Novel, potent, orally bioavailable and selective mycobacterial ATP synthase inhibitors that demonstrated activity against both replicating and non-replicating M. tuberculosis

Article abstract of DOI:10.1016/j.bmc.2014.12.060

The mycobacterial F₀F₁-ATP synthase (ATPase) is a validated target for the development of tuberculosis (TB) therapeutics. Therefore, a series of eighteen novel compounds has been designed, synthesized and evaluated against Mycobacterium smegmatis ATPase. The observed ATPase inhibitory activities (IC₅₀) of these compounds range between 0.36 and 5.45 μM. The lead compound 9d [N-(7-chloro-2-methylquinolin-4-yl)-N-(3-((diethylamino)methyl)-4-hydroxyphenyl)-2,3-dichlorobenzenesulfonamide] with null cytotoxicity (CC₅₀ >300 μg/mL) and excellent anti-mycobacterial activity and selectivity (mycobacterium ATPase IC₅₀ = 0.51 μM, mammalian ATPase IC₅₀ >100 μM, and selectivity >200) exhibited a complete growth inhibition of replicating Mycobacterium tuberculosis H37Rv at 3.12 μg/mL. In addition, it also exhibited bactericidal effect (approximately 2.4 log₁₀ reductions in CFU) in the hypoxic culture of non-replicating M. tuberculosis at 100 μg/mL (32-fold of its MIC) as compared to positive control isoniazid [approximately 0.2 log₁₀ reduction in CFU at 5 μg/mL (50-fold of its MIC)]. The pharmacokinetics of 9d after p.o. and IV administration in male Sprague-Dawley rats indicated its quick absorption, distribution and slow elimination. It exhibited a high volume of distribution (V_ss, 0.41 L/kg), moderate clearance (0.06 L/h/kg), long half-life (4.2 h) and low absolute bioavailability (1.72%). In the murine model system of chronic TB, 9d showed 2.12 log₁₀ reductions in CFU in both lung and spleen at 173 μmol/kg dose as compared to the growth of untreated control group of Balb/C male mice infected with replicating M. tuberculosis H37Rv. The in vivo efficacy of 9d is at least double of the control drug ethambutol. These results suggest 9d as a promising candidate molecule for further preclinical evaluation against resistant TB strains.

Full text of DOI:10.1016/j.bmc.2014.12.060

Bioorganic & Medicinal Chemistry xxx (2015) xxx–xxx
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
Novel, potent, orally bioavailable and selective mycobacterial ATP
synthase inhibitors that demonstrated activity against both
replicating and non-replicating M. tuberculosis q
Supriya Singh a, Kuldeep K. Roy a, Shaheb R. Khan b, Vivek Kr. Kashyap b, Abhisheak Sharma d,
Swati Jaiswal d, Sandeep K. Sharma b, Manju Yasoda Krishnan b, Vineeta Chaturvedi c, Jawahar Lal d,
Sudhir Sinha c, Arnab D. Gupta b, Ranjana Srivastava b, Anil K. Saxena a,
⇑
a Division of Medicinal and Process Chemistry, CSIR-Central Drug Research Institute, B.S. 10/1, Sector 10, Jankipuram Extension, Sitapur Road, Lucknow 226031, India
b Microbiology Division, CSIR-Central Drug Research Institute, B.S. 10/1, Sector 10, Jankipuram Extension, Sitapur Road, Lucknow 226031, India
c Division of Drug Target Discovery and Development, CSIR-Central Drug Research Institute, B.S. 10/1, Sector 10, Jankipuram Extension, Sitapur Road, Lucknow 226031, India
d Pharmacokinetics and Metabolism Division, CSIR-Central Drug Research Institute, B.S. 10/1, Sector 10, Jankipuram Extension, Sitapur Road, Lucknow 226031, India
a r t i c l e i n f o
a b s t r a c t
Article history:
The mycobacterial F0F1-ATP synthase (ATPase) is a validated target for the development of tuberculosis
(TB) therapeutics. Therefore, a series of eighteen novel compounds has been designed, synthesized and
evaluated against Mycobacterium smegmatis ATPase. The observed ATPase inhibitory activities (IC50) of
Received 8 September 2014
Revised 24 December 2014
Accepted 26 December 2014
Available online xxxx
these compounds range between 0.36 and 5.45
olin-4-yl)-N-(3-((diethylamino)methyl)-4-hydroxyphenyl)-2,3-dichlorobenzenesulfonamide] with null
cytotoxicity (CC50 >300 g/mL) and excellent anti-mycobacterial activity and selectivity (mycobacterium
ATPase IC50 = 0.51 M, mammalian ATPase IC50 >100 M, and selectivity >200) exhibited a complete
growth inhibition of replicating Mycobacterium tuberculosis H37Rv at 3.12 g/mL. In addition, it also
exhibited bactericidal effect (approximately 2.4 log10 reductions in CFU) in the hypoxic culture of non-
replicating M. tuberculosis at 100 g/mL (32-fold of its MIC) as compared to positive control isoniazid
[approximately 0.2 log10 reduction in CFU at 5 g/mL (50-fold of its MIC)]. The pharmacokinetics of 9d
lM. The lead compound 9d [N-(7-chloro-2-methylquin-
l
Keywords:
l
l
Tuberculosis
ATP synthase
Quinoline
Sulfonamide
Dormancy
l
l
l
after p.o. and IV administration in male Sprague–Dawley rats indicated its quick absorption, distribution
and slow elimination. It exhibited a high volume of distribution (Vss, 0.41 L/kg), moderate clearance
(0.06 L/h/kg), long half-life (4.2 h) and low absolute bioavailability (1.72%). In the murine model system
of chronic TB, 9d showed 2.12 log10 reductions in CFU in both lung and spleen at 173 lmol/kg dose as
compared to the growth of untreated control group of Balb/C male mice infected with replicating M.
tuberculosis H37Rv. The in vivo efficacy of 9d is at least double of the control drug ethambutol. These
results suggest 9d as a promising candidate molecule for further preclinical evaluation against resistant
TB strains.
Ó 2015 Published by Elsevier Ltd.
1. Introduction
immunodeficiency virus (HIV) epidemic proven to increase the risk
of developing active TB dramatically, (ii) the increasing emergence
Tuberculosis, primarily caused by Mycobacterium tuberculosis, is
one of the globally leading infectious disease that continues as glo-
bal epidemic with more than 9 million new cases and nearly 2 mil-
of multi-drug resistant TB (MDR-TB), extensively drug-resistant TB
(XDR-TB) and a recently identified TB state called total drug-resis-
tant TB (TDR-TB).1–5 Current TB chemotherapy is mainly based on
drugs that inhibit bacterial metabolism, and is characterized by
efficient bactericidal and extremely weak sterilizing activities.6
The dormant or latent or metabolically inactive bacilli are tolerant
to the anti-TB drugs that target cell division.7,8 Although an
extended chemotherapy may be partially successful in eradicating
such dormant bacilli, drug(s) with efficient sterilizing activity are
of current demand for properly tackling the dormant bacilli. In
such scenario, targeting the particular enzyme that are essential
1
lion deaths every year. In addition, over 2 billion people harbor
latent TB infection (LTBI), thus representing an enormous reservoir
of M. tuberculosis that can subsequently progress to active TB.1 The
main obstacles to the global TB control are: (i) the human
q
CDRI communication number allotted to this paper is 8882.
⇑
Corresponding author. Tel.: +91 (522) 2624273; fax: +91 (522) 2623938.
E-mail address: anilsak@gmail.com (A.K. Saxena).
http://dx.doi.org/10.1016/j.bmc.2014.12.060
0968-0896/Ó 2015 Published by Elsevier Ltd.
Please cite this article in press as: Singh, S.; et al. Bioorg. Med. Chem. (2015), http://dx.doi.org/10.1016/j.bmc.2014.12.060

2
S. Singh et al. / Bioorg. Med. Chem. xxx (2015) xxx–xxx
for the survival of both actively growing and latent/dormant bacilli,
is one of the most effective target-based drug discovery approach.2
The mycobacterial F0F1-ATP synthase (F0F1-ATPase) is a vali-
dated anti-TB target, which is responsible for the production of
adenosine 50-triphosphate (ATP) for energy homeostasis in differ-
ent mycobacterial species.9 Bedaquiline (TMC207, Sirturo™) is a
member of the diarylquinoline class of anti-TB drugs targeting this
new enzyme and has been approved by the U.S. Food and Drug
Administration (US-FDA) specifically for the treatment of MDR-
TB. This drug has shown promising efficacy against both the
drug-sensitive and drug-resistant TB, and has remarkable potential
for restricting the duration of TB treatment due to the fact that this
drug is bactericidal to both the replicating as well as non-replicat-
ing (dormant) bacilli.9–12 Bedaquiline (Sirturo™) has been reported
by the World Health Organization (WHO) to disturb the function of
the heart and liver in particular. Common observed side effects
include nausea, joint and chest pain, and headache.13 Bedaquiline
is proven to be a weak hERG (human Ether-à-go-go) potassium
(SAR) studies of these novel compounds proven as low micromolar
inhibitors of the mycobacterial F0F1-ATP synthase. Among the syn-
thesized series of 18 compounds, two lead compounds were bio-
logically explored in greater detail where one representative
compound was proven as an orally bioavailable and selective
mycobacterium ATP synthase inhibitor that demonstrated approx-
imately 2.12 log10 reductions in CFU in both lung and spleen at a
dose of 173 lmol/kg as compared to the growth in the untreated
control group of Balb/C male mice infected with M. tuberculosis
H37Rv. Like the drug bedaquiline (Sirturo™, TMC207), this com-
pound acts through a novel mechanism of action (ATP synthase),
and is active against both the replicating as well as non-replicating
M. tuberculosis H37Rv. This lead compound could be considered as
candidate molecule for further preclinical evaluation that could
lead to an effective TB therapeutics.
2. Materials and methods
(K+) channel blocker (in vitro IC50 = 0.2
lg/mL) that can cause pro-
2.1. Chemicals, reagents, culture media and test compounds
longation of the heart QT interval (the time between the start of
the Q wave and the end of the T wave in the heart’s electrical
cycle), thus disturbing normal function of the heart.14 The observed
efficacy of bedaquiline for the treatment of MDR-TB validates the
target mycobacterium F0F1-ATPase, and the observed adverse
events with this drug suggest further exploration of new chemical
entities as selective F0F1-ATPase inhibitors lacking any potential
adverse effects.
All of the chemicals, culture media, and reagents were pur-
chased from common commercial suppliers. Solvents were puri-
fied and dried by standard procedures, when required.
Chromatographic separations of the synthesized intermediates
and title compounds were performed on silica gel (Merck: 100–
200 mesh). Thin layer chromatography was used to monitor the
progress and/or completion of the reactions. Melting points
(uncorrected) were determined with Büchi 510 apparatus. Charac-
terization of the synthesized compounds was accomplished in the
sophisticated analytical instrument facility (SAIF) department of
CSIR Central drug Research Institute, Lucknow (India). The IR spec-
troscopy was carried out using Perkin-Elmer 881 spectrophotome-
ter and the values are expressed as vmax cmꢀ1. Mass spectra (MS)
were recorded on a Jeol (Japan) SX 102/DA-6000 Mass Spectrome-
ter. 1H NMR spectra were recorded on Bruker Spectrospin spec-
trometer at 300 MHz. The chemical shifts are reported in d scale
(ppm) and are relative to tetramethyl silane (TMS) as internal stan-
dard. The coupling constants J are given Hertz and spin multiplic-
ities are expressed s (singlet), d (doublet), t (triplet), dd (double
doublet), and m (multiplet). The purity of final compounds was
determined to be >95% by HPLC (high performance liquid
chromatography).
Structurally, the mycobacterial F0F1-ATPase is composed of a
membrane-embedded F0 portion comprising a1b2c10–15 subunits
and a hydrophilic F1 portion consisting of
a .
3b3de 15 Proton (H+)
migration through F0 triggers the rotation of rotary ring formed
by oligomeric subunits ‘c’ which is coupled to the rotation of the
‘c’ subunit resided within the (ab)3 hexamer of F1. This biochemi-
cal phenomenon thus drives the synthesis of ATP as a source of
energy for mycobacterium species. A high-resolution crystal struc-
ture of mycobacterial ATP synthase or its subunits is lacking till
date. Computational and biological studies have shown that
TMC207 binds at the interface of subunit ‘a’, and rotary subunits
‘c’.15–17 This drug is biochemically shown to interfere with the
rotary movement of subunit ‘c’ by mimicking a conserved basic
residue Arg186 in the subunit ‘a’ needed for the proton transfer
with Glu61 of the subunit ‘c’.16,17 Through detailed computational
studies, de Jonge and colleagues16 reported the chiral N,N-dimeth-
ylpropanol substructure of TMC207 to exhibit H-bond formation
with the carboxylate side chain of Glu61, and a favorable location
of quinoline substructure at the interface of subunits ‘a’ and ‘c’ that
might lead to blockade of the rotation of rotary ring, and thus nor-
mal interfering with the proton transfer cascade process. Haagsma
et al.15 have recently characterized the interaction between
TMC207 and mycobacterial ATP synthase using biochemical assays
and binding studies, where they have provided experimental sup-
port for ability of TMC207 to mimic a key residue Arg186 involved
in the proton transfer cascade, and thus blocking rotary movement
of subunits ‘c’ of the mycobacterial F0F1-ATP synthase.15,16
2.2. Bacterial strains and growth conditions
M. tuberculosis H37Rv was maintained on Lowenstein Jensen
slopes and freshly subcultured in the Middlebrook 7H9 medium
(BD) containing 10% oleic acid-albumin-dextrose-catalase (OADC)
and 0.05% Tween 80 maintained at 37 °C with shaking for
14–15 days to attain mid exponential phase (OD600 ꢁ0.6).
Mycobacterium smegmatis TMC 607 was grown in the Luria Bertani
broth containing 0.05% Tween 80 and 0.2% glycerol. These cultures
were used for downstream experiments.
In view of the success of mycobacterial F0F1-ATP synthase as
validated target for TB chemotherapy, we utilized the state-of-
the-art medicinal chemistry approach for the identification of
novel inhibitors of mycobacterial F0F1-ATP synthase. This effort
led to invention of quinoline class of aryl-sulfonamides as potent,
orally bioavailable and selective mycobacterial ATP synthase inhib-
itors that we have patented recently.18 We have recently reported
a short report on the biological screening of these compounds
against Mycobacterium smegmatis ATP synthase along with
in vitro testing against the dormant (non-replicating) M. tuberculo-
sis H37Rv.12 Herein, we report a full-length description of the
rational design, synthesis and structure-activity relationship
2.3. Preparation of inverted membrane vesicles
Inverted membrane vesicles of M. smegmatis were prepared
according to methods described elsewhere.10,12 The reversed mem-
brane fraction was resuspended in an appropriate volume of
50 mM MOPS buffer (pH 7.5) containing 2 mM MgCl2 and 1 mM
PMSF. The protein content was estimated using the Bradford
method. The vesicle preparation was stored as aliquots at ꢀ80 °C.
For the assay, membrane vesicles were diluted to a concentration
of 100
lg protein/mL with 50 mM MOPS buffer (pH 7.5) containing
10 mM MgCl2.
Please cite this article in press as: Singh, S.; et al. Bioorg. Med. Chem. (2015), http://dx.doi.org/10.1016/j.bmc.2014.12.060

S. Singh et al. / Bioorg. Med. Chem. xxx (2015) xxx–xxx
3
2.4. Assay for inhibition of mycobacterial ATP synthase
as indicator for oxygen depletion. The vials were incubated at 37 °C
on a magnetic stirrer with slow stirring for 20–22 days or until
methylene blue was decolorized. The two standard drugs isoniazid
The inhibition of ATP synthesis was measured according to
methods described elsewhere.10,12 The membrane vesicles were
preincubated with test compounds at two fold serial dilutions
(INH) and rifampicin (RIF) were tested at 0.5 and 5
thioridazine at 5 & 50 g/mL, metronidazole at 50
DCCD at 100 M. The test compounds were tested at the two doses
10 and 100 g/mL. All the drugs as well as compounds were tested
lg/mL, while
l
lg/mL and
ranging from 50
l
M to 0.39
l
M or N,N0-dicyclohexylcarbodiimide
M under stirring con-
l
l
(DCCD) at 100 M or isoniazid (INH) at 0.3
l
l
dition at room temperature for 10 min. Subsequently 2.5 mM
NADH was added and further incubated with vigorous shaking
for 1 min. The reaction was started by addition of 1 mM ADP and
10 mM potassium phosphate. After incubation for 1 h, aliquots
(0.05 mL) were added to 0.2 mL of stock solution (2 mM EDTA,
1% trichloracetic acid). 0.005 mL of this mixture was added to
0.1 mL tris acetate buffer (100 mM Tris, 2 mM EDTA, pH 7.75) in
a 96 well plate. After addition of 0.1 mL luciferase reagent (Pro-
mega Bactiter-Glo), luminescence was measured using a lumino-
meter (BMG POLARstar Galaxy Microplate reader). The
concentration causing 50% inhibition of ATP synthesis (IC50) was
determined after plotting concentration values of test sample(s)
against percent inhibition of ATP synthesis.
in duplicates. Appropriate concentration of the test compound/
standard drug was added to respective vials by injection through
septa and the vials were further incubated at 37 °C for 4 days.
The total CFU counts of serial dilutions were performed on Middle-
brook 7H10 (MB7H10) agar plates.
2.8. Inhibition of mitochondrial ATP synthase
Mouse liver mitochondria were isolated using the protocol
described by Frezza et al.21 In brief, an adult Swiss mouse
(3 months old) was starved overnight and then euthanized by cer-
vical dislocation prior to dissecting out the liver. The liver was
minced and homogenized (glass–teflon potter at 1600 rpm, 4
strokes) in ice cold buffer (200 mM sucrose, 10 mM Tris–MOPS
and 1 mM EGTA–Tris, pH7.4). The homogenate was then centri-
fuged at 600 rpm for 10 min at 4 °C to remove unbroken cells
and debris. The supernatant was further centrifuged at 7000 rpm
for 10 min at 4 °C. The pellet was washed once with ice-cold buffer.
The supernatant was discarded and the mitochondrial pellet was
resuspended in residual supernatant in the tube and stored at
ꢀ80 °C. The protein content was estimated by the Bradford
method. The ATP synthesis by the mitochondria was assayed on
the same day as isolation by method described elsewhere22 with
2.5. Determination of MICs against M. tuberculosis H37Rv
Agar dilution assay19 was used to determine the minimum
inhibitory concentration (MIC) of test compounds. Test compounds
were dissolved in dimethylsulfoxide (DMSO) to make 5 mg/mL
stock solutions. Serial dilutions from stocks were also made in
DMSO. Standard anti-TB drug INH was used as positive control
and the vehicle (DMSO) was used as negative control. An amount
of 0.1 mL of serially diluted test compounds or standard drugs
were added to 1.9 mL Middlebrook 7H10 agar medium (with OADC
supplement, in glass tubes). DMSO (0.1 mL/tube) was used as vehi-
cle control. The contents were mixed and allowed to solidify as
slants. Three-week old culture of M. tuberculosis H37Rv was har-
vested from Lowenstein-Jensen medium and its suspension
(1 mg/mL, equivalent to 108 bacilli) was made in normal saline
some modifications. Briefly, 100
bated at room temperature with vigorous stirring in the presence
of 100 M compounds/DCCD in 20 mM Tris/HCl, 0.15 M sucrose,
5 mM MgCl2, 100 M diadenosine pentaphosphate at pH 7.4. Reac-
lg of mitochondria were incu-
l
l
tions were started by adding 1 mM ADP, 20 mM Pi and 50 mM suc-
cinate and arrested after 1 h with 2 mM EDTA-1% TCA. The amount
of ATP formed was measured similar to the method described
above for mycobacterial ATP synthase assay.
containing 0.05% Tween-80. 10 lL of 1:10 dilution of this suspen-
sion (ꢁ105 bacilli) was inoculated into each tube and incubated
at 37 °C for 4 weeks. The lowest concentration of the compound
up to which there was no visible growth of bacilli was its MIC.
2.9. In vitro cytotoxicity
2.6. Estimation of ATP synthesis inhibition in whole M.
tuberculosis H37Rv
The compounds were tested in an in vitro model for cytotoxicity
with Vero monkey kidney cells using Resazurin assay.23 The Vero
cells (ATCC CCL-8 1) were seeded overnight at 1 ꢂ 104—3 ꢂ 106
cells per well in 96-well plates at 37 °C in RPMI supplemented with
10% heat-inactivated fetal bovine serum and 5% CO2. Cells were
exposed to dilutions of experimental and control drugs in triplicate
for 2 h with each compound at a range of concentrations from 100–
The culture of M. tuberculosis H37Rv was declumped by bead-
vortexing settling method and the resulting supernatant was
adjusted to turbidity equivalent to ꢁMcFarland #1 standard. This
was further diluted 100-fold and dispensed as 1 mL volumes
microfuge tubes containing 0.5 mm glass beads. Different concen-
trations of the test compounds were added and incubated for 18 h
at 37 °C. ATP synthesis in treated and untreated cultures were
determined by method described elsewhere,8 with some modifica-
tions. The samples were heated for 8–10 min and cooled on ice and
then lysed in a bead beater. Cell debris was removed by centrifuga-
tion and ATP content in the lysate was measured by Luciferase
based assay (Promega Bactiter-Glo).
1.56
trations. Each well had 100
descending concentrations. After 72 h of incubation, 10
l
g/mL. Rifampicin was used as a control at the same concen-
L of the test material in serially
L of Res-
l
l
azurin indicator solution (0.1%) was added and incubation was
continued for 4–5 h. Any color change from purple to pink or col-
orless was recorded as positive. Fluorescence was measured of
each sample with excitation wavelength at 530 nm and emission
wavelength at 590 nm using the BMG Polar Star Galaxy. The CC50
values (50% inhibitory concentrations) were calculated by plotting
fluorescence values using Microsoft excel template. The data from
this toxicity testing and MIC values were used to calculate a selec-
tivity index (SI), the ratio of CC50: MIC.
2.7. Determination of bactericidal activity in non-replicating
hypoxic culture of M. tuberculosis
The non-replicating hypoxic culture of M. tuberculosis H37Rv
(Wayne model) was prepared as described elsewhere.6,20 In brief,
a mid log phase culture of the bacilli in Middlebrook 7H9 broth
was diluted to approximately 105 cfu/mL and dispensed as 10 mL
volumes into head space vials of 20 mL capacity containing mag-
2.10. In vivo cytotoxicity
The compounds were also tested in an in vivo model for cyto-
toxicity with mouse bone marrow derived macrophages using
the method described elsewhere.24 [Please note: Prior undertaking
netic beads. Methylene blue 1.5 lg/mL was added to the two tubes
Please cite this article in press as: Singh, S.; et al. Bioorg. Med. Chem. (2015), http://dx.doi.org/10.1016/j.bmc.2014.12.060

4
S. Singh et al. / Bioorg. Med. Chem. xxx (2015) xxx–xxx
the in vivo experiments in mouse model system, a permission of
the institute’s ethics committee (CPCSEA registration number 34/
1999 dated 11.3.99, extended up to 2012) had been taken]. Mouse
was euthanized by exposure to CO2 and the femur bones were dis-
sected out. The bones were trimmed at each end, and the marrow
was flushed out (using 26-gauge needle) with 5 mL of Dulbecco’s
minimal essential medium (DMEM) supplemented with 10% FBS,
15% L-929 fibroblast conditioned supernatant (prepared as
described later), and non-essential amino acids. Cells were washed
twice and plated in 96-well tissue culture plates at a concentration
of 105 cells per well (100 mL) in supplemented DMEM. The mono-
layers were then incubated at 37 °C in 5% CO2. After 48 h, 20 mL of
MTS solution was added to each well and incubated for 2 h at 37 °C
in 5% CO2. Reading was taken at 490 nm using a plate reader.
Absorbance shown by DMSO containing wells was taken as 100%
survivors. A compound was considered toxic if it causes 50% inhi-
bition at a concentration 10-fold higher than its MIC.
D30, plated on MB7H10 containing OADC and incubated at 37 °C
for 4-6 weeks.
3. Results
3.1. Rational design and structure-based optimization
Through detailed computational studies, de Jonge and col-
leagues reported the important structural features in TMC207 gov-
erning its ATP synthase inhibitory activity.16 They reported the
chiral N,N-dimethylpropanol substructure to exhibit H-bond for-
mation with the carboxylate side chain of Glu61 residue, and the
favorable binding of the quinoline substructure at the interface of
subunits ‘a’ and ‘c’ of the mycobacterium F0F1-ATP synthase. In
view of this, we carried out the preliminary database screening
of the Asinex database keeping in view the three important struc-
tural features, namely quinoline, tertiary amine and hydroxyl func-
tion present in bedaquiline (TMC207), which led to the
identification of amodiaquine (BAS 00327385) (Fig. 1). Despite of
the presence of these features, amodiaquine is not known to inhibit
the ATP synthase enzyme. This knowledge prompted us to embark
on structure-based studies using the in house developed homology
model of mycobacterial ATP synthase using the NMR structure
Escherichia coli ATP synthase.26 Computational detail is given as
the Supplementary data. In brief, the molecular docking study
revealed that one of the possible reasons behind the lack of ATP
synthase inhibitory activity of amodiaquine may be the absence
of hydrophobic pocket attached to the NH group of amodiaquine,
corresponding to the naphthyl group of TMC207 that could occupy
the deep hydrophobic pocket (Fig. 2). Therefore, we docked several
new analogues containing hydrophobic groups like aryl (Ar),
ArCH2, ArCO, ArSO2, etc, substituted to the NH center. The docking
experiment revealed the particular suitability of aromatic sulfon-
amide substructure to favorably occupy this extended pocket
(Fig. 2D). The sulphonyl group aided further H-bond interactions
with the surrounding polar residues (described in the later sec-
tion). In addition, a methyl group as a replacement of the methoxy
group of the amodiaquine was done in order to simulate the steric
effect as suggested by the docking studies. With this new designed
prototype (compound 8), we synthesized a novel series of aryl sul-
fonamide derivatives (Table 1) for biological screening using both
in vitro enzyme-based as well as whole cell assays.
2.11. Pharmacokinetic (PK) studies
The PK studies was carried out in young and healthy male Spra-
gue–Dawley rats weighing 250 25 g. The study was conducted in
three rats per time point. In all experiments, euthanasia and dis-
posal of carcasses were carried out as per the guidelines of Local
Ethics Committee for animal experimentation. All pharmacokinetic
parameters were calculated by two compartmental models using
the WinNonlin program, version 5.1 (Scientific Consulting Inc.).
The experimental detail of the PK studies is provided in supporting
information.
2.12. In vivo anti-tubercular activity
Compounds were tested in a murine model of chronic TB,25
a
well known model for evaluation of new anti-TB drugs and combi-
nations. Balb/C mice (8–10 week-old) of average weight (18–20 g)
were infected intravenously with 0.2 mL of a freshly prepared bac-
terial suspension containing ꢁ5 ꢂ 106 CFU of M. tuberculosis
H37Rv. There were ten mice in each group and 10 mice were kept
as untreated control. The drugs were administered by an esopha-
geal cannula (gavage) six times weekly started on day 1 and con-
tinued for 4 weeks. The treated mice were administered one of
the following treatments INH at 25 mg/kg, ethambutol (EMB) at
100 mg/kg, compound 9d at 100 mg/kg and 50 mg/kg and com-
pound 9e at 100 mg/kg of body weight. The severity of infection
and the effectiveness of the treatments were assessed by the 30-
day survival rate, CFU per spleen and per lung. The last doses were
administered on day D28 and all surviving mice were sacrificed on
3.2. Chemistry
Compounds were prepared using steps as illustrated in
Schemes 1–3. The reaction of m-chloroaniline (1) with the
Figure 1. Scheme adopted for the design of new ATP synthase inhibitor.
Please cite this article in press as: Singh, S.; et al. Bioorg. Med. Chem. (2015), http://dx.doi.org/10.1016/j.bmc.2014.12.060

S. Singh et al. / Bioorg. Med. Chem. xxx (2015) xxx–xxx
5
Figure 2. (A) The binding pattern of the hit BAS 00327385 into the active site of homology modeled M. tuberculosis ATPase. (B) The comparison of the binding poses of the
potential hit BAS 00327385 (amber colored) and bedaquiline (TMC207) (green colored) into the active site of homology modeled ATPase. Arrow indicates the unoccupied
hydrophobic pocket beneath the NH group of the hit BAS 00327385, which is occupied by the naphthalene motif of the TMC207. The binding mode of (C) bedaquiline or
TMC207, (D) compound 9l, (E) compound 9d and (F) compound 9p into the active site of homology modeled ATP synthase. Important active site residues are labeled and
shown as grey colored sticks. The protein is depicted as cartoon. The letters ‘K or L’ in the parentheses of residue numbers depict the protein chain of the mycobacterial
ATPase.
ethylacetoacetate in the presence of catalytic amount of hydrogen
chloride (HCl), followed by heating under reflux using phenyl ether
afforded the intermediate compound 2. The intermediate 2 was
then treated with phosphorous oxychloride (POCl3) that afforded
the intermediate compound 3 (Scheme 1).
On the other hand, the reaction of p-anisidine (4) with acetic
anhydride in the presence of sodium acetate as a base, under
reflux, afforded the intermediate compound 5. This intermediate
5 upon treatment with molar solution of boron tribromide (BBr3)
in dichloromethane (DCM) afforded the intermediate compound
6. Finally, the reaction of the intermediate 6 with formaldehyde
and diethylamine under reflux in absolute ethanol using catalytic
amount of acetic acid afforded the desired intermediate compound
7 (Scheme 2).
intermediates (3 and 7) as outlined in Scheme 3. The reaction of
the intermediate compounds 3 and 7 in presence of 20% HCl fol-
lowed by the refluxing in absolute ethanol afforded the compound
8. Finally, the compound 8 was treated with different substituted/
unsubstituted aliphatic/alicyclic/aromatic/heteroaromatic sulfonyl
chloride in the presence of N,N-diisopropylethyl amine (DIPEA) as
inorganic base, and acetone as solvent to afford title compounds
9a–q.
3.3. Inhibition of M. smegmatis ATP synthase
The ATP synthase inhibitory activity of the synthesized title
compounds (8, 9a–q) was determined in the inverted membrane
vesicles isolated from M. smegmatis using the method described
elsewhere.10 The DCCD adduct was used as positive control,
In the next stage, syntheses of compound 8 and final title com-
pounds (9a–q) were accomplished using the above synthesized
Please cite this article in press as: Singh, S.; et al. Bioorg. Med. Chem. (2015), http://dx.doi.org/10.1016/j.bmc.2014.12.060

6
S. Singh et al. / Bioorg. Med. Chem. xxx (2015) xxx–xxx
Table 1
whereas INH was used as a negative control in this assay. The
experimentally determined M. smegmatis ATP synthase inhibitory
activities (IC50 M) of the tested compounds (8, 9a–q) along with
In vitro ATP synthase inhibitory activity (IC50, lM) of synthesized compounds (8, 9a–
q), DCCD (positive control) and INH (negative control) evaluated in the inverted
membrane vesicles isolated from M. smegmatis
,
l
positive and negative controls are summarized in Table 1.
OH
O
R
S
N
3.4. Structure–activity relationships
N
N
O
The synthesized compounds (8, 9a–q) exhibited potent inhibi-
tion of M. smegmatis ATP synthase activity with IC50 values ranging
Cl
from 0.36 to 5.45
99.5 3.53% inhibition at 100
the negative control INH did not show any significant inhibition
(0% inhibition at 0.3 M), thus validated the reliability of the assay
system (Table 1). The sulfonamide derivatives (9a–q) exhibited
potent inhibition of M. smegmatis ATP synthase better than the
non-sulfonamide precursor 8, and thus validated our computa-
tional insights of the suitability of hydrophobic groups linked to
the NH group at 4th position of quinoline for effective binding
within the hydrophobic pocket as shown by an arrow in Figure 2.
l
M, while the positive control DCCD showed
Title
R
M. smegmatis ATP synthase IC50 SEM (lM)
l
M concentration. In this assay,
8
N/A
5.45 0.264
l
9a
9b
0.53 0.015
0.74 0.021
Cl
N
-O
N+
9c
9d
9e
0.39 0.017
0.51 0.030
0.63 0.135
O
Cl
Cl
Cl
The compound 9p (IC50 = 0.36
2-sulfonyl group was the most potent compound in the series,
and the compound 9b ((IC50 = 0.39 M) possessing 4-nitrobenzene
lM) possessing 5-chlorothiophene-
l
Cl
sulfonyl group at the corresponding position exhibited almost
comparable ATP synthase inhibition.
Cl
Among the mono/di-substituted chlorobenzene sulfonyl deriva-
tives (9a, 9d–g), the three compounds 9a (IC50 = 0.53
(IC50 = 0.51 M) and 9e (IC50 = 0.63 M) exhibited about 2-fold
better inhibition than compounds 9f (IC50 = 1.34 M) and 9g
(IC50 = 1.26 M). The binding mode of the compound 9d at the
active site of ATP synthase (Fig. 2E) revealed that the 2,3-dichloro-
phenyl moiety occupy the deep hydrophobic pocket corresponding
to the naphth-1-yl substructure of the TMC207, where sulfonyl
oxygen form two direct H-bonds with the Asn190 and Gln227 res-
idues (located in the subunit ‘a’). Similar to TMC207, the proton-
ated tertiary amine and hydroxyl group of this compound
participated in H-bond formation with the carboxylate oxygen of
Glu61(L) and backbone amidic keto of the Val60(L) residues. Unlike
lM), 9d
9f
1.34 0.155
l
l
Cl
l
Cl
l
9g
9h
1.26 0.180
1.36 0.352
Cl
F
F
F
9i
9j
1.83 0.453
1.04 0.024
the compound 9p, compound 9o (IC50 = 1.57 lM) lacking chloro
group at 5th position of thiophene exhibited about 5-fold lower
ATPase inhibitory activity. This in turn substantiates particular
importance of the chlorothiophene group for stronger hydrophobic
interaction in this region for better binding and inhibition. This
inference was well substantiated by docking study, where the
chloro group of the compound 9p exhibited hydrophobic interac-
tion with Phe65(K) and Leu68(K) residues (Fig. 2F).
9k
3.29 0.091
Among the compounds 9l–n possessing bicyclic aromatic sulfo-
nyl groups (Table 1), the compounds 9l (IC50 = 1.00
(IC50 = 0.92 M) exhibited about 2-fold better inhibition than the
compound 9n (IC50 = 1.85 M). Although, the observed ATPase
inhibitory activities of these two compounds (9l and 9m) were
lower than compounds 9a–e and 9p possessing substituted mono-
cyclic aromatic sulfonyl groups, but still were comparable to the
lM) and 9m
9l
1.00 0.031
0.92 0.157
1.85 0.489
l
l
9m
9n
N
compound 9j (IC50 = 1.04
lM) comprising 2,4,6-trimethylbenzene
sulfonyl group. The compound 9q (IC50 = 3.41
l
M) having a fused
S
alicyclic sulfonyl group, exhibited poor ATPase inhibition com-
pared to those compounds possessing aromatic sulfonyl groups
(9a–p). This may be attributed to the greater bulk of the fused ali-
cyclic group, and thus suggested the importance of aromatic group
at this position for better accommodation into the aforementioned
extended hydrophobic pocket. Altogether, it may be concluded
that the most noticeable feature among the studied series of
aromatic sulfonamides essential for the ATP synthase inhibitory
9o
9p
1.57 0.300
0.36 0.303
Cl
S
O
9q
3.41 0.465
DCCD
99.5 3.53% inhibition at 100 lM
Please cite this article in press as: Singh, S.; et al. Bioorg. Med. Chem. (2015), http://dx.doi.org/10.1016/j.bmc.2014.12.060

S. Singh et al. / Bioorg. Med. Chem. xxx (2015) xxx–xxx
7
Cl
N
OH
i
ii
Cl
NH2
Cl
Cl
N
2
1
3
Scheme 1. Synthesis of the intermediate compound 3. Reagents and conditions: (i) (a) ethylacetoacetate, HCl; (b) phenyl ether, 190 °C, 1 h (ii) POCl3.
OH
O
O
O
O
i
ii
N
H
N
H
H2N
4
5
6
iii
OH
O
N
N
H
7
Scheme 2. Synthesis of the intermediate compound 7. Reagents and conditions: (i) acetic anhydride, sodium acetate, reflux; (ii) BBr3, DCM; (iii) formaldehyde, diethylamine,
acetic acid, ethanol.
OH
Cl
N
OH
O
i
N
N
HN
N
+
N
H
Cl
8
3
7
Cl
ii
OH
O
R
S
N
N
O
9a-q
Cl
N
Scheme 3. Synthesis of the compounds 8 and 9a–q. Reagents and conditions: (i) (a) 20% HCl (b) ethanol, reflux (ii) sulfonyl chloride, DIPEA, acetone.
activity is the presence of aromatic sulfonyl group attached to the
NH of the compound 8 for better accommodation and binding
within the an extended hydrophobic pocket, and hence needed
for better inhibition of mycobacterial ATP synthase enzyme.
3.12
9p exhibited complete inhibition at 2-fold higher concentration
of 6.25 g/mL. Other compounds were moderately active
(MIC P 12.5 M).
lg/mL each, whereas the other three compounds 9a, 9c and
l
l
3.5. Growth inhibition of whole M. tuberculosis H37Rv
3.6. Cytotoxicity
A set of eight compounds (9a–e, 9l, 9m and 9p) with proven
ATP synthase inhibitory activities (IC50 6 1 lM), were further
screened in the whole cell assay system against M. tuberculosis
H37Rv using the agar dilution assay. The antimycobacterial spec-
trum of these compounds against the whole M. tuberculosis
The four potent compounds (9a, 9c–e) were evaluated for any
cytotoxicity using an in vitro assay described by Anoopkumar-
Dukie et al.23 using the Vero monkey cells, and an in vivo assay
described by Mosmann24 using mouse bone marrow derived mac-
rophages. The experimentally determined CC50 (lg/mL) values and
H37Rv strain are summarized in Table 2, where MIC (
l
g/mL) rep-
selectivity index (SI) of these compounds using in vitro assay are
resents the minimum inhibitory concentration causing complete
(100%) growth inhibition of the M. tuberculosis H37Rv strain. The
two compounds 9d and 9e with promising ATP synthase inhibitory
summarized in Table 2. The CC50 values of the three compounds
9a, 9d and 9e were >300
lg/mL whereas for compound 9c was
55 g/mL. The selectivity index (SI = CC50/MIC) which is an indica-
l
activities (IC50 ꢁ0.5
l
M) exhibited the 100% (complete) inhibition
tor of therapeutic window, was highest for the compound 9d (>98)
and lowest for the compound 9c (17.55). The SI values for other
of M. tuberculosis H37Rv at the lowest concentration, that is,
Please cite this article in press as: Singh, S.; et al. Bioorg. Med. Chem. (2015), http://dx.doi.org/10.1016/j.bmc.2014.12.060

8
S. Singh et al. / Bioorg. Med. Chem. xxx (2015) xxx–xxx
Table 2
their MICs). These results indicate that these compounds do reduce
the amounts cellular ATP produced by ATP synthase and hence
cause ATP levels to fall faster than by INH /RIF.
Antimycobacterial spectrum (in vitro) of the selected compounds against whole M.
tuberculosis H37Rv strain
Title M. tuberculosis H37Rv growth
Vero cell CC50
g/mL)
Selectivity
index (SI)
inhibition MIC (
lg/mL)
(l
3.8. Selectivity towards mycobacterial ATP synthase
8
>50
ND
>300
ND
ND
>48
ND
18
>96
>48
ND
ND
ND
ND
9a
9b
9c
9d
9e
9l
9m
9p
INH
6.25
>12.5
6.25
3.12
3.12
12.5
>25
The ATP synthase is a universal enzyme, and hence the selectiv-
ity towards mycobacterial ATP synthase is crucial for an anti-TB
drug acting through this pathway.2 In order to elucidate the selec-
tivity of the lead compounds toward mycobacterial ATP synthase,
the lead compounds 9d and 9e along with DCCD, INH and RIF were
evaluated against mitochondrial ATP synthase obtained from
mouse liver. In this experiment, DCCD showed 83% inhibition,
while the lead compounds (9d and 9e) showed only 5–10% inhibi-
55
>300
>300
ND
ND
ND
6.25
0.1
ND
tion of the mitochondrial ATP synthase at 100 lM, thus suggesting
the identified lead compounds to be highly selective (selectivity
>155) towards mycobacterial ATP synthase (Table 3).
two compounds (9a and 9e) were greater than 48. In addition,
these compounds were non-toxic in the in vivo assay conducted
using the mouse bone marrow derived microphages. Therefore,
the two compounds 9d and 9e were selected as candidate mole-
cules for further investigation.
3.9. Bactericidal activity against non-replicating (Dormant) M.
tuberculosis
Since ATP synthase is crucial for the survival of mycobacteria
during non-replicating persistence, the two lead compounds (9d
and 9e) were also tested against non-replicating M. tuberculosis
3.7. Fall in cellular ATP level in whole M. tuberculosis
The culture of M. tuberculosis H37Rv was exposed to the test
compounds/drugs for 18 h in order to estimate the decline in total
cellular ATP content, which is shown in Figure 3A. Bacilli treated
H37Rv in Wayne Model under hypoxic/(oxygen depleted) condition
considering INH as negative and metronidazole (MTZ) as positive
controls. The lead compound 9d exhibited bactericidal effect in
the hypoxic culture of M. tuberculosis at 100 lg/mL (32-fold of its
MIC), and showed more than 2.3 log10 reductions in CFU while
the lead compound 9e showed about 1.5 log10 reductions in CFU
(Table 3). In this assay, the negative control INH showed only
with INH and RIF at 0.5
and 69.6% fall in total cellular ATP level respectively, while
100 M concentration of DCCD caused 30% fall in total ATP level,
lg/mL (5-fold of their MICs) led to 78.5%
l
most of which probably may be generated by substrate level phos-
phorylation in live cells. On the contrary, bacilli treated with the
two lead compounds 9d and 9e led to about 31 and 26% fall in total
approximately 0.2 log10 reduction in CFU at 5
its MIC), and MTZ showed approximately 1 log10 reduction in
CFU at 50 g/mL (Table 3). These results further support the
lg/mL (100 fold of
cellular ATP level, respectively at 50
lg/mL (approx. 15 fold of their
l
MICs), and approximately 60% and 54% fall at 12.5
lg/mL (4-fold of
Figure 3. (A) Effect of lead compounds on total cellular ATP content of M. tuberculosis H37Rv. (B) Concentration-time profile of the lead compound 9d after a single oral
(10 mg/kg) and intravenous (1 mg/kg) dose in male Sprague–Dawley rats. Bar represents the standard error of mean (SEM). (C) Survival rate of mice for 30 days after
intravenous infection of M. tuberculosis H37Rv. (D) Mean log10 CFU per spleen and lungs of mice after 30 days post-treatment with test compound 9d and drugs (INH and
EMB). Mice were infected intravenously with 5 ꢂ 106 CFU of M. tuberculosis H37Rv, and treatment was begun the next day after infection. There were 10 mice in each group.
Untreated control mice did not receive any treatment. Error bars represent standard deviations.
Please cite this article in press as: Singh, S.; et al. Bioorg. Med. Chem. (2015), http://dx.doi.org/10.1016/j.bmc.2014.12.060

S. Singh et al. / Bioorg. Med. Chem. xxx (2015) xxx–xxx
9
Table 3
Selective inhibition of mycobacterial versus mammalian ATP synthase and bactericidal activity against dormant bacilli by lead compounds 9d and 9e
Compd M. smegmatis ATPase IC50
(lM) or %
Mammalian ATPase IC50 or %
inhibition at 100
M. tuberculosis H37Rv MIC
(lg/mL) (ADA)
Bactericidal activity against non-
replicating M. tuberculosis H37Rv in
hypoxic culture
inhibition at 100
l
M
l
M
Concn
g/mL)
Fold of
MIC
Log10 reduction in
CFU/mL
(
l
9d
9e
0.51 0.030
0.63 0.135
>100
>100
l
M
M
3.12
3.12
100
10
100
10
5
50
32
3.2
32
3.2
100
NA
NA
2.34 0.41
1.14 0.17
1.64 0.14
0.79 0.24
0.21 0.08
1.16 0.16
4.68 0.08
l
INH
MTZ
DCCD
NA
NA
99.5%
NA
NA
83%
60.12
NA
NA
100
inhibitory property of the lead compounds (9d and 9e) for myco-
bacterial ATP synthase, as reflected by their effect on bacterial
viability under microaerophilic/dormant conditions.
Table 4
Pharmacokinetic parameters of the compound 9d in male Sprague–Dawley ratsa
Parameters
Compound 9d
Oral
Intravenous
3.10. Activity against bacteria and fungi
Cmax (ng/ml)
tmax (h)
AUC (ng h/ml)
t1/2 (h)
Vss (L/kg)
Clearance (L/h/kg)
Bioavailability (%)
616.09 58.11
0.5
2813
4.20
0.41
2416.89 92.46
0.0833
16393
8.84
0.84
0.07
—
The two lead compounds (9d and 9e) were evaluated up to
50 lg/mL concentration each against both bacteria [E. coli (ATCC-
9637), Staphylococcus aureus (ATCC-25923), Klebsiella pneumonia
(ATCC-27736), and Pseudomonas aeruginosa (ATCC BAA-427)] as
well as fungi [Candida albicans, Cryptococcus neoformans, Sporothrix
schenckii, Trichophyton mentagrophytes, Aspergillus fumigatus, Can-
dida parapsilosis (ATCC-22019)]. As expected, these lead com-
pounds did not show any inhibition of both bacteria as well as
0.06
1.72
a
Each value represents the average of three rats dosed orally (10 mg/kg) and
intravenously (1 mg/kg); values of Cmax are mean SEM; AUC = area under the
serum concentration-time curve, Cmax = serum peak concentration, tmax = time to
Cmax, t1/2 = elimination half-life, VSS = volume of distribution.
fungi at the maximum concentration of 50 lg/mL (details are given
in Supplementary material). This may be considered as an added
advantage for anti-TB drug which are given for longer duration
than other infections and hence, it is required that such molecule
should not have any antibacterial or antifungal activity which
may lead to the development of resistance against both bacteria
(Gram-positive and Gram-negative) and fungi.
each in the two separate groups of mice, whereas the two positive
controls (INH and EMB) were administered at fixed doses of 182
and 490 lmol/kg (i.e., 25 and 100 mg/kg, respectively) of body
weight respectively through an esophageal cannula (gavage) six
times weekly started on day 1 and continued for 4 weeks. The
severity of infection and the effectiveness of the treatments were
assessed by the 30-day survival rate, CFU per spleen and per lung.
The survival rates of mice for 30 days after intravenous infection
with 5 ꢂ 106 CFU of M. tuberculosis H37Rv are shown in Figure 3C.
The log10 CFU per spleen and lung of mice treated with the above
compound/drugs after intravenous infection with 5 ꢂ 106 CFU of
M. tuberculosis H37Rv is shown in Figure 3D.
3.11. Pharmacokinetic studies
Pharmacokinetic studies on the lead compound 9d revealed
that the animals well tolerated the treatment as no peculiarities
in the animal behavior were observed. This compound was rapidly
absorbed and slowly eliminated (Fig. 3B and Table 4). Following
After inoculation with 5 ꢂ 106 CFUs of M. tuberculosis H37Rv per
oral dose, the observed maximum plasma concentration (Cmax
and the time to reach the maximum plasma concentration (tmax
)
)
mice, all the mice in the untreated group died within 25 days after
infection, whereas all of the INH (182
vived till day 30. At day 30, the percent survival rates of mice trea-
ted with EMB (490 mol/kg), compound 9d (173 mol/kg) and
compound 9d (87 mol/kg) were 90%, 60%, and 40% respectively
(Fig. 3C). The compound 9d showed 2.12 log10 reduction in CFU
at 173 mol/kg and approximately 1 log10 reduction in CFU at
87 mol/kg both in the lungs as well as in spleen compared to
lmol/kg) treated mice sur-
values suggested that the compound is mainly absorbed from the
stomach. The volume of distribution (Vss, 0.41 L/kg) of compound
9d is greater than the total blood volume (0.054 L/kg)24 indicating
high extra-vascular distribution of the compound. The systemic
clearance of compound 9d was smaller than the hepatic blood flow
of the rat (2.9 L/h/kg),24 suggesting an insignificant amount of
extra-hepatic elimination of the compounds. Moreover, low clear-
ance (0.06 L/h/kg) of this compound suggests a potentially long
acting profile (6.69 h based on MRTp.o.).
l
l
l
l
l
the growth in the untreated control group of mice infected with
M. tuberculosis H37Rv (Fig. 3D). Although the in vivo efficacy of
the lead compound 9d is not better than the positive control drug
INH, however, it is at least double of the EMB. Importantly, this
lead compound act through a novel mechanism of action that
reflects this compound as candidate molecule for further biological
investigation.
3.12. In Vivo antimycobacterial activity
The lead compound 9d with potent ATP synthase inhibitory
activities (Table 1) and growth inhibitory activity against both rep-
licating as well as non-replicating whole M. tuberculosis (Table 2),
was evaluated in a murine model of chronic TB25 using the Balb/
C mice (8–10 week-old) of average weight (18–20 g). In this assay,
INH and EMB were used as positive controls. The compound 9d
was administered at the two dose levels, that is, 87 and
4. Discussions
Despite the availability of highly efficacious treatment(s) for
decades, TB is still a major global health problem mainly due to
the emergence of MDR-, XDR- and TDR-TBs, latent/dormant
173 lmol/kg (i.e., 50 and 100 mg/kg, respectively) of body weight
Please cite this article in press as: Singh, S.; et al. Bioorg. Med. Chem. (2015), http://dx.doi.org/10.1016/j.bmc.2014.12.060

10
S. Singh et al. / Bioorg. Med. Chem. xxx (2015) xxx–xxx
infection of TB and its synergy with HIV/AIDS. The mycobacterial
F0F1-ATP synthase is a well validated target for TB after the success
launch of bedaquiline (Sirturo™, TMC207) into the market as an
effective drug for MDR-TB. Bedaquiline has shown positive efficacy
against both the resistant (MDR and XDR) and latent/dormant TB
strains. However, this drug has been associated with some adverse
events like nausea, joint and chest pain, and headache. In addition,
bedaquiline is a weak hERG blocker leading to the prolongation of
the QT interval of the heart. Despite of these side effects, it is still
recommended by the US-FDA specifically for the treatment of
MDR-TB. Notably, there is a wide scope for further exploitation
of the medicinal chemistry tools in order to identify new chemical
entities targeting this validated anti-TB target that could provide
new potential treatments of the MDR-, XDR- as well as TDR-TB
cases. The pharmacophoric insights from the structure of bedaqui-
line (TMC207) have encouraged medicinal chemists to apply state-
of-the-art medicinal chemistry approaches to identify and explore
new chemical entities (NCEs) as potent and selective mycobacterial
ATP synthase inhibitor(s).
compounds have not shown any antibacterial or antifungal activity
up to 50 g/mL.
The pharmacokinetic studies conducted on the compound 9d
after p.o. and IV administration in young and healthy male Spra-
gue–Dawley rats weighing 250 25 g indicated its quick absorp-
l
tion, distribution and slow elimination. It exhibited
a high
volume of distribution (Vss, 0.41 L/kg), moderate clearance
(0.06 L/h/kg), long half life (4.2 h) and low absolute bioavailability
(1.72%). The observed maximum plasma concentration (Cmax) and
the time taken to reach the maximum plasma concentration (tmax
)
after oral administration of this compound suggested that the com-
pound is mainly absorbed from the stomach. Furthermore, the low
clearance (0.06 L/h/kg) of this compound suggests its long acting
profile (6.69 h based on MRTp.o.) (Table 4 and Fig. 3B).
In the murine model of the chronic TB, the lead compound 9d has
shown 2.12 log10 reductions in CFU at 173 lmol/kg dose in both the
lung and spleen of mice in comparison to the untreated control
group (Fig. 3C and D). The in vivo efficacy of 9d is at least double
of EMB, thus reflecting its effectiveness towards protection against
M. tuberculosis H37Rv. Altogether, the above detailed investigation
forecasts suitability of 9d as candidate drug for further preclinical
evaluation against both replicating and non-replicating TB. This
could lead to the new potential treatment of the resistant TB cases.
In an effort to identify novel potent and selective mycobacterial
ATP synthase inhibitors, a total of 18 new compounds designed
through the state-of-the-art medicinal chemistry approaches have
been synthesized and evaluated against M. smegmatis ATP synthase
(Table 1). The observed inhibitory activity (IC50) ranges from 0.36
to 5.45 lM. The better inhibitory activity of the sulfonamide deriv-
5. Conclusion
atives (9a–q) than the precursor non-sulfonamide 8 suggests the
particular importance of aromatic sulfonamide group in this class
of compounds. This further substantiates the importance of struc-
ture-based drug design approach as this approach forecasted the
particular importance of arylsulfonamide group for better binding
and inhibition of mycobacterial ATP synthase (Fig. 2). In addition,
the observed improvement in the ATP synthase inhibitory activity
of the compounds (9a–p) possessing arylsulfonyl group than the
compound 9q possessing alkylsulfonyl group signifies the better
suitability of the aromatic sulfonyl groups for potential inhibition
of mycobacterial ATP synthase (Table 1).
Despite the availability of highly efficacious treatment(s) for
decades, TB is still a major global health problem mainly due to
the emergence of MDR-, XDR- and TDR-TBs, latent/ dormant infec-
tion of TB and its synergy with HIV/AIDS. The mycobacterial F0F1-
ATP synthase is an anti-TB target for the bedaquiline (Sirturo™), a
drug recently approved by the US-FDA specifically for the treat-
ment of MDR-TB. In our efforts to identify novel compounds as
ATP synthase inhibitors, a total of eighteen new compounds
designed through the state-of-the-art drug design approaches has
been synthesized and evaluated against M. smegmatis ATP synthase
where the observed inhibitory activity (IC50) ranges from 0.36 to
The in vitro cell-based screening of a total of eight compounds
(ATPase IC50 6 1
H37Rv strain revealed two lead compounds (9d and 9e) with
MIC value of 3.12 g/mL (Table 2). Interestingly, these compounds
lM) against whole replicating M. tuberculosis
5.45
(ATPase IC50 6 1
revealed four compounds (9a, 9c–e) with MIC value 66.25
mL. Among these, the three compounds (9a, 9d and 9e) lacked
cytotoxicity (CC50 >300 g/mL) in the three different cytotoxicity
l
M. The in vitro cell-based screening of eight compounds
M) against whole M. tuberculosis H37Rv has
g/
l
l
l
lacked any cytotoxicity in both in vitro and in vivo assays con-
ducted using Vero monkey cells and bone marrow macrophages,
respectively. Notably, these two compounds did not show any
l
assay systems. The selectivity index for the compound 9d is the
cytotoxicity in vitro up to 300 lg/mL (CC50 >300 lg/mL), and the
highest (>98) and has shown ꢁ2.12 log10 reductions in CFU both
selectivity index for one of the compound 9d is the highest (>98)
(Table 2). Moreover, these inhibitors caused a greater fall in cellu-
lar ATP levels than by INH or RIF which act by different mecha-
nisms (Fig. 3A). Since these compounds seem to reduce ATP
levels, the bactericidal activity of these compounds were tested
on non-replicating M. tuberculosis H37Rv in hypoxic culture. As
summarized in Table 3, the compound 9d and 9e caused approxi-
mately 2.4 and 1.6 log10 reduction in CFUs, respectively at 32-fold
of their respective MICs as compared to control drug INH causing
only 0.2 log10 reduction in CFU even at 50-fold of its MIC value.
In this model, the negative control drug MTZ showed approxi-
in the lungs as well as in spleen at the dose of 173 lmol/kg as com-
pared to the growth in the untreated control group in the murine
model system. The in vivo efficacy of this compound is at least dou-
ble of the ethambutol, thus suggesting this compound to be better
in protection against M. tuberculosis H37Rv. In addition, this com-
pound 9d has exhibited high selectivity (>200) towards mycobac-
terium than mammalian ATP synthase and good bactericidal effect
(>2.3 log10 reductions in CFU) in the hypoxic culture of M. tubercu-
losis at 100
lg/mL (32 fold of its MIC) as compared to INH
[ꢁ0.2 log10 reduction in CFU at 5
lg/mL (100 fold of its MIC)].
The pharmacokinetics of 9d after p.o. and IV administration in rats
has indicated its quick absorption, distribution and slow elimina-
tion. It has exhibited a high volume of distribution (Vss, 0.41 L/
kg), moderate clearance (0.06 L/h/kg), long half-life (4.2 h) and
low absolute bioavailability (1.72%). Therefore, this lead compound
9d may be suitable as the candidate drug for further preclinical
evaluation against TB.
mately 1 log10 reduction in CFU at 50 lg/mL concentration.
Since ATP synthase is a universal enzyme, determining the
selectivity of compounds for the mycobacterial over mammalian
ATP synthase is a highly important task. For this, we used mouse
liver mitochondria as the source of the mammalian ATP synthase.
As summarized in Table 3, lead compounds 9d and 9e showed a
very low inhibition (between 5 and 15%) at 100 lM, while the posi-
Acknowledgments
tive control DCCD showed 83% inhibition of the mammalian ATP
synthase. Therefore, these two lead compounds 9d and 9e are
about 200-fold selective towards mycobacterial ATP synthase
The authors (S.S. and K.K.R.) are thankful to Ministry of Health
(MOH) and Council of Scientific and Industrial Research (CSIR),
(mammalian ATPase IC50 >100 lM). Furthermore, these lead
Please cite this article in press as: Singh, S.; et al. Bioorg. Med. Chem. (2015), http://dx.doi.org/10.1016/j.bmc.2014.12.060

S. Singh et al. / Bioorg. Med. Chem. xxx (2015) xxx–xxx
11
Chaffoy, D.; Huitric, E.; Hoffner, S.; Cambau, E.; Truffot-Pernot, C.; Lounis, N.;
Jarlier, V. Science 2005, 307, 223.
10. Koul, A.; Dendouga, N.; Vergauwen, K.; Molenberghs, B.; Vranckx, L.;
Willebrords, R.; Ristic, Z.; Lill, H.; Dorange, I.; Guillemont, J.; Bald, D.;
Andries, K. Nat. Chem. Biol. 2007, 3, 323.
11. Matteelli, A.; Carvalho, A. C.; Dooley, K. E.; Kritski, A. Future Microbiol. 2010, 5,
849.
12. Khan, S. R.; Singh, S.; Roy, K. K.; Akhtar, M. S.; Saxena, A. K.; Krishnan, M. Y. Int.
J. Antimicrob. Agents 2013, 41, 41.
New Delhi, respectively for financial assistance in terms of fellow-
ship. The authors thank Dr. P. K. Shukla and Ms. Pratiksha Singh of
Fermentation Technology Division, CDRI for screening lead com-
pounds against bacteria and fungi. The Sophisticated Analytical
Instrument Facility (SAIF) department is acknowledged for the
characterization of compounds. The technical assistances of Messrs
A. S. Kushwaha, D. N. Vishwakarma and Zahid Ali are also
acknowledged.
13. World Health Organization. The Use of Bedaquiline in the Treatment of
Multidrug-resistant Tuberculosis. Interim Policy Guidance (WHO/HTM/TB/
2013.6). Geneva, Switzerland, November 2013.
14. Anti-Infective Drugs Advisory Committee Meeting Report. The U.S. Food and
Drug Administration., November 28, 2012.
Supplementary data
15. Haagsma, A. C.; Podasca, I.; Koul, A.; Andries, K.; Guillemont, J.; Lill, H.; Bald, D.
PLoS ONE 2011, 6, e23575.
16. de Jonge, M. R.; Koymans, L. H.; Guillemont, J. E.; Koul, A.; Andries, K. Proteins
2007, 67, 971.
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.bmc.2014.12.060.
17. Rastogi, V. K.; Girvin, M. E. Nature 1999, 402, 263.
18. Singh, S.; Roy, K. K.; Khan, S. R.; Kashyap, V. K.; Sharma, S. K.; Krishnan, M. Y.;
Chaturvedi, V.; Sinha, S.; Srivastava, R.; Saxena, A. K. WO2013102936 (A4),
September 12, 2013.
References and notes
1. World Health Organization. Global Tuberculosis Report. WHO Report 2013;
WHO Press: Geneva, Switzerland, 2013.
19. McClatchy, J. K. Lab. Med. 1978, 9, 47.
2. Koul, A.; Arnoult, E.; Lounis, N.; Guillemont, J.; Andries, K. Nature 2011, 469,
483.
3. Migliori, G. B.; Centis, R.; D’Ambrosio, L.; Spanevello, A.; Borroni, E.; Cirillo, D.
M.; Sotgiu, G. Clin. Infect. Dis. 2012, 54, 1379.
4. Udwadia, Z. F. Thorax 2012, 67, 286.
5. Udwadia, Z. F.; Amale, R. A.; Ajbani, K. K.; Rodrigues, C. Clin. Infect. Dis. 2012, 54,
579.
6. Mitchison, D. A. Am. J. Respir. Crit. Care Med. 2005, 171, 699.
7. Vandiviere, H. M.; Loring, W. E.; Melvin, I.; Willis, S. Am. J. Med. Sci. 1956, 232,
30. passim.
8. Wayne, L. G.; Hayes, L. G. Infect. Immun. 1996, 64, 2062.
9. Andries, K.; Verhasselt, P.; Guillemont, J.; Gohlmann, H. W.; Neefs, J. M.;
Winkler, H.; Van Gestel, J.; Timmerman, P.; Zhu, M.; Lee, E.; Williams, P.; de
20. Lenaerts, A. J.; Gruppo, V.; Marietta, K. S.; Johnson, C. M.; Driscoll, D. K.;
Tompkins, N. M.; Rose, J. D.; Reynolds, R. C.; Orme, I. M. Antimicrob. Agents
Chemother. 2005, 49, 2294.
21. Frezza, C.; Cipolat, S.; Scorrano, L. Nat. Protoc. 2007, 2, 287.
22. Garcia, J. J.; Ogilvie, I.; Robinson, B. H.; Capaldi, R. A. J. Biol. Chem. 2000, 275,
11075.
23. Anoopkumar-Dukie, S.; Carey, J. B.; Conere, T.; O’Sullivan, E.; van Pelt, F. N.;
Allshire, A. Br. J. Radiol. 2005, 78, 945–947.
24. Mosmann, T. J. Immunol. Methods 1983, 65, 55.
25. Kelly, B. P.; Furney, S. K.; Jessen, M. T.; Orme, I. M. Antimicrob. Agents Chemother.
1996, 40, 2809.
26. Dmitriev, O. Y.; Abildgaard, F.; Markley, J. L.; Fillingame, R. H. Biochemistry
2002, 41, 5537.
Please cite this article in press as: Singh, S.; et al. Bioorg. Med. Chem. (2015), http://dx.doi.org/10.1016/j.bmc.2014.12.060