Discovery of 4,5,6,7-Tetrahydrobenzo[1,2- d ]thiazoles as Novel DNA Gyrase Inhibitors Targeting the ATP-Binding Site

Article abstract of DOI:10.1021/acs.jmedchem.5b00489

Bacterial DNA gyrase and topoisomerase IV are essential enzymes that control the topological state of DNA during replication and validated antibacterial drug targets. Starting from a library of marine alkaloid oroidin analogues, we identified low micromolar inhibitors of Escherichia coli DNA gyrase based on the 5,6,7,8-tetrahydroquinazoline and 4,5,6,7-tetrahydrobenzo[1,2-d]thiazole scaffolds. Structure-based optimization of the initial hits resulted in low nanomolar E. coli DNA gyrase inhibitors, some of which exhibited micromolar inhibition of E. coli topoisomerase IV and of Staphylococcus aureus homologues. Some of the compounds possessed modest antibacterial activity against Gram positive bacterial strains, while their evaluation against wild-type, impA and ΔtolC E. coli strains suggests that they are efflux pump substrates and/or do not possess the physicochemical properties necessary for cell wall penetration. Our study provides a rationale for optimization of this class of compounds toward balanced dual DNA gyrase and topoisomerase IV inhibitors with antibacterial activity.

Full text of DOI:10.1021/acs.jmedchem.5b00489

Subscriber access provided by UNIV OF MISSISSIPPI
Article
Discovery of 4,5,6,7-Tetrahydrobenzo[1,2-d]thiazoles as
Novel DNA Gyrase Inhibitors Targeting the ATP-Binding Site
Tihomir Tomasic, Sotirios Katsamakas, Ziga Hodnik, Janez Ilaš, Matjaz
Brvar, Tom J Solmajer, Sofia Montalvao, Päivi Tammela, Mihailo Banjanac,
Gabrijela Ergovic, Marko Anderluh, Lucija Peterlin Maši#, and Danijel Kikelj
J. Med. Chem., Just Accepted Manuscript • Publication Date (Web): 22 Jun 2015
Downloaded from http://pubs.acs.org on June 22, 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 71
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Discovery of 4,5,6,7ꢀTetrahydrobenzo[1,2ꢀ
d]thiazoles as Novel DNA Gyrase Inhibitors
Targeting the ATPꢀBinding 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
a,‡
a,b,‡
a
a
c
Tihomir Tomašič, Sotirios Katsamakas,
Žiga Hodnik, Janez Ilaš, Matjaž Brvar, Tom
c
d
d
e
e
Solmajer, Sofia Montalvão, Päivi Tammela, Mihailo Banjanac, Gabrijela Ergović, Marko
a
a
a,*
Anderluh, Lucija Peterlin Mašič and Danijel Kikelj
a
University of Ljubljana, Faculty of Pharmacy, Department of Medicinal Chemistry, Aškerčeva
7
, 1000 Ljubljana, Slovenia
b
Aristotle University of Thessaloniki, Faculty of Health Sciences, School of Pharmacy,
Department of Pharmaceutical Chemistry, 54124 Thessaloniki, Greece
c
National Institute of Chemistry, Laboratory for Biocomputing and Bioinformatics, Hajdrihova
ulica 19, 1001 Ljubljana, Slovenia
d
University of Helsinki, Faculty of Pharmacy, Centre for Drug Research, P.O. Box 56, FIꢀ00014
Helsinki, Finland
e
Galapagos Istrazivacki Ctr Doo, Prilaz Baruna Filipovica 29, HRꢀ10000 Zagreb, Croatia
KEYWORDS antibacterial, DNA gyrase, docking, SPR, topoisomerase IV
ACS Paragon Plus Environment
1

Journal of Medicinal Chemistry
Page 2 of 71
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
ABSTRACT. Bacterial DNA gyrase and topoisomerase IV are essential enzymes that control the
topological state of DNA during replication and validated antibacterial drug targets. Starting
from a library of marine alkaloid oroidin analogs, we identified low micromolar inhibitors of E.
coli DNA gyrase, based on the 5,6,7,8ꢀtetrahydroquinazoline and 4,5,6,7ꢀtetrahydrobenzo[1,2ꢀ
d]thiazole scaffolds. Structureꢀbased optimization of the initial hits resulted in low nanomolar E.
coli DNA gyrase inhibitors, some of which exhibited micromolar inhibition of E. coli
topoisomerase IV and of S. aureus homologs. Some of the compounds possessed modest
antibacterial activity against Gram positive bacterial strains, while their evaluation against wildꢀ
type, impA and ꢁtolC E. coli strains suggests that they are efflux pump substrates and/or do not
possess the physicoꢀchemical properties necessary for cell wall penetration. Our study provides a
rationale for optimization of this class of compounds towards balanced dual DNA gyrase and
topoisomerase IV inhibitors with antibacterial activity.
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
ACS Paragon Plus Environment
2

Page 3 of 71
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
INTRODUCTION
The increasing emergence of pathogenic bacteria resistant to antibacterial drugs in both
communityꢀ and hospitalꢀacquired infections is a serious threat to global health, since currently
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
1
available therapies will no longer be effective in treating these infections. Although the rate of
bacterial resistance to currently used antibacterials is on the rise, there have recently been only a
few successful examples of new antibacterial drugs with novel modes of action entering the drug
2
market.
Bacterial topoisomerases are enzymes that catalyze changes in DNA topology and are
3
clinically validated targets for antibacterial drug discovery. Bacterial DNA gyrase and
topoisomerase IV are type IIA topoisomerases that catalyze reactions involving the transient
4
break of both strands of DNA. Both enzymes modify the topological state of DNA, which is
vital to DNA replication, repair and decatenation and is thus essential for cell viability. DNA
gyrase is involved in negative supercoiling of DNA during replication, whereas topoisomerase
IV is involved in the decatenation of daughter DNA following DNA replication. DNA gyrase is a
heterotetrameric protein consisting of two GyrA and two GyrB subunits (A B ), while
2
2
topoisomerase IV is composed of two ParC and two ParE subunits (C E ) that are homologous to
2
2
GyrA and GyrB, respectively. The GyrA and ParC subunits are involved in DNA transit, while
5
the GyrB and ParE subunits contain ATPase domains.
The complexity of reactions in DNA replication catalyzed by DNA gyrase and topoisomerase
IV offers multiple opportunities for therapeutic intervention. Both enzymes are targets of
fluoroquinolones, a clinically important class of antibacterial drugs, which interact mostly with
6
GyrA and ParC subunits and stabilize the enzymeꢀDNA complex, and of aminocoumarins (e.g.
ACS Paragon Plus Environment
3

Journal of Medicinal Chemistry
Page 4 of 71
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
7
novobiocin and clorobiocin) that bind to GyrB and ParE. Growing resistance to
8
fluoroquinolones limits their therapeutic use and therefore stimulates the search for new
inhibitor classes that target different enzyme binding sites in order to avoid crossꢀresistance with
the fluoroquinolones.
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
Due to the structural similarities between DNA gyrase and topoisomerase IV, dual targeting is
possible in most bacteria, which prolongs the onset of resistance and makes bacterial type IIA
9
topoisomerases attractive targets for discovering novel antibacterial drugs. Their potential for
dual targeting is one of the main reasons for several currently onꢀgoing drug discovery programs
10
in the pharmaceutical industry. For example, GyrB and dual targeting GyrB/ParE inhibitors
1
1
12
have been identified among 5ꢀ(3ꢀpyrazolyl)thiazoles,
azaindoles,
pyrrolo[2,3ꢀ
1
3,14
15
16
d]pyrimidines,
imidazo[1,2ꢀa]pyridine ureas, benzimidazole ureas (Figure 1), and
17,18
pyrrolamides
(Figure 2a). Although these inhibitors target the ATPꢀbinding site of GyrB
and/or ParE, they usually display good selectivity profiles against eukaryotic ATPꢀbinding
1
9,20
proteins (e.g. kinases, human topoisomerase II).
Despite all the efforts to discover further
GyrB or dualꢀtargeting GyrB/ParE inhibitors, none have advanced into the clinic. Most of the
GyrB/ParE inhibitors display potent MICs values against Gram positive bacteria. However,
developing molecules with balanced dualꢀtargeting activity that possess the physicoꢀchemical
property profiles necessary to result in potent antibacterial activity against Gram negative
pathogens and to elude their multiꢀdrug efflux pumps remains very challenging, since only few
21
GyrB/ParE inhibitors possess cellular activity against Gram negative bacteria.
ACS Paragon Plus Environment
4

Page 5 of 71
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
Figure 1. Examples of selected structural types of GyrB or dualꢀtargeting GyrB/ParE inhibitors.
a)
b)
O
H
H
H
O
O
N
O
N
O
N
N
N
R1
NH2
S
N
OH
S
N
OH
H
HN
N
HN
N
H
NH
Cl
Cl
N
Cl
Cl
R2
O
F
O
O
pyrrolamides
clathrodin (R1 = R2 = H)
hymenidin (R1 = H, R2 = Br)
S. aureus GyrB IC50 = 4 nM
MIC (S. aureus) = 2 g/mL
MIC (E. coli) = >64 g/mL
S. aureus GyrB IC50 <10 nM
E. coli ParE IC50 = 73 nM
MIC (S. aureus) = 0.036 g/mL
MIC (E. coli) = 24 g/mL
oroidin (R = R2 = Br)
1
MIC (E. coli tolC) = 0.5 g/mL
MIC (E. coli tolC) = 0.94 g/mL
Figure 2. a) Pyrrolamideꢀbased GyrB inhibitors and b) 2ꢀaminoimidazole marine alkaloids
clathrodin, hymenidin and oroidin possessing the pyrrolamide moiety (in red).
ACS Paragon Plus Environment
5

Journal of Medicinal Chemistry
Page 6 of 71
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
Recently, we have been involved in design and synthesis of several series of analogs of marine
alkaloids clathrodin, oroidin and hymenidin (Figure 2b) displaying voltageꢀgated sodium channel
2
2ꢀ25
26
27
modulatory activity,
antimicrobial activity, inhibition of bacterial biofilm formation and
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
28
apoptosis inducing activity in HepG2 and THPꢀ1 cell lines. Some of these compounds
incorporate a bromo substituted pyrrolamide moiety and are therefore structurally similar to the
17,18
recently discovered pyrrolamideꢀbased GyrB inhibitors
(Figure 2). This observation
motivated us to test some of the synthesized clathrodin analogs for inhibition of DNA gyrase
from Escherichia coli. Our library of 120 clathrodin analogs was enriched by structureꢀbased
29
virtual screening, using the E. coli GyrB crystal structure (PDB entry: 4DUH ), selecting the
twenty top ranked compounds, and testing them for E. coli DNA gyrase inhibition. This resulted
in three initial hits with IC values between 12 ꢀM and 170 ꢀM (compounds 1-3, Figure 3). A
5
0
common structural feature of hits 1-3 is the hydrophobic 4,5ꢀdibromoꢀ1Hꢀpyrrole moiety, which
suggested testing compounds 4 and 5 (Figure 3) that contain the smaller 4ꢀbromoꢀ1Hꢀpyrrole and
the bulkier 1Hꢁindole moiety, respectively. Besides pyrrolamide moiety, the hydrophobic
character of the substituents on the pyrrole ring appears to play a crucial role, since among
compounds 3-5 4,5ꢀdibromoꢀ1Hꢀpyrroleꢀbased 3 was found to be the most active, with an IC50
value of 12 ꢀM, followed by the 1Hꢀindoleꢀbased compound 5 (IC = 21 ꢀM), while the 4ꢀ
5
0
bromoꢀ1Hꢀpyrroleꢀbased compound 4 possessed only weak activity (IC = 520 ꢀM).
5
0
We describe here the structural optimization of these initial hits toward potent E. coli DNA
gyrase inhibitors that also possess weak activity against Staphylococcus aureus DNA gyrase and
E. coli and S. aureus topoisomerase IV. Molecular docking binding mode studies of inhibitors in
the ATPꢀbinding pocket of E. coli GyrB guided the selection of compounds for synthesis and
provided a rationale for the observed structureꢀactivity relationship (SAR). Evaluation of
ACS Paragon Plus Environment
6

Page 7 of 71
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
antibacterial activity against Gram positive S. aureus and Enterococcus faecalis and Gram
negative E. coli and Pseudomonas aeruginosa highlighted some compounds possessing modest
antibacterial activity against Gram positive bacterial strains.
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
Br
NH2
O
N
Br
Br
H
S
H
NH2
NH
N
H
N
N
N
S
Br
N
Br
H
N
NH2
N
N
H
O
O
Br
1
2
3
E. coli DNA gyrase IC50 = 170
M
E. coli DNA gyrase IC50 = 120
M
E. coli DNA gyrase IC50 = 12
M
Br
S
S
H
N
H
NH2
NH2
N
N
N
N
N
H
H
O
O
5
4
E. coli DNA gyrase IC50 = 520
M
E. coli DNA gyrase IC50 = 21 M
Figure 3. Oroidin analogs identified as E. coli DNA gyrase inhibitors.
RESULTS AND DISCUSSION
Design. Several crystal structures of GyrBꢀinhibitor complexes have been shown to reveal
conserved binding modes of inhibitors in the GyrB ATPꢀbinding site. Inhibitors partially mimic
the binding of ATP by forming hydrogen bonds with Asp73 (E. coli GyrB numbering) and a
structurally conserved water molecule, both of which are in contact with the adenine moiety of
30
ATP. Inhibitors thus usually possess adjacent hydrogen bond acceptor and donor groups
(
Figure 4a). Common structural features of inhibitors are also additional functionalities
interacting with the Arg136 and/or Arg76 side chain.
ACS Paragon Plus Environment
7

Journal of Medicinal Chemistry
Page 8 of 71
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
O
O
H
H
N
N
O
Asp73
O
Asp73
O-
H
O-
H
O
H
O
H
H
H
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
Br
Br
N
N
N
O
H
N
HN
(S)
N
N
N
S
NH2
ATP
R
3
a)
b)
Figure 4. a) Schematic representation of the hydrogen bonding network between Asp73, the
conserved water molecule and the adenine moiety of ATP (left) or the pyrrolamide moiety of
compound 3 (right) in the E. coli GyrB ATPꢀbinding site. b) The FlexXꢀpredicted binding mode
29
of compound 3 (in yellow sticks) in the ATPꢀbinding site of E. coli GyrB (PDB entry: 4DUH ;
in green). Hydrogen bonds are displayed as black dashed bonds. The figure was prepared by
3
1
PyMOL.
Preliminary data from the virtual screening hits showed that in the tetrahydrobenzothiazole
series the additional methylene bridge between the carboxamide group and 2ꢀaminoꢀ4,5,6,7ꢀ
tetrahydrobenzo[1,2ꢀd]thiazole scaffold had a negative effect on E. coli DNA gyrase inhibition
(
compounds 2 vs 3). Consequently, the structures of compounds 3 and 5, which were found to be
the most active in the library of available oroidin analogs, and analog of compound 1 without the
methylene bridge were optimized (Figure 5). In the calculated binding mode of compound 3 in
the ATPꢀbinding site of E. coli GyrB (PDB entry: 4DUH), the 4,5ꢀdibromoꢀ1Hꢀpyrroleꢀ2ꢀ
carboxamide moiety is seen to be buried in the hydrophobic pocket and to form hydrogen bonds
with the Asp73 side chain and the conserved water molecule, while the 2ꢀamino group on the
thiazole ring is positioned in such a way as to enable substitution with functional groups for
interaction with Arg76 and/or Arg136 (Figure 4). The results of molecular docking calculations
ACS Paragon Plus Environment
8

Page 9 of 71
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
indicated that the oxalyl, malonyl and succinyl moieties attached to the 2ꢀamino group of the
5,6,7,8ꢀtetrahydroquinazoline or 4,5,6,7ꢀtetrahydrobenzo[1,2ꢀd]thiazole cores are of such lengths
as to enable the formation of hydrogen bond(s) with the Arg136 side chain. In addition, we have
also investigated the effects of (i) replacing the 4,5ꢀbromoꢀ1Hꢀpyrrole by the 4,5ꢀdichloroꢀ1Hꢀ
pyrrole or 1Hꢁindole moiety and (ii) chirality by preparing the R isomer of compound 3 (Figure
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).
Figure 5. Design strategy for optimization of initial hits 1 and 3.
Chemistry. The key intermediate 6 of the tetrahydroquinazoline series was synthesized as
32
described by Marinko and coꢀworkers. 5,6,7,8ꢀTetrahydroquinazolines 7, 8 and 9 were then
synthesized by acylation of 6 with the corresponding 2,2,2ꢀtrichloroꢀ1ꢀ(1Hꢀpyrrolꢀ2ꢀ
3
3
yl)ethanones in N,Nꢁdimethylformamide (DMF), using Na CO as base (Scheme 1). Analog 10
2
3
(Scheme 1) was synthesized with standard Oꢀ(benzotriazolꢀ1ꢀyl)ꢀN,N,N',N'ꢀtetramethyluronium
tetrafluoroborate (TBTU)ꢀpromoted coupling of 6 with the 1Hꢁindoleꢀ2ꢀcarboxylic acid at room
temperature.
With the aim to synthesize analogs 15 and 16, which would significantly contribute to our
envisaged SAR study, we acylated compounds 9 and 10 with methyl malonyl chloride in DMF to
yield compounds 11 and 12 (Scheme 1). The subsequent attempt at methyl ester cleavage with
1
M sodium or lithium hydroxide in THF or methanol resulted in an unusual amide bond
ACS Paragon Plus Environment
9

Journal of Medicinal Chemistry
Page 10 of 71
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
cleavage at position 2 and consequential isolation of compounds 9 and 10. Our further attempt to
cleave the methyl ester of compounds 11 and 12 with lithium iodide in pyridine at reflux resulted
in decarboxylation of these βꢀketocarboxylic acids. Since such a lithium iodideꢀpromoted
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
34
decarboxylation has already been described, we attempted the synthesis of benzylꢀprotected
analogs 13 and 14 to give the desired analogs 15 and 16. This was achieved by the TBTUꢀ
promoted coupling of compounds 9 and 10 with the 3ꢀ(benzyloxy)ꢀ3ꢀoxopropanoic acid at 40 ºC
(Scheme 1). Subsequent catalytic hydrogenation of compounds 13 and 14 again resulted in
decarboxylation of the βꢀketocarboxylic acid (e.g. compound 17), which suggested that the
desired acids 15 and 16 are not stable under the conditions used.
e
N
NH2
O
H
H
N
N
O
c
N
d
N
N
O
O
R
N
O
H
N
O
O
N
O
O
R
N
R
N
9
(R = 4,5-dibromopyrrol-2-yl)
H
H
10 (R = indol-2-yl)
1
1 (R = 4,5-dibromopyrrol-2-yl)
f
13 (R = 4,5-dibromopyrrol-2-yl)
14 (R = indol-2-yl)
g
12 (R = indol-2-yl)
H
H
N
N
OH
N
N
O
decarboxylation
O
N
O
O
N
O
R
N
R
N
H
H
15 (R = 4,5-dibromopyrrol-2-yl)
17 (R = indol-2-yl)
16 (R = indol-2-yl)
Scheme 1. Reagents and conditions: (a) for 7: 2,2,2ꢀtrichloroꢀ1ꢀ(4,5ꢀdichloroꢀ1Hꢀpyrrolꢀ2ꢀ
yl)ethanone, for 8: 2,2,2ꢀtrichloroꢀ1ꢀ(4ꢀbromoꢀ1Hꢀpyrrolꢀ2ꢀyl)ethanone, for 9: 2,2,2ꢀtrichloroꢀ1ꢀ
(
4,5ꢀdibromoꢀ1Hꢀpyrrolꢀ2ꢀyl)ethanone, Na CO , DMF, 35 ºC, 2.5ꢀ3 h; (b) 1Hꢁindoleꢀ2ꢀ
2 3
ACS Paragon Plus Environment
10

Page 11 of 71
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
carboxylic acid, TBTU, Et N, DMF, r.t., 2.5 h; (c) methyl malonyl chloride, Et N, DMF, r.t., 17
3
3
h; (d) 3ꢀ(benzyloxy)ꢀ3ꢀoxopropanoic acid, TBTU, Et N, DMF, 40 ºC, 4 h; (e) 1 M NaOH or
3
LiOH, THF or MeOH, r.t. or 0 °C; (f) LiI, pyridine, reflux; (g) H , Pd/C, MeOH/THF, r.t., 3 h.
2
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
Racemic 4,5,6,7ꢀtetrahydrobenzo[1,2ꢀd]thiazolꢀ2,6ꢀdiamine (18) was synthesized and optical
resolution of the racemate to (S)ꢀ18 and (R)ꢀ18 enantiomers was performed as described by
3
5
Schneider and Mierau. In the Xꢀray structure of Lꢀ(+)ꢀtartaric acid salt of (ꢀ)ꢀ18 chiral center in
35
position 6 was shown to have the S configuration. Since specific rotations of (S)ꢀ18 and (R)ꢀ18
3
5
were in agreement with the values reported in the literature, no additional experiments were
performed to determine the configuration of the compounds. Synthesis of the (S)ꢀ4,5,6,7ꢀ
tetrahydrobenzo[1,2ꢀd]thiazolꢀ2,6ꢀdiamines 3, 4 and 22 was achieved by standard synthetic
procedures by acylation of the (S)ꢀ4,5,6,7ꢀtetrahydrobenzo[1,2ꢀd]thiazoleꢀ2,6ꢀdiamine ((S)-18)
with the 2,2,2ꢀtrichloroꢀ1ꢀ(4,5ꢀdibromoꢀ1Hꢀpyrrolꢀ2ꢀyl)ethanone, 2,2,2ꢀtrichloroꢀ1ꢀ(4ꢀbromoꢀ
1
Hꢀpyrrolꢀ2ꢀyl)ethanone or 2,2,2ꢀtrichloroꢀ1ꢀ(4,5ꢀdichloroꢀ1Hꢀpyrrolꢀ2ꢀyl)ethanone in DMF,
2
2
respectively (Scheme 2). TBTUꢀpromoted coupling of (S)ꢀ18 with 1Hꢁindoleꢀ2ꢀcarboxylic
acid, 5ꢀfluoroꢀ1Hꢁindoleꢀ2ꢀcarboxylic acid or 5ꢀchloroꢀ1Hꢁindoleꢀ2ꢀcarboxylic acid afforded
compounds 5, 20 and 21 (Scheme 2).
Br
H N
R
2
S
NH
NH
H
N
H
N
Br
NH2
a
b
S
N
S
N
N
NH
2
NH
2
O
(S)-18
O
3
a
a
5, R = H
20, R = F
Cl
21, R = Cl
NH
NH
H
H
Br
Cl
N
N
S
N
S
NH2
NH2
O
O
N
4
22
Scheme 2. Reagents and conditions: (a) for 3: 2,2,2ꢀtrichloroꢀ1ꢀ(4,5ꢀdibromoꢀ1Hꢀpyrrolꢀ2ꢀ
yl)ethanone, for 4: 2,2,2ꢀtrichloroꢀ1ꢀ(4ꢀbromoꢀ1Hꢀpyrrolꢀ2ꢀyl)ethanone, for 22: 2,2,2ꢀtrichloroꢀ1ꢀ
ACS Paragon Plus Environment
11

Journal of Medicinal Chemistry
Page 12 of 71
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
(
4,5ꢀdichloroꢀ1Hꢀpyrrolꢀ2ꢀyl)ethanone, Na CO , DMF, 35 ºC, 2ꢀ3 h; (b) for 5: 1Hꢁindoleꢀ2ꢀ
2
3
carboxylic acid, for 20: 5ꢀfluoroꢀ1Hꢁindoleꢀ2ꢀcarboxylic acid, for 21: 5ꢀchloroꢀ1Hꢁindoleꢀ2ꢀ
carboxylic acid, TBTU, Et N, DMF, r.t., 2.5 h.
3
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
According to the general scheme for structural optimization of the initial hit 3 (Figure 5) and of
its 1Hꢀindoleꢀ, 4ꢀbromoꢀ1Hꢀpyrroleꢀ and 4,5ꢀdichloroꢀ1Hꢀpyrroleꢀbased analogs 5, 4 and 22, the
compounds were acylated at the amino group at position 2 of the (S)ꢀ4,5,6,7ꢀ
tetrahydrobenzo[1,2ꢀd]thiazole moiety with ethyl oxalyl chloride, methyl malonyl chloride or
methyl succinyl chloride in 1,4ꢀdioxane at room temperature (Scheme 3). The resulting products
were 4,5ꢀdibromoꢀ1Hꢁpyrroleꢀbased esters 23ꢀ25, 4ꢀbromoꢀ1Hꢁpyrroleꢀbased ester 27, 1Hꢁ
indoleꢀbased esters 28-30, and 4,5ꢀdichloroꢀ1Hꢁpyrroleꢀbased esters 31-33. Esters were
hydrolyzed to their carboxylic acids 34-36 and 38-44 using alkaline conditions (Scheme 3). To
further explore SAR, Rꢁisomers 19, 26 and 37 were prepared using the same conditions as
described for the synthesis of their S enantiomers 3, 24 and 35, starting from (R)-18 (Scheme 3).
To evaluate the importance of the terminal carboxyl group of compound 35 for DNA gyrase
inhibition, Nꢁacetyl compound 45 was prepared from 3 using acetyl chloride and triethylamine in
1
,4ꢀdioxane. Furthermore, compound 3 was also converted to the dihydropyrimidone analog 46
using oneꢀpot cyclization with acryloyl chloride, sodium carbonate and sodium iodide in
36
acetonitrile. Carboxamide 47 was prepared by aminolysis of the methyl ester 24, by bubbling
gaseous ammonia into solution of 24 in ethanol, with the aim of improving penetration of
compounds 24 and 35 through the membrane into the cytoplasm of bacteria.
ACS Paragon Plus Environment
12

Page 13 of 71
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
O
H
H
R1
N
R1
N
S
N
S
N
O
NH
NH
O
NH2
O
O
1
1
45 (R = 4,5-dibromopyrrol-2-yl)
47 (R = 4,5-dibromopyrrol-2-yl)
c
d
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
H
H
H
R1
N
R2
O
R1
N
R1
N
S
N
S
N
O
S
N
O
a
b
NH
O
NH
O
OH
NH2
O
O
n
O
n
3
(R = 4,5-dibromopyrrol-2-yl)
(R = 4-bromopyrrol-2-yl)
5 (R = indol-2-yl)
0 (R = 5-fluoroindol-2-yl)
1 (R = 5-chloroindol-2-yl)
4
1
2
1
2
2
2
3 (R = 4,5-dibromopyrrol-2-yl, R = Et, n = 0)
34 (R = 4,5-dibromopyrrol-2-yl, n = 0)
1
2
1
4 (R = 4,5-dibromopyrrol-2-yl, R = Me, n = 1)
35 (R = 4,5-dibromopyrrol-2-yl, n = 1)
2
2
1
2
1
5 (R = 4,5-dibromopyrrol-2-yl, R = Me, n = 2)
36 (R = 4,5-dibromopyrrol-2-yl, n = 2)
1
2
1
2
2 (R = 4,5-dichloropyrrol-2-yl)
27 (R = 4-bromopyrrol-2-yl, R = Me, n = 1)
38 (R = 4-bromopyrrol-2-yl, n = 1)
1
2
1
2
8 (R = indol-2-yl, R = Et, n = 0)
39 (R = indol-2-yl, n = 0)
1
2
1
29 (R = indol-2-yl, R = Me, n = 1)
40 (R = indol-2-yl, n = 1)
e
1
2
1
30 (R = indol-2-yl, R = Me, n = 2)
41 (R = indol-2-yl, n = 2)
H
1
2
1
R1
31 (R = 4,5-dichloropyrrol-2-yl, R = Et, n = 0)
42 (R = 4,5-dichloropyrrol-2-yl, n = 0)
N
S
N
1
2
1
3
2 (R = 4,5-dichloropyrrol-2-yl, R = Me, n = 1)
43 (R = 4,5-dichloropyrrol-2-yl, n = 1)
N
O
1
2
1
33 (R = 4,5-dichloropyrrol-2-yl, R = Me, n = 2)
44 (R = 4,5-dichloropyrrol-2-yl, n = 2)
O
1
4
6 (R = 4,5-dibromopyrrol-2-yl)
H
H
H
R1
N
R1
R1
S
N
N
S
N
O
N
S
N
O
a
b
NH2
NH
O
O
NH
O
OH
O
O
O
1
1
1
1
9 (R = 4,5-dibromopyrrol-2-yl)
26 (R = 4,5-dibromopyrrol-2-yl)
37 (R = 4,5-dibromopyrrol-2-yl)
Scheme 3. Reagents and conditions: (a) ethyl oxalyl chloride (for 23, 28 and 31) or methyl
malonyl chloride (for 24, 26-28 and 32) or methyl succinyl chloride (for 25, 30 and 33), Et N,
3
1
,4ꢀdioxane, r.t., 24 h; (b) 1 M NaOH, MeOH/H O, r.t., 24 h; (c) acetyl chloride, Et N, 1,4ꢀ
2
3
dioxane, r.t., 24 h; (d) NH , EtOH, r.t., 2 h; (e) acryloyl chloride, Na CO , CH CN, 0 °C, 2 h,
3
(g)
2
3
3
then NaI, 80 °C, 24 h.
In vitro Enzyme Inhibition. All synthesized compounds were tested for E. coli DNA gyrase
inhibitory activity, using the DNA gyrase supercoiling assay. Those that inhibited E. coli DNA
gyrase with IC values below 50 µM were also tested for inhibition of S. aureus DNA gyrase, as
50
well as topoisomerases IV from E. coli and S. aureus using the topoisomerase IV relaxation
ACS Paragon Plus Environment
13

Journal of Medicinal Chemistry
Page 14 of 71
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
assay (Tables 1ꢀ4). Results are presented as residual activities (RA) of the enzyme at 100 µM of
the tested compound or IC values for the more active compounds (RA < 50% at 100 µM).
50
Inhibitory activities of the 5,6,7,8ꢀtetrahydroquinazolineꢀ2,6ꢀdiamines 7-12 and 17 are
presented in Table 1. Compound 9, an analog of the initial hit 1 (IC = 170 µM) but lacking 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
5
0
methylene bridge, displayed, as expected, improved E. coli DNA gyrase inhibition, with an IC50
value of 2.9 µM, since similar difference in activity between compounds with and without the
methylene bridge was also observed in the case of the 4,5,6,7ꢀtetrahydrobenzo[1,2ꢀd]thiazoleꢀ2ꢀ
amines 2 and 3. Replacing the 4,5ꢀdibromoꢀ1Hꢀpyrrole moiety of 9 by 4ꢀbromoꢀ1Hꢀpyrrole (8,
IC = 35 µM), 1Hꢀindole (10, IC = 29 µM) or 4,5ꢀdichloroꢀ1Hꢀpyrrole (7, IC = 5.7 µM)
5
0
50
50
resulted in weaker activity. Contrary to expectations, introduction of the malonyl group on the 2ꢀ
amino group of the 5,6,7,8ꢀtetrahydroquinazolineꢀ2,6ꢀdiamine scaffold of 9 and 10 did not result
in improved E. coli DNA gyrase inhibition. NꢁMethylmalonylꢀ4,5ꢀdibromoꢀ1Hꢀpyrroleꢀbased
compound 11 (IC50 = 2.6 µM) possessed inhibitory activity similar to that of its N nonꢀ
substituted analog 9 (IC50 = 2.9 µM), while the indoleꢀbased compound 12 (RA = 77% at
100 µM) was devoid of this activity. Since, because of their instability due to decarboxylation,
our attempts to prepare carboxylic acid derivatives 15 and 16 were unsuccessful, only the
decarboxylated acetylated compound 17 was tested and found to possess weak activity (RA =
52% at 100 µM).
The weak antibacterial activity of these inhibitors (see below) prompted us to test them also
against E. coli topoisomerase IV as well as against S. aureus DNA gyrase and topoisomerase IV.
Compounds 7-11, that displayed inhibition of E. coli DNA gyrase with IC50 values between
2.6 µM and 35 µM, were found to be inactive at 100 µM in all three assays.
ACS Paragon Plus Environment
14

Page 15 of 71
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Table 1. Inhibition of Escherichia coli and Staphylococcus aureus DNA gyrase and
topoisomerase IV by the 5,6,7,8ꢀtetrahydroquinazolines 7-12 and 17.
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
DNA gyrase
topoisomerase IV
E. coli MIC
1
a
a
cpd
R
IC50 [ꢀM] or RA [%]
IC50 [ꢀM] or RA [%]
[ꢀg/mL]
E. coli
S. aureus
100%
E. coli
100%
11 µM
S. aureus
wt
ꢀ
tolC
impA
b
neg
ꢀ
ꢀ
100%
100%
ꢀ
ꢀ
ꢀ
ꢀ
c
NB
0.17 µM 0.040 µM
27 µM
ꢀ
d
7
5.7 ꢀM
35 ꢀM
90%
96%
97%
106%
105%
n.t.
n.t.
n.t.
n.t.
n.t.
8
9
100%
n.t.
2.9 ꢀM
29 ꢀM
2.6 ꢀM
96%
119%
71%
102%
97%
97%
98%
103%
98%
>256
n.t.
>256
n.t.
>256
n.t.
1
0
11
>256
64
256
1
1
2
7
77%
52%
n.t.
n.t.
n.t.
n.t.
n.t.
n.t.
n.t.
n.t.
n.t.
n.t.
n.t.
n.t.
a
b
c
Residual activity of the enzyme at 100 ꢀM of the tested compound.
1% DMSO was used as a negative control.
NB – novobiocin was used as a positive control.
ACS Paragon Plus Environment
15

Journal of Medicinal Chemistry
Page 16 of 71
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
d
n.t. – not tested.
Comparison of the E. coli DNA gyrase inhibitory activities of hits 3 and 4 (Table 2) revealed
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
that substituents on the pyrrole ring have an important effect on inhibition. Bromine atoms
presumably form hydrophobic interactions in the hydrophobic pocket of the ATPꢀbinding site. In
addition, they increase the acidity of the pyrrole amino group, thereby increasing the strength of
the hydrogen bond formed with the Asp73 side chain. This explains why the dibromoꢀsubstituted
pyrroleꢀbased compound 3 (IC = 12 ꢀM) exhibited more potent DNA gyrase inhibition than its
5
0
monobromopyrrole counterpart 4 (IC = 520 ꢀM) (Table 2). According to the calculated binding
5
0
mode of 3 in the ATPꢀbinding site of E. coli DNA gyrase (Figure 4), there is sufficient space for
introduction of a hydrogen bond acceptor on the 2ꢀamino group of the thiazole moiety for
interaction with the guanidine group of Arg136. Indeed esters 23, 24 and 25, containing the
oxalyl, malonyl and succinyl groups, displayed improved E. coli DNA gyrase inhibition with
IC values of 0.10 ꢀM, 0.096 ꢀM and 0.093 ꢀM, suggesting their interaction with Arg136. The
5
0
possibility of hydrogen bond formation between the carbonyl oxygens of these moieties with the
guanidine group of Arg136 is clearly indicated in the docking experiments (Figure 7).
Carboxylic acid derivatives 34-36 with IC values between 0.049 ꢀM and 0.069 ꢀM were not
50
substantially more potent E. coli DNA gyrase inhibitors than their ester counterparts with IC50
values between 0.093 ꢀM and 0.10 ꢀM. Their comparable activities indicate that the terminal
ester or acid groups of compounds 23-25 and 34-36 probably form a hydrogen bond, and not an
ionic/salt bridge interaction, with the guanidine group of Arg136. Alternatively, the carboxylic
acid could form a salt bridge with Arg136, but there is a higher desolvation penalty for taking the
carboxylate from the aqueous environment to the enzyme binding site than for esters, which
could explain comparable IC values of carboxylic acids and esters. Another hydrogen bond
5
0
ACS Paragon Plus Environment
16

Page 17 of 71
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
partner in the guanidineꢀcarboxylate interaction in the enzyme environment could be a
neighboring water molecule. A hydrogen bond interaction with Arg76 (Fig. 7b) could also be
responsible for the similar binding strengths of ester and carboxylic acid derivatives.
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 SAR was further explored by preparing R isomers 19, 26 and 37 of compounds 3, 24 and
35. The nonꢀsubstituted compound 19 (IC = 47 ꢀM) exhibited 4ꢀfold weaker activity than its S
5
0
isomer 3 (IC50 = 12 ꢀM). Changing the stereochemistry from S to R had an even more
pronounced effect in the cases of ester 26 (IC = 9.1 ꢀM vs 0.096 ꢀM for 24) and carboxylic
5
0
acid 37 (IC = 0.50 ꢀM vs 0.069 ꢀM for 35). Consequently, a conclusion can be made that in the
5
0
investigated series, S isomers are more potent inhibitors than the R optical antipodes.
Since the introduction of the Nꢁmalonyl moiety (24 and 35 vs 3) had a positive effect on
binding affinity, we also prepared Nꢁmalonyl compounds 27 (IC = 25 ꢀM) and 38 (IC =
5
0
50
14 ꢀM) that bear the 4ꢀbromoꢀ1Hꢀpyrrole moiety. Their activity was greater than that of 4 (IC50
=
520 ꢀM) but 200ꢀfold lower from the dibromo counterparts 24 and 35. This again highlights
the importance for potent inhibition of the bromo substituents on the pyrrole ring, since the
pyrrolamide moiety is buried deep in the ATPꢀbinding site and forms crucial hydrogen bonds
and hydrophobic interactions with the protein.
The importance of the terminal ester or carboxylic group of compounds 23-27 and 34-38 was
assessed by preparing the Nꢁacetyl derivative 45, which displayed E. coli DNA gyrase inhibition
with an IC value of 0.15 ꢀM, which is 3ꢀfold weaker than that observed for the most potent
5
0
compound 36 (IC = 0.049 ꢀM) but 80ꢀfold more potent than that of the parent compound 3
5
0
(
IC = 12 ꢀM) implying that the acetyl carbonyl group of 45 is, importantly, most probably
50
involved in hydrogen bond formation. In contrast, cyclization of 3 to the dihydropyrimidone
derivative 46 (IC = 12 ꢀM) did not result in improved inhibition.
5
0
ACS Paragon Plus Environment
17

Journal of Medicinal Chemistry
Page 18 of 71
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
The amide derivative 47, prepared from ester 24 (IC = 0.096 ꢀM), in an attempt to improve
5
0
the penetration through the membrane of bacteria into the cytoplasm (see below), displayed
decreased activity (IC = 1.7 ꢀM) as compared to its ester counterpart.
5
0
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
E. coli DNA gyrase inhibitors (IC < 50 ꢀM) bearing the bromoꢀsubstituted pyrrole moiety
5
0
were also tested for their inhibition of E. coli topoisomerase IV (Table 2). They were found to be
mostly inactive or only weak inhibitors (IC values between 73 ꢀM and 200 ꢀM for compounds
5
0
2
4, 34 and 35). The common structural feature of these active compounds is the presence of the
oxalyl or malonyl moiety. A similar trend was observed when these inhibitors were tested for
inhibition of S. aureus DNA gyrase and topoisomerase IV (IC50 values between 76 ꢀM and
2
70 ꢀM for compounds 23, 24, 34, 35, 36 and 45) (Table 2), which probably explains their weak
antibacterial activity against some Gram positive pathogens (see below).
Table 2. Inhibition of Escherichia coli and Staphylococcus aureus DNA gyrase and
topoisomerase IV by the 4,5,6,7ꢀtetrahydrobenzo[1,2ꢀd]thiazoles containing the 4,5ꢀdibromoꢀ1Hꢀ
pyrrole (3, 19, 23-26, 34-37, 45-47) or 4ꢀbromoꢀ1Hꢀpyrrole moiety (4, 27, 38).
DNA gyrase
topoisomerase IV
E. coli MIC
1
2
a
b
b
cpd
*
R
R
n
IC50 (K
d
) [ꢀM] or RA [%]
S. aureus
IC50 [ꢀM] or RA [%]
[ꢀg/mL]
E. coli
E. coli
100%
11 µM
S. aureus
wt
ꢀ
tolC
impA
c
neg
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
100%
100%
100%
ꢀ
ꢀ
ꢀ
ꢀ
d
NB
0.17 µM
12 ꢀM
0.040 µM
27 µM
ꢀ
3
S
ꢀ
ꢀ
90%
101%
102%
>256
64
64
a
(3 ꢀM )
ACS Paragon Plus Environment
18

Page 19 of 71
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
ec
4
S
R
S
S
ꢀ
ꢀ
ꢀ
520 ꢀM
47 ꢀM
n.t.
n.t.
96%
n.t.
92%
n.t.
n.t.
64
n.t.
64
1
9
ꢀ
118%
80 ꢀM
110 ꢀM
>256
>256
>256
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
23
OEt
OMe
0
1
0.10 ꢀM
74%
180 ꢀM
74%
128
16
>256
32
0
.096 ꢀM
2
4
86 ꢀM
a
(
0.072 ꢀM )
0
.093 ꢀM
2
2
2
3
3
5
6
7
4
5
S
R
S
S
S
OMe
OMe
OMe
OH
2
1
1
0
1
113%
64%
97%
101%
95%
99%
85%
>256
>256
n.t.
32
64
64
a
(0.15 ꢀM )
9.1 ꢀM
256
25 ꢀM
100%
120 ꢀM
86 ꢀM
104%
78 ꢀM
76 ꢀM
n.t.
n.t.
0.058 ꢀM
200 ꢀM
74 ꢀM
>256
>256
256
>256
>256
>256
0
.069 ꢀM
OH
a
(0.042 ꢀM )
.049 ꢀM
0
3
3
3
4
4
4
6
7
8
5
6
7
S
R
S
S
S
S
OH
OH
OH
ꢀ
2
1
1
ꢀ
270 ꢀM
108%
110%
26 ꢀM
n.t.
90%
96%
96%
98%
n.t.
110 ꢀM
102%
97%
>256
>256
n.t.
256
>256
n.t.
>256
>256
n.t.
a
(
0.010 ꢀM )
0.50 ꢀM
14 ꢀM
0.15 ꢀM
12 ꢀM
101%
n.t.
n.t.
n.t.
n.t.
ꢀ
ꢀ
>256
>256
>256
128
>256
>256
NH
1
1.7 ꢀM
71%
97%
98%
2
a
K as determined by surface plasmon resonance (see below).
d
ACS Paragon Plus Environment
19

Journal of Medicinal Chemistry
Page 20 of 71
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
b
c
Residual activity of the enzyme at 100 ꢀM concentration of the tested compound.
1% DMSO was used as a negative control.
d
NB – novobiocin was used as a positive control.
n.t. – not tested.
e
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
Structural optimization of the indoleꢀbased initial screening hit 5 was less successful, the E.
coli DNA gyrase inhibitory activities remaining in the micromolar range (Table 3). First, we
explored whether additional substitution of the indole moiety at position 5 is tolerated by the
enzyme active site. Introduction of a 5ꢀfluoro or 5ꢀchloro substituent was detrimental to the
activity, with 5ꢀfluoroꢀ1Hꢀindole 20 (IC = 270 ꢀM) and 5ꢀchloroꢀ1Hꢀindole 21 (RA = 94%)
5
0
displaying weaker inhibition than 5 (IC50 = 21 ꢀM). Following the optimization strategy
described for bromopyrroles, we synthesized the indoleꢀbased oxalyl, malonyl and succinyl
esters 28-30 and carboxylic acids 39-41. These inhibited E. coli DNA gyrase with IC values
5
0
between 7.6 ꢀM and 35 ꢀM, not significantly better than that of 5 (IC = 21 ꢀM), so SAR of
50
these series was not further explored. Indoleꢀbased compounds were also found to be devoid of
activity against E. coli topoisomerase IV and S. aureus homologs (Table 3).
Table 3. Inhibition of Escherichia coli and Staphylococcus aureus DNA gyrase and
topoisomerase IV by the 4,5,6,7ꢀtetrahydrobenzo[1,2ꢀd]thiazoles containing the 1Hꢁindole
moiety (5, 20, 21, 28-30, 39-41).
1
2
cpd
R
R
n
DNA gyrase
topoisomerase IV
E. coli MIC
ACS Paragon Plus Environment
20

Page 21 of 71
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
a
a
IC50 [ꢀM] or RA [%]
IC50 [ꢀM] or RA [%]
[ꢀg/mL]
E. coli
100%
S. aureus
100%
E. coli
100%
11 µM
S. aureus
100%
wt
ꢀ
tolC
impA
b
c
neg
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
NB
0.17 µM
0.040 µM
27 µM
ꢀ
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
5
ꢀ
ꢀ
21 ꢀM
270 ꢀM
94%
87%
n.t.
103%
n.t.
100%
n.t.
n.t.
n.t.
>256
n.t.
n.t.
>256
n.t.
2
2
0
1
ꢀ
ꢀ
>256
n.t.
ꢀ
ꢀ
n.t.
n.t.
n.t.
28
OEt
OMe
OMe
OH
OH
OH
0
1
2
0
1
2
10 ꢀM
35 ꢀM
9.1 ꢀM
7.7 ꢀM
7.6 ꢀM
8.1 ꢀM
76%
n.t.
101%
n.t.
99%
n.t.
n.t.
n.t.
n.t.
2
3
3
4
9
0
9
0
>256
n.t.
>256
n.t.
>256
n.t.
87%
79%
117%
95%
101%
47%
96%
99%
104%
99%
103%
79%
>256
>256
>256
>256
>256
>256
128
>256
>256
41
a
b
c
Residual activity of the enzyme at 100 ꢀM of the tested compound.
% DMSO was used as a negative control.
1
NB – novobiocin was used as a positive control.
d
n.t. – not tested.
Finally, we replaced the 4,5ꢀdibromoꢀ1Hꢁpyrrole moiety of 3, 23ꢀ25 and 34ꢀ36 by the 4,5ꢀ
dichloroꢀ1Hꢀpyrrole moiety (compounds 22, 31-33, 42-44, Table 4). The chlorine analogs
displayed less potent inhibition of E. coli DNA gyrase than their 4,5ꢀdibromoꢀ1Hꢀpyrrole
counterparts (Table 4), with malonyl acid 43 (IC50 = 0.13 ꢀM) being the most potent. The
ACS Paragon Plus Environment
21

Journal of Medicinal Chemistry
Page 22 of 71
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
inhibition of S. aureus DNA gyrase and E. coli and S. aureus topoisomerase IV by
dichloropyrroles 22, 31-33 and 42-44 followed the trends observed in the other two series, with
weaker activities in the micromolar range (IC values between 10 ꢀM and 320 ꢀM) as compared
5
0
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
to inhibition of E. coli DNA gyrase.
Table 4. Inhibition of Escherichia coli and Staphylococcus aureus DNA gyrase and
topoisomerase IV by the 4,5,6,7ꢀtetrahydrobenzo[1,2ꢀd]thiazoles containing the 4,5ꢀdichloroꢀ1Hꢀ
pyrrole moiety (22, 31-33, 42-44).
DNA gyrase
topoisomerase IV
E. coli MIC
1
2
a
b
b
cpd
R
R
n
IC50 (K
d
) [ꢀM] or RA [%]
S. aureus
IC50 [ꢀM] or RA [%]
[ꢀg/mL]
E. coli
100%
E. coli
100%
11 µM
S. aureus
wt
ꢀ
tolC
impA
c
neg
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
100%
100%
ꢀ
ꢀ
ꢀ
ꢀ
d
NB
0.17 µM
12 ꢀM
0.040 µM
27 µM
ꢀ
2
2
ꢀ
ꢀ
85%
97%
300 ꢀM
97%
98%
290 ꢀM
101%
93%
>256
>256
>256
>256
>256
>256
64
64
31
OEt
OMe
OMe
OH
0
1
2
0
1
0.40 ꢀM
320 ꢀM
63%
256
>256
32
0
.40 ꢀM
3
2
32
a
(
0.050 ꢀM )
33
0.89 ꢀM
86%
100%
82%
>256
>256
>256
>256
>256
>256
4
2
0.59 ꢀM
0.13 ꢀM
300 ꢀM
87%
170 ꢀM
102%
43
OH
98%
a
(
0.049 ꢀM )
ACS Paragon Plus Environment
22

Page 23 of 71
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
44
OH
2
0.30 ꢀM
10 ꢀM
72%
97%
>256
>256
>256
a
b
c
K as determined by surface plasmon resonance (see below).
d
Residual activity of the enzyme at 100 ꢀM of the tested compound.
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
1
% DMSO was used as a negative control.
d
NB – novobiocin was used as a positive control.
Superposition of the crystal structures of DNA gyrase and topoisomerase IV from S. aureus
and E. coli (Figure 6) provides a possible explanation for the observed differences in inhibition
of these enzymes. The hydrophobic pocket of the ATPꢀbinding site, that is predicted to bind the
pyrrolamide moiety of inhibitors, is smaller in S. aureus GyrB and E. coli ParE than in E. coli
GyrB or S. aureus ParE. For example, Val43, Met95 and Val167 in E. coli DNA gyrase are
replaced by the larger Ile51, Leu103 and Ile175 in S. aureus DNA gyrase (Figure 6). As
37
expected from the overlay of crystal structures, according to measurement by POVME (POcket
3
Volume MEasurer) the pocket volume of E. coli GyrB (V = 157 Å ) is larger than that of S.
3
3
3
aureus GyrB (V = 140 Å ), E. coli ParE (V = 149 Å ) and S. aureus ParE (V = 155 Å ). When
3
these pocket volumes are compared with those of the 1Hꢀindoleꢀ2ꢀcarboxamide (V = 143 Å ),
3
4,5ꢀdibromoꢀ1Hꢀpyrroleꢀ2ꢀcarboxamide (V = 135 Å ) and 4,5ꢀdichloroꢀ1Hꢀpyrroleꢀ2ꢀ
3
carboxamide (V = 126 Å ) (calculated by the Molinspiration property calculation service,
www.molinspiration.com), these moieties can be accommodated in the E. coli DNA gyrase ATPꢀ
binding site but are probably too large to fit the smaller hydrophobic pocket of S. aureus DNA
gyrase. A similar calculation can explain the weaker E. coli DNA gyrase inhibition by 20 and 21
than that by 5, since the volumes of 5ꢀfluoroꢀ1Hꢀindoleꢀ2ꢀcarboxamide (in 20) and 5ꢀchloroꢀ1Hꢀ
3
3
indoleꢀ2ꢀcarboxamide (in 21) moieties are 148 Å and 156 Å , being too bulky even for the
larger hydrophobic pocket of E. coli DNA gyrase. An alternative explanation for the lower
ACS Paragon Plus Environment
23

Journal of Medicinal Chemistry
Page 24 of 71
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
activities in the case of S. aureus DNA gyrase could be the loss of a favorable interaction with
Arg144 (Arg136 in E. coli GyrB, Fig. 7f).
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
9
Figure 6. Overlay of crystal structures of GyrB from E. coli (PDB entry: 4DUH, in cyan) and
17
13
S. aureus (PDB entry: 3TTZ, in green), ParE from E. coli (PDB entry: 4HZ0, in magenta)
3
8
and S. aureus (PDB entry: 4URL, in yellow). Pyrrolamideꢀbased GyrB inhibitor from the
crystal structure 3TTZ is represented in grey sticks. For clarity, only less conserved residues
defining the hydrophobic pocket of GyrB and ParE ATPꢀbinding sites are shown, except for Asp,
3
1
which is well conserved among various bacterial strains. Figure was prepared by PyMOL.
E.coli DNA Gyrase ATPase Assay. Compounds 3, 23, 34, 40 and 43 were also tested for
inhibition of E. coli DNA gyrase ATPase activity in a linked assay, which follows the hydrolysis
of ATP via the conversion of NADH to NAD+, to provide further experimental evidence on their
binding to the ATPꢀbinding pocket of the enzyme. Results presented in Table 5 show between
2.1ꢀ and 8.3ꢀfold weaker activities as those determined in the supercoiling assay. However, the
ATPase assay SAR, although with a limited number of compounds tested, follows the SAR
observed in the supercoiling assay, which further suggests the binding of 4,5,6,7ꢀ
ACS Paragon Plus Environment
24

Page 25 of 71
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
tetrahydrobenzo[1,2ꢀd]thiazoles 3, 23, 34, 40 and 43 and their analogs to the ATPꢀbinding site of
DNA gyrase.
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 5. Inhibition of the Escherichia coli DNA gyrase ATPase activity by the 4,5,6,7ꢀ
tetrahydrobenzo[1,2ꢀd]thiazoles 3, 23, 34, 40 and 44.
E. coli DNA gyrase
1
2
cpd
R
ꢀ
R
n
IC50 [ꢀM]
supercoiling assay ATPase assay
a
NB
ꢀ
ꢀ
ꢀ
ꢀ
0.17 µM
12 ꢀM
0.13 µM
25 ꢀM
3
23
Et
0
0
0.10 ꢀM
0.57 ꢀM
0.48 ꢀM
3
4
4
0
OH
0.058 ꢀM
OH
OH
1
1
7.6 ꢀM
22 ꢀM
43
0.13 ꢀM
0.66 ꢀM
a
NB – novobiocin was used as a positive control.
Surface Plasmon Resonance (SPR) Experiments. The ability of inhibitors 3, 24, 25, 32, 35,
6 and 43 to inhibit E. coli DNA gyrase in vitro was further evaluated by surface plasmon
3
resonance. In contrast to the functional DNA gyrase assay that was used for screening and
determining of IC values and comprises GyrA and GyrB subunits of DNA gyrase, the smaller
5
0
ACS Paragon Plus Environment
25

Journal of Medicinal Chemistry
Page 26 of 71
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
2
4 kDa Nꢀterminal part of the GyrB subunit (GyrB24) was used for SPR experiments. The
GyrB24 protein, immobilized on the CM5 chip, was used for testing using novobiocin as a
standard. Inhibitors were tested at eight concentrations in three parallels, depending on 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
response of the immobilized protein. K values were calculated using Origin software with a
d
steady state affinity binding model that assumes one site binding. The results revealed fast and
tight binding of inhibitors 3, 24, 25, 32, 35, 36 and 43 to the GyrB24 protein with K values in
d
agreement with the previously determined IC50 values in the functional DNA gyrase assay.
Measured K values for compounds 3, 24, 25, 32, 35, 36 and 43 were 3.0 µM, 0.072 µM,
d
0.15 µM, 0.050 µM, 0.042 µM, 0.010 µM and 0.049 µM, which are consistent with their IC50
values of 12 µM, 0.096 µM, 0.093 µM, 0.40 µM, 0.069 µM, 0.049 µM and 0.13 µM (Tables 2
and 4), suggesting the binding of these pyrrolamideꢀbased inhibitors to the ATPꢀbinding site of
E. coli DNA gyrase.
Antibacterial Activity. All prepared compounds that inhibited E. coli DNA gyrase were first
screened for their antibacterial activity at 50 ꢀM against two Gram negative (Escherichia coli
ATCC 25922, Pseudomonas aeruginosa ATCC 27853) and two Gram positive (Enterococcus
faecalis ATCC 29212, Staphylococcus aureus ATCC 25923) bacterial strains. Despite potent E.
coli DNA gyrase inhibition, all compounds were found to be inactive against both Gram negative
strains (Supplementary Figure S1). There are several possible explanations for the absence of
antibacterial activity against E. coli and P. aeruginosa. Achieving antibacterial activity against
Gram negative bacteria is challenging, since molecules have to be able to penetrate the cell
envelope and, in addition, to avoid excretion by the efflux pumps. Firstly, the absence of the
antibacterial activity can be attributed to excessive polarity of the compounds, responsible for
poor permeability across the cell wall. Secondly, it has been claimed that subꢀnanomolar activity
ACS Paragon Plus Environment
26

Page 27 of 71
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
19
is required to generate sufficient antibacterial potency on Gram negative strains. To assess
whether our inhibitors are substrates of the efflux pumps and/or have permeability problems, we
evaluated the compounds against permeabilized (impA) and efflux pump knockout (ꢂtolC)
strains of E. coli as well as the wildꢀtype strain (Tables 1ꢀ4). In general, it appears that the E. coli
DNA gyrase inhibitors presented in Tables 1ꢀ4 are substrates of the efflux pump tolC and/or
possess unfavorable physicoꢀchemical properties, since MIC values against ꢂtolC and impA E.
coli strains are lower for some compounds than for the wildꢀtype E. coli strain. In detail, nonꢀ
substituted compounds 3, 19 and 22 (IC values between 12 ꢀM and 47 ꢀM) displayed weak,
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
0
but still stronger antibacterial activity against ꢂtolC and impA E. coli strains with MICs of 64
g/mL than against the wildꢀtype strain. Oxalyl, malonyl and succinyl acids 34-37 and 42-44
ꢀ
were devoid of antibacterial activity against ꢂtolC and impA strains of E. coli, despite their
potent E. coli DNA gyrase inhibition (IC values between 0.049 ꢀM and 0.59 ꢀM), while their
5
0
ester analogs 23-25, 31 and 32 (IC values between 0.093 ꢀM and 0.40 ꢀM) showed stronger
50
antibacterial activities, with MICs between 16 ꢀg/mL and 256 ꢀg/mL. Compounds 24, 25 and 32,
that exhibited MICs against ꢂtolC E. coli strain below 32 ꢀg/mL, were among the most potent E.
coli DNA gyrase inhibitors in the series.
The antibacterial activities of compounds 3-5, 7-12, 17 and 19-47 against Gram positive S.
aureus and E. faecalis were surprisingly weak at 50 ꢀM, with growth inhibition less than 50%
for the majority of compounds (Supplementary Figure S1). The only exceptions were
compounds 24 (84.4% growth inhibition of E. faecalis), 25 and 32 (52.3% and 54.3% growth
inhibition of S. aureus). Confirmatory doseꢀresponse experiments showed that MICs for these
compounds were >125 µM (highest concentration tested). Therefore, their mode of action has
not been determined. In view of their weak antibacterial activity against Gram positive strains
ACS Paragon Plus Environment
27

Journal of Medicinal Chemistry
Page 28 of 71
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
the compounds were also tested against S. aureus DNA gyrase and topoisomerase IV (described
above, Tables 1ꢀ4). Since the compounds inhibited S. aureus enzymes in the micromolar range,
if at all, this probably explains the absence of growth inhibition of Gram positive bacteria by
these compounds. Indeed, in S. aureus strains IC values of less than 50 ꢀM against the S.
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
50
3
9
aureus GyrB are needed to result in MICs below 64 ꢀg/mL.
Molecular Modeling. Pyrrolamideꢀbased DNA gyrase inhibitors, such as natural products
4
0
38
17,18
clorobiocin and kibdelomycin and synthetic compounds
(Figure 2a) have been shown by
Xꢀray crystallography to bind to the ATPꢀbinding sites of DNA gyrase and topoisomerase IV.
Since our oroidinꢀderived DNA gyrase inhibitors also contain the pyrroleꢀ2ꢀcarboxamide moiety
and showed binding affinity for GyrB subunit of E. coli DNA gyrase in the SPR experiment, one
could anticipate that they bind to the ATPꢀbinding site of the enzyme. Therefore, the binding of
designed DNA gyrase inhibitors was studied by docking in the ATPꢀbinding site of E. coli GyrB
4
1,42
43
(PDB entry: 4DUH), using FlexX
as available in LeadIT (BioSolveIT GmbH). The docking
protocol was first validated by redocking of the inhibitor from the crystal structure in the defined
binding site. The use of FlexX reproduced the experimentally determined inhibitor conformation
with an all heavy atom rootꢀmean square distance (RMSD) of 1.2 Å, which makes this software
suitable for predicting binding modes of the novel DNA gyrase inhibitors and rationalization of
the observed SAR.
Docking experiments predicted similar binding modes of 4,5ꢀdibromoꢀ1Hꢀpyrroles, 1Hꢀ
indoles and 4,5ꢀdichloroꢀ1Hꢀpyrroles. A hydrogen bonding network was in all cases formed
between the pyrroleꢀ2ꢀcarboxamide or indoleꢀ2ꢀcarboxamide moiety of inhibitors, Asp73 side
chain and the highly conserved structural water, while in some cases additional interactions were
formed with Arg76 and/or Arg136 (Figure 7). For example, the potent E. coli DNA gyrase
ACS Paragon Plus Environment
28

Page 29 of 71
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
inhibitor 23 (IC50 = 0.10 ꢀM) and its carboxylic acid derivative 34 (IC50 = 0.058 ꢀM) are
predicted to form hydrogen bonds with Arg76 and Arg136 (Figure 7a and 7b). Both the carbonyl
of the aminocarbonyl group at position 2 and the ester group of 23 appear to be in contact with
the protein. This observation was confirmed with Nꢁacetyl derivative 45 (IC = 0.15 ꢀM), which
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
50
showed more potent E. coli DNA gyrase inhibition than its parent compound 3, but weaker than
the oxalyl derivative 23. Formation of a hydrogen bond between the acetamide carbonyl group
and the protein was also suggested by docking of 45 to E. coli GyrB (Figure 7c).
a
b
c
d
e
f
Figure 7. Calculated binding modes of inhibitors a) ester 23 (in magenta sticks), b)
corresponding carboxylic acid 34 (in yellow sticks), c) Nꢁacetyl derivative 45 (in orange sticks)
and d) malonic acid derivative 35 (in cyan sticks) in the ATPꢀbinding site of E. coli DNA gyrase
(PDB entry: 4DUH, in green cartoon); e) Overlay of the calculated binding modes of inhibitors
35 (in cyan lines) and 40 (in green sticks); f) Overlay of calculated binding modes of 35 in the
ATPꢀbinding site of E. coli (residues in grey sticks) and S. aureus (PDB entry: 3TTZ, residues in
3
1
blue sticks) DNA gyrase. Figures were prepared by PyMOL.
ACS Paragon Plus Environment
29