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 51
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
all atoms involved in the formation of hydrogen bonds are well defined by the electron
density. The following residues are present in the vicinity of the ligand at distances of 3.9 Å
or less: Asn46, Glu50, Asp73, Arg76, Gly77, Ile78, Pro79, Val120, Thr165, and Val167. The
weak electron density in this area indicates that the conformation of the carboxylate group of
the ligand is flexible.
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 4. Xꢀray crystal structure of E. coli DNA gyrase B in complex with ligand 18a (PDB
code: 4ZVI) shown in two orientations rotated by 90 degrees. The ligand and neighboring
protein side chains are shown as stick models, colored according to the chemical atom type
(C18a in cyan, CGyrB in grey, N in blue, O in red, and Br in brown). The conserved water
molecule is presented as a red ball. The ligand molecule is shown superimposed with the
refined 2FoꢀFc electron density map contoured at 1.0 σ. Hydrogen bonds are indicated as
2
5
green dotted lines. The Figure was prepared by PyMOL.
Molecular modeling. In the structure of E. coli DNA gyrase B in complex with inhibitor
18a the terminal carboxylate group of 18a is flexible and is not in contact with Arg136 side
2
6,27
chain, as designed and predicted by molecular docking (Figure 5a) using FRED software.
Since only weak electron density is present for the carboxylate group of 18a and because its
conformation is contrary to our expectations, we conducted a molecular dynamics simulation
to study the flexibility of the inhibitor in the binding site and to assess whether the
carboxylate group of 18a is able to interact with the Arg136 side chain during the simulation.
1
8
ACS Paragon Plus Environment

Page 19 of 51
Journal of Medicinal Chemistry
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
a
b
Figure 5. a) The binding mode of inhibitor 18a (in cyan sticks) in the ATP binding site of E.
coli DNA gyrase B (PDB code: 4DUH, in grey) predicted by docking, and b) the
conformation of the GyrB43ꢀ18a complex at the end of MD simulation (PDB code: 4ZVI).
For clarity, only side chains forming hydrogen bonds with inhibitor are presented. The
conserved water molecule is presented as red ball. Hydrogen bonds are indicated as black
2
5
dotted lines. The Figure was prepared by PyMOL.
First, in order to conduct the molecular dynamics study, the completeness of the Xꢀray
structure was improved by homology modeling of the missing loop (residues 99ꢀ117) and
residues with incomplete side chains, using E. coli topoisomerase IV ParE subunit structure
1
7a
28
(PDB entry: 4HZ0) as a template and MODELLER 9.13 as homology modeling software.
2
9
Amino acid sequences were aligned using TꢀCOFFEE server and the sequence alignment
used as obtained. Ten homology models were generated; the one used for further studies was
30
selected based on DOPE score and stereochemical quality as evaluated by Procheck and
3
1
ProSAꢀweb server.
The GyrB43ꢀ18a complex was first solvated using the TIP3P water model and then
subjected to a 12 ns molecular dynamics simulation consisting of a 1 ns equilibration run with
protein and inhibitor constrained, followed by a 1 ns equilibration run with protein and
inhibitor released; the final 10 ns production was then run. The volume, temperature and total
energy parameters of the system showed that these parameters were stable during the
1
9
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 51
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
production run, confirming the physical stability of the MD simulation (Figure 1S). The
RMSD of the protein backbone atoms, that provides information about the structural stability
of the system during the MD simulation, increased gradually during equilibration of the
system and then remained fairly stable at between 1.5 Å and 2.5 Å (Figure 6a). In contrast,
inhibitor 18a was more flexible than the protein backbone, with greater fluctuation of the
RMSD values, between 0.5 Å and 3.2 Å (Figure 6a). Visual inspection of MD trajectory
shows that the pyrrolamide moiety of the inhibitor is, as expected, the least flexible part of the
molecule, since it is buried deep in the hydrophobic pocket of the ATP binding site. Analysis
of the hydrogen bond formation between the pyrrole NH and carboxylate groups of Asp73
shows that the distance between these two groups remains approximately 2 Å throughout the
simulation, indicating the stability of this hydrogen bond (Figure 6b). In contrast, the
hydrogen bond formation between Arg136 side chain and the carboxylate group of 18a shows
a distribution of distances between 2.8 Å and 15.5 Å (Figure 6b), when monitored by
measuring the distances between Arg136NH1 and carboxylate O3 of 18a, Arg136NH1 and
carboxylate O4 of 18a, Arg136NH2 and carboxylate O3 of 18a, and Arg136NH2 and
carboxylate O4 of 18a (Figure 2S). This confirms the flexibility of the carboxylate group of
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
18a suggested by Xꢀray diffraction, and shows that it is in contact with the surrounding
solvent most of the simulation time. However, if 3.5 Å is taken as the distance threshold for
hydrogen bond formation, the carboxylate group of 18a is in contact with Arg136 side chain
for 34% of the simulation time (i.e. between 6.5 ns and 9 ns) (Figure 6b). The conformation of
the GyrB43ꢀ18a complex at the end of MD simulation, in which the carboxylate group of 18a
and the guanidine group of Arg136 interact, is shown in Figure 5b. It closely resembles that
obtained by docking (Figure 5a), since hydrogen bonds are formed with Asp73, Gly77 and
Arg136.
2
0
ACS Paragon Plus Environment

Page 21 of 51
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Thus, MD simulation of the GyrB43ꢀ18a complex confirms the flexibility of the terminal
carboxylate group of 18a, but also suggests the possibility of its interaction with Arg136,
which is not evident from the Xꢀray structure of the complex. Since the carboxylate group of
18a is, for most of the simulation time, in contact only with the solvent and not with Arg136
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
side chain, a less flexible moiety with a better directed hydrogen bond could improve the
binding affinity of this type of inhibitors.
a)
b)
Figure 6. a) Backbone RMSD (in blue) and ligand 18a RMSD (in red) vs time plots during
the MD simulation of E. coli DNA gyrase B in complex with 18a. b) Distance vs time plots
between selected GyrB residues and inhibitor 18a carboxylate and pyrrole NH groups. For
clarity, only 3 distance measurements are presented.
CONCLUSION
Twentyꢀone new Nꢀphenylꢀ4,5ꢀdibromopyrrolamides and Nꢀphenylindolamides were
designed and prepared as ATP competitive DNA gyrase B inhibitors and their inhibitory
activities on DNA gyrase from E. coli were measured. Five compounds displayed IC50 values
d
lower than 3 µM and the most active compound, 18a, had an IC50 of 450 nM. The K values
obtained from SPR for the binding of compounds 17b, 18a and 35 to the G24 protein that
contains an ATPase domain, were in good agreement with the determined IC50 values, which
confirmed that the tested compounds bind to the enzyme’s ATP binding site. The Xꢀray
crystal structure of compound 18a in complex with GyrB was resolved, revealing details of its
2
1
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 51
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 mode within the active site and confirming the rational of our design strategy. Even
though the crystal structure showed that the terminal carboxylate of 18a is flexible and is not
in contact with Arg136 side chain, the MD simulation of GyrB43ꢀ18a complex indicates the
possibility of this interaction. From the MD data it can be concluded that a less flexible
moiety with a better directed hydrogen bond could improve the binding affinity of inhibitors.
The inhibitory activities against DNA gyrase from S. aureus, topoisomerase IV from E. coli
and topoisomerase IV from S. aureus were generally weaker than those against E. coli DNA
gyrase, probably due to differences of the enzymes’ active sites. To obtain compounds with
stronger in vitro and in vivo antibacterial effectiveness, optimization of their physicoꢀ
chemical properties, e.g. by increasing the pKa value of the functionalities that interact with
Arg76 or Arg136, is suggested. Overall, the structureꢀactivity relationships derived from E.
coli DNA gyrase inhibitory activities and the information gained from the coꢀcrystal structure
of compound 18a in the active site of GyrB provide a good foundation for the further search
for new improved DNA gyrase inhibitors.
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
Determination of Inhibitory Activities on E. coli and S. aureus DNA Gyrase. The assay
for the determination of IC50 values (Inspiralis) was performed on the black streptavidinꢀ
coated 96ꢀwell microtiter plates (Thermo Scientific Pierce). The plates were first rehydrated
with the supplied wash buffer [20 mM TrisꢀHCl (pH 7.6), 137 mM NaCl, 0.01% (w/v) BSA,
0.05% (v/v) Tween 20] and biotinylated oligonucelotide was immobilized onto the wells. The
excess of oligonucleotide was then washed off, and the enzyme assay was carried out in the
wells. The final reaction volume of 30 ꢁL in buffer [35 mM Tris × HCl (pH 7.5), 24 mM KCl,
4 mM MgCl
2
, 2 mM DTT, 1.8 mM spermidine, 1 mM ATP, 6.5 % (w/v) glycerol, 0.1 mg/mL
albumin] contained 1.5 U of gyrase from E. coli or S. aureus, 0.75 ꢁg of relaxed pNO1
2
2
ACS Paragon Plus Environment

Page 23 of 51
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
plasmid, and 3 ꢁL of inhibitors solution in 10% DMSO and 0.008% Tween 20. Reactions
were incubated for 30 min at 37 °C and after the addition of the TF buffer [50 mM NaOAc
2
(pH 5.0), 50 mM NaCl and 50 mM MgCl ] which terminated the enzymatic reaction for
another 30 min at rt to allow the triplex formation (biotinꢀoligonucleotideꢀplasmid).
Afterwards, the unbound plasmid was washed off using TF buffer and the solution of
SybrGOLD stain in T10 buffer [10 mM Tris × HCl (pH 8.0) and 1 mM EDTA] was added.
After mixing the fluorescence (excitation: 485 nm, emission: 535 nm) was read using a
BioTek's Synergy H4 microplate reader. Preliminary screening was performed at inhibitor
concentrations of 100 and 10 ꢁM. For most potent compounds IC50 was determined with 7
concentrations of the inhibitors. IC50 values were calculated using GraphPad Prism software
and represent the concentration of inhibitor where the residual activity of the enzyme is 50%.
All compounds were assayed in three independent measurements and a final result is given as
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
2
their average value. Novobiocin [IC50 = 0,17 µM (lit. 0,08 µM) for E. coli gyrase and IC50
=
32
0
,041 µM (lit. 0,01 µM) for S. aureus gyrase] was used as the internal standard.
Determination of Inhibitory Activities on E. coli and S. aureus Topoisomerase IV. The
assay for the determination of IC50 values (Inspiralis) was performed on the black
streptavidinꢀcoated 96ꢀwell microtiter plates (Thermo Scientific Pierce). The plates were first
rehydrated with the supplied wash buffer [20 mM TrisꢀHCl (pH 7.6), 137 mM NaCl, 0.01%
(w/v) BSA, 0.05% (v/v) Tween 20] and biotinylated oligonucelotide was immobilized onto
the wells. The excess of oligonucleotide was then washed off, and the enzyme assay was
carried out in the wells. The final reaction volume of 30 ꢁL in buffer [40 mM HEPES KOH
(pH 7.6), 100 mM potassium glutamate, 10 mM magnesium acetate, 10 mM DTT, 1 mM
ATP, 0.05 mg/mL albumin] contained 1.5 U of topoisomerase IV from E. coli or S. aureus,
0.75 ꢁg of supercoiled pNO1 plasmid, and 3 ꢁL of inhibitors solution in 10% DMSO and
0.008% Tween 20. Reactions were incubated for 30 min at 37 °C and after the addition of the
2
3
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 51
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
TF buffer [50 mM NaOAc (pH 5.0), 50 mM NaCl and 50 mM MgCl ] which terminated the
2
enzymatic reaction for another 30 min at rt to allow the triplex formation (biotinꢀ
oligonucleotideꢀplasmid). Afterwards, the unbound plasmid was washed off using TF buffer
and the solution of SybrGOLD stain in T10 buffer [10 mM TrisꢀHCl (pH 8.0) and 1 mM
EDTA] was added. After mixing the fluorescence (excitation: 485 nm, emission: 535 nm) was
read using a BioTek's Synergy H4 microplate reader. Preliminary screening was performed at
inhibitor concentrations of 100 and 10 ꢁM. For most potent compounds IC was determined
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
with 7 concentrations of the inhibitors. IC50 values were calculated using GraphPad Prism
software and represent the concentration of inhibitor where the residual activity of the enzyme
is 50%. The assays for all compounds were run in triplicate and a final result is given as their
3
2
average value. Novobiocin [IC = 11 µM (lit. 10 µM) for E. coli topoisomerase IV and IC
50
50
32
=
27 µM (lit. 20 µM) for S. aureus topoisomerase IV] was used as the internal standard.
Surface Plasmon Resonance (SPR) Measurements. Surface plasmon resonance (SPR)
measurements were performed on a Biacore T100 machine using CM5 sensor chip (Biacore,
GE Healthcare). The system was primed twice with running buffer (10 mM Hepes, 150 mM
NaCl, 3 mM EDTA, 0.005% surfactant P20, pH 7.4). The G24 protein was immobilized on
the second flow cell of a sensor chip CM5 using standard amino coupling method. The
carboxymethylated dextran layer was activated with 7 min pulse of 1ꢀethylꢀ3ꢀ(3ꢀ
dimethylaminopropyl)ꢀcarbodiimide (EDC) and Nꢀhydroxysuccinimide (NHS) mixed in a 1:1
ratio. Protein was diluted to the final concentration of 50
µg/mL in 10 mM sodium acetate
(pH 4.5) and injected in two short pulses to reach the final immobilization level around
18.000 response units. Finally, the rest of the surface was deactivated with 7 min injection of
ethanolamine. The first flow cell was activated with EDC/NHS and deactivated with
ethanolamine and served as a reference cell for subtraction of nonspecific binding. The
d
activity of the chip was tested and confirmed using novobiocin as a standard. Measured K for
2
4
ACS Paragon Plus Environment

Page 25 of 51
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
2
3
novobiocin was 28 nM (lit. 19 nM). Analytes were prepared as DMSO 100× stock solutions
and were diluted with a running buffer prior to the injection. They were injected at a flow rate
of 30 ꢁL/min for 90 s, and dissociation was monitored for additional 120 s. Since the
dissociation of analytes from the ligand was rapid, no regeneration protocol was needed. For
the titration of analytes, the 1% of the DMSO was added to the running buffer in order to
diminish the difference in a refractive index between the samples and running buffer. All
three inhibitors (17b, 18a and 35) were tested in at least eight different concentrations in three
parallel titrations. The sensorgrams (Supporting Information, Figure 3S) were analyzed using
BiaEval software (Biacore, GE Healthcare). The equilibrium binding responses were
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
d
determined from the binding levels 5 s before the stop of the injection. K values were
determined by the fitting of the data to 1:1 steady state binding model as described in Results.
Determination of Antibacterial Activity. Clinical control strains of Enterococcus faecalis
(
Gramꢀpositive, ATCC 29212), Staphylococcus aureus (Gramꢀpositive, ATCC 25923),
Escherichia coli (Gramꢀnegative, ATCC 25922) and Pseudomonas aeruginosa (Gramꢀ
negative, ATCC 27853), were obtained from Microbiologics Inc. (St. Cloud, Minnesota,
USA). Bacterial cultures were initiated on MH agar (Becton Dickinson, Franklin Lakes, NJ,
USA) slants and prior to the assays, suspensions were prepared into MH II broth (Becton
Dickinson, Franklin Lakes, NJ, USA) and incubated at 37 °C for 16–20 h at 100 rpm.
Antimicrobial assays were performed by the broth microdilution method in 96ꢀwell plate
33
format according to the CLSI guidelines. Briefly, bacterial suspension was diluted with
5
MHB to obtain a final inoculum of 5 × 10 CFU/mL in the assay. An equal volume of
bacterial suspension and test compound solution diluted into assay media were mixed together
in the plate and incubated for 24 h at 37 °C. Absorbance values measured at 620 nm were
used for evaluating the antimicrobial effects of test compounds by comparing to untreated
controls and expressed as percentage inhibition of growth. Reference antibiotics were used as
2
5
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 51
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
positive controls on every assay plate (see Table 2S for details). Compounds were assayed at
final concentration of 50 µM (n = 3).
Crystallization, Preparation of the Inhibitor Complexes, and Data Collection was
performed by Proteros Biostructures GmbH, Martinsread, Germany.
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
Protein Production. Suitable constructs for GyrB expression had been previously
established and expression of GyrB was performed according to previously established
protocols. A purification protocol was established and homogeneous protein was produced in
preparative amounts. The protein was purified comprising affinity and gel filtration
chromatography steps. This procedure yielded homogenous protein with a purity greater 95%
as judged from Coomassie stained SDSꢀPAGE.
Crystallization. The purified protein was used in crystallization trials employing both, a
standard screen with approximately 1800 different conditions, as well as crystallization
conditions identified using literature data. Conditions initially obtained have been optimized
using standard strategies, systematically varying parameters critically influencing
crystallization, such as temperature, protein concentration, drop ratio, and others. These
conditions were also refined by systematically varying pH or precipitant concentrations.
Data Collection and Processing. A cryoꢀprotocol was established using PROTEROS
Standard Protocols. Crystals have been flashꢀfrozen and measured at a temperature of 100 K.
The Xꢀray diffraction data have been collected from complex crystals of GyrB with the ligand
18a at the SWISS LIGHT SOURCE (SLS, Villigen, Switzerland) using cryogenic conditions.
The crystals belong to space group C 2. Data were processed using the programs XDS and
XSCALE. The data collection and refinement statistics are summarized in Tables 4 and 5.
Table 4. Data collection and processing statistics for 18a.
Ligand
18a
a
Xꢀray source
Wavelength [Å]
Detector
PXI/X06SA (SLS )
1.00000
PILATUS 6M
100
Temperature [K]
2
6
ACS Paragon Plus Environment

Page 27 of 51
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Space group
Cell: a; b; c; [Å]
α; β; γ; [◦]
Resolution [Å]
Unique reflections
Multiplicity
C 2
117.88; 50.55; 71.34
90.0; 93.5; 90.0
2.20 (2.45ꢀ2.20)
20677 (5706)
2.6 (2.6)
95.9 (96.8)
4.5 (43.5)
b
Completeness [%]
R
R
sym [%]
meas [%]
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
5.5 (53.3)
13.86 (2.23)
c
Mean(I)/sd
a
SWISS LIGHT SOURCE (SLS, Villigen, Switzerland).
Values in parenthesis refer to the highest resolution bin.
Calculated from independent reflections.
b
c
a
Table 5. Refinement statistics for 18a.
Ligand
Resolution [Å]
Number of reflections (working/test) 19273/1403
18a
71.21ꢀ2.20
R
R
cryst [%]
free [%]
22.4
28.2
b
Total number of atoms:
Protein
Water
Ligand
Iodide
Deviation from ideal geometry:
Bond lengths [Å]
Bond angles [°]
Bonded B’s [Å ]
Ramachandran plot:
2784
76
23
2
c
0.007
1.10
2.9
2 d
e
Most favored regions [%]
91.0
8.3
0.0
Additional allowed regions [%]
Generously allowed regions [%]
Disallowed regions [%]
Values as defined in REFMAC5, without sigma cutꢀoff.
Testꢀset contains 6.8% of measured reflections.
Root mean square deviations from geometric target
0.6
a
b
c
values.
d
Calculated with MOLEMAN.
Calculated with PROCHECK.
e
Molecular Modeling. Computer Hardware. All of the computations were performed on a
workstation with four eightꢀcore AMD Opteron 6128 MagnyꢀCours 2.0 GHz processors, and
32 GB RAM, two 1 TB hard drives, and a 256 GB solidꢀstate drive, running 64ꢀbit Scientific
Linux, release 6.2.
Molecular Docking. The threeꢀdimensional model of compound 18a was built in
34
35
ChemBio3D Ultra 13.0. The geometries of the molecules were optimized using MMFF94
force field and partial atomic charges. The energy was minimized until the gradient value was
2
7
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 51
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
smaller than 0.001 kcal/(mol Å). The optimized structure was further refined with GAMESS
interface in ChemBio3D Ultra 13.0 using semiempirical PM3 method, QA optimization
34
algorithm and Gasteiger Hückel charges for all atoms for 100 steps. A library of 164
conformers of 18a was then generated by OMEGA 2.4.6. (OpenEye Scientific Software, Inc.,
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
36,37
Santa Fe, NM, USA; www.eyesopen.com).
MAKE RECEPTOR 3.0.1 software (OpenEye
Scientific Software, Inc., Santa Fe, NM, USA; www.eyesopen.com) was used for the protein
3
preparation. A box of 4408 Å was generated around the inhibitor from the crystal structure of
10
E. coli GyrB (PDB entry: 4DUH ). After molecular cavity detection algorithm for site
3
detection was used, a highꢀquality site shape potential of the final docking volume of 886 Å
was calculated. Of note, protonation state of the binding site residues and their conformations
were not modified, and no constraints were used. The OMEGAꢀgenerated library of 18a
conformers was docked into the prepared enzyme binding site using FRED 3.0.1 software
(OpenEye Scientific Software, Inc., Santa Fe, NM, USA; www.eyesopen.com) and the
2
6,27
Chemgauss4 scoring function.
The best scored docking conformation was visualized by
2
5
PyMOL.
Homology Modeling. Completeness of the GyrB43ꢀ18a complex was improved by creating
a homology model based on homologous E. coli topoisomerase IV ParE subunit (PDB entry:
1
7a
29
4
HZ0 ). Sequence alignment was performed using the TꢀCOFFEE server and the obtained
28
alignment was used as input for homology modeling software MODELLER 9.13 to model
the missing and incomplete residues. Ten models were generated and the best model for
further studies was selected based on DOPE score and stereochemical quality of the model as
3
0
31
evaluated by PROCHECK and ProSAꢀweb server.
38
Molecular Dynamics. The molecular dynamics package NAMD (version 2.9) and
3
9
CHARMM22 force field were used for molecular dynamics simulations of GyrB43ꢀ18a
complex. Molecular mechanics parameters for compound 18a were estimated using
2
8
ACS Paragon Plus Environment

Page 29 of 51
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
4
0
ParamChem tool. Steepest descent (10,000 steps) and adopted basis Newton–Raphson
10,000 steps) energy minimizations were first performed to remove atomic clashes and to
optimize the atomic coordinates of the GyrB43ꢀ18a complex. The structure of the energy
(
41
minimized complex for MD simulation was prepared using psfgen in VMD (version 1.9.1.).
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
The complex was then embedded in a box of water, which was modeled explicitly by a TIP3P
4
2
model. The system was neutralized by addition of KCl at 0.4 M concentration. The MD
simulation was carried out in the NPT ensemble employing periodic boundary conditions.
Langevin dynamics and Langevin piston methods were used for temperature (300 K) and
pressure (1 atm) control, respectively. Shortꢀ and longꢀrange forces were calculated every 1
and 2 time steps, respectively, with a time step of 2.0 ps. The smooth particle mesh Ewald
4
3
method was used to calculate electrostatic interactions. The shortꢀrange interactions were
cut off at 12 Å. All chemical bonds between hydrogen and heavy atoms were held fixed using
44
SHAKE algorithm. The simulation consisted of three consecutive steps: (i) solvent
equilibration for 1 ns with ligand and protein constrained harmonically around the initial
structure, (ii) equilibration of the complete system for 1 ns with ligand and protein released,
and (iii) an unconstrained 10 ns production run to allow the protein and the ligand to position
themselves according to physical forces between them. The trajectory of the equilibration and
production run was used for analysis in VMD. Distances between atoms (Figure 2S) were
measured using the distance tool in VMD, data exported and analyzed in MS Excel.
General Procedures - Chemistry. Chemicals were obtained from Acros Organics (Geel,
Belgium), SigmaꢀAldrich (St. Louis, MO, USA) and Apollo Scientific (Stockport, UK) and
used without further purification. Analytical TLC was performed on silica gel Merck 60 F254
plates (0.25 mm), using visualization with UV light and spray reagents. Column
chromatography was carried out on silica gel 60 (particle size 240–400 mesh). HPLC analyses
were performed on an Agilent Technologies 1100 instrument (Agilent Technologies, Santa
2
9
ACS Paragon Plus Environment