An optimised series of substituted N-phenylpyrrolamides as DNA gyrase B inhibitors

Article abstract of DOI:10.1016/j.ejmech.2019.02.004

ATP competitive inhibitors of DNA gyrase and topoisomerase IV have great therapeutic potential, but none of the described synthetic compounds has so far reached the market. To optimise the activities and physicochemical properties of our previously reported N-phenylpyrrolamide inhibitors, we have synthesized an improved, chemically variegated selection of compounds and evaluated them against DNA gyrase and topoisomerase IV enzymes, and against selected Gram-positive and Gram-negative bacteria. The most potent compound displayed IC₅₀ values of 6.9 nM against Escherichia coli DNA gyrase and 960 nM against Staphylococcus aureus topoisomerase IV. Several compounds displayed minimum inhibitory concentrations (MICs) against Gram-positive strains in the 1–50 μM range, one of which inhibited the growth of Enterococcus faecalis, Enterococcus faecium, S. aureus and Streptococcus pyogenes with MIC values of 1.56 μM, 1.56 μM, 0.78 μM and 0.72 μM, respectively. This compound has been investigated further on methicillin-resistant S. aureus (MRSA) and on ciprofloxacin non-susceptible and extremely drug resistant strain of S. aureus (MRSA VISA). It exhibited the MIC value of 2.5 μM on both strains, and MIC value of 32 μM against MRSA in the presence of inactivated human blood serum. Further studies are needed to confirm its mode of action.

Full text of DOI:10.1016/j.ejmech.2019.02.004

Accepted Manuscript
An optimised series of substituted N-phenylpyrrolamides as DNA gyrase B inhibitors
Davide Benedetto Tiz, Žiga Skok, Martina Durcik, Tihomir Tomašič, Lucija Peterlin
Mašič, Janez Ilaš, Anamarija Zega, Gábor Draskovits, Tamás Révész, Ákos Nyerges,
Csaba Pál, Cristina D. Cruz, Päivi Tammela, Dušan Žigon, Danijel Kikelj, Nace Zidar
PII:
S0223-5234(19)30114-X
DOI:
https://doi.org/10.1016/j.ejmech.2019.02.004
EJMECH 11096
Reference:
To appear in: European Journal of Medicinal Chemistry
Received Date: 14 November 2018
Revised Date: 18 January 2019
Accepted Date: 1 February 2019
Please cite this article as: D.B. Tiz, Ž. Skok, M. Durcik, T. Tomašič, L.P. Mašič, J. Ilaš, A. Zega, Gá.
Draskovits, Tamá. Révész, Á. Nyerges, C. Pál, C.D. Cruz, Pä. Tammela, Duš. Žigon, D. Kikelj, N. Zidar,
An optimised series of substituted N-phenylpyrrolamides as DNA gyrase B inhibitors, European Journal
of Medicinal Chemistry (2019), doi: https://doi.org/10.1016/j.ejmech.2019.02.004.
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT
GRAPHICAL ABSTRACT:
H
O
O
O
N
OH
O
Y =
N
N
N
S
N
N
H
H
N
NH
O
O
Y
O
OH
H
O
N
N
H
H
N
H
O
N
N
O
R
H
X
O
Cl
Cl
NH2+Cl-
X =
9
d
NH
NH
E. coli gyrase: IC50 = 6.9 nM

ACCEPTED MANUSCRIPT
An optimised series of substituted N-
phenylpyrrolamides as DNA gyrase B inhibitors
a
a
a
a
Davide Benedetto Tiz , Žiga Skok , Martina Durcik , Tihomir Tomašič , Lucija Peterlin Mašič
a
a
a
b
b
b
,
,
Janez Ilaš , Anamarija Zega , Gábor Draskovits , Tamás Révész , Ákos Nyerges , Csaba Pál
c
c
c
d
a
a, *
Cristina D. Cruz , Päivi Tammela , Dušan Žigon , Danijel Kikelj , Nace Zidar
a
b
Faculty of Pharmacy, University of Ljubljana, Aškerčeva cesta 7, 1000 Ljubljana, Slovenia
Synthetic and Systems Biology Unit, Institute of Biochemistry, Biological Research Centre of the
Hungarian Academy of Sciences, Szeged H-6726, Hungary
c
Drug Research Program, Division of Pharmaceutical Biosciences, Faculty of Pharmacy,
University of Helsinki, P.O. Box 56 (Viikinkaari 5 E), Helsinki FI-00014, Finland
d
Jožef Stefan Institute, Department of Environmental Science, Jamova cesta 39, 1000 Ljubljana,
Slovenia
ABSTRACT
ATP competitive inhibitors of DNA gyrase and topoisomerase IV have great therapeutic
potential, but none of the described synthetic compounds has so far reached the market. To
optimise the activities and physicochemical properties of our previously reported N-
phenylpyrrolamide inhibitors, we have synthesised an improved, chemically variegated selection
of compounds and evaluated them against DNA gyrase and topoisomerase IV enzymes, and
against selected Gram-positive and Gram-negative bacteria. The most potent compound
displayed IC50 values of 6.9 nM against Escherichia coli DNA gyrase and 960 nM against
Staphylococcus aureus topoisomerase IV. Several compounds displayed minimum inhibitory
concentrations (MICs) against Gram-positive strains in the 1-50 µM range, one of which
inhibited the growth of Enterococcus faecalis, Enterococcus faecium, S. aureus and
Streptococcus pyogenes with MIC values of 1.56 µM, 1.56 µM, 0.78 µM and 0.72 µM,
respectively. This compound has been investigated further on methicillin-resistant S. aureus
(MRSA) and on ciprofloxacin non-susceptible and extremely drug resistant strain of S. aureus
1

ACCEPTED MANUSCRIPT
(MRSA VISA). It exhibited the MIC value of 2.5 µM on both strains, and MIC value of 32 µM
against MRSA in the presence of inactivated human blood serum. Further studies are needed to
confirm its mode of action.
KEYWORDS: antibacterial; ATP competitive; DNA gyrase; GyrB; inhibitor; N-
phenylpyrrolamide; ParE; topoisomerase IV
ABBREVIATIONS
ACN, acetonitrile; ATCC, American type culture collection; ATR, attenuated total reflectance;
CFU, colony-forming unit; CLSI, Clinical and Laboratory Standards Institute; DCM,
dichloromethane; DMF, dimethylformamide; DIAD, diisopropylazodicarboxylate; GyrA, DNA
gyrase A; GyrB, DNA gyrase B; HepG2, human hepatocellular carcinoma; HPLC, high pressure
liquid chromatography; HUVEC, human umbilical vein endothelial cells; MIC, minimum
inhibitory concentration; MRSA, methicillin-resistant Staphylococcus aureus; MTS, 3-(4,5-
dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium; NMM, N-
methylmorpholine; PAβN, phenylalanine-arginine β-naphthylamide; ParC, topoisomerase IV
subunit A; ParE, topoisomerase IV subunit B; PC, principal component; RA, residual activity;
TBTU,
N,N,N′,N′-tetramethyl-O-(benzotriazol-1-yl)uroniumtetrafluoroborate;
THF,
tetrahydrofuran; topo IV, topoisomerase IV; VISA, vancomycin-intermediate Staphylococcus
aureus.
INTRODUCTION
Antibiotic-resistant infections kill an estimated 700,000 people each year [1]. Recently, the
World Health Organization (WHO) has issued a list of drug-resistant bacteria that pose the
greatest threat to human health and thus present a big challenge for antibacterial drug
development. Between 2010 and 2015, the US Food and Drug Administration (FDA) approved
only eight new antibacterials, and seven of them showed similar mechanism of action to the
previously approved antibiotics [2]. Thus, new antibacterial agents with novel mechanisms of
action are needed.
2

ACCEPTED MANUSCRIPT
Bacterial topoisomerases, DNA gyrase and topoisomerase IV (topo IV), are enzymes that
catalyse changes in DNA topology [3]. There are two types of topoisomerases: type I that
catalyse single-strand breaks, and type II that are responsible for the transient break of two
strands of DNA [4]. DNA gyrase and topoisomerase IV belong to type II topoisomerases and
display high structural and functional similarities. They are both heterotetrameric enzymes
composed of two pairs of subunits. DNA gyrase is composed of two GyrA and two GyrB
subunits, while topoisomerase IV is composed of two ParC and two ParE subunits. GyrA and
GyrB subunits of DNA gyrase are similar in amino acid sequence to ParC and ParE subunits of
topoisomerase IV, respectively [5]. The main function of GyrA/ParC is cleavage and reunion of
the DNA molecule, whereas GyrB/ParE binds ATP, providing through its hydrolysis, the energy
for the ligation process. DNA gyrase introduces negative supercoils in DNA during replication,
whereas topoisomerase IV is involved in the decatenation of the two daughter chromosomes after
replication [6].
Being absent in eukaryotes and at the same time an essential enzyme for bacterial survival, DNA
gyrase has become an attractive target for the development of antibacterials. Drugs targeting
DNA gyrase exert their antibacterial activity via two main mechanisms. The first is stabilisation
of the covalent enzyme–DNA complex, i.e. gyrase poisoning. This mechanism is typical of the
quinolone class of antibacterials, with ciprofloxacin as the most distinctive representative. The
second mechanism involves blocking the ATP binding site of DNA gyrase B. The
aminocoumarin natural products novobiocin and clorobiocin are examples of catalytic inhibitors
that target the ATP binding site on the GyrB/ParE subunit [4].
Fluoroquinolones have been used extensively in the clinic, resulting in the development of many
fluoroquinolone-resistant bacterial strains [7]. The growing resistance problem together with
some serious side effects has stimulated the search for new inhibitors which target ATP binding
site on DNA gyrase and topoisomerase IV. The only ATPase inhibitor that has been used in the
clinic is novobiocin, which was indroduced in the 1960s but was later withdrawn from the market
due to its toxicity and weak potency [6, 8]. More recently, many structurally diverse classes of
GyrB inhibitors have been discovered, such as pyrrolamides [9], benzothiazoles [10], ethyl ureas
[11], N-phenylpyrrolamides [12-15] and pyrimidoindoles [16] (Figure 1), which possess potent
on-target and antibacterial activities but have some liabilities, such as unfavourable
physicochemical properties or toxicity issues.
3

ACCEPTED MANUSCRIPT
Figure 1. Representative ATP competitive DNA gyrase and topoisomerase IV inhibitors, and
their IC and MIC values.
5
0
RESULTS AND DISCUSSION
Design. Using our previous N-phenylpyrrolamide inhibitors as starting points [12-15], and having
at disposal the crystal structure of E. coli GyrB in complex with one of our inhibitors (PDB code:
4
ZVI) [12], optimised N-phenylpyrrolamides were prepared with improved on-target and
antibacterial activities. The three structural types of compounds (types I-III, Figure 2) possess
modifications on the central part and right-hand sides of the molecules. In all three types of
compounds, a hydrogen bond donor-acceptor structural feature was maintained on the left-hand
side of the molecules. This part of the inhibitors is fundamental for establishing hydrogen
bonding with Asp73 (E. coli numbering) and a structurally conserved water molecule (Figure 2).
Additionally, the substituted pyrrole moiety makes important hydrophobic contacts with the
hydrophobic pocket composed of Val71, Val43, Val167, and Ala47, which is also the binding
site for the clorobiocin 5-methylpyrrole moiety [6]. Two differently substituted pyrrole-2-
carboxylic acids were used for preparing the inhibitors: 4,5-dibromopyrrole-2-carboxylic acid
4

ACCEPTED MANUSCRIPT
and 3,4-dichloro-5-methylpyrrole-2-carboxylic acid. The latter was selected preferentially
because the chlorine atoms are slightly smaller than the bromine atoms and thus fit better into the
adenine binding pocket not only of E. coli DNA gyrase, but also into the slightly smaller binding
pockets of S. aureus DNA gyrase and topoisomerase IV [17].
On the right-hand side of the molecules we introduced acidic groups (e.g. carboxylic acid-
containing groups, such as glycine and (S)- or (R)-alanine) capable of forming interactions with
the positively charged Arg136 residue, or moieties (e.g. 1,3,4-oxadiazol-2-thione and 5-
oxotetrazole) capable of forming cation-π interactions with the Glu50-Arg76 salt bridge and
hydrogen bonds with Arg136 side chain. The 1,3,4-oxadiazol-2-thione and 5-oxotetrazole rings
were introduced as bioisosteres of carboxylic acid group. The lower acidity of these moieties
compared to carboxylic acid was expected to improve entry into bacteria and, consequently,
result in better antibacterial activity [18].
Type I compounds contain an alkoxy substituent at the ortho position to the substituted amino
group of the central benzene ring (substituent X, Figure 1). The size of this substituent was varied
from the bulkier piperidin-4-yloxy, through pyrrolidin-3-yloxy to the smaller isopropoxy group.
An amino group-containing heterocycle was selected as substituent X for three reasons: i) the
possibility of forming hydrophobic contacts with the lipophilic floor of the enzyme (amino acid
residues Ile94, Ile78) [14], ii) the possibility of forming additional hydrogen bonds through the
ring NH group, and iii) the presence of a positive charge that has recently been associated with an
increased level of accumulation in E. coli [19]. It was previously shown by docking studies that
the nitrogen atom could form an H-bond with amino acid residue Ala100 that is a part of the
flexible loop between amino acid residues Gly97-Ser108 [15]. Moreover, because of the ionic
character of the amino group, the water solubility of these compounds is likely to improve. In the
case of isopropoxy substituent, only the interactions with the lipophilic floor are possible, but the
possibility of establishing H-bond with Ala100 is lost.
Type II compounds, unlike type I, do not contain substituents at the ortho position to the amino
group of the central benzene ring. The focus in this series was given on the right part of the
molecules, to which chemically diverse groups were introduced, such as sulphonamides, acyl
sulfonamides, aliphatic nitriles, malonates or a 5-oxotetrazole ring (Figure 2). Sulfonamides, acyl
sulphonamides and 5-oxotetrazole are well-known examples of carboxylic acid isosteres, and
have been of fundamental importance in the discovery of novel therapeutics [20] [21]. Nitriles are
5

ACCEPTED MANUSCRIPT
attractive moieties in medicinal chemistry because of their strong dipole moment and small size,
that facilitate polar interactions in which the nitrile group acts as a hydroxyl or carboxyl group
isostere [22]. Malonic ester derivatives were designed in order to investigate how two
carboxylate groups affect the interaction with the Arg136 residue.
In type III compounds, the central phenyl ring was replaced by a bicyclic scaffold to afford a
quinolone-like structure or a substituted indole group (Figure 2). To this heterocycle, a carboxylic
group was attached to establish ionic bonds with Arg136. With the quinolone derivative, our goal
was to prepare a molecular hybrid in which the left-hand side of the molecule is represented by
the N-phenylpyrrolamide DNA gyrase B-binding motif, and the right-hand side is represented by
the 4-oxo-quinoline-2-carboxlic acid moiety, a known DNA gyrase A binder (Figure 3).
Targeting both active sites with the same molecule could result in synergistic activity and in a
reduced probability of development of bacterial resistance [6].
6

ACCEPTED MANUSCRIPT
Figure 2. Design of three structural types (I-III) of N-phenylpyrrolamides as DNA gyrase B
inhibitors, based on the co-crystal structure of the inhibitor A – GyrB (E. coli) (PDB code: 4ZVI
[12]).
7

ACCEPTED MANUSCRIPT
prototype of GyrB inhibitor
nalidixic acid
O
GyrA inhibitor
O
O
OH
O
N
H
H
OH
N
O
N
H
N
N
Et
Cl
Cl
O
O
O
OH
H
N
N
N
H
Et
Cl
Cl
hybrid inhibitor
Figure 3. Design of a molecular hybrid inhibitor by merging the left side portion of N-
phenylpyrrolamide as a GyrB binder and nalidixic acid as a GyrA binder.
Chemistry. The synthesis of type I compounds (8a-d, 9a-d and 10a-b) is presented in Scheme
1
. In the first step, 3-hydroxy-4-nitrobenzoic acid (1) was reacted with thionyl chloride and
methanol to afford the corresponding methyl ester 2. The hydroxyl group of 2 was alkylated with
N-Boc-4-hydroxypiperidine under Mitsunobu conditions to afford the intermediate 3. The
alkaline hydrolysis of ester 3 gave the carboxylic acid 4 that was coupled with glycine methyl
ester, or (S)- or (R)-alanine methyl ester using N,N,N',N'-tetramethyl-O-(benzotriazol-1-
yl)uroniumtetrafluoroborate (TBTU) promoted coupling to yield compounds 5a-c. After catalytic
hydrogenation, the resulting anilines 6a-c were coupled with 4,5-dibromopyrrole-2-carboxylic
acid or 3,4-dichloro-5-methylpyrrole-2-carboxylic acid. The obtained esters 7a-d underwent
alkaline hydrolysis to afford the N-Boc protected compounds 8a-d that were, in turn, reacted with
1
M HCl in acetic acid or with 4 M HCl in 1,4-dioxane to give the target compounds 9a-d.
Additionally, esters 7c-d underwent the removal of Boc protecting group yielding the products
0a-b.
1
8

ACCEPTED MANUSCRIPT
COOH
COOCH3
COOCH3
NBoc
COOH
a
b
c
O2N
O2N
O2N
O2N
OH
OH
O
O
NBoc
1
2
3
4
O
R1
O
R1
O
R1
OCH3
OCH3
OCH3
N
N
O
N
*
*
*
H
H
H
H
d
e
f
O
O
N
O
O2N
H2N
N
R2
H
O
O
O
R4
R3
NBoc
a (R1 = H)
NBoc
NBoc
4
5
6a (R1 = H)
7a (R1 = H, R2 = R3 = Br, R = H)
1
7b (R = H, R2 = CH , R = R = Cl)
1
3
4
5
5
b (R1 = CH , S isomer)
c (R1 = CH , R isomer)
6b (R = CH , S isomer)
3
3
3
1
7c (R = CH , R = CH , R = R4 = Cl, S isomer)
d (R1 = CH , R = CH , R = R4 = Cl, R isomer)
1
2
3
6c (R = CH , R isomer)
3
3
3
3
2
3
7
3
3
O
R1
O
R1
OH
OH
O
N
O
N
*
*
H
H
H
H
g
h
N
O
N
O
N
N
R2
R2
H
H
O
O
R4
R4
R3
R3
+
NH2 Cl-
NBoc
4
8
8
8
8
a (R1 = H, R2 = R3 = Br, R4 = H)
9a (R1 = H, R2 = R3 = Br, R = H)
b (R1 = H, R2 = CH , R = R4 = Cl)
3
9b (R = H, R2 = CH , R = R = Cl)
1
3
4
3
3
c (R1 = CH , R2 = CH , R3 = R4 = Cl, S isomer)
9c (R = CH , R = CH , R = R4 = Cl, S isomer)
1
2
3
3
3
3 3
d (R1 = CH , R2 = CH , R3 = R4 = Cl, R isomer)
9d (R = CH , R = CH , R = R4 = Cl, R isomer)
1
2
3
3
3
3 3
O
R1
OCH3
O
N
*
H
N
H
i
O
7c-d
N
R2
H
O
R4
R3
+
-
NH2 Cl
1
1
0a (R1 = CH , R2 = CH , R3 = R4 = Cl, S isomer)
3 3
0b (R1 = CH , R2 = CH , R3 = R4 = Cl, R isomer)
3
3
Scheme 1. Reagents and conditions: (a) thionyl chloride, MeOH, 0 °C ‰ rt, 15 h; (b) PPh3,
DIAD, N-Boc-4-hydroxypiperidine, anhydrous THF, rt, 15 h; (c) 1 M NaOH, MeOH, rt, 15 h;
(d) glycine methyl ester hydrochloride (for the synthesis of 5a) or (S)- or (R)-alanine methyl
ester hydrochloride (for the synthesis of 5b-c), TBTU, NMM, anhydrous DMF, 60 °C, 12 h; (e)
H , Pd-C, MeOH, rt, 5 h; (f) i) 4,5-dibromopyrrole-2-carboxylic acid or 3,4-dichloro-5-
2
methylpyrrole-2-carboxylic acid, oxalyl chloride, anhydrous CH Cl , rt, 15 h, then ii) 6a-c,
2
2
anhydrous pyridine, anhydrous CH Cl , rt, 15 h; (g) 1 M NaOH, MeOH/THF, rt, 15 h (for the
2
2
synthesis of 8a-b) or 1 M LiOH, MeOH/THF, rt, 15 h (for the synthesis of 8c-d); (h) 1 M HCl
in acetic acid, DMF, rt, 8 h (for the synthesis of 9a-b) or 4 M HCl in 1,4-dioxane, rt, 4 h (for the
synthesis of 9c-d); (i) 4 M HCl in 1,4-dioxane, rt, 4 h.
9

ACCEPTED MANUSCRIPT
The synthesis of compounds 16, 17 and 18, containing a pyrrolidine moiety, is shown in Scheme
2
. The pathway is similar to the one described for the preparation of compounds 8a-d, 9a-d and
0a-b (Scheme 1), except that instead of N-Boc-4-hydroxypiperidine, (R)-N-Boc-3-pyrrolidinol
1
was used as the alcohol moiety for the Mitsunobu reaction in the synthesis of compound 11.
3
Scheme 2. Reagents and conditions: (a) PPh , DIAD, (R)-N-Boc-3-pyrrolidinol, anhydrous THF,
rt, 15 h; (b) 1 M NaOH, MeOH, rt, 15 h; (c) glycine methyl ester hydrochloride, TBTU, NMM,
anhydrous DMF, 60 °C, 12 h; (d) H , Pd-C, MeOH, rt, 5 h; (e) i) 3,4-dichloro-5-methylpyrrole-2-
2
carboxylic acid, oxalyl chloride, anhydrous CH
2 2
Cl , rt, 15 h, then ii) 14, anhydrous pyridine,
anhydrous CH Cl , rt, 15 h; (f) 1 M NaOH, MeOH/THF, rt, 15 h (g) 1M HCl in acetic acid,
2
2
DMF, rt, 8 h; (h) 4 M HCl in 1,4-dioxane, rt, 4 h.
The synthesis of compound 25 is shown in Scheme 3. The isopropoxy derivative 19, obtained
under Mitsunobu conditions from compound 2, underwent alkaline hydrolysis to afford the acid
1
0

ACCEPTED MANUSCRIPT
2
0. The coupling of 20 with (R)-alanine methyl ester hydrochloride, followed by catalytic
hydrogenation, gave the aniline 22 which was reacted with 3,4-dichloro-5-methylpyrrole-2-
carbonyl chloride to afford the ester 23. The hydrazinolysis of 23 yielded 24 which was reacted
with 1,1′-thiocarbonyldiimidazole to give the final product 25.
3
Scheme 3. Reagents and conditions: (a) PPh , DIAD, isopropanol, anhydrous THF, rt, 15 h; (b) 1
M NaOH, MeOH, rt, 15 h; (c) (R)-alanine methyl ester hydrochloride, TBTU, NMM, anhydrous
DMF, 60 °C, 12 h; (d) H , Pd-C, MeOH, rt, 5 h; (e) i) 3,4-dichloro-5-methylpyrrole-2-carboxylic
2
2 2
acid, oxalyl chloride, anhydrous CH Cl , rt, 15 h, then ii) 22, anhydrous pyridine, anhydrous
CH Cl , rt, 15 h; (f) hydrazine hydrate, dry MeOH/dry THF, reflux, 15 h; (g) 1,1′-
2
2
thiocarbonyldiimidazole, 1,4-dioxane/DMF, 100 °C, 15 h.
The synthesis of type II compounds 27, 29, 34 and 39 is presented in Scheme 4. Compounds 27
and 29 were obtained by coupling 3,4-dichloro-5-methyl-1H-pyrrole-2-carbonyl chloride, with
sulfanilamide (26) and sulfathiazole (28), respectively. For the preparation of compound 34, 4-
nitrobenzenesulfonyl chloride (30) was mixed with glycine methyl ester hydrochloride in the
presence of triethylamine to give 31, which underwent catalytic reduction to yield 32. The final
product 34 was obtained after alkaline hydrolysis of the ester 33, obtained by coupling of 32 and
3
,4-dichloro-5-methyl-1H-pyrrole-2-carbonyl chloride.
1
1

ACCEPTED MANUSCRIPT
For the synthesis of compound 39, methanesulfonyl chloride (35) was reacted with gaseous
ammonia to obtain methanesulfonamide (36). Compound 36 was then mixed with 4-nitrobenzoyl
chloride in the presence of triethylamine to obtain 37 which, after catalytic hydrogenation,
generated compound 38 that was coupled with 3,4-dichloro-5-methyl-1H-pyrrole-2-carbonyl
chloride to yield the acyl sulfonamide 39.
Scheme 4. Reagents and conditions: (a) i) 3,4-dichloro-5-methylpyrrole-2-carboxylic acid, oxalyl
chloride, anhydrous CH Cl , rt, 15 h, then ii) 26, anhydrous pyridine, anhydrous THF, rt, 15 h for
2
2
the synthesis of 27 or ii) 28, anhydrous pyridine, anhydrous THF, rt, 15 h for the synthesis of 29;
b) glycine methyl ester hydrochloride, triethylamine, CH Cl , 0 °C ‰ rt, 12 h; (c) H , Pd-C,
MeOH, rt, 3 h; (d) i) 3,4-dichloro-5-methylpyrrole-2-carboxylic acid, oxalyl chloride, anhydrous
CH Cl , rt, 15 h, then ii) 32, anhydrous pyridine, anhydrous THF, rt, 15 h; (e) 1 M NaOH,
(
2
2
2
2
2
MeOH/THF, rt, 15 h; (f) gaseous ammonia, anhydrous THF, 0 °C, 0.5 h; (g) 4-nitrobenzoyl
chloride, triethylamine, anhydrous CH Cl , 0 °C ‰ rt, 12 h; (h) i) 3,4-dichloro-5-methylpyrrole-
2
2
2
2 2
-carboxylic acid, oxalyl chloride, anhydrous CH Cl , rt, 15 h, then ii) 38, anhydrous pyridine,
anhydrous THF, rt, 15 h.
1
2

ACCEPTED MANUSCRIPT
The synthesis of compounds 43a-b, 46a-b and 50 is shown in Scheme 5. For the preparation of
the aliphatic nitriles 43a-b, aminoacetonitrile hydrochloride was firstly coupled with 4-nitro
benzoic acid (40) using TBTU methodology. The nitro group of the obtained intermediate 41 was
then reduced to give compound 42 which was reacted with 4,5-dibromo-1H-pyrrole-2-carbonyl
chloride or 3,4-dichloro-5-methyl-1H-pyrrole-2-carbonyl chloride to afford the final products
4
3a-b. The malonates 46a-b were prepared by reacting 40 with diethyl aminomalonate
hydrochloride via TBTU methodology. Compound 44 underwent catalytic hydrogenation to give
the substituted aniline 45 which was reacted with 4,5-dibromo-1H-pyrrole-2-carbonyl chloride or
3
,4-dichloro-5-methyl-1H-pyrrole-2-carbonyl chloride to afford the final compounds 46a or 46b.
The hydrolysis of ethyl ester groups of 46a-b was unsuccessful as decarboxylation was observed.
The 5-oxotetrazole derivative 50 was prepared by mixing 4-nitrobenzoyl chloride (47) with
azidotrimethylsilane to give compound 48 via initial acylazide formation and subsequent Curtius
rearrangement [23]. The following catalytic hydrogenation of 48 and the coupling of the amine
4
9 with 3,4-dichloro-5-methyl-1H-pyrrole-2-carbonyl chloride yielded the final compound 50.
1
3

ACCEPTED MANUSCRIPT
Scheme 5. Reagents and conditions: (a) aminoacetonitrile hydrochloride, TBTU, NMM,
anhydrous DMF, 60 °C, 12 h; (b) H , Pd-C, MeOH, rt, 3 h; (c) i) 4,5-dibromopyrrole-2-
2
carboxylic acid or 3,4-dichloro-5-methylpyrrole-2-carboxylic acid, oxalyl chloride, anhydrous
CH Cl , rt, 15 h, then ii) 42, anhydrous pyridine, anhydrous CH Cl , rt, 15 h (for the synthesis of
2
2
2
2
4
3a-b); (d) diethyl aminomalonate hydrochloride, TBTU, NMM, anhydrous DMF, 60 °C, 12 h;
(
e) H , Pd-C, EtOH, rt, 3 h; (f) i) 4,5-dibromopyrrole-2-carboxylic acid or 3,4-dichloro-5-
2
methylpyrrole-2-carboxylic acid, oxalyl chloride, anhydrous CH Cl , rt, 15 h, then ii) 45,
2
2
2 2
anhydrous pyridine, anhydrous CH Cl , rt, 15 h (for the synthesis of 46a-b); (g)
azidotrimethylsilane, 90 °C, 12h; (h) 3,4-dichloro-5-methylpyrrole-2-carboxylic acid, oxalyl
chloride, anhydrous CH Cl , rt, 15 h, then ii) 49, anhydrous pyridine, anhydrous CH Cl , rt, 15 h.
2
2
2
2
The synthesis of type III compounds 57 and 61 is presented in Scheme 6. For the synthesis of
compound 57 we employed the Gould-Jacobs route [24]. 3-Nitroaniline (51) was first mixed with
diethyl ethoxymethylenemalonate to give the derivative 52. This was then cyclized in diphenyl
ether as a solvent to afford compound 53, which was subsequently alkylated with ethyl bromide
to form 54. The nitro group of 54 was reduced with catalytic hydrogenation and the resulting
aniline 55 was coupled with 3,4-dichloro-5-methyl-1H-pyrrole-2-carbonyl chloride to give the
ester 56. Compound 56 underwent alkaline hydrolysis to yield the final quinolone 57. For the
synthesis of the indole derivative 61, compound 58 was first reduced to the corresponding amine
5
9, which was then reacted with 3,4-dichloro-5-methyl-1H-pyrrole-2-carbonyl chloride to give
the ester 60. Finally, the alkaline hydrolysis of ethyl ester group of 60 afforded the final product
1.
6
1
4

ACCEPTED MANUSCRIPT
Scheme 6. Reagents and conditions: (a) diethyl ethoxymethylenemalonate, 90 °C, 3 h; (b)
diphenyl ether, 260 °C, 2.5 h; (c) ethyl bromide, K
Pd-C, EtOH, rt, 3 h (e) i) 3,4-dichloro-5-methylpyrrole-2-carboxylic acid, oxalyl chloride,
anhydrous CH Cl , rt, 15 h, then ii) 55, anhydrous pyridine, anhydrous CH Cl , rt, 15 h; (f) 1 M
2 3 2
CO , anhydrous DMF, 70 °C, 12 h; (d) H ,
2
2
2
2
NaOH, EtOH/1,4-dioxane, rt, 5 h; (g) i) 3,4-dichloro-5-methylpyrrole-2-carboxylic acid, oxalyl
chloride, anhydrous CH Cl , rt, 15 h, then ii) 59, anhydrous pyridine, anhydrous CH Cl , rt, 15 h.
2
2
2
2
Inhibitory Activities Against DNA Gyrase and Topoisomerase IV. Thirty-five new
compounds were synthesized and evaluated for their inhibitory activity against E. coli DNA
gyrase in a supercoiling assay. The results are presented either as residual activities (RAs) of the
enzyme at 1 µM of compounds or as IC values (Tables 1-3). The most potent compounds were
5
0
additionally evaluated against S. aureus DNA gyrase, and against E. coli and S. aureus
topoisomerase IV (Table 4). To determine possible binding modes of compounds, molecular
docking of all tested compounds to the E. coli GyrB binding site was performed using GOLD
software [25] and the results are presented in Supporting information.
Thirteen compounds (8b, 8c, 9a-d, 10a-b, 16, 17, 18, 50 and 61) showed potent inhibition of E.
coli DNA gyrase with low nanomolar inhibitory values (IC50 < 100 nM). Twelve of these
1
5

ACCEPTED MANUSCRIPT
inhibitors contained the 3,4-dichloro-5-methyl-1H-pyrrole moiety in the left-hand side of the
molecule, which thus proved to be more optimal than 4,5-dibromo-1H-pyrrole group. In general,
carboxylic acid derivatives possessed stronger activity than the corresponding methyl ester
derivatives. As an example of this trend, the carboxylic acid 8b exhibited an IC value of 86 nM
5
0
whereas the ester 7b was inactive (RA = 100%). The reason for this difference stems from the
ability of carboxylic acids (and not of esters) to form salt bridge with Arg136 residue (Figure 1S
of Supporting information).
Type I inhibitors with carboxylic groups on the right-hand side of the molecule and amino group-
containing heterocycles attached at the ortho position to the amino group on benzene ring
(substituent X, Figure 1) (9a, 9b, 9c, 9d and 17) proved to be the most potent inhibitors of DNA
gyrase within the series (IC ≤ 31 nM). Piperidin-4-yloxy and pyrrolidin-3-yloxy substituents are
5
0
most probably forming favourable interactions with the hydrophobic floor of the enzyme.
Moreover, the amino group of these heterocycles can potentially form a hydrogen bond with
Ala100 that is part of the flexible loop between amino acid residues Gly97-Ser108, as predicted
by docking of compound 9d (IC = 6.9 nM) to the DNA gyrase B active site (Figure 1S), and as
5
0
reported by our research group for a related compound [15]. Interestingly, also some derivatives
containing a combination of N-Boc-protected amino group and free carboxylic acid group on the
right-hand side, e.g. compounds 8b, 8c, 8d and 16, showed good activity against E. coli gyrase
(
IC ≤ 100 nM). Also, the positively charged ester derivatives 10a, 10b and 18 showed good
50
activities on the enzyme with IC values of 40 nM, 48 nM and 74 nM, respectively. On the other
5
0
hand, compounds with simultaneous presence of the methyl ester group on the right-hand side
and N-Boc-protected heterocycle (7c, 7d and 15) were inactive (RA = 83%, 94% and 93%
respectively). The activities of pyrrolidine derivatives were comparable to the activities of their
piperidine-containing counterparts, as can be seen by comparing, e.g. pyrrolidine 17 (IC = 31
5
0
nM) to piperidine 9b (IC = 17 nM) and pyrrolidine 16 (IC = 92 nM) to piperidine 8b (IC =
5
0
50
50
8
6 nM). Among the isopropoxy derivatives, compound 25 with a 1,3,4-oxadiazol-2-thione ring
attached to the right-hand side, displayed a significantly higher activity (IC50 = 190 nM)
compared to its corresponding hydrazide 24 (RA = 89%) or ester analogue 23 (RA = 87%).
Compounds 9c (IC = 9.9 nM) and 9d (IC = 6.9 nM) with (S)- and (R)-alanine groups attached
5
0
50
to the right-hand side proved to be on the same order of potency as their glycine relative 9b (IC50
1
6

ACCEPTED MANUSCRIPT
=
17 nM). There were no marked differences in activities between the (S)- and (R)-alanine
isomers.
Overall, type II inhibitors were less potent than type I compounds, which is probably due to the
absence of substituents at the ortho position to the amino group on the benzene ring. The focus in
type II compounds was to study the effects of different substituents on the right-hand side of the
molecule. Three compounds - acyl sulfonamide 39, aliphatic nitrile 43b and 5-oxotetrazole 50 -
displayed good IC50 values of 160 nM, 410 nM and 77 nM, respectively. Owing to the acidity of
its NH group, acyl sulphonamides are commonly used as carboxylic acid isosteres [26].
Moreover, compounds containing this functional group often show improved oral availability,
metabolic stability and potency compared to the parent carboxylic acids [27]. The nitrile 43b and
its dibromo pyrrole analogue 43a were synthesised because of the important roles that nitrile
groups have as hydrogen bond acceptors [28]. The nitrile functionality can act as a carboxylic
acid isostere, which has been exploited in the design of vildagliptin [22]. Nitrile-containing
quinolones with good potency and safe therapeutic profiles were also reported [29]. Similarly, 5-
oxotetrazole derivative 50 was prepared as a carboxylic acid isostere. The NH group at the 4-
position of 5-oxotetrazole has a similar pKa value to the NH of tetrazole [21]. Given the smaller
size of compound 50 compared to type I inhibitors, it is expected that the 5-oxotetrazole ring is
making interactions with Arg76 rather than Arg136. Other compounds of the type II series, such
as sulfonamides 27, 29, 33 and 34 (RA = 82%, 100%, 87% and 87% respectively), and malonic
ester derivatives 46a (RA = 78%) and 46b (IC = 7700 nM), were not active on E. coli DNA
5
0
gyrase. This can perhaps be explained with the weaker acidity of sulfonamides (pKa range: 7.3-
.7 [30]) compared to carboxylic acids, while the weak inhibitory values of malonic ester
9
derivatives are in line with those of most esters in the type I series and additionally suggest that
the malonic functionality could be too bulky to establish strong interactions with Arg136.
The type III compounds contain a quinolone-like structure (56 and 57) or a substituted indole
moiety (60 and 61) and are more rigid than the compounds belonging to type I and II series.
Among the two quinolones, the ester 56 interestingly showed stronger activity (IC50 = 53 nM)
than the corresponding carboxylic acid analogue 57 (IC50 = 1400 nM). The opposite was
observed in the indole series, where the carboxylic acid 61 displayed excellent inhibitory activity
(
IC = 98 nM) whereas the ethyl ester analogue 60 was completely inactive (RA = 100%).
50
1
7

ACCEPTED MANUSCRIPT
Some of the prepared compounds displayed promising activities also on DNA gyrase from S.
aureus and on topoisomerase IV from E. coli and S. aureus, although the activities were
generally lower than those on E. coli DNA gyrase (Table 4). Compounds 9c and 9d, that were the
most potent on E. coli DNA gyrase (IC = 9.9 nM and IC = 6.9 nM, respectively), were also
5
0
50
among the most potent inhibitors of these three enzymes. In particular, derivative 9c displayed a
nanomolar activity against S. aureus DNA gyrase (IC = 99 nM) and a good activity against
5
0
topoisomerase IV from S. aureus (IC50 = 689 nM). Compound 9d, the
D-enantiomer of 9c,
showed slightly weaker activity on the S. aureus gyrase, but was comparably active on all other
tested enzymes. Similarly as for E. coli gyrase, zwitterionic compounds 9b-d and 17 were the
most potent in the series, which underlines the importance of the carboxylic moiety on the right-
hand side of the molecules and the amino group-containing heterocycles on the central benzene
ring for the binding affinity. Because of the slightly bigger size of bromine atoms compared to
chlorine atoms, and because of the smaller size of the ATP binding pocket of S. aureus gyrase
and of topo IV, the dibromopyrrole derivative 9a was less active against those enzymes. In
addition to zwitterionic compounds, also the methyl ester derivatives 10a, 10b and 18 showed
moderate activities, with L-alanine derivative 10a being the best. Derivatives containing an
alanine group attached to the right-hand side (8c, 8d, 9c, 9d, 10a-b and 25) were generally more
potent than their glycine counterparts, but glycine-containing compound 9b displayed the best
activity against E. coli topoisomerase IV among all tested compounds, with IC value of 7200
5
0
nM. Compound 25 with an isopropoxy group on the benzene ring, showed much weaker activity
than compounds with piperidin-4-yloxy (e.g. 8b-c, 9b-d and 10a) and pyrrolidin-3-yloxy (17 and
1
8) substituents. Type II and type III inhibitors, did not exhibit promising activities on the S.
aureus gyrase and on topo IV, which could be explained by the lack of the ortho substituents
capable of forming favourable hydrophobic interactions or hydrogen bonds with the enzyme.
Table 1. Inhibitory activities of type I compounds against DNA gyrase from E. coli.
a
b
IC50 (nM) or RA (%)
Compd.
R
1
X
R
2
Y
*
E. coli gyrase
1
8

ACCEPTED MANUSCRIPT
7
7
a
4,5-diBr
NHBoc
NHBoc
H
H
COOMe
COOMe
COOMe
COOMe
COOH
COOH
COOH
COOH
COOH
COOH
COOH
COOH
COOMe
COOMe
COOMe
COOH
COOH
COOMe
/
/
100%
100%
b
3,4-diCl-5-Me
3,4-diCl-5-Me
3,4-diCl-5-Me
4,5-diBr
7
c
CH
CH
H
3
3
L
D
/
83%
7d
8a
8b
94%
NHBoc
NHBoc
NHBoc
NHBoc
630 ± 100 nM
86 ± 18 nM
80 ± 14 nM
71 ± 2 nM
20 ± 3 nM
17 ± 5 nM
9.9 ± 2.9 nM
6.9 ± 0.5 nM
40 ± 22 nM
48 ± 4 nM
93%
3,4-diCl-5-Me
3,4-diCl-5-Me
3,4-diCl-5-Me
4,5-diBr
H
/
8
c
CH
CH
H
3
3
L
D
/
8d
9a
9b
3,4-diCl-5-Me
3,4-diCl-5-Me
3,4-diCl-5-Me
3,4-diCl-5-Me
3,4-diCl-5-Me
3,4-diCl-5-Me
3,4-diCl-5-Me
3,4-diCl-5-Me
3,4-diCl-5-Me
H
/
9
c
CH
CH
CH
CH
H
3
3
3
3
L
D
L
D
/
9
d
1
0a
0b
1
1
5
6
7
8
1
H
/
92 ± 26 nM
31 ± 3 nM
74 ± 12 nM
1
1
H
/
H
/
2
2
3
4
3,4-diCl-5-Me
3,4-diCl-5-Me
iPr
CH
CH
3
3
COOMe
D
D
87%
89%
iPr
iPr
COONHNH
2
2
5
3,4-diCl-5-Me
CH
3
D
190 ± 30 nM
170 ± 20 nM
novobiocin
a
b
Concentration of compound that inhibits the enzyme activity by 50%. Residual activity of the enzyme at 1 µM of the compound.
Table 2. Inhibitory activities of type II compounds against DNA gyrase from E. coli.
1
9

ACCEPTED MANUSCRIPT
a
b
IC50 (nM) or RA (%)
Compd.
R1
R2
E. coli gyrase
2
2
7
9
3,4-diCl-5-Me
82%
3,4-diCl-5-Me
3,4-diCl-5-Me
3,4-diCl-5-Me
3,4-diCl-5-Me
4,5-diBr
100%
87%
3
3
3
3
4
9
87%
160 ± 10 nM
960 ± 40 nM
410 ± 50 nM
4
3a
3b
4
3,4-diCl-5-Me
4
6a
6b
4,5-diBr
78%
4
3,4-diCl-5-Me
7700 ± 900 nM
5
0
3,4-diCl-5-Me
77 ± 10 nM
novobiocin
170 ± 20 nM
a
b
Concentration of compound that inhibits the enzyme activity by 50%. Residual
activity of the enzyme at 1 µM of the compound.
Table 3. Inhibitory activities of type III compounds against DNA gyrase from E. coli.
a
b
Compd.
Ar
IC50 (nM) or RA (%)
2
0

ACCEPTED MANUSCRIPT
E. coli gyrase
5
5
6
7
53 ± 26 nM
1
400 ± 200 nM
c
1
000 ± 200 nM
6
6
0
1
100%
98 ± 14 nM
novobiocin
170 ± 20 nM
a
Concentration of compound that inhibits the enzyme activity by 50%.
b
c
Residual activity of the enzyme at 1 µM of the compound. Result
obtained with ATPase assay.
Table 4. Inhibitory activities of selected compounds against DNA gyrase from S. aureus and
against topoisomerase IV from E. coli and S. aureus.
a
b
IC50 (nM) or RA (%)
Compd.
S. aureus gyrase
E. coli topo IV
S. aureus topo IV
1
008 ± 300 nM
8
b
10000 ± 2000 nM
7400 ± 800 nM
8
c
26%
68%
67%
88%
99%
830 ± 120 nM
79%
8d
9a
9b
100%
100%
960 ± 0.22 nM
99 ± 11 nM
31%
7200 ± 6700 nM
9100 ± 2200 nM
15000 ± 1000 nM
18000 ± 1000 nM
89%
3700 ± 2500 nM
690 ± 30 nM
960 ± 40 nM
3700 ± 300 nM
94%
9
c
9
d
1
0a
0b
1600 ± 300 nM
49%
1
1
6000 ± 2 nM
4300 ± 400 nM
1
7
290 ± 90 nM
1
8
5
9
0
7
950 ± 110 nM
59%
90%
92%
98%
94%
92%
94%
90%
2
3
5
5
55%
81%
47%
85%
90%
100%
novobiocin
41 ± 70 nM
11000 ± 2000 nM
27000 ± 7000 nM
2
1

ACCEPTED MANUSCRIPT
a
b
Concentration of compound that inhibits the enzyme activity by 50%. Residual activity
of the enzyme at 1 µM of the compound.
Antibacterial Activity. Antibacterial activities of the prepared inhibitors were evaluated against
the ESKAPE panel (Enterococcus faecium ATCC 35667, Staphylococcus aureus ATCC
2
1
9213, Klebsiella
pneumoniae
ATCC
700603, Acinetobacter
baumannii
ATCC
9606, Pseudomonas aeruginosa ATCC27853, and Enterobacter aerogenes ATCC 13408). All
the compounds were additionally tested against wild-type E. coli ATCC 25922 and two E. coli
mutant strains, JD17464 and JW5503. E. coli JD17464 is an lpxC deletion mutant with impaired
outer membrane, while E. coli JW5503 is a tolC deletion mutant with defective efflux pump.
Sixteen compounds were also evaluated against E. faecalis (ATCC 29212). The results are
presented in Table 1S and Table 2S (Supporting information) as growth inhibition (%). Minimum
inhibitory concentrations (MICs) were determined for compounds that inhibited one or more
bacterial strains by more than 60% (Tables 5 and 6).
Overall, compounds displayed much stronger inhibition of Gram-positive than Gram-negative
bacteria. Altogether five compounds (24, 25, 43b, 50 and 57) inhibited the growth of E. faecalis
by more than 85% at 50 µM concentration (Table 1S). Compound 25 displayed a better
antibacterial activity (MIC = 6.25 µM) than the corresponding hydrazide derivative 24 (MIC =
1
2.5 µM) (Table 5), probably because it is able to make stronger contacts with Arg136 through
its 1,3,4-oxadiazol-2-thione ring on the right-hand side. Additionally, lower polarity of compound
5 compared to the more polar hydrazide 24 (see ChemGPS analysis below [31]) is believed to
2
favour the permeation into the bacterial cells. Four compounds (7c, 8c, 25 and 57) inhibited the
growth of E. faecium by more than 60% and two compounds (50 and 57) inhibited the growth of
S. aureus by more than 95% at 50 µM concentration (Table 1S). Of particular interest was the
quinolone 57 that inhibited all the tested Gram-positive strains by more than 95%, with excellent
MIC values of 0.78 µM against S. aureus, and of 1.56 µM against E. faecalis and E. faecium
(Table 5). This result is especially interesting, since this compound did not possess strong activity
against DNA gyrase (E. coli IC = 1.4 µM, S. aureus RA = 90%) or topoisomerase IV (E. coli
5
0
RA = 92%, S. aureus RA = 100%) enzymes, which could suggest a different mode of action for
this compound. Interestingly, the zwitterionic compounds 9b-d and 17, which were the most
potent in the enzymatic assays, did not show good antibacterial activities. Probably these
2
2

ACCEPTED MANUSCRIPT
compounds have difficulties with passing through bacterial cell wall due to the presence of two
charged functional groups.
None of the compounds displayed significant inhibition of the wild-type E. coli but six of them
(10a, 10b, 18, 43b, 50 and 61) showed more than 80% inhibition of E. coli tolC deletion mutant
with defective efflux pump at 50 µM concentration (Table 2S), and for those compounds the MIC
values were between 3.13 and 25 µM (Table 6). These results indicate that some compounds are
subject to bacterial efflux, which is likely the main reason for their weaker activity against wild-
type Gram-negative bacteria. Among the compounds that were active against E. coli tolC deletion
mutant, three derivatives belonging to type I series (10a, 10b and 18) share similar structures, i.e.
a positively charged amino group-containing heterocycles on the central benzene ring and a
methyl carboxylate group on the right-hand side. This is in agreement with the recent reports, that
the overall positive charge of a molecule is an important factor to achieve higher level of
accumulation in E. coli [32]. Our results also confirm that the presence of a positively charged
amino group alone is not sufficient to obtain compounds with high accumulation in E. coli, but
that other parameters should be taken into account while designing antibacterials with Gram-
negative activity, such as high rigidity, low globularity and high amphiphilic moment [33]. In
addition to type I compounds 10a, 10b and 18, other three compounds that significantly inhibited
E. coli tolC deletion mutant were the nitrile 43b (MIC = 12.5 µM), the 5-oxotetrazole derivative
5
0 (MIC = 3.13 µM) and the indole derivative 61 (MIC = 12.5 µM) (Table 6).
Table 5. Minimum inhibitory concentrations (MICs) of compounds 7c, 8c, 24, 25, 43b, 50, and
5
7 against E. faecalis (ATCC 29212), E. faecium (ATCC 35667) and S. aureus (ATCC 29213).
a
MIC (µM)
Compd.
E. faecalis (ATCC 29212)
E. faecium (ATCC 35667)
S. aureus (ATCC 29213)
b
7
8
2
2
c
c
4
5
n.d.
>125
>125
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
50
n.d.
12.5
6.25
25
>125
n.d.
4
3b
5
0
7
12.5
1.56
n.d.
5
1.56
0.78
a
MIC (minimum inhibitory concentration that inhibits the growth of bacteria by ≥ 90%) values against E. faecalis, E.
faecium and S. aureus. Ciprofloxacin was used as a positive control for E. faecalis and S. aureus at MIC of 3.02 and
b
1
.51 µM, respectively. For E .faecium linezolid was used as positive control at MIC of 11.86 µM. Not determined.
2
3

ACCEPTED MANUSCRIPT
Table 6. Minimum inhibitory concentrations (MICs) of compounds 10a, 10b, 18, 43b, 50 and 61
against E. coli (JW5503).
a
MIC (µM)
Compd.
E. coli (JW5503)
1
0a
0b
6.25
6.25
25
1
1
8
4
3b
12.5
3
.13
5
0
1
1
2.5
6
a
MIC (minimum inhibitory concentration that
inhibits the growth of bacteria by ≥ 90%)
values against E. coli JW5503 (tolC deletion
mutant). Ciprofloxacin was used as a positive
control at a MIC of 0.015 µM.
Further Biological Evaluation of Compound 57. Overall, compound 57 was selected as the
most promising compound, since it displayed the lowest MIC values against Gram-positive
bacteria. Thus, for this compound we determined the MIC values against a panel of additional
Gram-positive and Gram-negative bacterial strains (Table 7). Compound 57 displayed promising
MIC values of 0.72 µM against vancomycin-resistant E. faecium (VRE, ATCC 700221) and
Streptococcus pyogenes (ATCC 19653), and MIC values of 2.5 µM against methicillin-resistant
S. aureus (MRSA, ATCC 43300), ciprofloxacin non-susceptible and extremely drug resistant
strain of S. aureus (MRSA VISA, ATCC 700699), and S. aureus (ATCC 25923) strains. The
activity against methicillin-resistant S. aureus (MRSA, ATCC 43300) in the presence of 50%
inactivated human blood serum was also promising, with MIC value of 32 µM. Compound 57
was additionally evaluated against the wild-type E. coli strains ATCC 25922 and MG1655 in the
presence of efflux pump substrate phenylalanine-arginine β-naphthylamide (PAβN), and the MIC
values were 2.3 µM and 2.5 µM, respectively. These results indicate that the inactivity against
wild-type Gram-negative organisms is probably associated with efflux. Also, MIC values against
P. aeruginosa (ATCC 27853), K. pneumoniae (ATCC 10031), Shigella flexneri (HNCMB
2
0018) and Salmonella enterica (LT2) were determined but were above 32 µM (Table 7).
2
4

ACCEPTED MANUSCRIPT
Given the quinolone substructure of compound 57, we decided to perform additional experiments
on E. coli and K. pneumoniae strains that possess mutations in the key amino acids of the GyrA
and GyrB binding sites, to investigate whether this compound binds preferably to the subunit A
or to the subunit B of the enzyme (Table 8). The results indicate that neither mutations at GyrA,
nor mutations at GyrB or ParC influence the MIC values of compound 57. This suggests a
different mode of action of compound 57.
To further investigate the mode of action of compound 57, we determined its inhibitory activity
on E. coli DNA gyrase using an ATP competitive assay (Table 3). This test measures the
competitive action of compounds against ATP and therefore provides experimental evidence that
compounds bind to the ATP binding pocket of the enzyme, located on the GyrB subunit. Since
the activity of compound 57, determined with ATPase assay (IC = 1.0 µM), was similar to the
5
0
activity determined with supercoiling assay (IC = 1.4 µM), its inhibition of DNA gyrase is most
5
0
likely due to binding to GyrB.
The possibility of compound 57 to bind to the fluoroquinolone binding site on the GyrA subunit
was additionally studied with molecular docking (Figure 2S, Supporting Information). The
docking pose of compound 57 in the S. aureus DNA gyrase (PDB entry: 2XCT [34]),
superimposed with the pose of ciprofloxacin, shows that compound 57 adopts a different
conformation as ciprofloxacin. In particular, compound 57 is not able to establish similar contacts
2
+
with Mn , which are crucial for the binding. The most likely reason is that the pyrrolamide
portion of the compounds is creating steric clashes, which prevent the compound to adopt the
same orientation as ciprofloxacin.
Table 7. Minimum inhibitory concentrations (MICs) of compound 57 against additional Gram-
positive and Gram-negative bacterial strains.
a
Compound 57
MIC (µM)
Gram-positive bacteria
E. faecium (VRE, ATCC 700221)
Streptococcus pyogenes (ATCC 19653)
S. aureus (MRSA, ATCC 43300)
0.72
0.72
2.5
S. aureus (MRSA VISA ATCC 700699)
2.5
2
5

ACCEPTED MANUSCRIPT
S. aureus (MRSA, ATCC 43300) + 50%
3
2
inactivated human blood serum
S. aureus (ATCC 25923)
2.5
Gram-negative bacteria
E. coli (ATCC 25922)
>
32
2.3
2.5
E. coli (ATCC 25922) + 50 µg/mL PAβN
E. coli (MG1655) + 50 µg/mL PAβN
P. aeruginosa (ATCC 27853)
K. pneumoniae (ATCC 10031)
S. flexneri (HNCMB 20018)
>
32
32
32
32
>
>
>
S. enterica (LT2)
a
MIC (minimum inhibitory concentration that inhibits the growth of
bacteria by >90%) measurements were performed according to the
EUCAST guidelines in three independent measurements.
Table 8. Minimum inhibitory concentrations (MICs) of compound 57 against Gram-negative
bacterial strains with mutated amino acids in the GyrA or GyrB active sites in the presence of
phenylalanine-arginine β-naphthylamide (PAβN).
a
MIC (µM)
Strain
Compd. 57 + 50 µg/ml PAβN
E. coli MG1655 (wild-type)
2.5
2.5
2.5
2.5
2.5
2.5
E. coli MG1655 GyrB R136C
E. coli MG1655 GyrA S83L
E. coli MG1655 GyrA S83L, D87N
E. coli MG1655 GyrA S83L, D87N; ParC S80I
E. coli MG1655 GyrA S83L, D87N; ParC S80I, E84G
K. pneumoniae (ATCC 10031) (wild-type)
K. pneumoniae (ATCC 10031) GyrA S83F
K. pneumoniae (ATCC 10031) GyrA S83Y
K. pneumoniae (ATCC 10031) GyrA S83L
K. pneumoniae (ATCC 10031) GyrA S83F, D87G
K. pneumoniae (ATCC 10031) GyrA S83F, D87G; ParC S80I
b
1.2
1.2
1.2
1.2
2.5
1.2
1.2
b
b
b
b
b
b
K. pneumoniae (ATCC 10031) GyrA S83F, D87G; ParC S80I, E84G
a
MIC (minimum inhibitory concentration that inhibits the growth of bacteria by > 90%) measurements
b
were performed according to the EUCAST guidelines in 3 independent measurements. MIC values
were determined in the presence of 10 µg/ml PAβN. Ciprofloxacin was used as a positive control. MIC
values of ciprofloxacin against E. coli MG1655 GyrB R136C and E. coli MG1655 GyrA S83L were
0
.01 µM, against E. coli MG1655 GyrA S83L, D87N was 0.19 µM, against E. coli MG1655 GyrA
S83L, D87N; ParC S80I was 8 µM and against E. coli MG1655 GyrA S83L, D87N; ParC S80I, E84G
was >32 µM.
2
6

ACCEPTED MANUSCRIPT
Physicochemical Properties of the Tested Compounds. To investigate the physicochemical
space that our molecules are populating, and to link this information to their antibacterial
activities, we performed a chemographic mapping of all the synthesised compounds using the
ChemGPS-NPWeb tool [35] (Figure 5). ChemGPS-NPWeb service is a useful tool that could
provide valuable information for designing compounds with broad-spectrum antibacterial activity
[
36]. For the compounds that were prepared as
c, 9c and 10a) were taken into consideration. For each compound, eight principal components
PCs) were generated (PC1-8, Table 3S of Supporting information). Each PC describes different
L- and D-enantiomers, only the L-enantiomers (7c,
8
(
physicochemical properties and, among those, the first three PCs (PC1-3) are the most
significant. PC1 accounts for size, shape and polarizability, PC2 comprises aromaticity and
conjugation properties, PC3 includes lipophilicity, polarity and H-bond formation capacity [37].
A description for PC4, PC5, PC6, PC7 and PC8 is given in Table 3S of Supporting information.
We plotted our compounds in a 3D diagram, setting PC1, PC2 and PC3 as the x, y and z axes. In
this study, the value of PC1 correlates with the size of the compounds, PC2 correlates with the
number of aromatic rings and the degree of conjugation in the molecule, while the value of PC3
correlates with their lipophilicity (cLogP). The chemographic plot shows that our compounds can
be divided by PC3 into three groups: lipohphilic derivatives (red-yellow spheres), moderately
lipophilic derivatives (green spheres) and more polar compounds (blue spheres). Compounds 24,
2
5, 43b, 50 and 57 that showed the most promising antibacterial activities against Gram-positive
bacteria all possessed moderately low values of PC3 (cLogP values were 1.07, 2.42, 1.34, 1.74
and 2.83, respectively) and are represented by green-blue colour in the graph. Additionally, these
compounds stay on the lower side of PC1 (MW = 456, 497, 351, 353 and 408, respectively),
indicating that they are small-sized and less bulky than type I compounds 7a-7c, 8a-c, 15 and 16
(MW = 658, 583, 597, 644, 569, 583, 569 and 555, respectively). The zwitterionic compounds
9
b, 9c and 17, which were the most potent in the enzymatic assays, show relatively low
lipophilicity (cLogP = 1.01, 1.31 and 0.89, respectively) and are medium sized (medium value of
PC1, MW = 490-510). Most probable reason for their inactivity in bacteria is the simultaneous
presence of two charged groups, which makes their entry into bacterial cell more problematic.
2
7

ACCEPTED MANUSCRIPT
Figure 5. Chemographic plot of the synthesised compounds. In the colour map displayed on the
right, the red-yellow band represents the lipophilic region, the green band represents the
moderately lipophilic region and the blue band represents the more polar region. The values of
cLogP described in the text were calculated by ChemOffice 2016 [38]. The graph was prepared
by Teraplot [39].
We performed a computational study on Lipinski's rule of five parameters for the most promising
compounds 24, 25, 43b, 50 and 57. The study was carried out using SwissADME free tool [40].
Lipinski’s parameters (molecular weight, MLOGP [41], number of H-bond donors and H-bond
acceptors) and calculated water solubility (LogS [42]) are reported in Table 9. All selected
compounds complied to Lipinski’s rules.
Table 9. Lipinski’s parameters (molecular weight, MLOGP, number of H-bond donors and H-
bond acceptors) and predicted water solubility (LogS) of compounds 24, 25, 43b, 50 and 57.
Compd.
MW (g/mol)
456.32
MLOGP
1.13
H-bond acceptor
H-bond donor
LogS
-3.91
-5.33
2
2
4
5
5
5
5
4
498.38
1.57
2
8