Design, synthesis, structural characterization by IR, 1H, 13C, 15N, 2D-NMR, X-ray diffraction and evaluation of a new class of phenylaminoacetic acid benzylidene hydrazines as pfenr Inhibitors

Article abstract of DOI:10.1111/cbdd.12118

Recent studies have revealed that plasmodial enoyl-ACP reductase (pfENR, FabI), one of the crucial enzymes in the plasmodial type II fatty acid synthesis II (FAS II) pathway, is a promising target for liver stage malaria infections. Hence, pfENR inhibitors have the potential to be used as causal malarial prophylactic agents. In this study, we report the design, synthesis, structural characterization and evaluation of a new class of pfENR inhibitors. The search for inhibitors began with a virtual screen of the iResearch database by molecular docking. Hits obtained from the virtual screen were ranked according to their Glide score. One hit was selected as a lead and modified to improve its binding to pfENR; from this, a series of phenylamino acetic acid benzylidene hydrazides were designed and synthesized. These molecules were thoroughly characterized by IR, ¹H, ¹³C, ¹⁵N, 2D-NMR (COSY, NOESY, ¹H-¹³C, ¹H-¹⁵N HSQC and HMBC), and X-ray diffraction. NMR studies revealed the existence of conformational/configurational isomers around the amide and imine functionalities. The major species in DMSO solution is the E, E form, which is in dynamic equilibrium with the Z, E isomer. In the solid state, the molecule has a completely extended conformation and forms helical structures that are stabilized by strong hydrogen bond interactions, forming a helical structure stabilized by N-H...O interactions, a feature unique to this class of compounds. Furthermore, detailed investigation of the NMR spectra indicated the presence of a minor impurity in most compounds. The structure of this impurity was deduced as an imidazoline-4-one derivative based on ¹H-¹³C and ¹H-¹⁵H HMBC spectra and was confirmed from the NOESY spectra. The molecules were screened for in vitro activity against recombinant pfENR enzyme by a spectrophotometric assay. Four molecules, viz. 17, 7, 10, and 12 were found to be active at 7, 8, 10, and 12?μm concentration, respectively, showing promising pfENR inhibitory potential. A classification model was derived based on a binary QSAR approach termed recursive partitioning (RP) to highlight structural characteristics that could be tuned to improve activity.

Full text of DOI:10.1111/cbdd.12118

Chem Biol Drug Des 2013
Research Article
Design, Synthesis, Structural Characterization by
1
13
15
IR, H, C, N, 2D-NMR, X-Ray Diffraction and
Evaluation of a New Class of Phenylaminoacetic Acid
Benzylidene Hydrazines as pfENR Inhibitors
Ramanuj P. Samal1, Vijay M. Khedkar1,
this class of compounds. Furthermore, detailed investi-
gation of the NMR spectra indicated the presence of a
Raghuvir R. S. Pissurlenkar1, Angela Gono
Bwalya2, Deniz Tasdemir2,3, Ramesh A. Joshi4,
minor impurity in most compounds. The structure of
this impurity was deduced as an imidazoline-4-one
P. R. Rajamohanan4, Vedavati G. Puranik4 and
derivative based on 1H-13C and 1H-15H HMBC spectra
Evans C. Coutinho1,*
and was confirmed from the NOESY spectra. The mol-
ecules were screened for in vitro activity against
recombinant pfENR enzyme by a spectrophotometric
assay. Four molecules, viz. 17, 7, 10, and 12 were
found to be active at 7, 8, 10, and 12 lM concentration,
1Department of Pharmaceutical Chemistry, Bombay
College of Pharmacy, Kalina, Santacruz (East), Mumbai,
400 098, India
2Department of Pharmaceutical & Biological Chemistry,
respectively, showing promising pfENR inhibitory
School of Pharmacy, University College London, 29–39
potential. A classification model was derived based on
Brunswick Square, London, WC1N 1AX, UK
a binary QSAR approach termed recursive partitioning
3School of Chemistry, National University of Ireland,
(RP) to highlight structural characteristics that could
Galway, Arts/Science Building, Room 213, University
be tuned to improve activity.
Road, Galway, Ireland, UK
4National Chemical Laboratory, Dr. Homi Bhabha Road,
Pune, 411008, India
Key words: ADMET, docking, enoyl-ACP reductase, FabI,
hydrogen bonding, NMR, phenylaminoacetic acid benzylidene
hydrazine, Plasmodium falciparum, Plasmodium falciparum
enoyl-ACP reductase, QSAR, recursive partitioning, X-ray dif-
fraction
*Corresponding author: Evans C. Coutinho,
evans@bcpindia.org
Recent studies have revealed that plasmodial enoyl-
ACP reductase (pfENR, FabI), one of the crucial
enzymes in the plasmodial type II fatty acid synthesis
II (FAS II) pathway, is a promising target for liver stage
malaria infections. Hence, pfENR inhibitors have the
potential to be used as causal malarial prophylactic
agents. In this study, we report the design, synthesis,
structural characterization and evaluation of a new
class of pfENR inhibitors. The search for inhibitors
began with a virtual screen of the iResearch database
by molecular docking. Hits obtained from the virtual
screen were ranked according to their Glide score.
Received 16 August 2012, revised 4 November 2012 and
accepted for publication 8 January 2013
Malaria continues to be one of the most endemic infec-
tious disease killing almost one million people per year. It
is estimated as the fifth leading cause of death globally
and about half the world’s population (3.3 billion) is at a
risk of contracting this disease. Among the four species of
the protozoan parasite which infect human, Plasmodium
falciparum, is the most dominant and the primary cause of
mortality (1). The emergence of drug resistance, even
against artemisinin-based combination therapies under-
scores the need to develop novel strategies to combat
malaria (2). There is no vaccine available and life-long pro-
phylaxis has not yet been achieved against malaria
because of multiple and sequential infection, particularly in
young children (3). One reason for this is the complex life
cycle of the parasite, which includes both the Anopheles
mosquito (vector) and human hosts. In humans, the para-
site goes through the initial liver (hepatocyte) stage which
is succeeded by the erythrocytic (blood) stage. The liver
stage (LS) of the parasite is currently drawing substantial
One hit was selected as
a lead and modified to
improve its binding to pfENR; from this, a series of
phenylamino acetic acid benzylidene hydrazides were
designed and synthesized. These molecules were thor-
oughly characterized by IR, 1H, 13C, 15N, 2D-NMR
(COSY, NOESY, 1H-13C, 1H-15N HSQC and HMBC), and
X-ray diffraction. NMR studies revealed the existence
of conformational/configurational isomers around the
amide and imine functionalities. The major species in
DMSO solution is the E, E form, which is in dynamic
equilibrium with the Z, E isomer. In the solid state, the
molecule has a completely extended conformation and
forms helical structures that are stabilized by strong
hydrogen bond interactions, forming a helical structure
stabilized by N-H…O interactions, a feature unique to
ª 2013 John Wiley & Sons A/S. doi: 10.1111/cbdd.12118
1

Samal et al.
attraction for the development of new vaccines (4). On the
other hand, because elimination of the parasites in the liver
prevents the initiation of the blood stage (BS) infection,
drugs targeting the LS represent significant potential in
causal prophylaxis of malaria. Primaquine is the only drug
specifically developed to inhibit the LS infection but suffers
from toxicity, poor compliance, and risk of hemolysis (5).
lidene hydrazine was identified as a lead. A set of phenyla-
minoacetic acid benzylidene hydrazine analogs were then
designed and those found to bind tightly to pfENR were
taken up for synthesis. The structures of these molecules
were confirmed by IR, 1H, 13C, 15N, 2D-NMR, and X-ray
diffraction. Further, Recursive Partitioning (RP) analysis
was performed on these compounds to develop a classifi-
cation model based on their activity using physicochemi-
cal, structural, electronic and thermodynamic properties.
The RP method is a specific case of the ‘binary QSAR’
approach which can classify analog activity data through
essential decisive elements that categorize the dataset into
groups of molecules based on the activity.
The liver phase has not been explored as a drug target,
mainly due to technical challenges in studying this stage.
Recent studies showed that de novo type II fatty acid bio-
synthesis (FAS-II) pathway, which was previously thought
to operate in BS, is only expressed in LS. Deletion of indi-
vidual FAS-II enzymes prevents the generation of erythro-
cytic merozoites, that is, the erythrocytic stage infection
cannot initiate (6,7). This data plus the fact that P. falcipa-
rum employs a dissociated type of pathway (type II FAS)
for fatty acid synthesis, whereas humans use an associ-
ated pathway (type I FAS) renders the FAS-II pathway an
attractive target. Plasmodium falciparum enoyl-ACP reduc-
tase (pfENR or FabI) (8,9) is a key enzyme in the type II
FAS pathway which reduces the trans-2 enoyl bond of
enoyl-ACP substrates to saturated acyl-ACPs and plays a
decisive role in completing the fatty acid elongation cycles.
Gene knockout of pfENR from P. falciparum has shown
that this enzyme is critical for development of the parasite
in the hepatic phase. Triclosan, [5-chloro-2-(2,4-dichloro-
phenoxy)phenol], is a wide spectrum antifungal and anti-
microbial agent which has been found to inhibit the fatty
acid synthesis pathway (FAS ΙΙ) by binding to enoyl-ACP
reductase (10,11). It has a Ki of 50 nM against pfENR and
has high nanomolar to low micromolar IC50 against both
drug sensitive and multidrug resistant organisms. How-
ever, due to its bioliability, the use of triclosan in humans
as an antimalarial drug is not yet suitable and several stud-
ies have reported triclosan resistance in E. coli. In addition,
triclosan presents poor bioavailability indicating a potential
limitation in the use of this compound as a therapeutic
agent. As a result, several attempts have been made to
modify triclosan (12,13) but have met with little success.
This underscores the need to develop novel strategies to
overcome existing barriers.
Materials and Methods
Virtual screening
Virtual Screening is an in silico approach that filters an
extensive library of compounds (available or virtual) to a
limited number of potentially promising compounds for the
target of interest by means of computational methods such
as molecular docking. High-throughput molecular docking
involves screening a large database of molecules against
the active site of the receptor followed by application of a
scoring function to estimate the possibility that the ligand
will bind to the protein with high binding affinity (15,16).
Molecular docking has a great advantage over 2D similarity
and 3D pharmacophore search techniques as it utilizes the
3D structure of the receptor in a quantitative way.
The program Glide (17,18) (Grid-Based Ligand Docking
with Energetics) incorporated in the Schrodinger molecular
modeling package (Schrodinger, Inc., USA) was used to
perform molecular docking. It works on the principle of
systematic search for conformations, orientations and
positions of the ligand in the receptor active site and elimi-
nates unwanted conformations by a scoring function. This
is followed by energy optimization.
Flexibility of the ligand/protein was considered during
docking, allowing for a flip of 5- and 6-membered rings. A
maximum of 10 poses per ligand were stored and these
were energy minimized postdocking. The receptor active
site was shaped by Tyr267, Gly313, Pro314, Ile323,
Phe368, Ile369, Ala372, Tyr277, Ala319, and Ala320; resi-
Our aim was to design agents that are more polar in char-
acter than triclosan and thus would not suffer from bioli-
ability or bioavailability issues and more importantly agents
with a same mechanism of action as triclosan, that is, an
inhibitor of pf ENR. A detailed investigation of the X-ray
crystal structure of pf ENR in complex with triclosan (14)
could provide a clear understanding of the active site and
the mode of interaction of triclosan with the target enzyme.
This information served as a guideline to initiate a rational
design approach to develop alternatives to triclosan that
target the enoyl-ACP reductase (ENR) of P. falciparum.
With this objective, a virtual screen of drug-like-molecules
from the iResearcha database, against the target (pfENR),
was carried out using molecular docking. Based on the
results of the virtual screen, phenylaminoacetic acid benzy-
ꢀ
dues in a 10A vicinity of triclosan. The molecules were
docked into the active site using the Glide Extra-Precision
(XP) docking algorithm. Prediction of the binding affinity
and rank ordering of ligands in the database was achieved
by the Glide Score scoring function.
An important step in virtual ligand screening (VLS) is the
docking protocol that is adopted to retrieve relevant hits
from the database. The docking protocol should be able
to correctly predict the binding modes of the molecules in
the database. The docking protocol adopted herein was
validated by reproducing the experimentally observed
2
Chem Biol Drug Des 2013

Synthesis and Evaluation of a New Class of pfENR Inhibitors
Figure 1: (A) Overlay of the experimental binding mode of triclosan over best pose obtained by molecular docking into the crystal
structure (PDB ID: 1nhg). (B) Binding mode of molecule 6 obtained by molecular docking.
binding pose of triclosan in complexation with the target
pf ENR (PDB code 1NHG). It was observed that the top
Table 1: Structures of the molecules designed as inhibitors of
pfENR along with their docking scores and in vitro pf ENR inhibi-
tory activity
ꢀ
10 docking solutions for triclosan were all within 1.0A
R
RMSD of the crystal structure. The best scoring pose for
triclosan obtained by docking has been compared with
the crystal structure in Figure 1A. A very good agreement
was observed between the pose of the inhibitor found by
docking and the crystal structure. This assured us that the
docking protocol would be good enough to correctly pre-
dict the binding modes of the molecules in the database
during virtual screening. With the receptor grid generated
for triclosan in the validation study, docking was performed
on drug-like molecules in the iResearch database compris-
ing of 14 million compounds against the target (pf ENR) to
search potential candidates.
H
N
R'
N
H
N
O
Mol. ID
R
R′
IC50 (lM)
Glide score
Triclosan
0.049
90
>159
>143
10
n.m
12
8
À8.60
À7.75
À6.75
À6.94
À7.68
À6.92
À8.33
À7.34
À7.53
À6.90
À8.23
À7.09
À8.07
À7.35
À6.95
À5.72
À8.34
À7.26
À6.58
À7.03
À6.31
À7.32
1
2
3
–
–
p-Cl
p-Cl
p-Cl
p-F
p-F
–
–
–
o-Cl
m,p-diOCH3
m,p-diOCH3
–
4
5
6
o-Cl
Chemistry
o-OH
p-N(CH3)2
–
m-Br
o-OH
p-N(CH3)2
o-OH
–
Synthesis was carried out using reported procedures (19
–21) and reactions were modified to improve the yield
and purity of the products. All solvents and reagents
used for synthesis were of general reagent grade. The
reactions were monitored by Thin Layer Chromatography
(TLC) with Merck precoated silica plates (GF 254) for
completion and for establishing the purity. Melting points
were recorded in open capillaries on an electrically
heated melting point apparatus and are uncorrected. IR
spectra were recorded (KBr disc method) on a Jasco FT-
IR 5300 spectrophotometer. NMR spectra were recorded
on a Bruker AV500 instrument using DMSO-d6 as well
as CDCl3 as solvents. 2D-NMR spectra were recorded
with standard routines available on the spectrometer.
Chemical shifts are reported in ppm down field from
tetramethylsilane (TMS) as the internal standard. Micro-
wave-assisted reactions were carried out using the
Discover microwave synthesizer (CEM, Buckingham,
UK). The molecules (Table 1) were synthesized in three
steps as described below (Scheme 1).
7
8
9
197
90
10
11
12
13
14
15
16
17
18
19
20
21a
>185
>168
>155
>163
29
>129
>164
7
–
m-Cl, p-F
m-Cl, p-F
m-Cl, p-F
m-Cl, p-F
p-Cl
p-Cl
p-F
m-Cl
m-Cl
m-Cl
o-Cl
m-Br
o-OH
m-Br
o-Cl
–
o-Cl
163
173
77
m-Br
163
n.m., not measurable due to self-absorbance at 340 nm.
aThe activity reported for molecule 21 is the cyclic imidazole-4-
one derivative (see Scheme 2 for structure and the text for more
details).
solid material obtained upon cooling was filtered and recrys-
tallized from hexane to yield ethyl phenylaminoacetate (iii).
General procedure for ethyl phenylaminoacetate
analogs (iii)
General procedure for the synthesis of
phenylaminoacetic acid hydrazideanalogs (iv)
Compound (iii) (10 mmol) was dissolved in hydrazine
hydrate (10 mmol) in a glass tube and placed in the
A mixture of aniline or its derivative (i) (10 mmol), ethyl chlo-
roacetate (10 mmol) (ii), and sodium acetate (20 mmol, dis-
solved in 2 mL of water) was stirred at 90 °C for 3 h. The
Chem Biol Drug Des 2013
3

Samal et al.
R
R
O
90°C
Cl
O
+
NH2
N
H
CH3COONa
O
O
(i)
(ii)
(iii)
NH2NH2
300 Watt/30 secs
R
R
R'
H
R'
N
H
N
+
N
H
N
300 Watt/60 secs
O
N
H
NH2
O
O
(v)
(vi)
(iv)
Scheme 1: Scheme for the synthesis of phenylaminoacetic acid benzylidene hydrazine analogs.
microwave synthesizer. After irradiation for 30 seconds
(300 Watts), the product solidified on cooling to room tem-
perature. This was purified by recrystallization from hot
ethanol to yield phenylaminoacetic acid hydrazide (iv).
binary splits for each descriptor and selects the optimal
one. At each non-terminal node in the decision tree, the
dataset is split into two groups by a particular descriptor
recursively until no further splitting is possible or a thresh-
old value of the termination rule is reached. The tree dis-
play of the SAR is clear, easy to interpret, and often
suggestive. The RP models can be generated very fast
compared with other grouping techniques and the output
provides a graphical tree, as well as a statistics table that
includes correctly classified percentages, the number of
false positives, and compound member information for
each node. Recursive partitioning model was generated
using the CERIUS2 (version 4.8; Accelrys Inc. USA)b soft-
ware. The detail procedure adopted for performing the PR
has been discussed in the supporting information.
General procedure for the synthesis of
phenylaminoacetic acid benzylidene hydrazine and
its derivatives (vi)
Compound (iv) (5 mmol) was mixed with benzaldehyde (v)
(5 mmol) in a microwave flask and the mixture was irradi-
ated in the Discover microwave synthesizer for 60 seconds
(300 Watts). After completion of the reaction as indicated
by TLC, the solid product was recrystallized from ethanol
to give the pure phenylaminoacetic acid benzylidene
hydrazine (vi).
The synthetic procedure for individual compounds and
their characterization (IR and NMR) has been discussed in
the supporting information.
RP model generation
Recursive partitioning model was generated using the CERI-
US2 (version 4.8, Accelrys Inc. USA)b software. The CSAR
RP method was used to perform an RP analysis, and the
decision tree was generated from the results of RP.
Numerical descriptors that encode topological, geometri-
cal/structural, electronic, and thermodynamic properties
were calculated using the CERIUS2 molecular modeling
package (Table 2). The antitubercular activity of com-
Recursive partitioning
Recursive partitioning (22–26) is a family of data analysis
techniques finding its root in ‘binary QSAR’ that seeks to
decode the elusive relationships in complex datasets
involving thresholds, interactions, and non-linearity. These
issues hinder an analysis based on assumptions of linearity
such as multiple linear regression, principal component
analysis (PCA), or partial least squares (PLS). RP catego-
rizes the dataset into groups of higher and lower
responses according to their appropriate descriptors. The
decision trees resulting from the RP analysis can be used
to qualitatively predict activities or activity classes in struc-
ture–activity relationships without considering complex sta-
tistical analysis. It uses the best chemical features to
categorize the large dataset into smaller and more homo-
geneous subgroups. The algorithm considers all possible
Table 2: Descriptors used to develop the classification model
Descriptor
Description
Structural
descriptor
Spatial
descriptors
Electronic
descriptors
Thermodynamic
descriptors
Number of H-bond donors and acceptors,
number of rotatable bonds, molecular weight
Radius of gyration, molecular volume, area,
density, principal moment of inertia
Charge, sum of atomic polarizabilities(Apol),
HOMO, LUMO, dipole moment
Atom-type-based logP (AlogP98), molar
refractivity (MolRef)
4
Chem Biol Drug Des 2013

Synthesis and Evaluation of a New Class of pfENR Inhibitors
pounds was described in terms of their IC50 values, and
the dataset was split into two classes – less active (0) and
active (1) based on these IC50 values. Class 1 (less active)
contains 13 compounds with IC50 values greater than
100 lM, while class 2 (active) contains 8 compounds
with IC50 values lower than or equal to 100 lM. Prior to
generating models, descriptors with zero variance as well
as those containing 95% zero values were eliminated
because these are not useful for explaining the variation in
the activity.
docking scores are given in Table 1 and a representative
docking pose is shown in Figure 1B.
It was observed that all 20 molecules adopted a binding
mode which is complementary with the pf ENR active
pocket. A very small difference in Glide score as well as
a low RMSD observed upon superposition of the binding
poses among all the 20 phenylamino acetic acid benzy-
lidene hydrazine analogs suggests an analogous binding
mode for all the candidates. The Glide docking score for
these analogs varied from À8.34 to À5.72 kcal/mol,
The correlation matrices for the descriptors were con-
structed and those descriptors with a cross-correlation
coefficient >0.5 were eliminated, because they contain
nearly the same information. The uncorrelated descriptors
formed the independent variables (X), while the biological
activity served as the dependent variable (Y) for the analy-
sis. The CARTTM (classification and regression trees)
method was used to generate the decision tree. The activ-
ity classes were weighted equally, and the tree was
formed by optimizing the Twoing rule scoring function. The
pruning factor was varied between 0 and 3. Also the val-
ues for maximum tree depth were varied from 5 to 10,
while default values were used for maximum number of
generic splits (30), and the number of knots per variable
(20). Internal validation was performed using cross-valida-
tion, setting the number of cross-validation groups to 10.
while triclosan has
a
score of À8.60 kcal/mol. This
means that these analogs could be developed as poten-
tial candidates for second-generation drug development
against pfENR.
NMR and conformational analysis
It is a well-known fact that small linear molecules are not
rigid in solution but have several degrees of internal
motional freedom. This results in several conformations
being populated at room or physiological temperature. As
a first step in understanding the conformational isomerism
in phenylaminoacetic acid benzylidine hydrazides,
a
1
detailed HNMR characterization was carried out. Phenyla-
minoacetic acid benzylidine hydrazides can in principle
exist in four different conformations arising from combina-
tions of E and Z geometrical/conformational isomers at the
imine and the amide functional groups. These configura-
tional and conformational isomers are depicted in Figure 3.
Results and Discussion
Design of substituted phenylamino acetic acid
benzylidenehydrazines
Of these structures, the E, Z and Z, Z configurations are
usually of higher energy, less stable and are not consid-
ered. The NMR spectra of these compounds in DMSO-d6
showed doubling of the signals indicating that only two of
these are present in solution. These two forms are likely to
be the Z, E and E, E isomers which differ in the position of
the phenylaminoacyl group relative to the amide C-N
From among the several hits obtained from virtual screen-
ing of drug-like molecules in the iResearch database,
ligand 45015 (Figure 2) was chosen as a candidate based
on its similarity to triclosan and also on the characteristics
of the enzyme active site to bind this ligand specifically.
This hit was used as a starting point to design a series of
potential candidates.
Ligand 45015 was iteratively optimized to generate a ser-
ies of substituted phenylaminoacetic acid benzylidene hy-
drazines. Several analogs in this series were identified and
subjected to molecular docking using the protocol defined
in the previous section. Based on their docking scores,
the best scoring molecules were taken up for synthesis.
The structures of the newly designed ligands and their
Figure 2: The hit obtained by virtual screening of the iResearch
database against the crystal structure of pfENR (PDB ID: 1nhg).
Figure 3: Configurational and conformational isomers observed
in phenyl amino acetic acid benzylidene hydrazides.
Chem Biol Drug Des 2013
5

Samal et al.
bond. The assignments of the signals for the Z, E and E, E
isomers can be made on the basis of just 1H and 13C
chemical shifts. In general, the azomethine protons and
carbons of the E, E isomer are known to be more shielded
compared with the Z, E isomer, while the amide carbonyl
shows the reverse trend, that is, they are shielded in the
Z, E isomer compared with E, E isomer (27–29).
studied. Incidentally, the crystal structure obtained for mol-
ecule 2 corresponds to the Z, E form (vide infra).
As mentioned earlier, molecule 21 was found to be differ-
ent from the other members in the series. Elucidating the
structure of molecule 21 was a difficult task, but never-
theless was achieved using 1H-13C and 1H-15H-HMBC
along with NOESY spectra. A careful analysis of the 1H
and 13C NMR data indicated that the CH2 protons
appear (d 4.49 and 4.24) as an AB part of an ABX spin
system with geminal coupling of ~15 Hz. Only one of the
CH2 protons shows very weak scalar coupling (2.7 Hz) to
the X proton at d 6.15. The 13C chemical shifts of these
carbons are 51.39 and 78.9, respectively. A deshielded
azomethine proton (d 9.49) and four additional aromatic
protons are also seen in the proton spectrum. The struc-
ture of this compound has been unambiguously eluci-
dated by detailed 1D and 2D NMR investigations by 1H,
13C, and 15N techniques (Figure S1–S9) as an imidazoli-
dine-4-one. 1H-13C HMBC correlations of the CH2 pro-
tons, the CH proton at d 6.15, the azomethine proton
and the CH carbon at d 78.9 were found to be very cru-
cial in arriving at the structure. For example, the CH car-
bon at d 78.9 shows correlations to the ortho protons of
the bromophenyl group and also to the CH2 protons.
While the CH proton at d 6.15 shows connectivities to
the carbon at d 164.56, the ipso carbons of the two aro-
matic rings (d 144.8 and 140.05) and to the two ortho
protons of one of the rings. The azomethine proton cor-
relations are seen mainly to the carbons of the third aro-
matic ring. The CH2 protons in turn show clear
correlations to the CH carbon at d 78.9, the quaternary
carbons at d 164.56, 140.05, and 144.8. All these obser-
vations could only be satisfied by considering a 1,2,3-tri-
substituted imidazolidine-4-one (Scheme 2). The HMBC
correlations are shown in Figures 6, 7 and S5–S7 along
In the present investigation, the major species present in
DMSO solution corresponds to the E, E form. The protons
and carbon atom of the CH2 group attached to the amide
function are also found to show considerable changes in
the chemical shifts in these two forms (Tables 3 and 4).
Further, the two forms present in solution are in dynamic
equilibrium. Evidence for this comes from the NOESY (Fig-
ure 4) where strong positive exchange cross-peaks are
seen between the major and minor signals of the same
proton. The NOESY spectrum also indicated a weak
cross-peak between the CH2 protons and the ortho pro-
tons of the Ar2 ring, indicating a favor for the E, E form.
The high field CH2 carbons of the major E, E form show a
1H-13C HMBC correlation to the low field NH (major) pro-
ton, whereas no correlation is seen from the minor CH2 to
the minor NH under the conditions of measurement
3
(Figure 5). In the E, E form the JC-C-N-H is expected to be
larger than that in the Z, E form. Thus, the predominance
of the E, E configuration can be envisaged in all the cases
Table 3: 1H and 13C chemical shifts distinguishing the major and
minor isomers of the imine group in different compounds
Imine (major) (d,
Imine (minor) (d,
ppm)
ppm)
Compound
HC=N-
HC=N-
with
a schematic representation of the correlations
observed.
4
1H
8.02
144.05
7.76
145.56
7.89
144.90
7.89
1H
8.25
147.38
7.92
149.30
8.08
148.25
8.09
13C
1H
13C
1H
2
7
This product has been formed by reaction of the aldehyde
(3-bromobenzaldehyde) with the NH of the phenylaminoa-
cyl moiety (iii, Scheme 1) followed by cyclization through
dehydration (Scheme 2). A reaction of phenayaminoacyl
hydrazone with excess aromatic aldehyde (1:2) indeed
gave this product. The proposed structure is also sup-
13C
1H
13C
1H
13C
1H
13C
1H
11
13C
144.88
13C
148.21
Table 4: 1H and 13C resonances
of the major and minor isomers
around the amide bond
Amide (major) (d, ppm)
Amide (minor) (d, ppm)
Compound
NH
CH2
CO
Ar NH
NH
CH2
CO
Ar NH
4
11.52
4.24
–
5.98
11.48
3.81
–
6.22
–
44.30
4.33
45.60
4.16
44.77
5.66
44.28
171.64
–
171.90
–
171.24
–
171.31
–
–
–
5.65
–
–
46.21
3.95
48.61
3.74
–
167.14
–
166.96
–
166.67
–
166.76
–
–
–
5.92
–
2
7
10.17
–
11.23
–
11.23
–
10.02
–
10.02
–
11.13
–
11
4.19
–
3.76
46.49
5.94
–
6
Chem Biol Drug Des 2013

Synthesis and Evaluation of a New Class of pfENR Inhibitors
Figure 4: NOESY spectrum of molecule 4 recorded at 298K (A) D1: 3.6–11.6 ppm, D2: 3.6 – 11.6 ppm; (B) D1: 3.74 – 4.34 ppm, D2:
5.8 –7.8 ppm.
Figure 5: (A) 13C HMBC spectrum of molecule 4 recorded at 298K; D1: 11.0–11.5 ppm, D2: 43.2–47.1 ppm (B) 15N HMBC spectrum of
molecule 4 recorded at 298K; D1: 5.6–11.9 ppm, D2: 0–400 ppm.
ported by the 1H-15N HMBC (Figure 7) and the NOESY
correlations (Figures S9–S10). The most conclusive evi-
dence for the NOESY data are the weak nOe’s between
the protons on the various rings. For example, the one
between the ortho protons of the aromatic rings A and B
and between the protons on the imidazolidine-4-one ring
(D) with protons on rings A and B.
more stable E configuration. The nOe data are also in
agreement with this as no nOes are observed between the
protons of the aromatic ring (7.53d, 7.27d, and 7.79d) on
the imine carbon and the CH2 group in the five-membered
ring. Besides, weak nOes are observed between the imine
CH at ~9.5d and the inequivalent CH2 protons at 4.19 and
4.44d (Figure S12). During the formation of molecule 21,
an asymmetric center is created on the five membered
ring. It very likely that the product formed is racemic in
nature. We have not made any attempt to identify the
chirality of the molecule because this molecule is found
to be of less importance in the context of the studies
conducted.
1
The H NMR spectrum of molecule 21 (Figure S11) shows
only a single signal for the imine CH proton at ~9.5d,
indicating the presence of only one geometrical isomer of
the imine. Considering the bulkiness of the substituents,
we assume this molecule will have a preference for the
Chem Biol Drug Des 2013
7

Samal et al.
H
N
H
N
N
C
H
Br
O
Cl
O
O
Cl
CH
H
N
NH
N
OH
Br
Br
Br
O
N
CH
N
Cl
N
Br
Scheme 2: Formation of
the imidazolidine 4-one derivative
(molecule 21).
Br
Figure 7: 1H-15N HMBC spectrum of molecule 21 recorded at
298K; D1: 3.6–9.9 ppm, D2: 0–400 ppm.
Figure 6: 1H-13C HMBC spectrum of molecule 21 recorded at
298K; D1: 3.6–9.9 ppm, D2: 40–170 ppm.
0.15 mm3 were used for data collection on a Bruker
SMART APEX CCD diffractometer using Mo Ka radiation
with fine focus tube with 50 kV and 30 mA. Details about
data acquisition are as follows: crystal to detector distance
6.05 cm, 512 9 512 pixels/frame, Hemisphere data
acquisition, total scans 3, total frames 1271, oscillation/
frame À0.3°, exposure/frame 10.0 sec/frame, maximum
detector swing angle À30.0°, beam center (260.2, 252.5),
in plane spot width 1.24, SAINT integration, h range 1.85–
25.00°, completeness to h of 25.00° is 99.7%. SADABS
correction applied, C17 H19N3O3, M = 313.35. Crystals
Incidentally in the NMR spectra of molecules 1–20, we
could detect small amounts of the cyclized imidazole-4-
one derivates as an impurity.
X-ray crystal structure analysis
Single crystals of the molecule were grown by slow evapo-
ration of the solution mixture of molecule 2. Colorless nee-
dle like crystals of approximate size 0.51 9 0.30 9
8
Chem Biol Drug Des 2013

Synthesis and Evaluation of a New Class of pfENR Inhibitors
H…O interactions. The hydrogen bond distances are N(2)-
belong to Orthorhombic, space group Pbca, a = 8.419 (1),
ꢀ
ꢀ3
ꢀ
ꢀ
b = 18.234 (2), c = 22.063(3) A, V = 3386.8(8) A , Z = 8,
Dc = 1.433 g/cc, l (MoKa) = 0.367 per mm, T = 293 (2)
K, 13003 reflections measured, 2965 unique [I > 2r(I)], R
value 0.0492, xR2 = 0.1118. All the data were corrected
for Lorentzian, polarization, and absorption effects.
SHELX-97 (ShelxTL) (30) was used for structure solution
and full matrix least squares refinement on F2. Hydrogen
atoms were included in the refinement as per the riding
model. Data collection and refinement parameters are
listed in Table S1.
H…..O(3) = 2.836(3)A and N(3)-H(3)….O(1) = 3.084(3) A
giving rise to a H-bonded helical structure as depicted in
Figure 9 which is seen as an infinite one-dimensional chain
(Base Vector: [1 0 0]). Analysis of potential hydrogen
bonds is given in Table 5.
pfENR inhibition activity (FabI Assay)
The pf ENR protein was cloned, expressed, and purified
by methods reported earlier (31,32). All measurements
were carried out on a Perkin Elmer Lambda 25 UV/VIS
spectrophotometer in a buffer solution containing 20 mM
HEPES, pH 7.4 and 150 mM NaCl. Solutions for assay
were prepared by dissolving compounds in DMSO (max.
final concentration was 1%). For measuring pfENR inhibi-
tion, 1 lL of 100 lM NADH cofactor (43420; Sigma-
Aldrich, St.Louis, MO, USA) was added to 1 lg of enzyme
and then 1 lL of varying compound concentration in a
total of 1 mL buffer. The reaction was started by adding
1 lL of 50 lM crotonyl-CoA substrate (C6146; Sigma) and
the absorbance of the mixture was read spectrophotomet-
rically for 1 min at 340 nm. For comparison purposes, tri-
closan was chosen as the reference standard and it
exhibited an IC50 of 0.04 lm. IC50 values were estimated
from graphically plotted dose–response curves (Table 1).
Crystal data, structure refinement, atomic co-ordinates,
bond lengths, bond angles and torsion angles anisotropic/
isotropic displacement parameters are given in the sup-
porting information (Tables S2–S6).
The ORTEP diagram of molecule 2 with the atom-number-
ing scheme is displayed in Figure 8. The important bond
lengths, bond angles, and torsion angles are given in
Tables S2–S6. As is evident from the table of torsion
angles, the backbone of the molecule is nearly extended
except for the torsion angle C11-N3-C12-C13 which
describes the folding of the Ar2 ring in a gauche arrange-
ment. The observed bond parameters indicate that the
molecule forms hydrogen bonded helical structure via N-
H….O interactions (Figure 9). This is a unique attribute for
this class of compounds and to the best of our knowl-
edge, helical structures of this type have not been
reported earlier. There are two intermolecular H-bond N-
Eight of the 20 molecules were found to be active against
pfENR with an IC50 value below 100 lm, while rest of the
analogs displayed moderate to low activity. These analogs
Figure 8: An ORTEP diagram of
molecule 2 with the atom-
numbering scheme (Ellipsoids are
drawn at 50% probability).
Chem Biol Drug Des 2013
9

Samal et al.
Figure 9: Hydrogen bonded helical
structure for molecule 2 viewed
along ‘a’-axis showing intermolecular
hydrogen bonding between amide
functionality of one molecule and
carbonyl oxygen of second molecule
forming a unit cell.
Table 5: Analysis of potential hydrogen bonds
cyclized structure is not more potent than the linear phe-
nylaminoacetic acid benzylidene hydrazine scaffold, and
hence, no further attempts were made in exploring the
cyclized structure for the other molecules (1–20) in the
series.
D-H….A
No (A)
D-H
H….A D….A
Angle
D-H…A (o)
ꢀ
ꢀ
ꢀ
ꢀ
(A)
(A)
(A)
1
2
N(2)-H(2N)…O(3)i
N(3)-H(3)…O(1)ii
0.86 2.04
0.86 2.26
2.836(3) 154
3.084 161
i
ii
Equivalent Position Code = ½ + x, y, ½ Àz and = Àx, ½ + y, ½
Àz.
ADMET predictions
The 20 molecules in Table 1 were evaluated for ADME
and toxicity profiles using the in silico ADME/Tox web-box
v3.5 (33) (Table 6). Most of the molecules have a pka for
the basic group around 3.0 suggesting they are weak
bases. The molecules have an average plasma protein
binding of 80–96%. The molecules are predicted to have a
reasonably well oral bioavailability with few exceptions
where an oral bioavailability is predicted to be 30–70%. It
should be noted that ADME/Tox web-box v3.5 does not
predict clearance values and as such the bioavailability
predictions may be based only on the predicted solubility
and permeability of the molecules. Most of the molecules
are also predicted to have a high oral LD50 dose, indicat-
ing their safety. Also these molecules show a very low
probability for genotoxicity as predicted by the Ames test.
Overall, the predicted starting pharmacokinetic and toxicity
profiles for the entire set of molecules seems favorable for
hit/lead optimization.
bear a common scaffold, that is, phenylaminoacetic acid
benzylidene hydrazine with variations on the two aromatic
rings (R and R′). Molecules 17 and 7 were found to be the
most potent analogs with IC50 values of 7 and 8 lm,
respectively. It was observed that substitution at the para
position of the phenyl ring (R) with an electron with draw-
ing group such as Cl or F correlates with greater inhibition
as evident for molecules 4, 6, and 17. However, additional
substituents (R) lead to a significant rise in the IC50 value
(molecules 12, 13, 14, 15). For the R group, para substitu-
tion was observed to be more favored over the meta sub-
stitution as regards to activity (compare molecules 7, 17,
19, 20). A similar trend was observed for the R′ group,
with an electron withdrawing group being favored over an
electron releasing group. As for the R-group, mono substi-
tution is preferred over di substitution for R′. The difficulty
of crossing the complex membranes enveloping the
malaria parasite could be a reason for the poor activity
observed for some of these compounds.
Recursive partitioning analysis
The RP model was generated by variation of parameters
discussed in materials and methods section, so as to
improve the following parameters: ‘Class% Observed Cor-
The activity reported for molecule 21 in Table 1 is for the
cyclic imidazole-4-one derivative. It was observed that the
10
Chem Biol Drug Des 2013

Synthesis and Evaluation of a New Class of pfENR Inhibitors
Table 6: ADME and toxicity prediction data of the phenylaminoacetic acid benzylidine hydrazine derivatives
Acute
toxicity
LD50
Acute
toxicity
LD50
Rat
Physicochemical
determinants
Solubility
Distribution
Genotoxicity
Ames test
Mouse
Water
(logSw)
Buffer (pH 7.4)
(logS)
logD
Plasma protein
binding
Mol. ID
(pH 7.4)
logP
Vd (L/kg)
Oral mg/kg
Oral mg/kg
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
À3.73
À4.05
À5.05
À3.81
À4.13
À3.75
À4.08
À3.16
À4.05
À2.92
À3.51
À3.78
À4.41
À4.59
À4.85
À3.85
À4.29
À4.03
À3.73
À4.05
À3.73
À4.05
À5.05
À3.81
À4.13
À3.71
À4.08
À3.16
À4.05
À2.88
À3.51
À3.75
À4.41
À4.59
À4.85
À3.82
À4.29
À4.03
À3.73
À4.05
2.62
2.22
2.92
3.02
3.32
2.55
2.60
2.33
3.30
2.46
2.50
3.33
3.20
3.50
4.18
3.16
3.40
2.71
3.14
3.44
2.62
2.22
2.92
3.02
3.32
2.59
2.60
2.33
3.30
2.49
2.50
3.37
3.20
3.50
4.18
3.19
3.40
2.71
3.14
3.44
89.67
81.18
93.07
92.60
96.43
82.29
88.31
80.10
92.56
81.74
88.04
87.08
93.06
96.66
97.65
86.48
92.74
89.90
93.17
96.72
1.8
0.91
0.73
0.72
0.82
0.91
0.75
0.93
0.82
0.82
0.84
0.96
0.71
0.67
0.82
0.65
0.84
0.70
0.85
0.82
0.91
1300
1200
1000
1500
1300
1600
1200
1300
1800
1600
1000
1100
1100
950
1500
1700
1600
1200
1400
1100
1400
1900
1700
1900
1400
2100
1300
1300
830
2300
1100
1300
1300
940
1400
2300
1500
1300
1900
1500
1.55
1.90
1.93
2.03
1.03
1.79
1.58
2.03
0.81
1.63
1.18
1.99
2.10
2.37
1.15
2.06
1.83
1.97
2.08
Table 7: Statistical results of recursive partitioning
Table 8: Activity prediction by recursive partitioning (PR)
Descriptors
Class%
Overall%
Number of
Class compounds
observed predict
RP-
predicted of
Activity activity gyration Density PMI-mag AlogP98
Radius
%
correct
91.66
correct
Enrichment
Mol.
ID
1
2
12
8
60.00
100
1.61
2.33
40.00 100
88.89
1
2
1
0
0
1
0
1
1
0
1
0
0
0
0
1
0
0
1
0
0
1
1
0
0
1
0
1
1
0
1
0
0
0
0
1
0
1
1
0
0
1
4.06
3.58
3.81
3.21
3.83
3.36
3.94
3.43
3.39
3.95
3.66
3.83
4.73
3.23
4.94
3.27
3.54
3.50
3.72
3.92
1.14
1.08
1.14
1.14
1.20
1.14
1.08
1.06
1.29
1.09
1.04
1.20
1.18
1.25
1.39
1.16
1.33
1.18
1.14
1.20
778.40
686.43
889.95
527.98
772.97
572.96
858.91
495.86
675.57
728.74
680.66
819.95
1405.55
662.18
2319.27
599.26
1140.02
756.70
690.83
1096.43
3.01
2.31
2.98
3.01
3.67
2.31
2.71
2.34
3.09
2.10
2.51
2.97
3.21
3.88
3.96
2.77
3.30
3.21
3.01
3.67
3
rect,’ ‘Overall% Predict Correct,’ and ‘the enrichment fac-
tor’. Here, ‘Class% Observed Correct’ is the intraclass
prediction wherein only the molecules in the corresponding
class are being predicted. Information on false negatives
as well as false positives can be derived from this parame-
ter. On the other hand, the term ‘Overall% Predict Correct’
is the so-called overall prediction wherein all the molecules
in the set are being predicted. It provides information on
the accuracy of the prediction when the whole set is being
predicted with the model. The enrichment factor is the
percentage of compounds correctly predicted to belong to
that class over the original percentage of compounds
belonging to that class. The statistical outcomes of the RP
model and activity predictions are summarized in Tables 7
and 8, respectively.
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
In class 1, that is, the class of ‘less actives,’ 11 of 12
compounds were correctly classified as class 1. The
model misclassified compound 16 (Table 8) as active. For
class 2, which is the class of ‘actives,’ all eight com-
pounds were correctly classified representing 100% intra-
class predictivity. The enrichment factor of class 1 is 1.61
and 2.33 for class 2. The number of true positives among
the predictions in each activity class is listed as the term
‘Overall% Predict Correct’. The enrichment factor (1.61 for
class 1 and 2.33 for class 2, respectively) also indicates
Chem Biol Drug Des 2013
11

Samal et al.
Figure 10: Recursive partitioning tree
generated with moderate pruning;
score splits using Twoing rule for
prediction of antitubercular activity
classes. Those marked 1:0, 2:1, 3:0,
4:1, 5:0 and 6:1 correspond to terminal
nodes 1–6 from the top to bottom and
each terminal node corresponds to the
value of 1 (active) or 0 (less active).
Table 9: Cross-validation statistics of the recursive partitioning
model
the statistical significance of the RP model for classifying a
new library of compounds.
Class%
observed predict
correct
Overall%
Figure 10 shows the optimized 6-leaf recursive partitioning
decision tree generated using the above-mentioned 2D
descriptors to classify the compounds. The structure of
the decision tree consists of six terminal and five non-ter-
minal nodes. At each decision point (node), molecules are
split into groups, higher and lower responses, according
to their descriptors. The terminal nodes 1, 3, and 5 are
class 1 (less active), while the terminal nodes 2, 4, and 6
are class 2 (active). The final tree model correctly classified
92 and 100% of the molecules in class 1 and 2 using
radius of gyration, principle moment of inertia, density, and
AlogP descriptors as decision points. A true response to
any given split follows the branch to the downside and a
false response to any given split follows the branch to the
upside in the decision tree.
Number of
Class compounds
%
correct
Enrichment
1
2
12
8
60.00 46.15
40.00 62.50
66.67
41.67
1.08
1.09
terminal node 4 contains two compounds with density
<1.14 which is grouped in class 2 (active), while the
remaining 6 compounds with density >1.14 are placed on
the non- terminal node.
The fourth split appears on AlogP98, an atom-type-based
logP method calculated using the latest published set of
parameters (34). It is a thermodynamic descriptor that can
be used to relate chemical structure to observed lipophilic-
ity and its value is 3.26. It reflects the hydrophobic charac-
ter of the molecule. The terminal node 3 contains three
compounds with AlogP98 <3.26 which are grouped in
class 1 (less active), while the remaining three compounds
with AlogP98 >3.26 are placed on the 4th non-terminal
node.
The first primary split is observed on radius of gyration and
its value is 3.41. It is a spatial descriptor related to a mea-
sure of the size of a molecule. The terminal node 6 con-
tains five compounds with radius of gyration value <3.41
which is shown as class 2 (active), while the remaining 15
compounds have radius of gyration above 3.41 which
form the non-terminal node.
The last split occurs again on density and its value is 1.34.
The terminal node 2 contains two compounds with density
value <1.34 which are categorized in class 2 (active), while
the remaining one compound has the density above 1.34
which forms the 1st terminal node.
The second split is observed on principal moment of iner-
tia, another spatial descriptor. In terminal node 5, seven
molecules have principal moment of inertia values <775.69
and are grouped in class 1 (less active).
The third split is observed on density, which is also a spa-
tial descriptor and its value is 1.14. It reflects the type of
atoms and how tightly they are packed in a molecule and
can also be related to transport and melt behavior. The
To avoid over fitting and to improve generalization of the
classification models, internal validation was performed
using 10-fold cross-validation which resulted in 67% and
42% correct predictions for classes 1 and 2, respectively.
12
Chem Biol Drug Des 2013

Synthesis and Evaluation of a New Class of pfENR Inhibitors
Considering the limitation of the size of the dataset, the
prediction rate is acceptable. The statistical results of
cross-validation are shown in Table 9. These predictive
parameters indicate that our classification model will be
able to classify and predict the activity (class) of new can-
didates.
Acknowledgments
The authors are grateful to the All India Council for Technical
Education (AICTE), New Delhi, the Department of Science
and Technology (DST), New Delhi through their FIST pro-
gram (SR/FST/LSI-163/2003), the Department of Biotech-
nology (BT/PR11810/BRB/10/690/2009), the Council of
Scientific and Industrial Research (01/23/99/10-EMR-II) and
the University of Mumbai (APD/237/663/2010) for financial
support and computational facilities at Bombay College of
Pharmacy. Angela Gono Bwalya thanks the Commonwealth
Scholarship Commission for financial support.
Conclusions
Recognizing the importance of enoyl-ACP reductase
(pfENR) in the type II fatty acid synthesis pathway in
Plasmodium, efforts were initiated to design some novel
inhibitors of pfENR. The top 10 hits obtained from virtual
screening of the iResearch database were ranked with
the Glide scoring function and one of them was selected
and modified based on binding site analysis and syn-
thetic adaptability to construct a series of phenylamino-
acetic acid benzylidene hydrazides. 1D and 2D
multinuclear NMR studies on these molecules revealed
the existence of configurational isomers around the amide
and imine functionalities. In the solid state, the molecule
has a completely extended conformation but forms helical
structures that are stabilized by strong hydrogen bond
interactions. Detailed NMR investigation showed the pres-
ence of a minor impurity in most compounds. The struc-
ture of this impurity was deduced as an imidazoline-4-
one derivative based on 1H-13C and 1H-15H-HMBC which
was further confirmed from the NOESY spectra. The mol-
ecules showed promising in vitro activity against pfENR
with IC50 values in the low micromolar range and merit
further testing against LS malaria parasites. Recursive
References
1. Davidson E.A., Gowda D.C. (2001) Glycobiology of
Plasmodium falciparum. Biochimie;83:601–604.
2. Yeung S., Socheat D., Moorthy V.S., Mills A.J. (2009)
Artemisinin resistance on the Thai-Cambodian border.
Lancet;374:1418–1419.
3. Schofield L. (2007) Rational approaches to developing
an anti-disease vaccine against malaria. Microbes
Infect;9:784–791.
4. VanBuskirk K.M., O’Neill M.T., De La Vega P., Maier
A.G., Krzych U., Williams J., Dowler M.G. et al. (2009)
Preerythrocytic, live-attenuated Plasmodium falciparum
vaccine candidates by design. Proc Natl Acad Sci
USA;106:13004–13009.
5. Kappe S.H., Vaughan A.M., Boddey J.A., Cowman
A.F. (2010) That was then but this is now: malaria
research in the time of an eradication agenda. Sci-
ence;328:862–866.
partitioning was employed to develop
a classification
model and to gain an insight into the structure–activity
relationships for these molecules. In spite of a high
degree of structural similarity in the dataset, the classifi-
cation model has a considerable discriminative power.
The current RP model will be improved continuously by
enlarging the library. The model is being used to design
candidates with even better activity and selectivity pro-
files.
6. Vaughan A.M., O’Neill M.T., Tarun A.S., Camargo N.,
Phuong T.M., Aly A.S., Cowman A.F., Kappe S.H.
(2009) Type II fatty acid synthesis is essential only for
malaria parasite late liver stage development. Cell
Microbiol;11:506–520.
7. Yu M., Kumar T.R., Nkrumah L.J., Coppi A., Retzlaff S., Li
C.D., Kelly B.J. et al. (2008) The fatty acid biosynthesis
enzyme FabI plays a key role in the development of liver-
stagemalarialparasites.CellHostMicrobe;4:567–578.
8. Heath R.J., White S.W., Rock C.O. (2001) Lipid biosyn-
thesis as a target for antibacterial agents. Prog Lipid
Res;40:467–497.
9. Smith S. (1994) The animal fatty acid synthase: one
gene, one polypeptide, seven enzymes. FASEB
J;8:1248–1259.
10. Singh A.P., Surolia N., Surolia A. (2009) Triclosan inhi-
bit the growth of the late liver-stage of Plasmodium.
IUBMB Life;61:923–928.
The objective of this endeavor was to design molecules
that posses more polar character than triclosan and thus
would not suffer from bioliability or bioavailability issues.
That we had partially achieved our goals is evident from
the in silico pharmacokinetic profile described in terms of
solubility and bioavailability (>70%) as reflected in the AD-
MET data. While we did not arrive at a candidate that is
more potent than triclosan, we have discerned areas of
the phenylaminoacetic acid scaffold where modifications
can be made to improved the pharmacokinetic profile, as
well as characteristics necessary to improve inhibitor
potency. This novel molecular scaffold presents an
opportunity for further optimization to more potent pfENR
inhibitors which hopefully may possess better oral bio-
availability.
11. Pidugu L.S., Kapoor M., Surolia N., Surolia A., Suguna
K. (2004) Structural basis for the variation in triclosan
affinity to enoyl reductases. J Mol Biol;343:147–155.
12. Freundlich J.S., Anderson J.W., Sarantakis D., Shieh
H.M., Yu M., Valderramos J.C., Lucumi E., Kuo M.,
Jacobs W.R. Jr, Fidock D.A, Schiehser G.A, Jacobus
Chem Biol Drug Des 2013
13

Samal et al.
D.P, Sacchettini J.C. (2005) Synthesis, biological activ-
study and recursive partitioning of heterocyclic quinone
derivatives with antifungal activity. Bioorg Med
Chem;14:1608–1617.
ity, and X-ray crystal structural analysis of diaryl ether
inhibitors of malarial enoyl acyl carrier protein reduc-
tase. Part 1: 4’-substituted triclosan derivatives. Bioorg
Med Chem Lett;15:5247–5252.
25. Young S.S., Hawkins D.M. (2004) Using recursive par-
titioning analysis to evaluate compound selection
methods. Methods Mol Biol;275:317–334.
26. Rusinko A. 3rd, Young S.S., Drewry D.H., Gerritz S.W.
(2002) Optimization of focused chemical libraries using
recursive partitioning. Comb Chem High Throughput
Screen;5:125–133.
27. Mamatha N., Murtuja S.B., Varaprasad B.F.M., Reddy
L.V., Bhattacharya A., Helliwell M., Mukherjee A.K.,
Beevi S.S., Mangamoori L.N., Mukkanti K., Pal S.
(2010) Naproxen and ibuprofen based acyl hydrazone
derivatives: synthesis, structure analysis and cytotoxic-
ity studies. J Chem Pharm Res;2:393–409.
28. Ershov A.Y., Lagoda I.V., Yakimovich S.I., Pakal’nis
V.V., Zerova I.V., Dobrodumov A.V., Shamanin V.V.
(2009) Tautomerism and conformational isomerism of
mercaptoacetylhydrazones of aliphatic and aromatic
aldehydes. Russ J Org Chem;45:660–666.
13. Mishra S., Karmodiya K., Parasuraman P., Surolia A.,
Surolia N. (2008) Design, synthesis, and application of
novel triclosan prodrugs as potential antimalarial and
antibacterial agents. Bioorg Med Chem;16:5536–5546.
14. Perozzo R., Kuo M., Sidhu A.S., Valiyaveettil J.T., Bitt-
man R., Jacobs W.R. Jr, Fidock D.A., Sacchettini J.C.
(2002) Structural elucidation of the specificity of the
antibacterial agent triclosan for malarial enoyl acyl car-
rier protein reductase. J Biol Chem;277:13106–13114.
15. Khedkar V., Verma J., Coutinho E. (2009) Small-mole-
cule screening techniques in drug discovery. Bio-
bytes;4:4–7.
16. Kitchen D.B., Decornez H., Furr J.R., Bajorath J.
(2004) Docking and scoring in virtual screening for
drug discovery: methods and applications. Nat Rev
Drug Discov;3:935–949.
17. Friesner R.A., Banks J.L., Murphy R.B., Halgren T.A.,
Klicic J.J., Mainz D.T., Repasky M.P., Knoll E.H., Shel-
ley M., Perry J.K., Shaw D.E., Francis P., Shenkin P.S.
(2004) Glide: a new approach for rapid, accurate
docking and scoring. 1. Method and assessment of
docking accuracy. J Med Chem;47:1739–1749.
18. Halgren T.A., Murphy R.B., Friesner R.A., Beard H.S.,
Frye L.L., Pollard W.T., Banks J.L. (2004) Glide: a new
approach for rapid, accurate docking and scoring. 2.
Enrichment factors in database screening. J Med
Chem;47:1750–1759.
29. Palla G., Giovanni P., Domiano P. (1986) Conforma-
tional behaviour an E/Z isomerization of N-Acyl and N-
aroylhydrazones. Tetrahedron;42:3649–3654.
€
30. Hubschle C.B., Sheldrick G.M., Dittrich B.J. (2011)
ShelXle: a Qt graphical user interface for SHELXL. J
Appl Crystallogr;44:1281–1284.
31. Tasdemir D., Lack G., Brun R., Ruedi P., Scapozza L.,
Perozzo R. (2006) Inhibition of Plasmodium falciparum
fatty acid biosynthesis: evaluation of FabG, FabZ, and
FabI as drug targets for flavonoids.
J
Med
Chem;49:3345–3353.
19. Ramamurthy B., Bhatt M.V. (1989) Synthesis and antit-
ubercular activity of N-(2-naphthyl)glycine hydrazide
analogues. J Med Chem;32:2421–2426.
32. Tasdemir D., Topaloglu B., Perozzo R., Brun R., O’Ne-
ill R., Carballeira N.M., Zhang X., Tonge P.J., Linden
€
A., Ruedi P. (2007) Marine natural products from the
20. Jain A.G., Gupta P., Ganesan G., Pande A., Malhotra
R. (2007) Rapid solvent-free synthesis of aromatic hy-
drazides under microwave irradiation. Def Sci J;57:267
–270.
Turkish sponge Agelas oroides that inhibit the enoyl re-
ductases from Plasmodium falciparum, Mycobacterium
tuberculosis and Escherichia coli. Bioorg Med
Chem;15:6834–6845.
21. Awasthi S.R., Gupta P., Rao A., Ganesan K., Malhotra
R. (2007) Synthesis, characterization and spectral
studies of various newer long chain aliphatic acid (2-
hydroxy benzylidene and 1H-indol-3-ylmethylene) hy-
drazides as mosquito para-pheromones. J Korean
Chem Soc;51:506–512.
22. Blower P., Fligner M., Verducci J., Bjoraker J. (2002)
On combining recursive partitioning and simulated
annealing to detect groups of biologically active com-
pounds. J Chem Inf Comput Sci;42:393–404.
33. Zmuidinavicius D., Japertas P., Petrauskas A., Didzia-
petris R. (2003) Progress in toxinformatics: the chal-
lenge of predicting acute toxicity. Curr Top Med
Chem;3:1301–1314.
34. Ghose A.K.V., Viswanadhan V.N., Wendolowski J.J.
(1998) Prediction of hydrophobic (lipophilic) properties
of small organic molecules using fragmental methods:
an analysis of ALOGP and CLOGP methods. J Phys
Chem;102A:3762.
23. Strobl C., Malley J., Tutz G. (2009) An introduction to
recursive partitioning: rationale, application, and charac-
teristics of classification and regression trees, bagging,
and random forests. Psychol Methods;14:323–348.
24. Choi S.Y., Shin J.H., Ryu C.K., Nam K.Y., No K.T.,
Park Choo H.Y. (2006) The development of 3D-QSAR
Notes
a(1999–2009) iResarch Library. In: editor. iResarch Library.
ChemNavigator.com. Inc.; p.
bAccelrys I. (1998) Cerius2. In: editor. Cerius2. San Diego,
CA, USA; p.
14
Chem Biol Drug Des 2013

Synthesis and Evaluation of a New Class of pfENR Inhibitors
1
Figure S7. Important H-15N HMBC correlations observed
Supporting Information
for the molecule 21 recorded at 298K.
Figure S8. H and 13C assignments of molecule 21.
1
Figure S9. NOESY spectra of molecule 21 recorded at
298K.
Additional Supporting Information may be found in the
online version of this article:
Figure S10. Schematic showing nOe peaks that support
the imidazolidine-4-one structure.
Figure S11. 400 MHz 1H NMR spectrum of molecule 21
in CDCl3.
Figure S1. 500 MHz 1H NMR spectrum of molecule 21
recorded at 298 K.
Figure S2. 100 MHz 13C DEPT and 1H decoupled spec-
tra of molecule 21 recorded at 298K.
Figure S12. 500 MHz NOESY spectrum of molecule 21 in
CDCl3. assignments are given in the inset.
Table S1. Crystal data and structure refinement.
Table S2. Atomic coordinates (9104) and equivalent iso-
Figure S3. 400 MHz COSY spectrum of molecule 21
recorded at 298K; D1: D1:3.5 – 9.9 ppm and D2:3.5 –
9.9 ppm.
ꢀ2
tropic displacement parameters (A 9 103) for rupe-
Figure S4. 1H-13C HSQC spectrum of molecule 21
recorded at 298K; D1: 3.3 – 10.2 ppm, D2: 0 – 165 ppm.
Figure S5. Important 1H-13C HMBC correlations used for
structural elucidation of molecule 21 recorded at 298K; (a)
D1:9.36 – 9.63 ppm, D2: 49 – 139 ppm (b) 6.07 –
6.22 ppm, D2: 119 – 168 ppm (c) 4.0 – 4.6 ppm, D2: 78
– 168 ppm.
shr_0 m. U(eq) is defined as one-third of the trace of the
orthogonalized Uij tensor.
ꢀ
Table S3. Bond lengths [A] and angles [°].
Table
ꢀ2
S4.
(A 9 103).
Anisotropic
displacement
parameters
Table S5. Hydrogen coordinates (9104) and isotropic dis-
placement parameters.
Figure S6. Important 1H-13C HMBC correlations in mole-
cule 21 recorded at 298K; D1: 4.1 – 7.9 ppm, D2: 75 –
83 ppm.
Table S6. Torsion angles [°].
Chem Biol Drug Des 2013
15