Indole Based Weapons to Fight Antibiotic Resistance: A Structure-Activity Relationship Study

Article abstract of DOI:10.1021/acs.jmedchem.5b01219

Antibiotic resistance represents a worldwide concern, especially regarding the outbreak of methicillin-resistant Staphylococcus aureus, a common cause for serious skin and soft tissues infections. A major contributor to Staphylococcus aureus antibiotic resistance is the NorA efflux pump, which is able to extrude selected antibacterial drugs and biocides from the membrane, lowering their effective concentrations. Thus, the inhibition of NorA represents a promising and challenging strategy that would allow recycling of substrate antimicrobial agents. Among NorA inhibitors, the indole scaffold proved particularly effective and suitable for further optimization. In this study, some unexplored modifications on the indole scaffold are proposed. In particular, for the first time, substitutions at the C5 and N1 positions have been designed to give 48 compounds, which were synthesized and tested against norA-overexpressing S. aureus. Among them, 4 compounds have NorA IC₅₀ values lower than 5.0 μM proving to be good efflux pump inhibitor (EPI) candidates. In addition, preliminary data on their ADME (absorption, distribution, metabolism, and excretion) profile is reported.

Full text of DOI:10.1021/acs.jmedchem.5b01219

Subscriber access provided by NEW MEXICO STATE UNIV
Article
Indole based weapons to fight antibiotic
resistance: a structure-activity relationship study
Susan Lepri, Federica Buonerba, Laura Goracci, Irene Velilla, Renzo Ruzziconi,
Bryan D. Schindler, Susan M. Seo, Glenn W Kaatz, and Gabriele Cruciani
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.5b01219 • Publication Date (Web): 12 Jan 2016
Downloaded from http://pubs.acs.org on January 13, 2016
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 84
Journal of Medicinal Chemistry
1
2
3
Indole based weapons to fight antibiotic resistance: a structure-activity relationship
study
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
‡
§
‡§
§
§
§
Susan Lepri,
Federica Buonerba,
Laura Goracci, Irene Velilla, Renzo Ruzziconi, Bryan D.
ς
ς
¥ς*
§*
Schindler, Susan M. Seo, Glenn W. Kaatz, and Gabriele Cruciani.
§
Department of Chemistry, Biology and Biotechnology, University of Perugia, 06123 Perugia, Italy.
ςThe John D. Dingell Department of Veterans Affairs Medical Center, Detroit, MI 48201, USA
¥
Department of Internal Medicine, Division of Infectious Diseases, Wayne State University School of
Medicine, Detroit, MI 48201, USA
TITLE RUNNING HEAD: Fight MRSA antibiotic resistance
ABSTRACT
Antibiotic resistance represents a worldwide concern, especially regarding the outbreak of methicillinꢀ
resistant Staphylococcus aureus, a common cause for serious skin and soft tissues infections. A major
contributor to Staphylococcus aureus antibiotic resistance is the NorA efflux pump, which is able to
extrude selected antibacterial drugs and biocides from the membrane, lowering their effective
concentrations. Thus, the inhibition of NorA represents a promising and challenging strategy that
would allow recycling of substrate antimicrobial agents. Among NorA inhibitors, the indole scaffold
proved particularly effective and suitable for further optimization. In this study, some unexplored
modifications on the indole scaffold are proposed. In particular, for the first time, substitutions at the
C5 and N1 positions have been designed to give 48 compounds, which were synthesized and tested
against norAꢀoverexpressing S. aureus. Among them, 4 compounds have NorA IC values lower than
5
0
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 84
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
5
.0 µM proving to be good efflux pump inhibitors (EPI) candidates. In addition, preliminary data on
their ADME (absorption, distribution, metabolism, and excretion) profile is reported.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
INTRODUCTION
Antibiotic resistance represents a worldwide threat. According to a new report issued by the Centers for
1
Disease Control and Prevention each year more than two million people in the United States get
infections that are resistant to antibiotics and at least 23,000 people die as a result. In particular,
methicillin resistant S. aureus (MRSA) is a major communityꢀacquired as well as nosocomial pathogen
causing skin and soft tissues infections, respiratory disease and more serious illness like pneumonia,
2
endocarditis and sepsis.
The most important factor in antibiotic multiꢀdrug resistance (MDR) is the irresponsible use of
antibiotics. Since Fleming's revolutionary discovery of penicillin G in 1928, a great effort has been
lavished on the development of novel antibiotic classes (e.g. sulfonamides, quinolones,
3
oxazolidinones) but with time their utility became compromised by the emergence of resistant
bacteria. This led the medical community to ask for urgent development of novel drugs to fight
resistant strains.
Bacterial resistance can rise through three main mechanisms: drug target modification, drug
inactivation or export by membraneꢀbased efflux proteins. While the fight against drug target
modification or drug inactivation entails developing new classes of antibiotics with novel mechanisms
of action, the inhibition of efflux pump systems represents a valid option for both its innovative
character and the intrinsic advantage of recycling antimicrobial agents that are efflux pump substrates.
Efflux pumps are transport proteins involved in the extrusion of toxic substrates (including antibiotics)
4
from within cells into the external environment. The genes expressing many pump proteins are present
5
in both antibiotic susceptible and resistant bacteria. Whether resistance is related to increased pump
ACS Paragon Plus Environment

Page 3 of 84
Journal of Medicinal Chemistry
1
2
3
gene expression or an amino acid substitution, which improves the efficiency of the pump protein, the
end result is a reduced concentration of the antimicrobial drug at its active site. In particular, the S.
aureus NorA efflux pump, which belongs to the major facilitator superfamily (MFS) of transport
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
6
proteins, can contribute to quinoloneꢀbased drug resistance by removal of these drugs thus lowering
7
,8
their effective concentrations.
Increased interest has been devoted towards molecules able to inhibit the NorA efflux pump, whose
action is considered an important factor for selected drug and biocide resistance. The major obstacle in
designing specific inhibitors of the NorA pump is that the three dimensional structure of the protein is
not yet known. Some molecules are wellꢀknown inhibitors of NorA, such as reserpine, which is
frequently used as reference compound in inhibition studies. Unfortunately, although reserpine is an
approved drug for other targets, it cannot be used for NorA inhibition because of its neurotoxic effect at
9
the required concentrations. In the past, a number of NorA inhibitors belonging to different chemical
1
0
11
classes have been developed including flavones and flavonolignans, hydroxyquinoline derivates,
1
2
13,14
15ꢀ17
piperines, pyridines and phenylpiperidine,
chalcones and alkenamides.
Their activities are
mainly evaluated in terms of ethidium bromide (EtBr) efflux inhibition, or for their capacity of
restoring the antibacterial activity of an antibiotic such as ciprofloxacin (CPX).
The indole moiety has been shown to be promising with respect to the EPI activity of a compound. The
9
compound 5ꢀnitroꢀ2ꢀphenylꢀ(1H)ꢀindole (1), known as INF55, is one of the first identified indoleꢀ
based inhibitors of NorA and is capable of producing a 4ꢀfold increase in the susceptibility of S. aureus
to ciprofloxacin at a concentration of 1.5 µg/mL (Figure 1). Since the discovery of 1, a number of
indoleꢀbased analogues were synthesized and tested for EPI activity and the most potent of these are
1
8ꢀ20
illustrated in Figure 1 (2ꢀ6).
Structural modifications were introduced at the C2, C3 and C5
positions; in particular, in the first discovered compounds, an electron withdrawing group was always
1
8,19
preserved (a nitro group in 1, 3-4 or a cyano group in 2).
In addition, an aromatic moiety with
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 84
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
various substituents was preserved in all active compounds (1-5). More recently, indoles bearing
halogen atoms in the C5 position and a peculiar nitrone moiety in the C3 position have been evaluated
2
0
for their good EPI properties (compounds 6a-c, Figure 1). However, to the best of our knowledge 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
NH group in the indole scaffold has yet to be derivatized.
Figure 1. Structures of indoleꢀbased compounds discovered to date that target NorA.
In this study, we report on a novel series of compounds bearing an indole scaffold in which
modifications were performed at the N1, C3, and C5 positions of the indole nucleus. Staphylococcus
aureus strains SAꢀ1199 (wild type) and SAꢀ1199B (overexpressing norA) were used to evaluate the
21
inhibition effect. As a result, twenty novel EPIs with various scaffolds inhibited EtBr efflux by more
than 90%. Among them, four compounds exhibited an IC lower than 5.0 µM. Our results show that
50
ACS Paragon Plus Environment

Page 5 of 84
Journal of Medicinal Chemistry
1
2
3
tuning of the molecular features can result in pure NorA inhibitors that lack any significant
antimicrobial effect.
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
METHODOLOGY
Inhibitor Design
Inhibitor design for the NorA efflux pump is a challenging issue due to the lack of a crystal structure of
the target. However, a “ligandꢀbased” approach can be performed utilizing known inhibitors as
templates. Nꢀsubstituted indoles were designed by inspecting compounds in Figure 1. Our strategy was
to remove the phenyl group at the C2 position and to synthesize an Nꢀsubstituted indole that could
possibly mimic the C2ꢀphenyl hydrophobic interaction. Toward this end, the Nꢀbenzyl moiety was
designed; indeed, the Nꢀbenzyl indole and the 2ꢀphenyl indole were superimposed using the FLAP
2
2
software to compare them in size, shape and hydrophobic interactions. The latter ones were evaluated
2
3
by comparing the GRID molecular interaction fields generated by the DRY probe (see Materials and
Methods). Based on this simulation the two compounds share similar hydrophobic interactions as well
as size and shape, as shown in Figure 2.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 84
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
Figure 2. Overlay of 2ꢀphenylindole (in yellow sticks) and Nꢀbenzylindole (green sticks), obtained
2
2
23
using the FLAP software. GRID hydrophobic molecular interaction fields are shown in yellow and
green color, respectively.
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
From Figure 1 we also noticed that the presence of an electronꢀwithdrawing group (EWG) such as
NO , CN, and halogen atoms in the C5 position was highly conserved. As such, we elected to maintain
2
the nitro moiety at this position. In addition, the most recent series of indole derivatives (6) reported by
20
Hequet et al. suggested that the presence of a polar substituent in the C3 position might be important
for an inhibitory effect. Thus, an ester group was placed at C3 of the indole scaffold. Based on these
considerations, compound 7 in Table 1 was the first designed compound. Finally, aiming at examining
substituents other than EWGs, we also explored substitutions at the C5 position. Consequently,
compounds 8ꢀ13 were designed (Table 1). Trifluoromethylated compound 8 still bears an EWG group
at C5, but differs from NO ꢀ and CNꢀ moieties by its greater hydrophobicity. In Compound 9, a
2
hydroxyl group was positioned at C5. This compound was also the starting material for further
functionalizations. Indeed, reaction with acetyl chloride gave the acetyl derivative 10, and the reaction
with phenethyl bromide gave the correspondent phenethyl ether 11. Finally, the reaction with 2ꢀchloroꢀ
N,Nꢀdimethylethylamine or 1,2ꢀepoxybutane allowed the synthesis of the amino ether and the hydroxyl
ether 12 and 13, respectively.
In particular, with respect to structures in Figure 1, selected modifications allowed to include a weak
acid group (compounds 9 and 13) and a basic center (compound 12), also varying the number of Hꢀ
bond donor (HBD) and Hꢀbond acceptor (HBA) atoms in the series. Of course, both the protonation
state and the number and disposition of HBD and HBA groups might have an impact on the interaction
with the target. The logP values for all compounds are in the range between 4 and 5, and the most
lipophilic compounds are trifluoromethyl 8 and phenethyl derivative 11. The presence at physiological
pH of protonated species should be also taken in consideration: for example, according to the pKa
ACS Paragon Plus Environment

Page 7 of 84
Journal of Medicinal Chemistry
1
2
3
24,25
values predicted by Moka
(Table 1), for 12 the protonated form is 93% abundant, while for 9 the
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
neutral form prevails over the deprotonated one (99%).
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
Table 1. Designed NorA inhibitors bearing modification at the C5 position.
%
EtBr efflux
a
b
MIC d
IC50e
Comp.
X
LogP
pKa
inhibition @50
c
(µg/ml)
(µM)
µM
f
7
8
9
NO
2
4.43
5.43
4.07
4.41
6.42
4.51
4.88
̶
̶
9.1
36.6
1.0
ND
ND
ND
ND
ND
ND
22.5
ND
CF
ND
>100
ND
3
OH
9.43 (a)
1
1
1
1
0
1
2
3
̶
20.4
4.7
̶
>100
>100
>100
8.63 (b)
15.58 (a)
98.2
71.1
Biological data were generated employing SAꢀ1199B, which overexpresses norA; a) predicted by
26
24,25
Volsurf; b) predicted by Moka;
c) % of ethidium bromide (EtBr) efflux inhibition with compound
at a concentration of 50 µM; d) intrinsic antimicrobial activity of test compounds; e) IC (inhibitor
5
0
concentration producing a 50% reduction in EtBr efflux) calculated from doseꢀresponse experiments.
See Biology section for details; f) ND, not determined.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 84
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
Chemistry
The studied compounds were obtained through a number of diverse approaches, summarized in
Schemes 1−4.
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
7
2
ꢀMethylꢀ5ꢀnitroindole 14, prepared by direct nitration of 2ꢀmethylindole, was benzylated at the
2
8
nitrogen to obtain the Nꢀbenzylindole 15 (Scheme 1). Trifluoroacetylation with TFAA gave the 3ꢀ
trifluoroacetylated indole 16 which was hydrolyzed to the corresponding acid 17 with 20% aqueous
NaOH. The acyl chloride obtained by refluxing the acid 17 in thionyl chloride was then transformed to
the corresponding ethyl ester 7 by treatment with ethanol.
a
Scheme 1
NO
2
H
H
i
N
ii
N
N
O N
2
1
4
15
iii
OEt
OH
CF
3
O
O
O
NO2
NO2
NO
2
iv
v
N
N
N
7
17
16
a
Reagents and conditions: (i) HNO , H SO , 0 °C; (ii) NaH, BnBr, DMF; (iii) TFAA, DMF; (iv) NaOH
3
2
4
aq. 20%, 70 °C; (v) 1. SOCl reflux; 2. EtOH, r.t.
2
ACS Paragon Plus Environment

Page 9 of 84
Journal of Medicinal Chemistry
1
2
3
The 5ꢀtrifluoromethylindole 18 was prepared oneꢀpot from 2ꢀiodoꢀ4ꢀtrifluoromethylaniline and ethyl
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
9
acetoacetate, according to the Suzuki’s protocol (Scheme 2). The resulting ester 18 was then alkylated
at Nꢀ1 position by simply treatment with sodium hydride and benzyl bromide in DMF.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
a
Scheme 2
a
29
Reagents and conditions: (a) NaH, CuI, DMF; (b) NaH, BnBr, DMF.
The hydroxyindole 9 was the common precursor of the target indole derivatives 11-13. The substituted
5ꢀhydroxyindole 9 was synthesized according to a modified Nenitzescu reaction by refluxing 1,4ꢀpꢀ
30
benzoquinone, zinc chloride, and aminocrotonate 19 in dichloromethane. 19 was prepared from
condensation of ethyl acetoacetate and benzylamine (Scheme 3).
Thus, the reaction of 9 with acetyl chloride in acetone in the presence of pyridine gave ester 10; in
addition, reaction of 9 with 2ꢀphenethylbromide in DMF at 50 °C, in the presence of K CO , gave
2
3
phenethoxyindole 11. The amino derivative 12 was obtained from 9 by nucleophilic substitution on 2ꢀ
chloroꢀN,Nꢀdimethylamine hydrochloride in the presence of K CO ; eventually, it gave the hydroxyl
2
3
ether 13 in 70 % yield by reaction with 1,2ꢀepoxybutane in DMF (Scheme 3).
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 84
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
a
Scheme 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
a
Reagents and conditions: (i) Ethyl acetoacetate, pꢀTsOH, benzene reflux; (ii) pꢀbenzoquinone, ZnCl2,
30
CH Cl reflux; (iii) acethyl chloride, pyridine 45 °C; (iv) 2ꢀphenylethylbromide, K CO , DMF 50 °C;
2
2
2
3
(
v) 2ꢀchloroꢀN,Nꢀdimethylamine, K CO , ethanol/toluene 1:1, reflux; (vi) 1ꢀbutene oxide, NaH, DMF,
2 3
100 °C.
Oꢀalkylation of 5ꢀhydroxyindole 9 afforded halo intermediates 20a-c, which were converted into the
secondary or tertiary amino compounds 21-31 (for further details see Experimental section); the halo
intermediates 20a-c were also used to prepare the azides 32a-c, which were converted by
triphenylphosphine in THF/water mixture into the final primary amines 33ꢀ35 (Scheme 4).
ACS Paragon Plus Environment

Page 11 of 84
Journal of Medicinal Chemistry
1
2
3
4
5
a
Scheme 4
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
a
31
Reagents and conditions: (i) suitable dibromoꢀ or bromochloroꢀalkane, K CO , ethanol reflux; (ii) 1.
2
3
31
32
NaI, acetonitrile reflux; 2. R R NH, K CO , reflux; (iii) NaN , DMF 80 °C; (iv) PPh , THF/H O
1
2
2
3
3
3
2
33
r.t;
Oꢀalkylation of the hydroxyindole 9 with epibromohydrin afforded the epoxide 36, which was
converted into the desired amino alcohols 37-39 by reaction with the proper aliphatic secondary
amines. In particular, primary amino alcohols 41-42 were obtained passing through the azido
intermediate 40 which gave amino alcohol 41 by reduction with tin chloride or its regioisomer 42 using
triphenylphosphine in tetrahydrofuran/water mixture (Scheme 5).
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 84
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
a
Scheme 5
OEt
OEt
OEt
O
O
O
OH
O
OH
O
O
NR R
1 2
i
ii
N
N
N
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
9
36
37-39
iii
OEt
OEt
OEt
O
O
OH
O
NH2
OH
O
N3
O
OH
O
NH2
v
iv
N
N
N
41
40
42
a
34
Reagents and conditions: (i) epibromohydrin, K CO , acetone reflux; (ii) LiClO , R R NH,
2
3
4
1
2
35
36,37
.
2
38
acetonitrile reflux; (iii) NaN , LiClO , acetonitrile reflux;
(iv) SnCl 2 H O, MeOH, r.t; (v)
3
4
2
3
3
PPh , THF/H O r.t.
3
2
Three fluorinated compounds were also investigated (Scheme 6); in particular, upon exposure to
DAST, hydroxyl compound 37 was converted into two regioisomers 43ꢀ44. In addition, the NaI
mediated reaction of chloro derivative 20b gave difluorinated compound 45 after treatment with 4,4ꢀ
difluoropiperidine in the presence of K CO as a base.
2
3
a
Scheme 6
ACS Paragon Plus Environment

Page 13 of 84
Journal of Medicinal Chemistry
1
2
3
OEt
OEt
OEt
O
OH
O
F
O
CHF2
N
4
O
N
O
N
O
5
6
7
8
9
i
N
N
+
N
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
3
7
43
44
F
OEt
OEt
F
O
O
O
Cl
O
N
ii
N
N
2
0b
45
a
Reagents and conditions: (i) DAST, DCM, 0 °C; (ii) 1. NaI, acetonitrile reflux; 2. 4,4ꢀ
difluoropiperidine hydrochloride, K CO , reflux.
2
3
Ethyl 5ꢀhydroxyꢀ2ꢀmethylꢀ1Hꢀindoleꢀ3ꢀcarboxylate 46 was obtained according to Nenitzescu’s method
3
9
and further converted into final amine 47 (Scheme 7). From the (1H)ꢀindole 47, a phenethyl
derivative 48 was obtained by treatment with phenethyl bromide in the presence of NaH.
a
Scheme 7
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 84
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
a
39
31
Reagents and conditions: (i) AcOH, r.t.; (ii) 3ꢀchloroꢀ1ꢀbromopropane, K CO , ethanol reflux; (iii)
2
3
3
1
1
. NaI, acetonitrile reflux; 2. piperidine, K CO , reflux. (iv) NaH, phenethyl bromide, DMF, 50 ºC.
2 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
40
Finally, different substituted benzylindoles 49 were synthesized by CAN induced cyclization or by
4
1
triethylamine catalysis (Scheme 8). Intermediates 49 were converted, similarly to hydroxyindole 9, to
4
1
final compounds 50-63. Nitro derivatives were obtained by a different route or triethylamine
a
Scheme 8
a
40
Reagents and conditions: (i) pꢀbenzoquinone, ethyl acetoacetate, CAN, EtOH, r.t.; (ii) 1. ethyl
41
acetoacetate, Et N, r.t. 2. pꢀbenzoquinone, CH NO r.t.; (iii) 3ꢀchloroꢀ1ꢀbromopropane, K CO ,
3
3
2,
2
3
3
1
31
ethanol, reflux; (iv) 1. NaI, acetonitrile reflux; 2. piperidine, K CO , reflux.
2
3
The strategy adopted for the synthesis of the aminoether 69 is illustrated below (Scheme 9). The Fisher
condensation of pꢀmethoxyphenylhydrazine with ethyl levulinate in acetic acid gave the ethyl
indolacetate 64 which was transformed into the corresponding Nꢀbenzyl derivative 65 by reaction with
NaH and benzyl bromide. Treatment of the ester 65 with BBr in dichloromethane at ꢀ78 °C afforded
3
ACS Paragon Plus Environment

Page 15 of 84
Journal of Medicinal Chemistry
1
2
3
the hydroxyacid 66 which was converted in the hydroxyester 67 by ethanolysis of corresponding acyl
chloride. Treatment of the ester 67 with 3ꢀchloroꢀ1ꢀbromopropane in refluxing acetone, in the presence
of K CO gave the chloro derivative 68 which, in turn, was converted into the target aminoether 69 as
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
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
previously reported.
a
Scheme 9
2
EtO C
2
HO C
NHNH2
OCH3
OH
EtO C
2
i
ii
iii
OCH
3
.
N
N
HCl
N
H
OCH3
64
6
5
66
iv
EtO C
2
EtO C
2
2
EtO C
O
Cl
OH
O
N
vi
v
N
N
N
69
68
67
a
Reagents and conditions: (i) ethyl levulinate, AcONa, AcOH, reflux; (ii) NaH, BnBr DMF; (iii) BBr3,
3
1
DCM, ꢀ78 °C; (iv) 1. SOCl ; 2. EtOH; (v) 3ꢀchloroꢀ1ꢀbromopropane, K CO , ethanol, reflux; (vi) 1.
2
2
3
3
1
NaI, acetonitrile reflux; 2. piperidine, K CO , reflux.
2
3
Aminoether 73, was obtained by decarboxylation of 46 and further converted into final amine 72
Scheme 10). From this 1Hꢀindole, the benzyl derivative 73 was obtained by treatment with benzyl
(
bromide in the presence of NaH.
a
Scheme 10
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 84
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
a
39
31
Reagents and conditions: (i) 2N HCl, reflux; (ii) 3ꢀchloroꢀ1ꢀbromopropane, K CO , ethanol, reflux;
2
3
31
(
iii) 1. NaI, acetonitrile reflux; 2. piperidine, K CO , reflux. (iv) NaH, BnBr DMF.
2 3
Finally, indoleꢀ3ꢀcarboxylic acid 76 was obtained by hydrolysis of 9 and further converted into final
amine 76 (Scheme 11).
a
Scheme 11
ACS Paragon Plus Environment

Page 17 of 84
Journal of Medicinal Chemistry
1
2
3
O
EtO
HO2C
HO C
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
OH
O
Cl
OH
ii
i
N
N
N
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
9
74
75
iii
HO C
2
O
N
N
7
6
a
42
Reagents and conditions: (i) 1. 2N NaOH, tetrahydrofurane, reflux; 2. 2N HCl; (ii) 3ꢀchloroꢀ1ꢀ
3
1
31
bromopropane, K CO , ethanol, reflux; (iii) 1. NaI, acetonitrile reflux; 2. piperidine, K CO , reflux.
2
3
2
3
Biology
Ethidium bromide efflux inhibition
All 48 synthesized compounds were evaluated for their ability to inhibit efflux of ethidium bromide,
43
(
EtBr) by SAꢀ1199B. This strain overexpresses norA, resulting in an avid efflux phenotype. EtBr was
used because it is an excellent NorA substrate and has convenient fluorescent properties that allow
qualitative monitoring of its intracellular concentration. The efflux inhibition assay employed a realꢀ
44
time fluorometric approach essentially as described previously. SAꢀ1199B was grown overnight in
cationꢀsupplemented MuellerꢀHinton broth (SMHB, BD Biosciences, Sparks, MD) and then was
diluted 25ꢀfold into fresh SMHB. Cultures were incubated at 35 °C with shaking until an optical
density at 600 nm (OD600) of 0.7–0.8 was achieved. Cells were then pelleted and reꢀsuspended at
OD600 = 0.8 in 0.5 mL aliquots of SMHB containing EtBr plus carbonyl cyanide mꢀ
chlorophenylhydrazone to “load” cells with EtBr (final concentrations, 25 and 100 µM, respectively).
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 84
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
Following gentle agitation for 20 min at room temperature, cells were pelleted and stored on ice. Pellets
were warmed at room temperature for 5 min and then reꢀsuspended in 1 mL of fresh SMHB and 200
µL aliquots were immediately transferred into the wells of opaque 96ꢀwell flatꢀbottom plates (Corning
Inc., Corning, NY) containing or lacking a 50 µM concentration of each test compound. Fluorescence
was monitored continuously using a BioTek FLx800 microplate reader (BioTek Instruments Inc.,
Winooski, VT) at excitation and emission wavelengths of 485 nm and 645 nm, respectively, for 5 min.
Experiments were performed in triplicate with two technical replicates per biological replicate.
Efflux activity of SAꢀ1199B was expressed as percent fluorescence decrease over a 5ꢀmin time course.
Inhibition of this efflux by test compounds was determined using the equation [efflux in the absence]ꢀ
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
[
efflux in the presence of test compound]/[efflux in absence of test compound] × 100, giving the
percent efflux inhibition observed. If a 50 µM concentration of test compound achieved at least 80%
efflux inhibition, a series of concentrations were tested to quantify potency by determining the 50%
inhibitory concentration (IC ). IC50 determinations also were performed for selected compounds
50
having less than 80% efflux inhibition.
Intrinsic antimicrobial activity
In addition to IC determinations, compounds with at least 80% efflux inhibition were evaluated for
5
0
intrinsic antimicrobial activity. This was accomplished by determining minimal inhibitory
concentrations (MIC) employing a microdilution method according to Clinical and Laboratory
4
5
Standards Institute (CLSI) guidelines. The antimicrobial activity of selected compounds having less
than 80% efflux inhibition were also determined. Compounds having an MIC value greater than 100
µg/ml were considered antimicrobially inactive. In these instances, efflux inhibitory activity was
considered to be unrelated to any direct antimicrobial effect on the test strain.
ACS Paragon Plus Environment

Page 19 of 84
Journal of Medicinal Chemistry
1
2
3
Lead optimization
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
Based on data in Table 1 (to be discussed in detail subsequently), the most promising EPIs were the
dimethylamino ether 12 and the hydroxylether 13, inhibiting EtBr efflux by 98.2 and 71.1 %,
respectively, at the 50 µM screening concentration. Indeed, the EWG group in the C5 position appears
nonꢀessential to preserve the inhibitory effect, and the target seems to be able to accommodate larger
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
8,19
compounds with respect to data previously reported in the literature.
Thus, we decided to perform additional structural modifications involving both basic and acid
functionalities. In particular, we decided to synthesize and test a novel series of compounds with a twoꢀ
,
three or fourꢀcarbon atom linker carrying different amino groups. A series of compounds having the
hydroxylic group together with different amines was also explored and, finally, a compound presenting
the amine and hydroxylic group in inverted positions was synthesized (Figures 3 and 4).
Figure 3. Scheme of the modulation of the indoleꢀbased compounds reported in this study. Blue
spheres represent basic centers, while red spheres represent weak acid centers.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 84
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
Subsequently, different spacers (twoꢀ, threeꢀ and four carbons) were used to evaluate the appropriated
distance between the aromatic portion and the basic functionality. As well as the length of the chain,
the different terminal amines were investigated; the aim was to determine the effect of a primary amine
compared to the tertiary one. Furthermore fiveꢀ or sixꢀmembered rings were used, and additional
hydrogen bond donors (HBD) or acceptors (HBA) were introduced (e.g. using morpholine or
piperazine instead of piperidine). As a refinement three fluorinated derivatives, together with indoles
bearing different benzyl groups or substituted at C3, were synthesized and investigated (Figure 4).
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
Even though the mentioned compounds differ for slight modifications in the NMe replacement, the
2
basic pK values of the novel amine groups (see Table 2) fall in a broad range from 4.12 (28) to 10.35
a
(35). For example, a morpholinic moiety (compounds 22, 27, and 39) has a notable effect on the pKa
value when compared with a piperidine group (derivatives 21, 25, and 37). In addition, a piperazine
moiety might be important to modulate the interaction with the targeted protein due to the possible
protonation of both nitrogen atoms. Similarly, the LogP (see Table 2) is also heavily affected by the
proposed modifications, ranging from 2.85 (41) to 6.45 (26). In general, the greater the spacer and the
ring size, the higher the liphophilicity. The presence of fluorinated rings or chains also have an impact
on the basicity of the amino group, with compound 45 being one of the less basic in our series.
ACS Paragon Plus Environment

Page 21 of 84
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 4. Detailed structural modification of lead compound 12 involving the replacement of the
terminal dimethylamine moiety (block I), ethyl linker (block II), benzyl group (block III), and ethyl
carboxylate (block IV).
RESULTS AND DISCUSSION
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 84
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
All new indoleꢀbased compounds were evaluated using EtBr efflux inhibition assays and MIC
determinations. These data are provided in Tables 2 and 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
Table 2. Side chain effect on inhibition of EtBr efflux in S. aureus 1199B
% EtBr
a
b
efflux
inhibition
MIC
(µg/ml)
IC50
(µM)
Comp.
X
LogP
pKa
@
50 µM
2
1
2
5.54
4.11
3.79
4.34
6.03
6.45
4.60
4.83
4.28
4.77
8.82
6.83
100
>100
>100
>100
>100
>100
>100
>100
>100
11.6
16.8
15.5
10.1
6.0
2
90.9
99.2
99.7
98.5
77.4
98.6
100
23
24
4.64; 9.51
4
8
.14;
.71
25
26
27
9.59
9.49
7.60
c
ND
10.6
11.9
12.5
ND
4
.12;
28
8
.69
5
.23;
2
9
0
100
9
.80
100
5
.23;
3
71.9
>100
9.80
31
5.98
8.81
45.4
ND
ND
ACS Paragon Plus Environment

Page 23 of 84
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
3
3
3
3
4
5
3.49
3.98
4.47
4.90
9.07
10.13
10.35
67.7
100
ND
>100
100
ND
12.5
6.7
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
93.7
99.3
14.46*;
37
>100
5.8
8
.91
14.46*;
3
8
9
5.32
92.8
100
6.8
8.80
OH
O
1
1
1
4.46*;
6.92
3
3.47
2.85
2.96
5.41
99
100.0
100
>100
>100
>100
>100
17.0
17.1
10.4
5.1
O
N
3.74*;
41
9
.34
3.81*;
.31
4
2
3
8
4
7.91
7.29
96.7
44
5.32
38.5
>100
ND
45
6.09
5.50
25.9
ND
ND
26
24,25
a) predicted by Volsurf; b) predicted by Moka;
c) ND, not determined; *relative to the acidity of
OH group.
Looking at the activities shown in Table 1 and 2, it appears that while the hydroxylic group is not
essential the terminal amine moiety is fundamental for activity retention. Thus, all other compounds
having a different terminal amino group (for example 37ꢀ39 and 41), containing or lacking the
hydroxyl group in C2’ position (26ꢀ29, 34), and having a shorter (2 carbon atoms, 12, 21-24) or longer
spacer (4 carbon atoms, 30 and 35) resulted very strong inhibitors (% inhibition > 89 %). Among the
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 84
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
different acyclic (dimethylamine 12, or amino 33-35, and 41-42) and cyclic amines examined, such as
morpholine (22, 27, 39), azepane (26 and 38), and piperazines (23ꢀ24 and 28ꢀ29), the best results were
achieved with the piperidineꢀcontaining compounds (25 and 37), showing IC values of 6 and 5.8 µM,
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
respectively. In particular, comparing compounds 13 (Table 1), 34 and 41 (Table 2) is possible to
notice that the lacking of the amino group produced an inhibition activity decrease from 100%
(compound 41) to 71.1% (compound 13); while the lacking of the hydroxyl group in compound 34
respect to compound 41 did not affect at all the inhibition ability that is 100%.
Concerning the length of the spacer, three and four carbon atoms seems to positively affect the
inhibitory activity. For example, between the two piperidine containing compounds 25 and 21 the first
(
containing the three carbon atom chain) exhibits a 2ꢀfold IC increase with respect to 21 (containing a
50
two carbons linker). Moreover, trying to increase hydrophobicity by using a benzylamine (such as in
compound 31) was detrimental for EtBr efflux inhibition activity.
In addition, inspection of the three fluorinated analogues of the most active 25 (pKa = 9.60 predicted
2
4,25
by Moka)
revealed that introduction of two fluorine atoms in 4ꢀposition of the piperidine ring
24,25
(
compound 45, pK = 5.54 predicted by Moka)
were detrimental for the inhibitory effect. On the
a
contrary, the presence of one fluorine in the C2’ position of the spacer gave the most active compound
2
4,25
4
3 (pK = 7.92 predicted by Moka,
IC = 5.1 ꢁM). Finally, the fluoromethyl compound 44, which
a
50
24,25
is a regioisomer of compound 43 (pK = 7.28 predicted by Moka,
resulted in a weak inhibitor. The
a
inhibition behavior correlated well with the pK value in that the greater the pK , the higher the
a
a
possibility for the protonated compound to bind the target. However, comparison between the
regioisomers 43 and 44, having similar pK values but bearing a different spacer length, suggests that
a
the interaction with the NorA cavity is sensitive to the distances between the indole scaffold and the
basic center.
ACS Paragon Plus Environment

Page 25 of 84
Journal of Medicinal Chemistry
1
2
3
The importance of Nꢀbenzyl moiety in the piperidineꢀcontaining derivative 25 was also explored (see
Table 3). The replacement performed slightly affected the predicted LogP values (in the range of 4.51
6.97). A benzyl replacement with a phenethyl moiety (48) produced a 2ꢀfold activity decrease based on
IC values, while the substitution with a hydrogen resulted in a compound with much reduced activity
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
̶
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
5
0
indicating that benzyl is more advisable. An attempt to make the analogue of 25 by replacing the
benzyl group with a phenyl ring was made. However, the compound instability prevented further
investigations. Different benzylꢀsubstituted compounds were synthesized and tested by the insertion of
a methyl group in the ortho, meta and para positions (50-52). These changes resulted in less active
compounds in terms of the 50 µM screening concentration, but all were antimicrobially active. This
effect might be attributable to an increased hydrophobicity rather than an increased steric hindrance.
Indeed, methoxyꢀ (53ꢀ55) and nitroꢀsubstituted (61ꢀ63) compounds maintained good EPI without any
appreciable antimicrobial activity. Noteworthy, fluoro derivatives 56ꢀ58 turned out to be very good
candidates (IC values of 4.8
̶
8.7 ꢁM), while their trifluoromethyl counterpart 59ꢀ60 showed a twofold
50
increase in IC values.
5
0
Table 3. Benzyl effect on EtBr efflux inhibition
% EtBr
efflux
inhibition
a
MIC
(µg/ml)
IC50
(µM)
Comp.
Y
LogP
@
50 µM
2
4
4
5
7
8
benzyl
H
6.03
4.51
6.39
98.5
>100
6
b
34.2
99.3
ND
ND
13.7
phenethyl
>100
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 84
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
50
oꢀmethylbenzyl
mꢀmethylbenzyl
pꢀmethylbenzyl
6.41
6.49
6.47
5.89
6.11
6.09
6.10
6.23
6.22
6.97
6.96
5.87
5.97
5.55
54
25
4.8
6.1
6.7
5
5
1
70.9
58.3
87.9
77.1
77.7
89
50
52
25
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
3
oꢀmethoxylbenzyl
mꢀmethoxylbenzyl
pꢀmethoxylbenzyl
oꢀfluorobenzyl
> 100
> 100
> 100
> 100
> 100
> 100
> 100
> 100
> 100
> 100
> 100
54
7.3
7.1
4.8
6.1
8.7
14.2
14.1
6.3
4.4
6.5
5
5
56
5
7
mꢀfluorobenzyl
89.7
91.1
12.9
77.7
91.8
85.1
89.9
58
pꢀfluorobenzyl
5
9
mꢀtrifluoromethylbenzyl
pꢀtrifluoromethylbenzyl
oꢀnitrobenzyl
60
6
1
62
mꢀnitrobenzyl
6
3
pꢀnitrobenzyl
2
6
a) predicted by Volsurf; b) ND, not determined.
Finally, we investigated the effect of modification at the C3 position of the most potent indole 25. In
particular, a further investigation on the role of the ester group by varying the substituent at C3 (69) or
completely removing it (73). Eventually, an acidic moiety (76) was explored (pK = 2.65 predicted by
a
2
4,25
Moka).
ACS Paragon Plus Environment

Page 27 of 84
Journal of Medicinal Chemistry
1
2
3
Table 4. Effect of substitution in C3 position on EtBr efflux inhibition in S. aureus 1199-B
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
% EtBr efflux
a
b
MIC
µg/ml)
IC50
(µM)
Comp.
Z
n
LogP
pKa
inhibition
(
@
50 µM
25
69
73
76
CO
2
Et
1
2
1
1
6.03
6.31
5.55
5.25
9.59
9.49
98.5
>100
100
6
CH
CO
H
Et
98.4
95.6
0
6.1
6.6
ND
2
2
9.59
0.63
2.65 (a);
c
COOH
ND
9
.59
24,25
26
a) predicted by Moka;
b) predicted by Volsurf; ND, not determined.
Data in Table 4 show that removing the ester function at C3 (73) does not affect the % inhibition and
IC values compared to 25 but does strongly increase antimicrobial activity. This antimicrobial activity
5
0
could contribute to the low IC value observed, unrelated to any pump inhibitory effect that may be
5
0
present. Indeed, further studies would be necessary to discriminate between EPI and intrinsic
4
6
antimicrobial activity, as previously reported for Totarol. Similarly, the substitution of the ethyl ester
group with a methyleneꢀethyl ester function did not alter the inhibition properties (69), showing that the
position C3 is not critical for interaction with the NorA pump cavity.
On the contrary, a carboxylic acid in the C3 position had a detrimental effect with a complete loss of
inhibition (76). This data are noteworthy, because the ester functionality usually undergoes hydrolysis
to some extent in a biological environment. Thus, although many drugs possess ester groups to improve
4
7
ADME properties, a valuable test in our study could be the total removal of the ester function making
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 84
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
compound 73 a good candidate for further development although its intrinsic antimicrobial activity
may pose problems to separate the antimicrobial from the EPI activities.
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
Preliminary pharmacokinetic studies
Finally, preliminary ADME studies (see Table 5) were carried out for the best compounds 23, 25, 29,
3
4, 43, 53, 61, and 73 in order to experimentally evaluate their metabolic stability (in vitro assays with
1
48
HLM) and water solubility (by H NMR experiments in D O solution using TSP as internal standard).
2
In addition, permeability in CaCo2 cells has been evaluated in silico (Table 5).
Table 5. ADME study on best compounds.
% substrate
Solubility
b
CaCo2
Permeability
pKa
a
after 30 min
Metabolites analysis
c
(
mM)
b
(HLM)
4
9
.64;
.51
Mꢀ90 (Nꢀdealkylation);
Mꢀ28 (Oꢀdealkylation)
2
2
2
3
3
5
9
4
0.19
72
76
82
87
52
66
89
100
0.81
1.28
0.82
0.46
1.36
1.21
1.00
1.53
Mꢀ90 (Nꢀdealkylation);
Mꢀ28 (Oꢀdealkylation)
9.59
0.023
0.02
5.23;
9.80
Mꢀ90 (Nꢀdealkylation);
Mꢀ28 (Oꢀdealkylation)
Mꢀ90 (Nꢀdealkylation);
Mꢀ28 (Oꢀdealkylation)
10.13
7.91
9.59
9.59
9.59
0.017
0.020
0.025
0.014
0.070
Mꢀ90 (Nꢀdealkylation),
Mꢀ68 (Nꢀdealkylation)
43
Mꢀ120 (Nꢀdealkylation);
Mꢀ14 (Oꢀdemethylation)
53
61
73
stable
stable
ACS Paragon Plus Environment

Page 29 of 84
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
48
a) evaluated by HꢀNMR in D O solution (containing 0.4% of DMSOꢀd ) at 25 °C; b) see
2
6
experimental section for further details; c) predicted by Volsurf (the value +1 is assigned to
ꢀ6
ꢀ1 26
permeable compounds, with Papp. ≥ 8 x 10 cm s ).
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
All tested indoles exhibit a medium to high metabolic stability. While compound 23, 25, 29, 34, 43,
and 53 underwent Nꢀdealkylation (loss of the benzyl group), nitrobenzylindole 61 was quite stable with
no metabolite detected. This finding suggests that the presence of an electronꢀwithdrawing group (such
a nitro group) enhances the compound metabolic stability. On the contrary, the presence of electronꢀ
donor group (i.e. 53) induced a slight metabolic stability drop, also due to the presence of a further
metabolic reaction by direct involvement of the methoxy group. To be noted that 25 is fully protonated
under physiologic conditions. The different behavior shown by fluorinated compound 43 with respect
49
to its analogue 25 has been rationalized with the aid of MetaSite, a software predicting the Site of
Metabolism (SoM) of a compound with different CYP450 using reactivity and molecular interaction
4
9
fields information. According to MetaSite, as reported in Figure 5a the charged piperazinyl group of
2
5 is linked to Glu183 thus the ester group and the benzyl group (the two experimental sites of
oxidation) are in close proximity of the heme. The Oꢀdealkylation is faster while the benzylic oxidation
leading to Nꢀdealkylation is slower, but both contribute to the parent disappearance. A fluorine atom β
to the amino group (43, Figure 5b), makes the pK of the piperazinyl group drop significantly leading
a
to a majority of neutral form. The CYP recognition is now different, and the main recognition
anchorage is produced by the benzyl group (that was the less reactive in molecule 25). This binding
mode makes the exposure of the piperazinyl group towards the heme more favourable, leading to a fast
oxidation. The molecular stability decreases and the main metabolite of 43 becomes the Mꢀ68 (Nꢀ
dealkylation twice in the ring).
ACS Paragon Plus Environment