Structure-Guided Discovery of Antitubercular Agents That Target the Gyrase ATPase Domain

Article abstract of DOI:10.1002/cmdc.201500556

In this study we explored the pharmaceutically underexploited ATPase domain of DNA gyrase (GyrB) as a potential platform for developing novel agents that target Mycobacterium tuberculosis. In this effort a combination of ligand- and structure-based pharmacophore modeling was used to identify structurally diverse small-molecule inhibitors of the mycobacterial GyrB domain based on the crystal structure of the enzyme with a pyrrolamide inhibitor (PDB ID: 4BAE). Pharmacophore modeling and subsequent in vitro screening resulted in an initial hit compound 5 [(E)-5-(5-(2-(1H-benzo[d]imidazol-2-yl)-2-cyanovinyl)furan-2-yl)isophthalic acid; IC₅₀=4.6±0.1 μm], which was subsequently tailored through a combination of molecular modeling and synthetic chemistry to yield the optimized lead compound 24 [(E)-3-(5-(2-cyano-2-(5-methyl-1H-benzo[d]imidazol-2-yl)vinyl)thiophen-2-yl)benzoic acid; IC₅₀=0.3±0.2 μm], which was found to display considerable in vitro efficacy against the purified GyrB enzyme and potency against the H₃₇Rv strain of M. tuberculosis. Structural handles were also identified that will provide a suitable foundation for further optimization of these potent analogues.

Full text of DOI:10.1002/cmdc.201500556

DOI: 10.1002/cmdc.201500556
Full Papers
Structure-Guided Discovery of Antitubercular Agents That
Target the Gyrase ATPase Domain
Variam U. Jeankumar,[a] Shalini Saxena,[a] Rahul Vats,[a] Rudraraju Srilakshmi Reshma,[a]
Renuka Janupally,[a] Pushkar Kulkarni,[b, c] Perumal Yogeeswari,[a] and Dharmarajan Sriram*[a]
In this study we explored the pharmaceutically underexploited
ATPase domain of DNA gyrase (GyrB) as a potential platform
for developing novel agents that target Mycobacterium tuber-
culosis. In this effort a combination of ligand- and structure-
based pharmacophore modeling was used to identify structur-
ally diverse small-molecule inhibitors of the mycobacterial
GyrB domain based on the crystal structure of the enzyme
with a pyrrolamide inhibitor (PDB ID: 4BAE). Pharmacophore
modeling and subsequent in vitro screening resulted in an ini-
tial hit compound 5 [(E)-5-(5-(2-(1H-benzo[d]imidazol-2-yl)-2-cy-
anovinyl)furan-2-yl)isophthalic acid; IC50 =4.6Æ0.1 mm], which
was subsequently tailored through a combination of molecular
modeling and synthetic chemistry to yield the optimized lead
compound 24 [(E)-3-(5-(2-cyano-2-(5-methyl-1H-benzo[d]imida-
zol-2-yl)vinyl)thiophen-2-yl)benzoic acid; IC50 =0.3Æ0.2 mm],
which was found to display considerable in vitro efficacy
against the purified GyrB enzyme and potency against the
H37Rv strain of M. tuberculosis. Structural handles were also
identified that will provide a suitable foundation for further op-
timization of these potent analogues.
Introduction
The currently available arsenal of anti-tuberculosis (TB) drugs
remains insufficient to combat the emergence of drug resist-
ance, necessitating research into the development of novel
agents that lack cross-resistance mediated by mutations in the
bacterial targets.[1,2] With drug discovery programs becoming
more focused and rational, extending to the genomic level,
the identification and selection of the appropriate target has
become a fundamental prerequisite.
(GyrB) of bacterial DNA gyrase remains pharmaceutically un-
derexploited, despite being genetically demonstrated to be
a bactericidal target in M. tuberculosis.[3] Inhibitors of the GyrB
subunit competitively inhibit its ATPase activity, thereby de-
priving bacteria of the energy source required for maintaining
the topological state of DNA in the cell and affecting processes
such as DNA replication. The development of new inhibitors
that target the GyrB subunit offers an excellent opportunity to
create an effective treatment for TB that is not affected by
target-mediated cross-resistance associated with previously re-
ported drugs.[3–10] Furthermore, the increasing amount of struc-
tural information available for bacterial DNA gyrases has in-
creased their potential as drug targets and has opened the
possibility of using structure-based drug design (SBDD) in this
context.[5,6,8]
Moxifloxacin, the most advanced drug for the treatment of
TB, targets the GyrA segment of DNA gyrase—the sole type II
topoisomerase present in Mycobacterium tuberculosis—which
introduces negative supercoils in an ATP-dependent manner.
Moxifloxacin has demonstrated excellent potency against the
drug-sensitive and drug-resistant strains of M. tuberculosis, vali-
dating the druggability of DNA gyrase as a potential antituber-
cular drug target; however, the emergence of fluoroquinolone-
resistant strains underscores the genuine urgency in the search
for new therapeutic agents to combat TB. The ATPase domain
This study involved an SBDD approach to identify potential
inhibitors of the M. tuberculosis gyrase ATPase domain. The ini-
tial hit obtained by pharmacophore screening of a commercial
chemical database was subsequently remodeled by using a me-
dicinal chemistry approach to generate a lead molecule that
shows a 10-fold enhancement in potency over the starting hit
compound. The binding affinity at the enzyme level was fur-
ther corroborated by biophysical characterization techniques.
Early pharmacokinetic (PK) evaluations of the optimized ana-
logue were encouraging, demonstrating the suitability of this
compound class for development as leads in the fight against
TB.
[a] Dr. V. U. Jeankumar, Dr. S. Saxena, Dr. R. Vats, R. S. Reshma, R. Janupally,
Prof. P. Yogeeswari, Prof. D. Sriram
Department of Pharmacy, Birla Institute of Technology & Science – Pilani
Hyderabad Campus, Shameerpet, R.R. District
Hyderabad 500078, Andhra Pradesh (India)
E-mail: dsriram@hyderabad.bits-pilani.ac.in
[b] Dr. P. Kulkarni
Dr. Reddy’s Institute of Life Sciences, University of Hyderabad Campus
Gachibowli, Hyderabad 500046 (India)
[c] Dr. P. Kulkarni
Zephase Therapeutics (an incubated company at Dr. Reddy’s Institute of
Life Sciences), University of Hyderabad Campus
Gachibowli, Hyderabad 500046 (India)
Results and Discussion
We took an e-pharmacophore approach, which uses energy-
and ligand-based techniques to identify putative ligands that
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/cmdc.201500556.
ChemMedChem 2016, 11, 539 – 548
539
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
bind to the active site of the mycobacterial ATPase domain. As
a template, we used the crystal structure of the Mycobacterium
smegmatis GyrB ATPase domain in complex with a pyrrolamide
(PDB ID: 4BAE).[5] The pharmacophore hypotheses based on
mapping of the energetic terms from the Glide extra precision
scoring function (Glide XP; Glide version 5.7, Schrçdinger LLC,
New York, NY, USA) onto atom centers was used to derive
pharmacophore sites. These were based on structural and
energy information between the protein and ligand using
PHASE (version 3.3, Schrçdinger LLC). The initial number of
pharmacophore sites was set at 10 for the crystal structure.
The docking result (Xpdes) was then imported to find the
structure-based pharmacophoric features, which would aid in
finding the best-featured functional groups. The maximum
number of pharmacophore sites derived for the reference
ligand was four, with two aromatic rings (R), one negative ion-
izable group (N), and one donor (D) (Figure S3, Supporting In-
formation).
À9.15 to À6.50 kcalmolÀ1. Final short-listing of possible hits
were based on visual inspection of the important amino acid
residues in the active site cleft involved in binding that includ-
ed hydrogen bonds to Asp79, Arg141, and p-stacking interac-
tions with Arg82. The top-12 compounds retrieved from the
Asinex library in the Glide XP docking study were procured
and experimentally screened for their in vitro GyrB inhibitory
potential using a malachite green assay adapted to a 96-well
plate format, as described previously.[5–7] Novobiocin, which
has been demonstrated to be a potent inhibitor of GyrB, was
used as a positive control in the study. Compounds were also
tested in the presence of Brij-35, a non-anionic detergent, to
ascertain whether inhibition is an artifact due to sequestration
of the enzyme by drug aggregates. Other artifacts such as
auto-absorbance of the drug, were also ruled out by nullifying
its absorbance during the reaction. Two-dimensional structures
of the top hits identified, together with their GyrB IC50 values,
mapped features, fitness, docking scores, and the significant
hydrogen bonding interactions are provided in Figure S5 and
Table S2 of the Supporting Information.
The four-point pharmacophore was then used for virtual
screening of a commercial chemical database (Asinex) contain-
ing 500000 compounds, using a protocol summarized in
Figure 1. The methodology used for pharmacophore genera-
tion and docking is provided in detail in the Supporting Infor-
mation.
The most promising compounds from the in vitro GyrB
assay, 5-{5-[(1E)-2-(1H-1,3-benzodiazol-2-yl)-2-cyanoeth-1-en-1-
yl]furan-2-yl}benzene-1,3-dicarboxylic acid, (5) exhibited a GyrB
inhibitory IC50 value of 4.6Æ0.1 mm. The compound also
showed a well-correlated supercoiling inhibitory IC50 value of
5.1Æ0.2 mm and an in vitro minimum inhibitory concentration
(MIC) of 7.82 mm against the M. tuberculosis H37Rv strain.
To provide a structural basis for further ligand improvement
the protein–ligand interactions of compound 5 were analyzed
in greater detail. Compound 5 exhibited a good binding
energy with a Glide score of À8.55 kcalmolÀ1. The molecule
was also found to be involved in three prominent hydrogen
bonding interactions with Asp79, Gly83/Arg82, and Arg82, and
Compounds retrieved by the e-pharmacophore model using
PHASE with a fit value >1.0 were regarded as potential hits
and were carried forward for high-throughput virtual screening
(HTVS). Top compounds from this screen resulting in a score of
ꢀÀ5.0 kcalmolÀ1 and docked with one or more hydrogen
bonds analogous to those observed in the crystal structure
were subjected to another round of docking by Glide XP,
which combines accurate, physics-based scoring terms and
thorough sampling, and the results gave scores ranging from
Figure 1. Workflow used for hit identification and lead optimization of putative ligands as mycobacterial DNA GyrB inhibitors.
ChemMedChem 2016, 11, 539 – 548
www.chemmedchem.org
540
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
was found in the vicinity of Asn52, Gly81, Ala53, Val77, Val49,
Ile71, Val123, Val125, Ile84, Pro85, Val99, Pro85, and Gly168,
where it was observed to be stabilized by hydrophobic interac-
tions and some polar contacts with the protein. A closer look
at the binding pattern of the molecule showed the benzimida-
zole core to be buried deeply into the hydrophobic cavity, en-
gaging in important hydrophobic interactions with Ala53,
Val77, Val49, Ile71, Val123, Val125, and Ile84. Furthermore, the
benzimidazole N1 atom was found to be involved in a hydro-
gen bonding interaction with Asp79, analogous to that ob-
served in the crystal structure complex. This interaction is be-
lieved to be critical for activity. The carboxylic acid group pres-
ent at position 3 of the phenyl side chain of the furanyl ring
undergoes a hydrogen bonding interaction with Arg82 instead
of a p-stacking interaction, as observed in the case of the refer-
ence pyrrolamide ligand (PDB ID: 4BAE).[5] An additional hydro-
gen bonding interaction was also found between the terminal
nitrogen atom of the acetonitrile side chain and Gly83.
of obtaining a lead series with tractable structure–activity
relationships (SAR) and potencies better than those of the
identified initial screening hits. A series of 24 compounds
(Table 1) was designed and taken up for synthesis by introduc-
ing chemical diversity at positions which, in addition to the in-
teractions observed with the hit molecule, could further
engage in much more robust interactions with enzyme resi-
dues in the vicinity.
Synthesis of the designed ligands was achieved by a straight-
forward methodology as depicted in Scheme 1. The synthesis
began by fusing the respective phenylenediamine with ethyl-
2-cyanoacetate at 2008C. The so-obtained 2-cyanomethylben-
zimidazole derivative was then functionalized at the acetoni-
trile chain via Knoevenagel condensation with the desired al-
dehyde in ethanol at reflux using piperidine or 10% methanol-
ic KOH as base.[11]
The 2-cyanomethylbenzimidazole core was unaltered in the
preliminary hit amelioration studies because of the critical in-
teractions that the benzimidazole NH group (N1) and the ter-
minal cyano group undergo with Asp79 and Gly83. The syn-
thetic chemistry refinements were directed mostly toward the
With an understanding of the critical handles and interac-
tions essential for retaining activity, we set out to search for
a more potent molecule by designing a library with the goal
Table 1. Biological evaluations of synthesized analogues 3–26.
Compd
R
X
R1
R2
R3
GyrB IC50 [mm]
MIC [mm][c]
Cytotox. [%][d]
ATPase[a]
Supercoil[b]
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
H
H
H
H
H
H
Cl
Cl
Cl
O
O
O
S
S
S
O
O
O
S
S
S
O
O
O
S
S
S
O
O
O
S
S
S
COOH
H
COOH
COOH
H
COOH
COOH
H
COOH
COOH
H
COOH
COOH
H
COOH
COOH
H
COOH
COOH
H
COOH
COOH
H
H
COOH
H
H
COOH
H
H
COOH
H
H
COOH
H
H
COOH
H
H
COOH
H
H
COOH
H
H
COOH
H
H
H
COOH
H
H
COOH
H
H
COOH
H
H
COOH
H
H
COOH
H
H
COOH
H
3.1Æ0.6
26.8Æ3.2
4.6Æ0.1
1.8Æ0.6
13.1Æ1.2
2.4Æ0.4
9.8Æ0.4
42.9Æ3.1
3.4Æ0.2
5.9Æ0.3
19.1Æ1.8
8.9Æ0.1
0.9Æ0.1
19.1Æ1.2
2.5Æ0.2
0.7Æ0.1
4.9Æ1.4
0.4Æ0.03
1.0Æ0.02
5.5Æ1.2
2.3Æ0.1
0.3Æ0.2
1.1Æ0.9
0.9Æ0.1
0.046Æ0.010
ND
9.6Æ0.7
25Æ5.2
5.1Æ0.2
2.2Æ0.5
14.9Æ0.5
3.9Æ0.2
8.2Æ1.2
20.9Æ3.2
4.3Æ0.7
6.2Æ0.9
21.9Æ4.2
8.2Æ0.3
1.1Æ0.1
21.6Æ1.4
3.1Æ0.7
0.8Æ0.1
3.0Æ0.1
0.9Æ0.07
1.7Æ0.1
4.3Æ1.1
2.1Æ0.4
0.5Æ0.2
1.7Æ0.1
0.6Æ0.08
0.180Æ0.039
ND
17.59
35.18
7.82
31.21
19.56
15.32
26.55
11.04
23.32
33.67
22.99
32.69
23.67
22.35
26.95
37.45
43.88
47.66
34.47
27.75
41.75
27.28
34.74
32.35
12.46
15.75
23.46
ND
33.5
33.65
15.04
16.03
64.14
14.4
15.39
30.8
27.8
15.61
31.22
14.06
7.5
7.5
13.57
16.9
16.92
15.12
8.11
16.2
7.2
ND[e]
0.66
0.23
2.16
Cl
Cl
Cl
NO2
NO2
NO2
NO2
NO2
NO2
CH3
CH3
CH3
CH3
CH3
CH3
H
COOH
H
H
COOH
COOH
novobiocin
isoniazid
rifampicin
ofloxacin
ND
ND
ND
ND
ND
ND
ND
[a] M. smegmatis GyrB ATPase activity; values are the meanÆSD (n=3). [b] M. tuberculosis GyrB DNA supercoiling activity; values are the meanÆSD (n=3).
[c] In vitro anti-Mtb activity. [d] Cytotoxicity (percent inhibition) determined at 100 mm against RAW 264.7 cells. [e] ND: not determined.
ChemMedChem 2016, 11, 539 – 548
www.chemmedchem.org
541
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
lar IC50 values (compounds 15, 18, 20, 24, and 26). Of the ex-
plorations attempted, compound 24 emerged as the lead com-
pound in in vitro studies, exhibiting a GyrB IC50 value of 0.3Æ
0.2 mm, a 10-fold improvement in potency from the hit com-
pound 5 (IC50: 4.6Æ0.1 mm). To our delight, replacement of the
central furanyl core (compounds 3–5, 9–11, 15–17, and 21–23)
with bioisosteric thiophene (6–8, 12–14, 18–20, 24–26) in the
ligand-expansion step brought about a two- to threefold im-
provement in activity, in most cases. This, as understood from
the protein–ligand interaction (Figure 2), was probably due to
an extra p-stacking interaction that the thiophene ring was
able to establish with Arg82, analogous to that observed with
the reference pyrrolamide ligand (PDB ID: 4BAE).[5] Substitution
of the carboxylic acid moiety on the phenyl ring off the hetero-
aryl nucleus also played a critical role in retaining bioactivity.
Shifting the carboxylic acid group from the meta (compounds
3, 5, 6, 8, 9, 11, 12, 14, 15, 17, 18, 20, 21, 23, 24, and 26) to
the para position (compounds 4, 7, 10, 13, 16, 19, 22, and 25)
was found to be detrimental to bioactivity, resulting in less
potent molecules. This observation highlights the importance
of a carboxyl group at this position and is consistent with the
tight hydrogen bonding interactions with the protein observed
with hit compound 5 and the pyrrolamide ligand in the tem-
plate crystal structure used for this study. Substitution attempt-
ed at the 5-position of the 2-(1H-benzo[d]imidazol-2-yl)acetoni-
trile core also played a notable role in determining activity.
Only methyl and nitro groups were well accommodated at this
position, as the compounds in this set (15 and 17–26) exhibit-
ed potent GyrB inhibitory potency; other substituents attempt-
ed did not bring any significant enhancement in potency rela-
tive to hit compound 5. The interaction profiles of hit com-
Scheme 1. Synthetic strategy for generating the designed ligands 3–26. Re-
agents and conditions: a) ethyl 2-cyanoacetate, 2008C; b) corresponding al-
dehyde, piperidine or methanolic KOH (10%), EtOH, 808C.
right-hand heteroaryl core, as shown in Table 1. We also intro-
duced various substituents at the 5-position of the 2-cyano-
methyl benzimidazole core to understand the ideal site for in-
troducing chemical diversity on the benzimidazole ring. Molec-
ular modeling studies also revealed that substitution at the 5-
position could be beneficial for improving the activity relative
to positions 4 and 7 of the benzimidazole ring. Notably, due to
the two possible tautomers of the benzimidazole ring, substi-
tution at positions 5 and 6 becomes indistinguishable and
could only be differentiated by disubstitution, which has not
been explored at this stage of the study.
The medicinal chemistry explorations attempted in the
ligand expansion step gave a comprehensive analysis of SAR
that are important determinants of inhibitory potency. In the in
vitro GyrB assay, 15 compounds exhibited GyrB inhibitory IC50
values of ꢁ5 mm, and four compounds exhibited sub-micromo-
Figure 2. Interaction profile diagram of hit compound 5 and the optimized leads 24 and 26.
ChemMedChem 2016, 11, 539 – 548
www.chemmedchem.org
542
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
pound 5 and two of the more potent lead compounds 24 and
26 is shown in Figure 2 for a better illustration.
supercoiling-inhibitory IC50 value of 0.5Æ0.2 mm. In the M. tu-
berculosis supercoiling assay, 14 compounds exhibited IC50
values <5 mm, and four showed sub-micromolar IC50.
The binding affinity of the most potent analogue was evalu-
ated by measuring the thermal stability of the protein–ligand
complex using biophysical differential scanning fluorimetry
(DSF) experiments (Figure 3).[12] Compound 24 displayed a Tm
shift of 3.68C (Tm =47.68C) relative to that of the native protein
(Tm =448C), a reflection of strong ligand binding to the protein
which is well correlated with its GyrB IC50 value of 0.3Æ0.2 mm.
The DSF curves for the inactive compound 10 along with the
active compound 24 is shown in Supporting Information Fig-
ure S8.
The antimycobacterial potency of all the new molecules
were evaluated by in vitro microplate alamar blue assay
(MABA) against the M. tuberculosis H37Rv strain.[13] All com-
pounds exhibited reasonably good antitubercular potency. The
optimized analogue 24 showed an in vitro M. tuberculosis MIC
of 8.11 mm in the MABA. However, a lack of translation of the
potent in vitro enzyme activity to antimycobacterial potency,
comparable to some of the standards tested, could be due to
variations in cell-wall permeability or efflux pump transport.
Compound 24 was also evaluated for its potency against
the dormant M. tuberculosis (non-replicating persistent) model
using the nutrient starvation protocol as established by Betts
et al.[14] Compound 24 brought ~2.4 and 2.9 log bacterial re-
duction at 10 and 25 mgmLÀ1, respectively, as shown in
Figure 4. This is similar to the effects displayed by the first-line
antitubercular drugs isoniazid and rifampicin, which showed
~1.7 and 2.1 log bacterial reductions, respectively, at
10 mgmLÀ1; however, 24 was found to be less effective than
moxifloxacin, which brought about a ~2.9 log bacterial reduc-
These molecules were also evaluated for their in vitro capaci-
ty to inhibit gyrase-mediated DNA supercoiling.[7] In general,
a good correlation was observed between in vitro GyrB poten-
cy and in vitro supercoiling activity, concordant with the obser-
vations that any inhibition of ATPase activity conferred by the
DNA GyrB subunit should also inhibit the supercoiling activity
performed by the GyrA domain. The most potent compound
in the GyrB assay, compound 24, showed a well-synchronized
tion at 10 mgmLÀ1
.
Considering the severe cardiotoxicity and hERG affinity of
some of the previously reported gyrase inhibitors,[15,16] the
potent analogous compounds in this study were also evaluat-
ed for their hERG channel inhibition by assessing their arrhyth-
mogenic potential on the zERG gene, which is orthologous to
the hERG gene.[17,18] The more potent analogues (18, 24, and
26) were tested at a concentration range of 1–30 mm with
0.1% DMSO as vehicle (Figure 5). Zebrafish embryo heart rate
(HR) variations and atrioventricular (AV) ratios were monitored
using a protocol described in detail in the experimental section
of the Supporting Information. The compounds were found to
be relatively safe relative to the positive control (20 mm terfe-
nadine), not showing any significant cardiotoxicity below
30 mm. Significant changes in HR and AV ratio relative to the
Figure 3. DSF curves for the native GyrB protein (red) and GyrB in complex
with compound 24 (blue); a Tm shift of 3.68C indicates an increase in ther-
mal stability for the GyrB–24 complex relative to GyrB alone.
Figure 4. Bactericidal effects of the most active compound 24, the moderately active compound 3, and the inactive compound 10 on dormant M. tuberculosis
cells. Bacterial log reduction is highest for 24 in comparison with other compounds. Cells were treated with reference drugs or test compounds as indicated
for seven days at 378C. The viability of both treated and untreated (vehicle control) cells was determined by MPN assay (see the Supporting Information for
details). Data are the meanÆSEM of n=2 experiments performed in duplicate.
ChemMedChem 2016, 11, 539 – 548
www.chemmedchem.org
543
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
Table 2. Pharmacokinetic parameters of compound 24 following intrave-
nous (1 mgkgÀ1) administration to Wistar rats.
Parameter
Value[a]
AUC0Àt [hngmLÀ1
]
948.03Æ79.15
1004.13Æ83.21
3.10Æ0.04
AUC0À1 [hngmLÀ1
]
t1/2 [h]
CLtotal [mLhÀ1 kgÀ1
]
995.78Æ79.89
1257.83Æ89.78
Vd [mLkgÀ1
]
[a] Values are the meanÆSD (n=5).
Table 3. Plasma protein binding of compound 24.
Matrix
Binding [%][a]
Human
Rat
96.1Æ1.3
97.2Æ1.6
[a] Plasma protein binding was determined by an ultrafiltration method
using microfilters (Amicon Ultra, Mr cutoff: 10 kDa); values are the
meanÆSD (n=3).
Figure 5. a) Atrial and ventricular heart rates, and b) atrioventricular heart
rate ratios of zebrafish embryos treated with test compounds 18, 24, and 26
at the indicated concentrations, with 20 mm terfenadine (T) as positive refer-
ence; values are the meanÆSEM of n=6 embryos. Statistical significance
between vehicle control (C) and test groups was determined by two- and
one-way ANOVA followed by Bonferroni and Dunnet post-tests (*p<0.05,
**p<0.01, ***p<0.001). See the Supporting Information for details.
In order to predict hepatic clearance and oral bioavailability
in humans, we performed in vitro microsomal clearance studies
in human liver microsomes. The results predicted slow hepatic
clearance (8.5 mLminÀ1 kgÀ1) with moderate oral bioavailability
(~40%) of compound 24 in humans.
These studies demonstrate promising PK properties of com-
pound 24, encouraging further in vivo studies and consider-
ation of compound 24 as a drug candidate. Nevertheless, ex-
tensive pre-clinical toxicological and pharmacological data
would be needed to support the safety and efficacy of the
molecule before extrapolation of the current information to
human use.
control group were also not observed, showing that these
compounds are relatively safe in this regard.
We also evaluated the toxicity profile of all compounds
against the mouse macrophage RAW 264.7 cell line at 100 mm
using the MTT assay.[19] Most of the test compounds did not
show any significant inhibition toward the macrophages
(Table 1), making them relatively safe to host cells in compari-
son with their gyrase inhibitory values. Only the nitro ana-
logues (compounds 15–20) showed slight inhibition at this
tested concentration, with compound 17 showing the greatest
inhibition (47.6%), but this can be considered relatively insig-
nificant at this stage of the study, as the gyrase inhibitory IC50
values are uniformly <20 mm.
Finally, the pharmacokinetic parameters of potent analogue
24 were assessed in male Wistar rats following intravenous
(1 mgkgÀ1) administration. Post-bolus administration, com-
pound 24 showed longer half-life, low plasma clearance and
extensive distribution outside of vascular system (Table 2). For
anti-TB treatment, PK attributes such as extensive distribution
and slow elimination are essential to ensure prolonged drug
availability at infection sites that are remotely located (lungs,
liver, bone marrow, and brain) in the body.
Conclusions
This study presents the discovery of a small-molecule inhibitor
of the Mycobacterium tuberculosis gyrase ATPase domain, iden-
tified through a structure-based drug design approach. Phar-
macophore modeling and subsequent in vitro screening result-
ed in an initial hit compound 5; hit expansion of 5 by chemical
synthesis led to the identification of five sub-micromolar inhibi-
tors of gyrase ATPase. Compound 24 emerged as the most
potent lead, with an ATPase inhibitory IC50 value of 0.34Æ
0.18 mm while showing good antimicrobial potency against
both replicative and dormant stains of M. tuberculosis. The pre-
liminary toxicity and PK and evaluations are encouraging, with
characteristics of promising lead candidates for use in rational
drug design against M. tuberculosis ATPase from a pharmaceuti-
cal point of view.
Furthermore, to asses the protein binding properties of com-
pound 24, plasma protein binding studies were conducted
with human and rat plasma. Data from these studies revealed
an extensive capacity (>95%, n=3) for this compound to
bind plasma protein (Table 3). Considering the fact that only
the free drug concentration in plasma is responsible for anti-
bacterial effect, the data obtained from the in vivo PK studies
should be viewed sagaciously.
Experimental Section
Chemistry
All commercially available chemicals and solvents were used with-
out further purification. Thin-layer chromatography (TLC) was per-
ChemMedChem 2016, 11, 539 – 548
www.chemmedchem.org
544
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
formed on alumina-backed silica gel 40 F254 plates (Merck, Darm-
stadt, Germany). Compound homogeneity was monitored by TLC,
visualized by UV light and KMnO4 treatment. Flash chromatogra-
phy was performed on a Biotage Isolera instrument with prepack-
nitrile (0.25 g, 1.6 mmol), 5-(5-formylfuran-2-yl)isophthalic acid
(0.41 g, 1.6 mmol), and piperidine (0.027 g, 0.32 mmol) to afford 5
1
(0.4 g, 63%) as a buff-colored solid. H NMR ([D6]DMSO): d=7.11–
7.69 (m, 7H), 8.36 (s, 1H), 8.87 (s, 2H), 12.43 (brs, 2H), 12.78 ppm
(brs, 1H); 13C NMR ([D6]DMSO): d=171, 155.8, 150.3, 144.8, 141.3,
138.6, 136.1, 130.8, 130.6, 129.9, 122.8, 119.1, 114.9, 113.4, 110.3,
106.6 ppm; ESI-MS m/z 398.2 [MÀH]+; Anal. calcd for C22H13N3O5: C
66.17, H 3.28, N 10.52, found: C, 66.11, H 3.35, N 10.59.
1
aged disposable normal-phase silica columns. H (300.12 MHz) and
13C NMR (75.12 MHz) spectra were recorded on a Bruker AM-300
NMR spectrometer (Bruker BioSpin Corp., Germany). Chemical
shifts (d) are reported in ppm with reference to internal TMS stan-
dard. Signals are designated as follows: s, singlet; d, doublet; dd,
doublet of doublets; t, triplet; m, multiplet; br, broad. Molecular
weights of the synthesized compounds were determined by LC–
MS (Agilent Technology 6100B series instrument). Elemental analy-
ses were carried out on an automatic Flash EA 1112 Series CHN An-
alyzer (Thermo Scientific). Final compound purity was determined
by HPLC (Shimadzu, Japan) using a Phenomenex C8 double end-
capped RP-HPLC column (150”4.6 mm, 5 mm, 100 Š), and in all
cases was >95%. The synthesis of scaffold compounds 2a–2d is
provided in the Supporting Information. Final library compounds
3–26 were assembled either by following a published procedure,[11]
or as outlined below. 3-(5-Formylfuran-2-yl)benzoic acid, 4-(5-for-
mylfuran-2-yl)benzoic acid, 5-(5-formylfuran-2-yl)isophthalic acid, 3-
(5-formylthiophen-2-yl)benzoic acid, 4-(5-formylthiophen-2-yl)ben-
zoic acid, and 5-(5-formylfuran-2-yl)isophthalic acid, required for
the synthesis of compounds 3–26, were generated by following
the well-known Meerwein arylation protocol.[20–22]
(E)-3-(5-(2-(1H-benzo[d]imidazol-2-yl)-2-cyanovinyl)thiophen-2-
yl)benzoic acid (6): The compound was synthesized according to
the above general procedure using 2-(1H-benzo[d]imidazol-2-yl)a-
cetonitrile (0.25 g, 1.6 mmol), 3-(5-formylthiophen-2-yl)benzoic acid
(0.37 g, 1.6 mmol), and piperidine (0.027 g, 0.32 mmol) to afford 6
(0.45 g, 76%) as an off-white solid; mp: 231–2338C. 1H NMR
([D6]DMSO): d7.21–8.09 (m, 10H), 8.67 (s, 1H), 12.72 (brs, 1H),
13.03 ppm (brs, 1H); 13C NMR ([D6]DMSO): d=170.4, 143.1, 141.8,
141.4, 139.1, 137.6, 133.4, 131.8, 130.9, 130.8, 130.2, 129.9, 125.8,
122.8, 118.7, 115.1, 113.9 ppm; ESI-MS m/z 370.1 [MÀH]+; Anal
calcd for C21H13N3O2S: C 67.91, H 3.53, N 11.31, found: C, 68.02, H
3.49, N 11.28.
(E)-4-(5-(2-(1H-benzo[d]imidazol-2-yl)-2-cyanovinyl)thiophen-2-
yl)benzoic acid (7): The compound was synthesized according to
the above general procedure using 2-(1H-benzo[d]imidazol-2-yl)a-
cetonitrile (0.25 g, 1.6 mmol), 4-(5-formylthiophen-2-yl)benzoic acid
(0.37 g, 1.6 mmol), and piperidine (0.027 g, 0.32 mmol) to afford 7
(0.34 g, 58%) as an off-white solid; mp: 179–1818C. 1H NMR
([D6]DMSO): d=7.23–8.16 (m, 11 H), 12.58 (brs, 1H), 12.83 ppm
(brs, 1H); 13C NMR ([D6]DMSO): d=169.9, 143.1, 141.8, 141.6, 139.1,
138.9, 137.4, 130.2, 130.0, 128.9, 127.6, 127.3, 122.8, 118.7, 115.0,
113.9 ppm; ESI-MS m/z 370.1 [MÀH]+; Anal calcd for C21H13N3O2S:
C 67.91, H 3.53, N 11.31, found: C, 67.83H, 3.59, N 11.27.
General procedure for the synthesis of library analogues 3–26:
To a warm solution of the appropriate (benzimidazol-2-yl)acetoni-
trile (1 mmol) in absolute EtOH (8 mL) was added the correspond-
ing aldehyde (1 mmol) and a catalytic amount of piperidine
(0.2 mmol). The reaction mixture was then stirred and heated at
808C for 0.5–1 h. Reaction completion was monitored by TLC and
LC–MS. The precipitate formed was collected by vacuum filtration
and recrystallized from EtOH to give the desired product in good
yield, as detailed below.
(E)-4-(5-(2-(1H-benzo[d]imidazol-2-yl)-2-cyanovinyl)thiophen-2-
yl)isophthalic acid (8): The compound was synthesized according
to the above general procedure using 2-(1H-benzo[d]imidazol-2-
yl)acetonitrile (0.25 g, 1.6 mmol), 4-(5-formylthiophen-2-yl)isoph-
thalic acid (0.44 g, 1.6 mmol), and piperidine (0.027 g, 0.32 mmol)
to afford 8 (0.51 g, 77%) as a buff-colored solid; mp: 198–2008C.
1H NMR ([D6]DMSO): d7.27–8.43 (m, 10H), 12.68 (brs, 2H),
13.12 ppm (brs, 1H); 13C NMR ([D6]DMSO): d=170.2, 168.7, 143.0,
141.9, 141.6, 139.8, 138.9, 137.8, 132.6, 131.4, 130.0, 129.7, 128.6,
128.3, 127.5, 122.9, 118.6, 114.9, 113.8 ppm; ESI-MS m/z 414.1
[MÀH]+; Anal calcd for C22H13N3O4S: C 63.61, H 3.15, N 10.12,
found: C, 63.55, H 3.17, N 10.16.
(E)-3-(5-(2-(1H-benzo[d]imidazol-2-yl)-2-cyanovinyl)furan-2-yl)-
benzoic acid (3): The compound was synthesized according to the
above general procedure using 2-(1H-benzo[d]imidazol-2-yl)aceto-
nitrile (0.25 g, 1.6 mmol), 3-(5-formylfuran-2-yl)benzoic acid (0.34 g,
1.6 mmol), and piperidine (0.027 g, 0.32 mmol) to afford 3 (0.36 g,
64% yield) as a buff-colored solid; mp: 244–2468C. 1H NMR
([D6]DMSO): d=6.89–8.23 (m, 10H), 8.76 (s, 1H), 12.72 (brs, 1H),
13.29 ppm (brs, 1H); 13C NMR ([D6]DMSO): d=170.2, 156.3, 150.8,
144.8, 141.6, 139.0, 130.9, 130.8, 130.2, 130.0, 129.8, 125.9, 124.1,
118.6, 115.4, 113.8, 110.2, 106.1 ppm; ESI-MS m/z 354.2 [MÀH]+;
Anal calcd for C21H13N3O3: C 70.98, H 3.69, N 11.83, found: C, 70.86,
H 3.65, N 11.88.
(E)-3-(5-(2-(5-chloro-1H-benzo[d]imidazol-2-yl)-2-cyanovinyl)fur-
an-2-yl)benzoic acid (9): The compound was synthesized accord-
ing to the above general procedure using 2-(5-chloro-1H-benzo[d]i-
midazol-2-yl)acetonitrile (0.25 g, 1.3 mmol), 3-(5-formylfuran-2-yl)-
benzoic acid (0.28 g, 1.3 mmol), and piperidine (0.022 g,
0.26 mmol) to afford 9 (0.27 g, 53%) as an off-white solid; mp:
(E)-4-(5-(2-(1H-benzo[d]imidazol-2-yl)-2-cyanovinyl)furan-2-yl)-
benzoic acid (4): The compound was synthesized according to the
above general procedure using 2-(1H-benzo[d]imidazol-2-yl)aceto-
nitrile (0.25 g, 1.6 mmol), 4-(5-formylfuran-2-yl)benzoic acid (0.34 g,
1.6 mmol), and piperidine (0.027 g, 0.32 mmol) to afford 4 (0.31 g,
55%) as an off-white solid; mp: 289–2918C. 1H NMR ([D6]DMSO):
d=6.91–7.31 (m, 4H), 7.53–7.59 (m, 3H), 8.12–8.24 (m, 4H), 12.32
(brs, 1H), 12.98 ppm (brs, 1H); 13C NMR ([D6]DMSO): d=170.1,
156.1, 150.7, 145.3, 141.7, 139.0, 135.7, 130.4, 127.7, 123.5, 123.2,
118.9, 115.4, 113.8, 109.7, 106.5 ppm; ESI-MS m/z 354.1 [MÀH]+;
Anal calcd for C21H13N3O3: C 70.98, H 3.69, N 11.83, found: C, 70.87,
H 3.63, N 11.89.
1
259–2618C. H NMR ([D6]DMSO): d=6.89–8.37 (m, 9H), 8.71 (s, 1H),
12.36 (brs, 1H), 12.71 ppm (brs, 1H); 13C NMR ([D6]DMSO): d=
170.1, 156.2, 150.7, 144.8, 141.5, 140.4, 136.9, 130.8, 130.7, 130.3,
130.1, 130.0, 129.2, 125.9, 124.2, 118.9, 116.6, 115.8, 113.9, 110.0,
106.5 ppm; ESI-MS m/z 388.1 [MÀH]+; Anal calcd for C21H12ClN3O3:
C 64.71, H 3.10, N 10.78, found: C, 64.78, H 3.18, N 10.71.
(E)-4-(5-(2-(5-chloro-1H-benzo[d]imidazol-2-yl)-2-cyanovinyl)fur-
an-2-yl)benzoic acid (10): The compound was synthesized accord-
ing to the above general procedure using 2-(5-chloro-1H-benzo[d]i-
midazol-2-yl)acetonitrile (0.25 g, 1.3 mmol), 4-(5-formylfuran-2-yl)-
benzoic acid(0.28 g, 1.3 mmol), and piperidine (0.022 g, 0.26 mmol)
to afford 10 (0.21 g, 41%) as a buff-colored solid; mp: 229–2308C.
(E)-5-(5-(2-(1H-benzo[d]imidazol-2-yl)-2-cyanovinyl)furan-2-yl)iso-
phthalic acid (5): The compound was synthesized according to the
above general procedure using 2-(1H-benzo[d]imidazol-2-yl)aceto-
ChemMedChem 2016, 11, 539 – 548
www.chemmedchem.org
545
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
1H NMR ([D6]DMSO): d=6.85–8.31 (m, 10H), 12.33 (brs, 1H),
12.67 ppm (brs, 1H); 13C NMR ([D6]DMSO): d=170.2, 155.9, 150.8,
145.1, 141.7, 140.4, 136.8, 135.3, 130.4, 129.3, 127.8, 124.3, 123.3,
118.8, 116.8, 115.9, 113.8, 110.0, 106.7 ppm; ESI-MS m/z 388.1
[MÀH]+; Anal calcd for C21H12ClN3O3: C 64.71, H 3.10, N 10.78,
found: C, 64.78, H 3.06 N, 10.72.
afford 15 (0.28 g, 57%) as solid; mp: 289–2918C. 1H NMR
([D6]DMSO): d=6.91–7.11 (m, 2H), 7.53–8.39 (m, 7H), 8.71 (s, 1H),
12.74 (brs, 1H), 13.18 ppm (brs, 1H); 13C NMR ([D6]DMSO): d=
170.5, 156.4, 150.8, 145.1, 144.9, 144.3, 141.7, 140.1, 130.8, 130.6,
130.4, 130.3, 130.1, 126.3, 118.9, 118.7, 115.9, 113.8, 113.0, 110.1,
106.3 ppm; ESI-MS m/z 399.1 [MÀH]+; Anal calcd for C21H12N4O5: C
63.00, H 3.02, N 13.99, found: C, 63.09, H 3.09, N 14.05.
(E)-5-(5-(2-(5-chloro-1H-benzo[d]imidazol-2-yl)-2-cyanovinyl)fur-
an-2-yl)isophthalic acid (11): The compound was synthesized ac-
cording to the above general procedure using 2-(5-chloro-1H-ben-
zo[d]imidazol-2-yl)acetonitrile (0.25 g, 1.3 mmol), 5-(5-formylfuran-
2-yl)isophthalic acid (0.34 g, 1.3 mmol), and piperidine (0.022 g,
0.26 mmol) to afford 11 (0.26 g, 46%) as a buff-colored solid; mp:
(E)-4-(5-(2-cyano-2-(5-nitro-1H-benzo[d]imidazol-2-yl)vinyl)furan-
2-yl)benzoic acid (16): The compound was synthesized according
to the above general procedure using 2-(5-nitro-1H-benzo[d]imida-
zol-2-yl)acetonitrile (0.25 g, 1.24 mmol) and 4-(5-formylfuran-2-yl)-
benzoic acid (0.26 g, 1.24 mmol), and piperidine (0.021 g,
0.25 mmol) to afford 16 (0.36 g, 73%) as solid; mp: 229–2318C.
1H NMR ([D6]DMSO): d6.87–8.33 (m, 10H), 12.57 (brs, 1H),
12.97 ppm (brs, 1H); 13C NMR ([D6]DMSO): d=170.3, 155.9, 150.3,
145.2, 144.8, 144.6, 141.8, 140.0, 135.6, 130.3, 127.3, 122.9, 119.1,
118.7, 116.0, 113.5, 113.0, 110.2, 106.3 ppm; ESI-MS m/z 399.1
[MÀH]+; Anal calcd for C21H12N4O5: C, 63, H 3.02, N 13.99, found: C,
62.91, H 3.07, N 14.06.
1
289–2918C. H NMR ([D6]DMSO): d=7.01–7.53 (m, 5H), 8.34 (s, 1H),
8.46 (s, 1H), 8.89 (s, 2H), 12.64 (brs, 2H), 12.89 ppm (brs, 1H);
13C NMR ([D6]DMSO): d=171.1, 156.2, 150.7, 145.2, 141.7, 140.5,
137.1, 136.2, 130.7, 130.5, 129.9, 129.1, 124.1, 118.9, 116.7, 115.9,
113.8, 109.8, 106.3 ppm; ESI-MS m/z 432.1 [MÀH]+; Anal calcd for
C22H12ClN3O5: C 60.91, H 2.79, N 9.69, found: C, 60.83, H 2.72, N
9.61.
(E)-3-(5-(2-(5-chloro-1H-benzo[d]imidazol-2-yl)-2-cyanovinyl)thio-
phen-2-yl)benzoic acid (12): The compound was synthesized ac-
cording to the above general procedure using 2-(5-chloro-1H-ben-
zo[d]imidazol-2-yl)acetonitrile (0.25 g, 1.3 mmol), 3-(5-formylthio-
phen-2-yl)benzoic acid (0.3 g, 1.3 mmol), and piperidine (0.022 g,
0.26 mmol) to afford 12 (0.34 g, 64%) as solid; mp: 253–2558C.
1H NMR ([D6]DMSO): d=7.32–8.18 (m, 8H), 8.37 (s, 1H), 8.68 (s,
1H), 12.39 (brs, 1H), 12.73 ppm (brs, 1H); 13C NMR ([D6]DMSO): d=
170.6, 143.2, 141.8, 141.6, 140.4, 137.6, 136.9, 133.8, 131.7, 130.8,
130.6, 130.2, 130.1, 129.2, 128.8, 126.2, 124.1, 118.9, 116.7, 115.9,
113.8 ppm; ESI-MS m/z 404.1 [MÀH]+; Anal calcd for
C21H12ClN3O2S: C 62.15, H 2.98, N 10.35, found: C, 62.19, H 2.91, N
10.28.
(E)-5-(5-(2-cyano-2-(5-nitro-1H-benzo[d]imidazol-2-yl)vinyl)furan-
2-yl)isophthalic acid (17): The compound was synthesized accord-
ing to the above general procedure using 2-(5-nitro-1H-benzo[d]i-
midazol-2-yl)acetonitrile (0.25 g, 1.24 mmol), 5-(5-formylfuran-2-yl)i-
sophthalic acid (0.32 g, 1.24 mmol), and piperidine (0.021 g,
0.25 mmol) to afford 17 (0.37 g, 67%) as solid; mp: 287–2898C.
1H NMR ([D6]DMSO): d=6.89–8.61 (m, 7H), 8.92 (s, 2H, 12.84 (brs,
2H), 13.04 ppm (brs, 1H); 13C NMR ([D6]DMSO): d=170.9, 156.1,
150.3, 145.1, 144.9, 144.6, 141.7, 140.0, 136.1, 130.8, 130.5, 130.0,
118.9, 118.6, 115.8, 113.5, 112.9, 110.0, 106.2 ppm; ESI-MS m/z 443.1
[MÀH]+; Anal calcd for C22H12N4O7: C, 59.47, H 2.72, N 12.61, found:
C, 59.38, H 2.78, N 12.53.
(E)-3-(5-(2-(5-nitro-1H-benzo[d]imidazol-2-yl)-2-cyanovinyl)thio-
phen-2-yl)benzoic acid (18): The compound was synthesized ac-
cording to the above general procedure using 2-(5-nitro-1H-ben-
zo[d]imidazol-2-yl)acetonitrile (0.25 g, 1.24 mmol), 3-(5-formylthio-
phen-2-yl)benzoic acid (0.29 g, 1.24 mmol), and piperidine (0.021 g,
0.25 mmol) to afford 18 (0.31 g, 61%) as solid; mp: 231–2348C.
1H NMR ([D6]DMSO): d=7.39–8.11 (m, 8H), 8.43 (s, 1H), 8.74 (s, 1H),
12.63 (brs, 1H), 13.18 ppm (brs, 1H); 13C NMR ([D6]DMSO): d=
170.5, 145.2, 144.4, 143.0, 141.8, 141.6, 140.0, 137.7, 133.8, 131.7,
130.8, 130.6, 130.2, 130.0, 128.6, 126.2, 118.7, 118.6, 116.2, 113.8,
113 ppm; ESI-MS m/z 415.1 [MÀH]+; Anal calcd for C21H12N4O4S: C
60.57, H 2.90, N 13.45, found: C, 60.51, H 2.83, N 13.53.
(E)-4-(5-(2-(5-chloro-1H-benzo[d]imidazol-2-yl)-2-cyanovinyl)thio-
phen-2-yl)benzoic acid (13): The compound was synthesized ac-
cording to the above general procedure using 2-(5-chloro-1H-ben-
zo[d]imidazol-2-yl)acetonitrile (0.25 g, 1.3 mmol), 4-(5-formylthio-
phen-2-yl)benzoic acid (0.3 g, 1.3 mmol), and piperidine (0.022 g,
0.26 mmol) to afford 13 (0.40 g, 76%) as solid; mp: 277–2798C.
1H NMR ([D6]DMSO): d=7.23–8.13 (m, 10H), 11.96 (brs, 1H),
12.78 ppm (brs, 1H); 13C NMR ([D6]DMSO): d=169.8, 143.2, 141.8,
141.6, 140.7, 140.0, 137.8, 136.9, 130.3, 130.1, 129.4, 128.7, 127.9,
127.6, 123.9, 118.6, 116.7, 116.0, 113.9 ppm; ESI-MS m/z 404.1
[MÀH]+; Anal calcd for C21H12ClN3O2S: C, 62.15, H 2.98, N 10.35,
found: C, 62.10, H 3.08, N 10.31.
(E)-4-(5-(2-(5-nitro-1H-benzo[d]imidazol-2-yl)-2-cyanovinyl)thio-
phen-2-yl)benzoic acid (19): The compound was synthesized ac-
cording to the above general procedure using 2-(5-nitro-1H-ben-
zo[d]imidazol-2-yl)acetonitrile (0.25 g, 1.24 mmol), 4-(5-formylthio-
phen-2-yl)benzoic acid (0.29 g, 1.24 mmol), and piperidine (0.021 g,
0.25 mmol) to afford 20 (0.3 g, 58%) as solid; mp: 211–2138C.
1H NMR ([D6]DMSO): d=7.34–8.10 (m, 9H), 8.31 (s, 1H), 12.78 (brs,
1H), 12.92 ppm (brs, 1H); 13C NMR ([D6]DMSO): d=170.2, 145.3,
144.3, 143.1, 141.8, 141.6, 140.0, 138.9, 137.6, 130.3, 130.1, 128.9,
127.5, 127.2, 118.7, 118.5, 116.2, 113.8, 113.2 ppm; ESI-MS m/z 415.1
[MÀH]+; Anal calcd for C21H12N4O4S: C 60.57, H 2.90, N 13.45,
found: C, 60.52, H 2.84, N 13.41.
(E)-4-(5-(2-(5-chloro-1H-benzo[d]imidazol-2-yl)-2-cyanovinyl)thio-
phen-2-yl)isophthalic acid (14): The compound was synthesized
according to the above general procedure using 2-(5-chloro-1H-
benzo[d]imidazol-2-yl)acetonitrile (0.25 g, 1.3 mmol), 4-(5-formylth-
iophen-2-yl)isophthalic acid (0.36 g, 1.3 mmol), and piperidine
(0.022 g, 0.26 mmol) to afford 14 (0.34 g, 57%) as solid; mp: 287–
1
2898C. H NMR ([D6]DMSO): d=7.34–8.39 (m, 9H), 12.38 (brs, 2H),
13.12 ppm (brs, 1H); 13C NMR ([D6]DMSO): d=170.6, 169.2, 143.0,
141.8, 141.5, 140.2, 139.6, 137.4, 136.8, 132.9, 131.2, 130.3, 130.1,
129.3, 128.7, 128.2, 127.4, 124.2, 118.9, 116.7, 115.9, 113.7 ppm; ESI-
MS m/z 448.1 [MÀH]+; Anal calcd for C22H12ClN3O4S: C 58.74, H
2.69, N 9.34, found: C, 58.71, H 2.61, N 9.27.
(E)-4-(5-(2-(5-nitro-1H-benzo[d]imidazol-2-yl)-2-cyanovinyl)thio-
phen-2-yl)isophthalic acid (20): The compound was synthesized
according to the above general procedure using 2-(5-nitro-1H-ben-
zo[d]imidazol-2-yl)acetonitrile (0.25 g, 1.24 mmol), 4-(5-formylthio-
phen-2-yl)isophthalic acid (0.34 g, 1.24 mmol), and piperidine
(0.021 g, 0.25 mmol) to afford 20 (0.41 g, 72%) as solid; mp: 199–
(E)-3-(5-(2-(5-nitro-1H-benzo[d]imidazol-2-yl)-2-cyanovinyl)furan-
2-yl)benzoic acid (15): The compound was synthesized according
to the above general procedure using 2-(5-nitro-1H-benzo[d]imida-
zol-2-yl)acetonitrile (0.25 g, 1.24 mmol), 3-(5-formylfuran-2-yl)ben-
zoic acid (0.26 g, 1.2 mmol), and piperidine (0.021 g, 0.25 mmol) to
ChemMedChem 2016, 11, 539 – 548
www.chemmedchem.org
546
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
1
2018C. H NMR ([D6]DMSO): d=7.37–8.45 (m, 9H), 12.79 (brs, 2H),
0.29 mmol) to afford 25 (0.37 g, 66%) as solid; mp: 243–2458C.
1H NMR ([D6]DMSO): d=2.27 (s, 3H), 7.13–8.21 (m, 10H), 12.16
(brs, 1H), 12.68 ppm (brs, 1H); 13C NMR ([D6]DMSO): d=170.3,
143.1, 141.7, 141.3, 139.0, 138.6, 137.8, 136.1, 132.4, 130.3, 129.9,
128.5, 127.6, 127.2, 126.0, 119.0, 115.7, 115.3, 113.9, 21.4 ppm; ESI-
MS m/z 384.1 [MÀH]+; Anal. calcd for C22H15N3O2S: C, 68.55, H 3.92,
N 10.90. Found: C, 68.48, H 3.97, N 10.95.
13.11 ppm (brs, 1H); 13C NMR ([D6]DMSO): d=170.6, 168.9, 145.2,
144.3, 143.1, 141.8, 141.6, 140.0, 139.6, 137.6, 132.9, 131.3, 130.2,
130.0, 128.8, 128.6, 127.4, 118.9, 118.7, 116.2, 113.8, 113.1 ppm; ESI-
MS m/z 459.1 [MÀH]+; Anal calcd for C22H12N4O6S: C 57.39, H 2.63,
N 12.17, found: C, 57.31, H 2.68, N 12.11.
(E)-3-(5-(2-(5-methyl-1H-benzo[d]imidazol-2-yl)-2-cyanovinyl)fur-
an-2-yl)benzoic acid (21): The compound was synthesized accord-
ing to the above general procedure using 2-(5-methyl-1H-benzo[-
d]imidazol-2-yl)acetonitrile (0.25 g, 1.46 mmol), 3-(5-formylfuran-2-
yl)benzoic acid (0.32 g, 1.46 mmol), and piperidine (0.025 g,
0.29 mmol) to afford 21 (0.44 g, 81%) as solid; mp: 261- 2638C.
1H NMR ([D6]DMSO): d=2.36 (s, 3H), 6.95–7.16 (m, 3H), 7.41–8.5
(m, 6H), 8.73 (s, 1H), 12.38 (brs, 1H), 12.69 ppm (brs, 1H); 13C NMR
([D6]DMSO): d=170.3, 156.3, 150.7, 145.0, 141.7, 138.9, 136.0,
132.7, 130.8, 130.7, 130.3, 130.1, 129.9, 126.3, 125.9, 118.9, 115.4,
115.2, 113.8, 110.0, 106.6, 21.7 ppm; ESI-MS m/z 368.1 [MÀH]+;
Anal calcd for C22H15N3O3: C 71.54, H 4.09, N 11.38, found: C, 71.48,
H 4.02, N 11.44.
(E)-4-(5-(2-(5-methyl-1H-benzo[d]imidazol-2-yl)-2-cyanovinyl)th-
iophen-2-yl)isophthalic acid (26): The compound was synthesized
according to the above general procedure using 2-(5-methyl-1H-
benzo[d]imidazol-2-yl)acetonitrile (0.25 g, 1.46 mmol), 4-(5-for-
mylthiophen-2-yl)isophthalic acid (0.4 g, 1.46 mmol), and piperidine
(0.025 g, 0.29 mmol) to afford 26 (0.43 g, 68%) as solid; mp: 240–
2428C. 1H NMR ([D6]DMSO): d=2.27 (s, 3H), 7.19–8.41 (m, 9H),
11.96 (brs, 2H), 12.53 ppm (brs, 1H); 13C NMR ([D6]DMSO): d=
170.3, 168.8, 143.2, 141.8, 141.6, 139.6, 138.9, 137.6, 136.2, 132.8,
132.6, 131.2, 130.0, 129.8, 128.6, 128.3, 127.2, 126.0, 118.9, 115.2,
115.0, 113.8, 21.6 ppm; ESI-MS m/z 428.1 [MÀH]+; Anal calcd for
C23H15N3O4S: C 64.33, H 3.52, N 9.78, found: C, 64.28, H 3.57, N
9.84.
(E)-4-(5-(2-(5-methyl-1H-benzo[d]imidazol-2-yl)-2-cyanovinyl)fur-
an-2-yl)benzoic acid (22): The compound was synthesized accord-
ing to the above general procedure using 2-(5-methyl-1H-benzo[-
d]imidazol-2-yl)acetonitrile (0.25 g, 1.46 mmol), 4-(5-formylfuran-2-
yl)benzoic acid (0.32 g, 1.46 mmol), and piperidine (0.025 g,
0.29 mmol) to afford 22 (0.24 g, 45%) as solid; mp: 214–2168C.
1H NMR ([D6]DMSO): d=2.34 (s, 3H), 6.89–7.49 (m, 6H), 8.09–8.23
(m, 4H), 11.87 (brs, 1H), 12.38 ppm (brs, 1H); 13C NMR ([D6]DMSO):
d=170.4, 156.2, 150.7, 144.9, 141.5, 139.1, 136.1, 135.7, 132.8,
130.3, 127.7, 126.0, 123.4, 118.7, 115.4, 115.1, 113.8, 109.8, 106.7,
21.5 ppm; ESI-MS m/z 368.1 [MÀH]+; Anal calcd for C22H15N3O3: C
71.54, H 4.09, N 11.38, found: C, 71.58, H 4.14, N 11.32.
Supporting Information
The Supporting Information contains detailed experimental proce-
dures used for docking, the synthesis of scaffolds 2a–2d, biologi-
cal and biophysical evaluations, safety profiling, and pharmacoki-
netics assays.
Acknowledgements
D.S. acknowledges the University Grant Commission, India for
a UGC Researcher award.
(E)-5-(5-(2-(5-methyl-1H-benzo[d]imidazol-2-yl)-2-cyanovinyl)fur-
an-2-yl)isophthalic acid (23): The compound was synthesized ac-
cording to the above general procedure using 2-(5-methyl-1H-ben-
zo[d]imidazol-2-yl)acetonitrile (0.25 g, 1.46 mmol),5-(5-formylfuran-
2-yl)isophthalic acid (0.38 g, 1.46 mmol), and piperidine (0.025 g,
0.29 mmol) to afford 23 (0.46 g, 76%) as solid; mp: 276–2788C.
1H NMR ([D6]DMSO): d=2.31 (s, 3H), 6.99–7.58 (m, 6H), 8.43 (s,
1H), 8.86 (s, 2H), 12.24 (brs, 2H), 12.58 ppm (brs, 1H); 13C NMR
([D6]DMSO): d=171, 156.2, 150.5, 144.9, 141.3, 138.7, 136.0, 135.9,
132.8, 130.7, 130.5, 130.0, 126.0, 119.0, 115.3, 115.1, 113.9, 110.2,
106.2, 21.6 ppm; ESI-MS m/z 412.1 [MÀH]+; Anal calcd for
C23H15N3O5: C 66.83, H 3.66, N 10.16, found: C, 66.77, H 3.63, N
10.10.
Keywords: ATPase domain · DNA gyrase · e-pharmacophores ·
Mycobacterium tuberculosis · Structure-based drug design
[1] C. Lienhardt, M. Raviglione, M. Spigelman, R. Hafner, E. Jaramillo, M.
Hoelscher, A. Zumla, J. Gheuens, J. Infect. Dis. 2012, 205, S241–249.
[2] M. Zignol, W. van Gemert, D. Falzon, C. Sismanidis, P. Glaziou, K. Floyd,
M. S. Raviglione, Bull. World Health Organ. 2012, 90, 111 –119D.
[3] S. Chopra, K. Matsuyama, T. Tran, J. P. Malerich, B. Wan, S. G. Franzblau,
S. Lun, H. Guo, J. Antimicrob. Chemother. 2012, 67, 415–421.
[4] K. Mdluli, Z. Ma, Infect. Disord.: Drug Targets 2007, 7, 159–168.
[5] P. S. Hameed, S. Solapure, K. Mukherjee, V. Nandi, D. Waterson, R. Shan-
dil, M. Balganesh, V. K. Sambandamurthy, A. K. Raichurkar, A. Desh-
pande, A. Ghosh, D. Awasthy, G. Shanbhag, G. Sheikh, H. McMiken, J.
Puttur, J. Reddy, J. Werngren, J. Read, M. Kumar, M. R, M. Chinnapattu,
P. Madhavapeddi, P. Manjrekar, R. Basu, S. Gaonkar, S. Sharma, S. Hoff-
ner, V. Humnabadkar, V. Subbulakshmi, V. Panduga, J. Antimicrob. Che-
mother. 2014, 58, 61–70.
[6] P. S. Shirude, P. Madhavapeddi, J. A. Tucker, K. Murugan, V. Patil, H. Basa-
varajappa, A. V. Raichurkar, V. Humnabadkar, S. Hussein, S. Sharma, V. K.
Ramya, C. B. Narayan, T. S. Balganesh, V. K. Sambandamurthy, ACS Chem.
Biol. 2013, 8, 519–523.
[7] V. U. Jeankumar, R. Janupally, V. K. Pulla, V. Soni, J. P. Sridevi, P. Suryade-
vara, M. Shravan, R. Medishetti, P. Kulkarni, P. Yogeeswari, D. Sriram, Int.
J. Antimicrob. Agents 2014, 43, 269–278.
(E)-3-(5-(2-cyano-2-(5-methyl-1H-benzo[d]imidazol-2-yl)vinyl)th-
iophen-2-yl)benzoic acid (24): The compound was synthesized ac-
cording to the above general procedure using 5-methyl-((2-benzi-
midazolyl)acetonitrile) (0.25 g, 1.46 mmol), 3-(5-formylthiophen-2-
yl)benzoic acid (0.34 g, 1.46 mmol), and piperidine (0.025 g,
0.29 mmol) to afford 24 (0.30 g, 54%) as solid; mp: 267–2698C.
1H NMR ([D6]DMSO): d=2.31 (s, 3H), 7.19–8.71 (m, 10H), 12.56
(brs, 1H), 12.83 ppm (brs, 1H); 13C NMR ([D6]DMSO): d=170.6,
143.2, 141.9, 141.3, 138.6, 137.6, 136.2, 133.8, 132.9, 131.8, 130.6,
130.3, 129.8, 129.6, 128.2, 125.8, 125.3, 118.9, 115.6, 115.3, 114.0,
21.6 ppm; ESI-MS m/z 384 [MÀH]+; Anal. calcd for C22H15N3O2S: C,
68.55, H 3.92, N 10.90. Found: C, 68.51, H 3.97, N 10.86.
[8] M. RouØ, A. Agrawal, C. Volker, D. Mossakowska, C. Mayer, B. D. Bax,
Acta Crystallogr. Sect. F 2013, 69, 679–682.
[9] M. G. Kale, A. Raichurkar, P. S. H. D. Waterson, D. McKinney, M. R. Manju-
natha, U. Kranthi, K. Koushik, L. K. Jena, V. Shinde, S. Rudrapatna, S.
Barde, V. Humnabadkar, P. Madhavapeddi, H. Basavarajappa, A. Ghosh,
V. K. Ramya, S. Guptha, S. Sharma, P. Vachaspati, K. N. Kumar, J. Giridhar,
(E)-4-(5-(2-cyano-2-(5-methyl-1H-benzo[d]imidazol-2-yl)vinyl)th-
iophen-2-yl)benzoic acid (25): The compound was synthesized ac-
cording to the above general procedure using 5-methyl-((2-benzi-
midazolyl)acetonitrile) (0.25 g, 1.46 mmol), 4-(5-formylthiophen-2-
yl)benzoic acid (0.34 g, 1.46 mmol), and piperidine (0.025 g,
ChemMedChem 2016, 11, 539 – 548
www.chemmedchem.org
547
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
J. Reddy, V. Panduga, S. Ganguly, V. Ahuja, S. Gaonkar, C. N. Kumar, D.
Ogg, J. A. Tucker, P. A. Boriack-Sjodin, S. M. de Sousa, V. K. Sambanda-
murthy, S. R. Ghorpade, J. Med. Chem. 2013, 56, 8834–8848.
[10] R. R. Kale, M. G. Kale, D. Waterson, A. Raichurkar, S. P. Hameed, M. R.
Manjunatha, B. K. Kishore Reddy, K. Malolanarasimhan, V. Shinde, K.
Koushik, L. K. Jena, S. Menasinakai, V. Humnabadkar, P. Madhavapeddi,
H. Basavarajappa, S. Sharma, R. Nandishaiah, K. N. Mahesh Kumar, S.
Ganguly, V. Ahuja, S. Gaonkar, C. N. Naveen Kumar, D. Ogg, P. A. Boriack-
Sjodin, V. K. Sambandamurthy, S. M. de Sousa, S. R. Ghorpade, Bioorg.
Med. Chem. Lett. 2014, 24, 870–879.
[16] F. Reck, R. Alm, P. Brassil, J. Newman, B. DeJonge, C. J. Eyermann, G.
Breault, J. Breen, J. Comita-Prevoir, M. Cronin, H. Davis, D. Ehmann, V.
Galullo, B. Geng, T. Grebe, M. Morningstar, P. Walker, B. Hayter, S. Fisher,
J. Med. Chem. 2011, 54, 7834–7847.
[17] V. U. Jeankumar, R. R. Srilakshmi, J. Renuka, S. Saxena, J. P. Sridevi, B.
Medapi, P. Kulkarni, P. Yogeeswari, D. Sriram, Org. Biomol. Chem. 2015,
13, 2423–2431.
[18] S. W. Mittelstadt, C. L. Hemenway, M. P. Craig, J. R. Hove, J. Pharmacol.
Toxicol. Methods 2008, 57, 100–105.
[19] D. Gerlier, N. Thomasset, Immunol. Methods 1986, 94, 57–63.
[20] M. Y. Kornilov, E. M. Ruban, V. N. Fredchuk, E. V. Starinskya, M. V. Buznik,
Zh. Org. Khim. 1973, 9, 2577–2582.
[11] F. Saczewski, P. Reszka, M. Gdaniec, R. Grunert, P. J. Bednarski, J. Med.
Chem. 2004, 47, 3438–3449.
[12] F. H. Niesen, H. Berglund, M. Vedadi, Nat. Protoc. 2007, 2, 2212–2221.
[13] S. G. Franzbalu, R. S. Witzig, J. C. McLaughlin, P. Torres, G. Madico, A.
Hernandez, M. T. Degnan, M. B. Cook, V. K. Quenzer, R. M. Ferguson,
R. H. Gilman, J. Clin. Microbiol. 1998, 36, 362–366.
[21] S. Malinowski, Polish J. Chem. 1953, 27, 54–61.
[22] L. Racan›, V. Tralic-Kulenovic, D. W. Boykin, G. Karminski-Zamola, Mole-
cules 2003, 8, 342–349.
[14] J. C. Betts, P. T. Lukey, L. C. Robb, R. A. McAdam, K. Duncan, Mol. Micro-
biol. 2002, 43, 717–731.
Received: November 26, 2015
[15] A. J. Alexandrou, R. S. Duncan, A. Sullivan, J. C. Hancox, D. J. Leishman,
H. J. Witchel, J. L. Leaney, Br. J. Pharmacol. 2006, 147, 905–916.
Published online on January 25, 2016
ChemMedChem 2016, 11, 539 – 548
www.chemmedchem.org
548
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim