In Silico Exploration for New Antimalarials: Arylsulfonyloxy Acetimidamides as Prospective Agents

Article abstract of DOI:10.1021/acs.jcim.5b00392

A strategy is described to identify new antimalarial agents to overcome the drug resistance and/or failure issues through in silico screening of multiple biological targets. As a part of this, three enzymes namely CTPS, CK, and GST were selected, from among 56 drug targets of P. falciparum, and used them in virtual screening of ZINC database entries which led to the design and synthesis of arylsulfonyloxy acetimidamides as their consensus inhibitors. From these, two compounds showed good activity against sensitive (3D7; IC₅₀, 1.10 and 1.45 μM) and resistant (K1; IC₅₀, 2.10 and 2.13 μM) strains of the parasite, and they were further investigated through docking and molecular dynamics simulations. The findings of this study collectively paved the way for arylsulfonyloxy acetimidamides as a new class of antimalarial agents. (Chemical Presented).

Full text of DOI:10.1021/acs.jcim.5b00392

Article
pubs.acs.org/jcim
In Silico Exploration for New Antimalarials: Arylsulfonyloxy
Acetimidamides as Prospective Agents
Saroj Verma,† Utsab Debnath,† Pooja Agarwal,‡ Kumkum Srivastava,‡ and Yenamandra S. Prabhakar*,†
‡
†Medicinal and Process Chemistry Division, Parasitology Division, CSIR-Central Drug Research Institute, Jankipuram Extension,
Lucknow-226 031, India
S
* Supporting Information
ABSTRACT: A strategy is described to identify new antimalarial agents to overcome the
drug resistance and/or failure issues through in silico screening of multiple biological
targets. As a part of this, three enzymes namely CTPS, CK, and GST were selected, from
among 56 drug targets of P. falciparum, and used them in virtual screening of ZINC
database entries which led to the design and synthesis of arylsulfonyloxy acetimidamides as
their consensus inhibitors. From these, two compounds showed good activity against
sensitive (3D7; IC50, 1.10 and 1.45 μM) and resistant (K1; IC50, 2.10 and 2.13 μM) strains
of the parasite, and they were further investigated through docking and molecular dynamics simulations. The findings of this
study collectively paved the way for arylsulfonyloxy acetimidamides as a new class of antimalarial agents.
feasible options to deal with the situation, the strategies mulling
over “multiple drugs against single or multiple targets” (or
combination therapy)2 and “single drug against multiple
targets”3,4 are the notable ones. While the former one is fairly
explored in the chemotherapy of infectious diseases with limited
success, the later one holds promise for exploration. The main
causative agent of human malaria, P. falciparum, with its 14
chromosomes encodes more than five thousand proteins.7 It is a
huge reservoir to explore all probable targets of antimalarial
drugs. Even though every organism tanks large number of
enzymes and proteins as part of its biochemical machinery, its
survival and sustenance depends on the function of a few key
ones in the regulatory network. In this backdrop a computational
approach is contemplated to identify multiple enzymes to design
single consensus chemical entity (CCE) as their inhibitors/
modulators to overcome the treatment failure and emergence of
drug resistance in malaria.
Some connectedness among possible drug targets together
with their functional importance in biochemical/metabolic paths
may offer scope for the selection of potential enzymes and
thereby the design of CCEs as their inhibitors/modulators. The
concept is explored through a distance matrix approach by
identifying three key enzymes of the parasite which have vital
functions in the biochemical/metabolic network. This has
resulted in cytidine triphosphate synthetase (CTP synthetase
or CTPS),8 choline kinase (CK),9 and glutathione S-transferase
(GST)10 as potential enzymes for exploring the concept of a
single drug against multiple targets or CCE. Furthermore, the
pathways of these enzymes, as indicated in Brenda enzyme
database, suggested their significance to the parasite’s
sustenance.11 They are used in virtual screening of ZINC
database entries12 to identify the CCEs. This has led to
1. INTRODUCTION
Worldwide malaria maps with poverty. This protozoan infection
is a leading cause of morbidity and mortality in tropical and
subtropical regions. Globally it alone causes a disease burden of
about 200 million infections and more than 500 000 deaths
annually.1 Different Plasmodium species which include P.
falciparum, P. malariae, P. vivax, and P. ovale are the causative
agents. Among these, P. falciparum (Pf) is the most lethal and
widely spread in human beings. Its clinical management depends
on several drugs.2 They include chloroquine (CQ), mefloquine,
atovaquone, proguanil, sulphadoxine−pyrimethamine combina-
tion, artemisinins, artemisinin combination therapy, etc. Apart
from the foregoing, some widely tried procedures for discovering
fresh leads/drugs against this parasite include screening of
diverse chemicals, new combinations of two (or more) existing
drugs to act on same or different enzyme targets and strategies to
counter the drug resistance.2−4 In the face of these interventional
measures malarial parasite, with impunity, has managed to keep
the drugs at bay in controlling its rise and transmission.2 The
treatment failures are often attributed to a multitude of reasons
which include mutation of targeted enzyme(s), amplification of
parasite’s transport proteins in pumping out or expelling the
drug(s), and/or opening up of alternative biochemical/
metabolic paths in the organism for its survival.5,6 The situation
calls for rethinking and conceptualization of strategies to
interfere with the deep-seated pathways of the parasite to
restrain its resurgence and control it with little scope for the
emergence of drug-resistant strains. In this outset, exploring the
parasite’s biological machinery for hitherto unexplored bio-
chemical pathways may open up fresh and tactical avenues for
new chemicals to combat the malaria.
The therapeutic strategies aiming at single biological target (of
the pathogen) with limited or closely related chemical entities are
often attributed to be a leading cause for the drug failure and/or
emergence of drug resistant strains of pathogens.5,6 Among the
Received: March 14, 2015
© XXXX American Chemical Society
A
DOI: 10.1021/acs.jcim.5b00392
J. Chem. Inf. Model. XXXX, XXX, XXX−XXX

Journal of Chemical Information and Modeling
Article
arylsulfonyloxy acetimidamides as model compounds for
synthesis and biological evaluation. Some of these compounds
showed encouraging activity against P. falciparum (3D7, K1).
The most active compound from these is further studied through
docking13,14 and molecular dynamic (MD) simulations15 to shed
some light on possible molecular interactions with the selected
enzymes. The details are discussed here.
analysis, for the selection of three enzymes from among the
considered targets by a distance matrix approach and by
examination of their functional roles in the metabolic pathways.
2.2. Selection of Target Enzymes. The study has
proceeded with segregation of all the targets into six distinct
enzyme classes (Table 1). The FASTA sequences of all the
targets were retrieved from NCBI and UniProt. The BLAST
(Basic Local Alignment Search Tool) analysis of FASTA
sequences of all targets was carried out to identify the
connectedness and proximity of the targets. The sequence
similarity of each target was determined against all others in
terms of e-values. These were used to compute the total distance
(TD) and average distance (AD) of targets, between different
classes (inter; BC) as well as within each class (intra; C), as
described below.
2. MATERIALS AND METHODS
2.1. Enzyme Database. The study has involved fifty-six drug
targets of P. falciparum reported in recent literature (Table 1).
The availability of primary structures (FASTA sequences) and a
precedence/proposal for them as probable drug targets was the
criteria for the consideration of these enzymes. It is assumed that
from among these enzymes, the “connected and proximately
placed” ones give scope to design a CCE to modulate their
activity. Thus, we formulated a strategy, akin to phylogenic
TDBC
ADBC
=
e
i,j
∑
∑
=
e /(nm)
i,j
Table 1. P. falciparum Enzymes Considered in the Exploration
where, TDBC is between classes total distance, ADBC is between
classes average distance, ei,j is the e-value between target i (i = 1−
n; n is total targets in p-class) of p-class and target j (j = 1−m; m is
total targets in q-class) of q-class.
S.
S.
a
a
no.
targets
no.
targets
1
2
cytochrome c oxidase
29 farnesyltransferase
1-deoxy-D-xylulose 5-phosphate
reducto-isomerase
30 glutathione S-transferase
TD =
e
i,j
∑
∑
C
3
4
5
deoxyhypusine synthase
dihydrofolate reductase
31 β-ketoacyl-ACP synthase
32 N-myristoyltransferase
AD =
e /((n2 − n)/2)
dihydroorotate dehydrogenase
33 cholinephosphate
cytidylyltransferase
C
i,j
where, TDC is within class total distance, ADC is within class
average distance, ei,j is the e-value between targets i and j with i
varying from 1 to n and j varying from i + 1 to n; n is the total
targets in the class.
6
enoyl-ACP-reductase
34 hypoxanthine guanine
phospho ribosyl-transferase
7
8
9
ferredoxin NADP reductase
glutamate dehydrogenase
glutathione reductase
35 choline kinase
36 dihydropteroate synthase
37 hexokinase
In the first iteration, the enzyme class with smallest average in-
between class distances (ADBC) was considered and assigned as
closely placed to all others (complete data is provided as part of
the Supporting Information). The first enzyme of intended
multiple targets was selected from among the members of this
class using the within class distances as well as its relevance in the
metabolic pathway for the parasite’s survival and sustenance. In
the second iteration, the first target was taken as reference to
compute the distances of remaining classes from it and the class
which is closely placed to it was considered for further analysis. In
view of the importance of “cleaning up” or “detoxification”
processes in parasite milieu, P. falciparum glutathione S-
transferase (GST) is adopted as third enzyme of the intended
multiple targets for inhibition/modulation.
2.3. Structural Modeling. The FASTA sequence of target
protein was retrieved from the NCBI database and its primary
properties were assessed using different servers.16,17 The
template protein structure(s) consensus to target sequence was
searched from the protein data bank with PSI-BLAST,18
PHYRE19 and JPred320 under default parameter settings. The
templates showing low e-value and high sequence identity with
the target sequence were selected as model template(s). The
alignment of target sequence with the template was done using
Modeler 9.10.21 On alignment the disordered regions of target
were checked in DisEMBL.22 Giving due consideration to the
conserved motifs, the disordered regions of the aligned target was
rectified in Bio-Editor23,24 by editing the sequence to minimize
the gaps/insertions. The “rectified” target sequence was
realigned with the template in Modeler 9.10 to generate the
initial models. The model(s) was assessed for the stereochemical
quality scores which include DOPE score, MOLPDF, GA341
10 lactate dehydrogenase
38 homospermidine synthase
11 ubiquinol-cytochrome-c
reductase
39 S-adenosylmethionine
synthetase
12 ribonucleotide reductase
13 succinate dehydrogenase
14 thioredoxin reductase
15 DNA gyrase
40 aspartate
carbamoyltransferase
41 1-deoxy-D-xylulose-5-
phosphate synthase
42 purine nucleoside
phosphorylase
43 plasmepsin I
44 plasmepsin II
16 peptidyl-prolyl cis−trans
isomerase
17 topoisomerase I
45 sphingomyelinase
46 adenosine deaminase
47 histone deacetylases
48 S-adenosyl-L-homocysteine
49 peptide deformylase
50 helicase
18 topoisomerase II
19 triose phosphate isomerase
20 adenylosuccinate lyase
21 β-hydroxyacyl-ACP dehydratase
22 chorismate synthase
23 DNA-(apurinic/apyrimidinic
site) lyase
51 dihydroorotase
24 fructose 1,6 biphosphate aldolase 52 acyl-CoA synthetase
25 2-C-methyl-D-erythritol-2,4-
cyclodiphosphate synthase
53 CTP synthase
26 orotate
54 carbamoyl phosphate
synthetase
phosphoribosyltransferase
27 RNA polymerase
28 DNA polymerase
55 dihydrofolate synthase
56 acetyl-CoA carboxylase Ligase
a
The enzymes listed under serial numbers 1−14 are oxidoreductases
(Ox), 15−19 are isomerases (Is), 20−25 are lyases (Ly), 26−42 are
transferases (Tr), 43−51 are hydrolases (Hy), and 52−56 are ligases
(Li). Complete bibliographic details are provided in the Supporting
Information.
B
DOI: 10.1021/acs.jcim.5b00392
J. Chem. Inf. Model. XXXX, XXX, XXX−XXX

Journal of Chemical Information and Modeling
Article
a
Table 2. Average Distance of Enzyme Classes Based on the e-Values between Targets of P. falciparum
enzyme class
oxidoreductase
isomerase
lyase
transferase
hydrolase
ligase
oxidoreductase
isomerase
lyase
2.56
2.95
2.89
2.32
2.40
2.18
3.40
2.71
2.61
2.67
2.51
1.47
2.18
2.46
2.30
transferase
hydrolase
ligase
2.25
2.09
1.86
2.60
1.86
1.11
a
All enzymes and corresponding data are provided in the Supporting Information.
score, Ramachandran plot, Errat plot, and Prosa plot, and the
best scored one was selected for refinement.21,25,26 This was
further refined for relieving steric clashes and improper contacts
by energy minimization using GROMOS96 force field27 in Swiss
PDB viewer 4.0.1,28 followed by iterative loop refinement in
Modeler 9.10 (ERRAT plot showing confidence limits to reject
residue regions that exceed ≥95%) to give the homology model
of the target. Followed by this, the active site(s) (binding pocket)
of selected targets were identified with due consideration to the
bound ligands of template protein(s) and the conserved domain
by involving the CASTp (Computed Atlas of Surface Topology
of Proteins), Site-finder of MOE, and protomol of SYBYLX 1.3.
2.4. Molecular Dynamics (MD) Simulations. In
NAMD2.715 MD simulations were carried out on the homology
model to obtain equilibrated protein structure. This was done
under periodic boundary conditions using the solvated (water
box) protein by applying appropriate force field, constant
molecular number, pressure (1 atm), temperature (310 K;
NPT), cut-offs for nonbonded atom interactions (12 Å), and
Particle mesh Ewald (PME) algorithm for long-range electro-
static forces between the atoms. The system (protein in water
box) was adequately minimized followed by equilibrated for
enough length of time. The integration time step of dynamics
was set to 2 fs, and the trajectories were collected at regular
intervals. The results were analyzed and protein stability was
assessed in VMD 1.91 (Visual molecular dynamics),29
VEGAZZ,30 and Video match31 by computing different
parameters which include, energies, radius of gyration, and
root-mean-square deviation (RMSD) of the trajectories.
Furthermore, the MD simulations were also suitably carried
out on the selected enzymes and CCE complexes to study the
stability of these systems and dynamics of molecular interactions
between them.
2.5. Virtual Screening. From ZINC database, using the
structural similarity between its entries and the known
substrates/inhibitors of Pf-CTPS, CK, and GST, 8635 molecules
were identified for virtual screening. The identified molecules
were screened against the selected targets in the Surflex-Dock
Geomax (SFXC) module of Sybyl-X 1.3.13 The corresponding
proteins were prepared following the default procedure set in the
module by applying the Gasteiger−Huckel charges and subjected
to staged energy minimization using Tripos force field. The
residue based protomols (binding pockets) were generated from
the prepared proteins for the docking experiments. The
screening of the identified molecules was done by successively
docking them into the generated binding pockets. The affinities
of the molecules to the respective enzymes were analyzed using
crash, polar, and total scores obtained from the docking
experiments.
2.6.1. Synthesis of N′-Hydroxy-2-arylacetimidamides (2).
Hydroxylamine hydrochloride (2.4 mmol) was dissolved in
methanol. Methanolic sodium methoxide (2.5 mmol) was added
to this solution and refluxed for 30 min. 4-Substituted
benzeneacetonitraile (2 mmol) (1) was added to this solution
and further refluxed the contents for 12 hours to yield
corresponding N′-hydroxy-2-arylacetimidamide (2). The pro-
gress of the reaction was monitored by TLC (10% MeOH/
CHCl3). On completion of the reaction, methanol was
evaporated by rota vapor under reduced pressure. The residue
was dissolved in dichloromethane and washed with brine. The
organic layer was collected and dried with anhydrous sodium
sulfate to obtain crude compound 2 (yield 75−79%). This was
further purified through column chromatography and the
structure of N′-hydroxy-2-arylacetimidamide (2) was confirmed
by analyzing NMR and mass spectra.
2.6.2. Synthesis of Arylsulfonyloxy Acetimidamides (3). In a
round-bottom flask to 30 mL dichloromethane, 1 mmol of
compound 2 followed by 1.2 mmol triethayl amine were added
and the solution was cooled to 0 °C. To this, 1.2 mmol
arylsulphonyl chloride was added with constant stirring. The
stirring of reaction mixture was continued for 1−2 h to complete
the sulfonylation of 2 to afford corresponding arylsulfonyloxy
acetimidamide (3). The completion of reaction was monitored
through TLC (40% EtOAc/hexane). On completion of reaction,
the mixture was washed with brine and dried with anhydrous
sodium sulfate followed by evaporated dichloromethane under
reduced pressure to obtain crude arylsulfonyloxy acetimidamide
(3) (yield 70−85%). This was further purified through column
chromatography and the structure of arylsulfonyloxy acetimida-
mide (3) was confirmed by analyzing 1H NMR, 13C, ES-MS, and
HRMS spectra.
2.7. Biological Screening. The synthesized compounds
were evaluated for antimalarial activity against 3D7 (CQ-
sensitive) and K1 (CQ-resistant) strains of P. falciparum using
Malaria SYBR Green I nucleic acid staining dye based
fluorescence (MSF) assay as mentioned by Singh et al.33 The
stock (10 mM) solution was prepared in DMSO and test
dilutions were prepared in culture medium (RPMI-1640-FBS).
Chloroquine-diphosphate (SIGMA) was used as reference drug.
Briefly the procedures are described below.
2.7.1. Antimalarial Assay. 50 μL of culture medium was
dispensed in 96 well plate followed by addition of 50 μL of
highest concentration (in subsequent screens with 2-fold serial
dilutions) of test compounds (in duplicate wells) in row B.
Following this, 50 μL of 2.0% parasitized cell suspension
containing 0.8% parasitaemia (Asynchronous culture containing
more than 80% ring stages) was added to each well except four
wells in row “A” which received matching volume of non
parasitized erythrocyte suspension. The plates were incubated at
37 °C in CO2 incubator in an atmosphere of air mixture with 5%
CO2 for 72 h. Following this 100 μL of lysis buffer containing 2×
2.6. Synthesis. The arylsulfonyloxy acetimidamides, the
designed common inhibitors of three targets, are prepared using
the following synthetic scheme.32
C
DOI: 10.1021/acs.jcim.5b00392
J. Chem. Inf. Model. XXXX, XXX, XXX−XXX

Journal of Chemical Information and Modeling
Article
concentration of SYBR Green-I (Invitrogen) was added to each
well and the plates were incubated for one more hour at 37 °C.
The plates were examined at 485 20 nm (excitation) and 530
20 nm (emission) for relative fluorescence units (RFUs) per well
using the fluorescence plate reader (FLX800, BIOTEK). The
IC50 values were obtained by Logit regression analysis of dose
response curves using a preprogrammed Excel spreadsheet.
2.7.2. Cytotoxicity Assay. Cytotoxicity of the compounds was
determined using Vero cell line (C1008; Monkey kidney
fibroblast) following the method as mentioned in the work of
Sharma et al.34 The cells were incubated with compound-
dilutions for 72 h and MTT (SIGMA) was used as reagent for
detection of cytotoxicity. Podophyllotoxin (SIGMA) was used as
the reference drug. The 50% cytotoxicity concentration (CC50)
was determined using nonlinear regression analysis of dose
response curves using a preprogrammed Excel spreadsheet.
choline) followed by to phosphatidylcholine. Interestingly, for
the parasite ethanolamine kinase (EK) is another enzyme in its
pathway for the production of phosphatidylethanolamine via
CTP.
For all organisms/living systems, cleaning up or detoxification
of its milieu is an important task. In the parasite, glutathione S-
transferase (GST) is one enzyme involved in the detoxification
process via ceramide. As a part of downstream regulation of
events sphingomyelin synthase (SS) oversees the formation of
sphingomylein from ceramide and phosphatidylcholine (from
CK/CTPS pathway) and pave the way for detoxification/
apoptosis signaling in the parasite. In the process, parasite’s GST
facilitates the glutathione-toxin (ceramide/xenobiotics) adduct
formation and helps in detoxification of the milieu. Thus,
inhibition of GST may help in the retention of ceramide/
xenobiotic substances thereby make the parasite milieu toxic.
Considering this we have adopted parasite’s GST as third
enzyme of the intended multiple targets for inhibition. The
interconnections between the biochemical pathways of selected
enzymes are shown in Figure 2, thus justifying them as possible
targets against malaria.
3. RESULTS AND DISCUSSION
3.1. Target Selection. The average in-between class
distances of six distinct enzyme classes comprising fifty-six
drug targets of P. falciparum are shown in Table 2 (Figure 1).
3.2. Pf-CTPS Homology Model. In the absence of the 3D
structure of Pf-CTPS, its homology model has been developed
from the FASTA sequence (860 residues) with accession code
AAC36385 retrieved from the NCBI database.8 It is a 98.85 kDa
protein with isoelectric point 6.381. Its secondary structure
membrane topology showed that the residues are distributed in
the cytoplasmic (1−520), transmembrane (521−539), and
extra-cyoplasmic (540−860) regions. It has two substrate
binding domains namely synthetase domain in vicinity of N-
terminal and glutaminase domain in vicinity of C-terminal. Using
the glutaminase domain, Pf-CTPS transforms uridine-5′-
triphosphate (UTP) into cytidine triphosphate (CTP).
A standard search of protein data bank with the Pf-CTPS
FASTA sequence led to the X-ray structures of CTPS from
sulfolobus solfataricus (PDB code: 3NVA; number of residues,
535; e-value, 7.00 × 10−107; sequence identity, 39%)35 and
Thermus thermophilus (PDB code: 1VCO; number of residues,
550; e-value, 9.00 × 10−102; sequence identity, 36%)36 as best
consensus templates for its homology modeling. A further
examination of Pf-CTPS sequence in PHYRE, a fold recognition
method, suggested 3NVA as best suited template for the
structural modeling of this protein. In Modeler 9.10 the
alignment of target FASTA sequence with the 3NVA template
followed by the editing of undefined portions with due care for
the conserved regions and on iterative refinement led to the
optimized homology model of Pf-CTPS (Figure 3). We also
made attempts to model the Pf-CTPS using 3NVA, 1VCO, and
other pdb templates, through multiple sequence alignment;
however, they did not give any better results than using 3NVA
alone. In this scenario, we carried out the further study using the
Pf-CTPS model generated from 3NVA. The model is further
refined through MD simulations for 5 ns. The structure has
reached an equilibrium conformation approximately after 0.75 ns
of dynamics simulations. The last frame has shown RMSD 4.87 Å
(for frames from 0.75 to 5 ns, average RMSD SD is 4.39 0.28
Å). On reaching equilibrium, the model has retained its shape
without much change in its structure (Figure 3).
Figure 1. Interclass distances (e-value) of P. falciparum targets.
This clearly showed that ligase is “closely” placed to all other
classes (∑ADBC = 10.70). This class has five enzymes in it. The
within class distances of each enzyme with all other targets led to
CTP synthase (CTPS) as closely placed one to remaining
members of this class. In the parasite’s metabolic pathway, CTPS
is involved in the synthesis of CTP, a high-energy molecule like
ATP. The parasite uses CTP in the synthesis of glycerophos-
pholipids and glycosylation of proteins and as a substrate in RNA
synthesis. In view of this, CTPS has been selected as one of the
three intended targets for inhibition/modulation.
For the selection of next target, CTPS has been taken as the
focal point to compute the distances between CTPS and
remaining five enzyme classes. Herein, the transferases have
emerged as close to CTPS. This enzyme class has 17 targets in it.
The search for second enzyme in this class in conjunction with
CTPS has led to choline kinase (CK) as potential target. In the
parasite CK facilitates the biosynthesis of phospholipid (PL)
through the phosphorylation of choline. In the metabolic
pathway by making use of CTP, choline monophosphate gets
converted into corresponding diphosphate nucleotides (CDP-
3.3. Target Sites. In Pf-CTPS, as the glutaminase domain is
important for the transformation of UTP into CTP, in the model
it is determined by making use of glutamine bound X-ray
cocrystal of CTPS (PDB code 1VCO) of Thermus thermophilus
HB8 together with 3NVA in multiple sequence alignment
D
DOI: 10.1021/acs.jcim.5b00392
J. Chem. Inf. Model. XXXX, XXX, XXX−XXX

Journal of Chemical Information and Modeling
Article
Figure 2. Interconnections between the pathways of CTPS, CK, and GST of P. falciparum (CDP, cytidine diphosphate; CMP, cytidine monophosphate;
CPCT, choline-phosphate cytidyltransferase; EPCT, ethanolamine-phosphate cytidyltransferase; CPT, cholinephosphotransferase; EPT, ethanolamine
phosphotransferase; SS, sphingomyelin synthase; SM, sphingomyelinase; SD, sphingomyelin deacylase).
Figure 3. (a) 3D structure of Pf-CTPS materialized from CTPS template 3NVA through homology modeling; (b) homology model of Pf-CTPS at the
end of 5 ns molecular dynamics simulations in NAMD; (c−e) RMSD, radius of gyration, and different components of molecular energy profiles of Pf-
CTPS trajectories, respectively. The data in the box shows some formal characteristics of Pf-CTPS model as developed from homology modeling
procedure and after subjecting it to 5 ns MD simulations.
method. This has led to the Pf-CTPS’s glutaminase site as
composed of residues TYR54, GLY384, GLY385, PHE386,
CYS412, LEU413, GLN416, GLU435, ARG583, HIS584,
ARG585, TYR586, and HIS631.36 These residues are part of
conserved ones in CTPS enzyme from different species. Among
these, LEU413, GLU416, GLY385, PHE386, GLU435, and
E
DOI: 10.1021/acs.jcim.5b00392
J. Chem. Inf. Model. XXXX, XXX, XXX−XXX

Journal of Chemical Information and Modeling
Article
ARG585 are key residues for interaction with glutamine.
Furthermore, the residues MET51 to GLY67 of loop region
are part of the gateway for the transfer of ammonia released from
the glutaminase domain to the synthetase domain.36 The results
from CASTp, Site-finder of MOE, and protomol of SYBYLX 1.3
also supported the identified pocket.37,38
In Surflex-Dock using default settings all these 8635 molecules
were screened by docking them in to each target site. For every
molecule the best docked pose (for each target) is selected based
on its total dock score with due consideration to crash and polar
scores. For simplicity only one dock pose, corresponding to
maximum total dock score, per molecule is considered. The
molecules’ total dock scores for each enzyme were analyzed, in
conjunction with the total dock scores from remaining two
enzymes, by successively arranging them in descending order to
identify the chemical entities consensus to all three enzymes.
This has led to several ZINC database entries bearing
substructures of aryl sulfonamide, aryl sulfonate, aryl acetamide
(molecular weight spread 313−470), and peptidomimetics
(molecular weight ∼600+) containing glutamine, alanine and/
or cysteine moieties, and combinations thereof as agreeable with
consensus docking scores to the selected enzymes. From among
these, the ZINC entries with substructures of aryl sulfonamide,
aryl sulfonate, aryl acetamide are considered in view of their
relatively simple scaffolds and low molecular weights. Moreover,
the molecular weights of these entries are within the limits
suggested by Lipinski’s rule of five. In most of these molecules
OH, SO3, SO2, NH2, NH, and/or CONH functionalities with
aryl groups were found to be involved in binding to the target
sites. Briefly, the molecular structures of these ZINC entries
represent a functionalized core extended by hydrophobic
moieties on one or more sides.
Figure 5 shows the interactions and docking scores of selected
ZINC database entries. In Pf-CTPS, the SO2/ SO3 groups of
ZINC00627448 and ZINC79315625 showed H-bond with
ARG585. In ZINC49575668, its SO2 group showed H-bond
with GLY387 of Pf-CTPS. Apart from these the ZINC entries
showed H-bonds with MET51, SER52, VAL59, GLY57,
GLY385, GLU441, ARG583, and TYR586 through their NH,
NH2, and OH groups. In CK, ZINC49575668 and
ZINC79315625 interacted with the enzyme through the SO2/
SO3 groups by H-bonding with ARG399. In case of
ZINC00627448, its SO2 group showed H-bond with GLY114
and ASN117 of the enzyme. The other important residues for H-
bonding with NH, NH2, and OH of ZINC entries are found to be
THR116, ASP305, GLN290, GLU307, and TYR308. In GST,
the SO2 group of ZINC49575668 and ZINC00627448 showed
H-bond with GLN73. In ZINC79315625, its SO3 group showed
H-bond with ASP105 of GST. The other key residues for H-
bonding with NH, NH2, and OH of the compounds are found to
be LYS15, GLN71, CYS101, and GLN104.
The target sites of remaining two enzymes CK and GST are
adopted from their enzyme−ligand cocrystal structure PDB
codes 3FI8 and 2AAW, respectively.39,40 However, the reported
3D structure of CK has gaps at several places. For the purpose of
MD simulations the structure is rectified by appropriate
modeling techniques. The CK target site residue numbers in
the original pdb and rectified pdb are ASP288 (210, residue
number in rectified pdb), GLN290 (212), ASP305 (227),
GLU307 (229), TYR308 (230), GLU324 (246), TYR329 (251),
TRP392 (314), TRP395 (317), TYR414 (336), and ARG418
(340). The GST target site is mainly composed of residues
TYR9, GLY14, LYS15, PHE45, GLN58, VAL59, PRO60,
GLN71, SER72, GLN104, HIS107, ASN111, PHE116, and
TYR211. In CK, among the identified residues, ASP210 and
GLN212 play important role in binding the substrate, whereas in
GST, the residues TYR9, LYS15, and GLN71 are important for
glutathione binding. These target sites are used in virtual
screening to identify the CCEs.
3.4. Virtual Screening. In order to acquire the chemical
entities for virtual screening (VS) the substrates/inhibitors of the
three enzymes were used as query molecules to filter the
matching structures from ZINC database. Here, glutamine and 6-
diazo-5-oxo-L-norleucine (DON) of CTPS, choline, N-[2-[[3-
(4-bromophenyl)-2-propenyl]amino]ethyl]-5-isoquinoline-
sulfonamide (H89), and 2-amino-1-butanol of CK, and
glutathione, 1-amino-4-{[4-({4-chloro-6-[(2-sulfophenyl)-
amino]1,3,5-triazin-2-yl}amino)-3-sulfophenyl]amino}-9,10-
dioxo-9,10-dihydro-2-anthracenesulfonic acid of GST were used
as query molecules (Figure 4).41−44 The filtering of the ZINC
database entries for the VS structures is carried out in
conjunction with query molecules (Figure 4) by using the
substructure based similarity and “tight filtering” with 50%
similarity as search criteria. This has resulted in 8635 molecules
for virtual screening (details are in the Supporting Information).
As the ZINC entries selected from the virtual screening are
found to interact with the envisaged enzyme targets using their
OH, SO3, SO2, NH2, NH, and/or CONH functionalities and aryl
moieties, we looked for a scaffold satisfying these features. Also,
malaria being an infection prevalent in the lower strata of socio-
economic population, its chemotherapy demands compounds
from simple and efficient synthetic procedures for exploration.
This has led to arylsulfonyloxy acetimidamides, which has high
functional similarity with the selected ZINC entries, for
exploration. These compounds are easy to synthesize and meet
the economic criteria of new drug explorations in malaria
chemotherapy.
3.5. Synthesis. The synthesis of target compounds
arylsulfonyloxy acetimidamides (3) was accomplished using
previously reported method.32 Following Scheme 1, they were
synthesized from different 4-substituted benzeneacetonitrailes
(1). Compound 1 was refluxed with hydroxylamine hydro-
chloride dissolved in methonolic sodium methoxide for about 12
Figure 4. Query structures used in acquiring the ZINC database entries
for virtual screening.
F
DOI: 10.1021/acs.jcim.5b00392
J. Chem. Inf. Model. XXXX, XXX, XXX−XXX

Journal of Chemical Information and Modeling
Article
Figure 5. Surflex docked poses of selected ZINC database entries in the target sites of Pf-CTPS, CK, and GST with dock scores in parentheses.
Scheme 1. Synthetic Scheme of Arylsulfonyloxy Acetimidamides
h to yield corresponding N′-hydroxy-2-arylacetimidamide (2).
This in dichloromethane in the presence of triethayl amine
on reaction with arylsulphonyl chloride at 0 °C afforded
corresponding arylsulfonyloxy acetimidamides (Table 3) in
good yield (70−85%). The compounds were purified by column
chromatography and the structures were confirmed by analyzing
their 1H and 13C NMR, mass, and HRMS spectra.
3.6. In Vitro Antimalarial Activity and Cytotoxicity. All
the compounds (Table 3) were tested for their antimalarial
activity against 3D7 (CQ-sensitive) and K1 (CQ-resistant)
strains of P. falciparum and for cytotoxicity in Vero cell line.
Among these, four compounds (3a, 3d, 3e, and 3h) showed
response against sensitive strain (IC50, 1.10−4.36 μM) as well as
resistant strain (IC50, 2.13−4.28 μM) (Table 3). From these,
compounds 3a and 3d showed IC50 values of 1.10 and 1.45 μM,
respectively, against sensitive strain. They also respectively
showed IC50 values of 2.13 and 3.10 μM against the resistant
strain. The remaining two compounds (3e and 3h) showed IC50
values of 3.02 and 4.36 μM against sensitive strain and 3.34 and
4.28 μM against resistant strain. For other compounds the IC50
values against one or both the strains is above 5.00 μM. Among
these, compound 3o showed activity only against sensitive strain
(IC50 = 1.69 μM) and compounds 3b and 3k showed activity
only against resistant strain (IC50, 2.13 and 4.51 μM). In Vero cell
line none of the compounds showed any cytotoxicity within the
prescribed limits of testing concentrations (10 times the IC50
values) (Table 3).
Furthermore, the structures of all synthesized compounds
were created in Sybyl-X 1.3 and screened in Surflex-Dock by
docking them in each target site following the same protocol as
implemented in case of ZINC entries. While the compound’s
response against 3D7 and K1 strains indicate its net in vitro
antimalarial activity, the dock scores for Pf-CTPS, CK, and GST
indicate its affinity for the respective targets. Since the
G
DOI: 10.1021/acs.jcim.5b00392
J. Chem. Inf. Model. XXXX, XXX, XXX−XXX

Journal of Chemical Information and Modeling
Article
Table 3. Arylsulfonyloxy Acetimidamides and Their in Vitro Antimalarial Activity, Cytotoxicity, and Surflex-Dock Scores for Pf-
CTPS, CK, and GST
a
IC50 (μM)
dock score
CK
b
compd
R1
p-Cl
R2
3D7
K1
CC50 (μM)
CTPS
GST
3a
3b
3c
3d
3e
3f
2,4,6-tri-iPr
p-CH3
p-NO2
2,4,6-tri-iPr
p-CH3
p-NO2
2,4,6-tri-iPr
p-iPr
1.10
>5.00
>5.00
1.45
2.10
2.13
57.20
34.00
157.50
>200
6.52
6.21
5.18
7.71
7.31
7.72
7.27
5.59
6.85
6.07
5.76
5.00
6.29
6.15
6.19
9.17
6.43
5.29
7.91
6.73
5.82
8.51
6.88
5.97
5.81
5.88
6.02
5.65
6.48
7.29
6.93
5.40
4.84
7.13
6.45
5.62
5.99
6.71
5.72
5.62
5.57
6.47
4.89
5.65
5.75
p-Cl
p-Cl
>5.00
3.10
p-OCH3
p-OCH3
p-OCH3
3,5-di-Cl
3,5-di-Cl
3,5-di-Cl
3,5-di-Cl
3,5-di-Cl
3,5-di-Cl
3,5-di-Cl
3,5-di-Cl
p-F
3.02
3.34
>200
>5.00
>5.00
4.36
>5.00
>5.00
4.28
>200
3g
3h
3i
117.40
54.08
138.70
>200
p-CH3
p-F
>5.00
>5.00
>5.00
>5.00
>5.00
>5.00
1.69
>5.00
>5.00
4.51
3j
3k
3l
p-I
>200
H
>5.00
>5.00
>5.00
>5.00
77.33
125.80
>200
3m
3n
3o
p-NO2
p-OCH3
p-CH3
83.98
a
b
3D7, chloroquine sensitive strain; K1, chloroquine resistant strain. Vero cell line.
Figure 6. Surflex docked poses of compound 3a in the target sites of (a) Pf-CTPS, (b) CK, and (c) GST.
antimalarial activity of these compounds is expected due to their
combined affinity to Pf-CTPS, CK, and GST, the cumulative
dock scores of these three targets is considered as an index of the
in vitro antimalarial activity. The in vitro antimalarial activities of
the compounds, excepting 3g, are in agreement with the
cumulative dock scores of the same. On excluding compound
3g, the correlation between IC50 of 3D7 (compounds with IC50 >
5.00 are considered as 5.00) and cumulative dock scores of the
compounds is 0.69 (n = 14, r2 = 0.69); in the case of IC50 of K1, it
is less but still stands at significant level (n = 14, r2 = 0.41). Even
though in terms of dock scores compound 3g appeared very
good, in biological test milieu the combination of R1 and R2
groups of this may not be appropriate due to other parameters.
Thus, the dock-scores of these compounds with Pf-CTPS, CK,
and GST are largely in sync with the observed antimalarial
activities (Table 3).
residues GLY384 and GLU441 through its NH and O of
SO3 moieties, respectively (Figure 6a). In the case of CK, the
compound’s O (of SO3) group formed H-bonds with
ASN292, TYR329, and SER330 and the NH group formed
H-bond with GLN290 (Figure 6b). In the target site of GST, the
O of SO3 and NH of compound 3a showed H-bond
interaction with VAL59 (Figure 6c). In spite of best efforts we
could not procure/find economical commercial source of
selected enzymes; therefore, the compounds could not be tested
against them.
3.7. Structure−Activity Relations. The structure−activity
profile of the compounds (Table 3) are examined by using the
hydrophobicity (π), molar refractivity (MR), and electronic
properties (polarity F, resonance R, and Hammett’s sigma
constant σp) of R1 and R2 substituent groups (data provided in
the Supporting Information).45 As the number of compounds
with definite activity are limited, the inferences are deduced by
comparing molecular pairs sharing common substituent features.
The structure−activity profile of these compounds (Table 3)
indicate that para-substitution on both the aryl moieties is critical
Since compound 3a (Table 3) showed best activity, its docking
interactions are analyzed further. Figure 6 shows the docked
poses (Surflex) of compound 3a in the target sites of Pf-CTPS,
CK, and GST. In Pf-CTPS, it has formed H-bonds with the
H
DOI: 10.1021/acs.jcim.5b00392
J. Chem. Inf. Model. XXXX, XXX, XXX−XXX

Journal of Chemical Information and Modeling
Article
Figure 7. Docking (AUTODOCK) poses of compound 3a in the target sites of (a) Pf-CTPS, (b) CK, and (c) GST, graphs of RMSDs (d−e), radii of
gyration (RG) (g−i), and nonbonded interaction energies (NBEs) (j−l) from the MD simulation profile of the Pf-CTPS/CK/GST-compound 3a
complex.
for the antimalarial activity against sensitive as well as resistant
strains. An examination of the physicochemical characteristics of
R1 and R2 groups of the compounds indicated a preference to
hydrophobicity (π) and/or molar refractivity (MR) for activity
against both the strains (compounds 3a and 3b vs 3c; 3d and 3e
vs 3f; 3h vs 3i, 3j, and 3l−3n). Furthermore, the properties (π
and/or MR) of R2 substituent(s) showed some degree of
cumulative influence on the activity (compounds 3a vs 3b; 3d vs
3e). Here, 3g and 3h are exceptions; it may be that the absence of
a para-substitution (R1) on the aryl ring altered the scenario. It is
also pointed out that the sensitive strain is more desirous of
hydrophobic groups when compared to that of the resistant
strain (compounds 3a vs 3b; 3d vs 3e). Compounds lacking π
and/or MR R1 and R2 groups did not show high order of activity
in either strain (compounds 3c, 3f, 3j, and 3l−3n). The
electronic substituent constants of R1 and R2 groups have not
shown a clear trend in inferring the activity profile of the
compounds. Nevertheless, this does not rule out their influence
on the antimalarial activity of the compounds.
3.8. MD Simulations. Sybyl’s Surflex and AUTODOCK are
two widely used tools for molecular docking in drug discovery
studies. Between these two AUTODOCK is very popular as it is
available in public domain and perceived to be very rigorous.
With this view we repeated the docking analysis of the most
active analogue 3a (Table 3) and taken it further for molecular
dynamics in NAMD. In AUTODOCK, the best docked pose of
3a resulted in −9.31, −10.88, and −8.65 kcal/mol as binding
energies with Pf-CTPS, CK, and GST respectively (Figure 7a−
c). Here the top ranked poses of 3a occupied the same locations
of the binding pockets as those from Surflex dock. Furthermore,
the docked poses of 3a in Pf-CTPS and GST target sites from
Surflex and AUTODOCK showed comparable orientations, but
the docked poses of 3a in CK target site showed some deviation
in their orientation (RMSDs between Surflex and AUTODOCK
docked poses of 3a in Pf-CTPS, CK and GST are 2.919, 4.587
and 2.861, respectively) (Figures 6 and 7; the superimposed
docked poses are provided in the Supporting Information). We
have no explanation for the deviation, but it is not unusual in
docking exercises. Nevertheless, as in case of Surflex, the docked
poses of 3a from AUTODOCK indicated that NH and NH of
acetimidamide, O of SO3 and aryl moieties of the compound
(3a) play key role in binding to the active sites of the enzymes
through one or more H-bonds and/or arene interactions. Some
key interactions observed in AUTODOCK poses are described
below.
In Pf-CTPS, compound 3a with its NH, NH (of
acetimidamide), and O of SO3 moieties anchored on to the
binding pocket residues GLY384, GLY387, and GLU441
through H-bonds (Figure 7a). The back bones of GLY384 and
GLY387 acted as H-bond acceptor and H-bond donor for NH
and O of SO3 moieties of the compound, respectively. The
GLU441 with its side chain acted as hydrogen bond acceptor for
the NH moiety of the compound. In CK, the compound used its
O of SO3 moiety for H-bonding with the side chains of
I
DOI: 10.1021/acs.jcim.5b00392
J. Chem. Inf. Model. XXXX, XXX, XXX−XXX

Journal of Chemical Information and Modeling
Article
GLN212 and ASN214 (Figure 7b). Apart from this, the
chlorobenzene moiety present in compound 3a showed arene
interactions with the indole moiety of TRP317 (Figure 7b). In
the docking experiment with GST, compound 3a showed H-
bond interactions with TYR9, VAL59, and TYR211. In this, the
side chains of TYR9 and TYR211 acted as hydrogen bond
donors for O of SO3 moiety, and the backbone of VAL59 acted
as hydrogen bond acceptor for NH (of acetimidamide) moiety of
the compound (Figure 7c).
systems. In a 10 ns MD simulation, the RMSDs of GST-
compound 3a trajectories indicated that while the protein
maintained comparable 3D conformation all through the
dynamics, the compound showed significant changes in its
conformation after 2.5 ns and remained stable thereafter (Figure
7f). Also, the radius of gyration indicated that all through the
dynamics the protein trajectories swirled close to each other;
whereas, that of the compound showed a clear shift within the
binding pocket (Figure 7i). The shift(s) in the compound
conformation(s) is also reflected in the nonbonded energies of
the system (Figure 7l). As the trajectories of the compound have
undergone conformational variations during the simulations, it is
also noticed that the compound forming H-bonds with different
pocket residues of the protein trajectories. In addition to the
prior mentioned interactions, it is observed that the compound’s
NH and O of SO3 moieties showing hydrogen bonds with
GLN58 and ASN111, respectively. Interestingly, in most of the
sampled trajectories of GST-3a, the compound anchored onto
the target site by H-bond with TYR9 and VAL59 through its O
of SO3 and/or NH functional groups. An examination of the
video of the trajectories of GST-compound 3a complex from
MD simulations showed that in the binding pocket the
chlorobenzene moiety of the compound flipped during later
part of dynamics and remained in that pose until the termination
of MD simulation (i.e., 10 ns). Conformational changes in the
docked ligands during MD simulations are known in the
literature.46 As the GST-compound 3a complex showed a fast
conformational change after 2.5 ns of MD simulation, a fresh
simulation of GST complexed with another docked pose of 3a
(second best docked pose from AUTODOCK) has been carried
out for 10 ns to examine the influence of starting pose on the
outcome of the results.47 For this MD simulation all the
parameters are kept same as in case of previous GST-3a run. We
noticed that even though the starting 3a conformations are
different in these experiments, during the course of MD
simulation, the conformation of 3a shifted toward a pose
comparable to that of one emerged from previous run. In the
second experiment also the chlorobenzene moiety of the
compound almost attained same orientation during later part
of dynamics (figures and data are in the Supporting
Information).
In multitarget approach primarily the drug should be
sufficiently tuned to bind to the pathogen’s designated enzymes,
and at the same time should avoid binding to human proteins. In
view of this, the relevance of prototype compound 3a to Pf-
CTPS, CK, and GST, in comparison to the matching human
proteins, is examined by BLAST search of the respective target
sequences (for similarity) and by comparing the docking scores
of the compound 3a against parasite and human targets. An
examination of relevant protein sequences of parasite and human
revealed that while the binding pocket residues of CTPS and CK
are conserved in them, in GST the same are less conserved
(<50%). Furthermore, in the BLAST search, the CTPS, CK, and
GST from parasite and human sources showed sequence identity
between 29 to 38%. These may hint at the dissimilarity of parasite
and human proteins. Interestingly, in Surflex-Dock compound 3a
showed higher affinity toward the parasite enzymes when
compared to that of human enzymes (Dock scores of Pf-
enzymes: CTPS, 6.52; CK, 9.17; GST, 6.93. Dock scores of
human enzymes: CTPS, 3.74; CK, 5.89; GST, 5.65). These are
very preliminary and in vivo studies give the final call on the
matter.
While docking studies provide static frames of protein−ligand
interactions, the trajectories from molecular dynamics (MD)
simulations give scope to investigate the protein−ligand
interactions and their energitics during the course of time.
Hence, the trajectories of enzyme (Pf-CTPS/CK/GST)-
compound 3a complexes from the molecular dynamics (MD)
simulations are analyzed to determine the quality of compound’s
binding and its affinity to the enzymes through their RMSDs
from the docked state (Figure 7d−f), radius of gyration (RG;
Figure 7g−i), nonbonded interaction energies (NBEs; Figure
7j−l), and intermolecular interactions. The RMSDs of Pf-CTPS-
compound 3a and CK-compound 3a complexes suggest that
during entire course of MD simulations, the protein and
compound trajectories remained close to each other and retained
similarity with their docked orientations (Figure 7d−e).
Furthermore, their nonbonded energies remained almost
uniform (Figure 7j−k) and RGs confirmed that these systems
followed a harmonious swirl throughout the dynamics (Figure
7g−h). For the sake of analysis and discussion of intermolecular
interactions, the poses of trajectories of enzyme-compound 3a
complexes were sampled at time intervals of 0.5, 2.0, 3.0, 5.0, 7.5,
and 10 ns. The timeline figures of these trajectory poses are
provided as part of the Supporting Information. They showed
that, in Pf-CTPS-3a and CK-3a trajectories, the compound has
maintained many of the prior observed H-bonds with these
enzymes.
All through the dynamics of Pf-CTPS-3a complex, GLU441
has acted as acceptor in anchoring 3a; GLY384 and GLY387 are
other important residues in anchoring 3a into the target site. In
most of the trajectories of Pf-CTPS-3a, the O of SO3 of the
compound has shown H-bond interactions with GLU441,
GLY385, and GLY387; the NH moiety of the compound also
formed H-bond with GLU441. Also, 2,4,6-tri-isopropylbenzene
moiety of the compound participated in arene interaction with
the PHE386. Apart from these H-bonds, in some other
trajectories, the NH moiety of compound 3a showed hydrogen
bond with the backbone of GLY385 of Pf-CTPS.
In the MD simulation of CK-3a complex, GLN212 is one
important residue acted as acceptor as well as donor in anchoring
3a. In the samples of CK-3a trajectories, the O of SO3 of the
compound showed H-bonds with GLN212 and GLU246. In
addition to this, GLN212 showed interaction with the NH
moiety of the compound. In some trajectories of CK, the side
chain of TYR251 participated in H-bond with NH and O of
SO3 moieties of compound 3a. Furthermore, the TRP314 (with
its indole moiety) and TYR251 (with its phenyl moiety) have
respectively provided arene interactions with the aryl moieties of
the compound. The videos of these MD simulations showed that
in Pf-CTPS and CK, compound 3a maintained the binding
orientation comparable to that of its docked poses in the
respective enzymes suggesting the durability and favorability of
its interactions with the binding pocket residues.
The picture that emerged from the MD simulation of GST-3a
complex is somewhat different from that of the Pf-CTPS and CK
J
DOI: 10.1021/acs.jcim.5b00392
J. Chem. Inf. Model. XXXX, XXX, XXX−XXX

Journal of Chemical Information and Modeling
Article
Notes
These findings collectively suggested that prototype com-
pound 3a with SO3, NH, and/or NH functionalities flanked
by aryl groups may be critical for binding to Pf-CTPS, CK, and
GST and suitable for the exploration/development as CCE
against the said enzymes. For these systems free energy
simulations48 would have provided more reliable estimates of
protein−ligand binding affinities and energetics of conforma-
tional changes. However, we could not implement them as they
are computationally expensive and beyond our reach. All the
same, this prototype compound is worth exploring and may
come to use in overcoming the clinical issues associated with
routinely used antimalarial drugs.
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
The authors thank Dr S.K. Puri, Scientist G, CSIR-CDRI, for the
helpful discussions. The authors thankfully acknowledge M/S
Gromada.com for the VideoMach software. S.V. and U.D.
thankfully acknowledge the financial support, in the form of
Senior Research Fellowships, of the Department of Science and
Technology (DST-INSPIRE), New Delhi, and the Council of
Scientific and Industrial Research, New Delhi, respectively. The
SAIF, CSIR-CDRI, is thankfully acknowledged for the analytical
and spectral data. This paper is dedicated to Dr. Arun K. Shaw,
Medicinal & Process Chemistry Division, CSIR-CDRI,
Lucknow, on his 60th birthday as a mark of respect for his
commitment to scientific quest. CDRI Communication No.
9039.
4. CONCLUSIONS
This study has explored a strategy to design new class of
antimalarials through identifying consensus inhibitor against
multiple enzyme targets. For this, Pf-CTPS, CK, and GST were
selected from a database of fifty-six drug targets of P. falciparum
based on their connectedness and functional importance in
biochemical/metabolic pathways. In the absence of 3D structure
of Pf-CTPS, it has been developed from the FASTA sequence by
employing protein structural modeling techniques. The target
sites of the selected enzymes were utilized in the virtual screening
of chemical structures from ZINC database to identify consensus
chemical entity to inhibit them. This has led to the design and
synthesis of arylsulfonyloxy acetimidamides as new class of
compounds against malaria. Some of these compounds showed
good activity against CQ-sensitive (3D7) and resistant (K1)
strains of P. falciparum. In docking studies these compounds
showed binding affinity toward the selected enzymes.
Furthermore, the molecular dynamics simulations carried out
on these enzymes complexed with compound 3a ascertained the
importance of SO3, NH, and/or NH functionalities flanked by
aryl groups in bind to Pf-CTPS, CK, and GST through one or
more H-bonds and/or arene interactions. The findings
collectively paved the way for arylsulfonyloxy acetimidamides
as new class of antimalarial agents. As new chemical scaffolds are
urgently needed and are in high demand for the clinical
management of malaria, this is a step toward opening a window
for the exploration of new antimalarial agents.
REFERENCES
■
(1) World malaria report 2014; World Health Organization: Geneva,
2014; pp 1−227; http://www.who.int/malaria/publications/world_
malaria_report_2014/wmr-2014-no-profiles.pdf.
(2) Flannery, E. L.; Chatterjee, A. K.; Winzeler, E. A. Antimalarial Drug
Discovery - Approaches and Progress Towards New Medicines. Nat.
Rev. Microbiol. 2013, 11, 849−862.
(3) Njogu, P. M.; Chibale, K. Recent Developments in Rationally
Designed Multitarget Antiprotozoan Agents. Curr. Med. Chem. 2013, 20,
1715−1742.
(4) Li, K.; Schurig-Briccio, L. A.; Feng, X.; Upadhyay, A.; Pujari, V.;
Lechartier, B.; Fontes, F. L.; Yang, H.; Rao, G.; Zhu, W.; Gulati, A.; No, J.
H.; Cintra, G.; Bogue, S.; Liu, Y. L.; Molohon, K.; Orlean, P.; Mitchell,
D. A.; Freitas-Junior, L.; Ren, F.; Sun, H.; Jiang, T.; Li, Y.; Guo, R. T.;
Cole, S. T.; Gennis, R. B.; Crick, D. C.; Oldfield, E. Multitarget Drug
Discovery for Tuberculosis and other Infectious Diseases. J. Med. Chem.
2014, 57, 3126−3139.
(5) White, N. J. Antimalarial Drug Resistance. J. Clin. Invest. 2004, 113,
1084−1092.
(6) Bloland, P. B. Drug Resistance in Malaria; World Health
Organization: Geneva, 2001; pp 1−27; http://www.who.int/csr/
resources/publications/drugresist/malaria.pdf.
(7) Gardner, M. J.; Hall, N.; Fung, E.; White, O.; Berriman, M.;
Hyman, R. W.; Carlton, J. M.; Pain, A.; Nelson, K. E.; Bowman, S.;
Paulsen, I. T.; James, K.; Eisen, J. A.; Rutherford, K.; Salzberg, S. L.;
Craig, A.; Kyes, S.; Chan, M. S.; Nene, V.; Shallom, S. J.; Suh, B.;
Peterson, J.; Angiuoli, S.; Pertea, M.; Allen, J.; Selengut, J.; Haft, D.;
Mather, M. W.; Vaidya, A. B.; Martin, D. M.; Fairlamb, A. H.; Fraunholz,
M. J.; Roos, D. S.; Ralph, S. A.; McFadden, G. I.; Cummings, L. M.;
Subramanian, G. M.; Mungall, C.; Venter, J. C.; Carucci, D. J.; Hoffman,
S. L.; Newbold, C.; Davis, R. W.; Fraser, C. M.; Barrell, B. Genome
Sequence of the Human Malaria Parasite Plasmodium Falciparum.
Nature 2002, 419, 498−511.
ASSOCIATED CONTENT
* Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.jcim.5b00392.
■
S
Further detail on work flow, druggable enzyme database,
selection of target enzymes, structural modeling, virtual
screening, synthesis, docking and dynamics, and refer-
ences (PDF)
(8) Hendriks, E. F.; O’Sullivan, W. J.; Stewart, T. S. Molecular Cloning
and Characterization of the Plasmodium Falciparum Cytidine
Triphosphate Synthetase Gene. Biochim. Biophys. Acta, Gene Struct.
Expression 1998, 1399, 213−218.
(9) Ancelin, M. L.; Vial, H. J. Quaternary Ammonium Compounds
Efficiently Inhibit Plasmodium Falciparum Growth in Vitro by
Impairment of Choline Transport. Antimicrob. Agents Chemother.
1986, 29, 814−820.
(10) Hiller, N.; Fritz-Wolf, K.; Deponte, M.; Wende, W.;
Zimmermann, H.; Becker, K. Plasmodium Falciparum Glutathione S-
transferase–Structural and Mechanistic Studies on Ligand Binding and
Enzyme Inhibition. Protein Sci. 2006, 15, 281−289.
(11) Schomburg, I.; Chang, A.; Placzek, S.; Sohngen, C.; Rother, M.;
Lang, M.; Munaretto, C.; Ulas, S.; Stelzer, M.; Grote, A.; Scheer, M.;
Schomburg, D. BRENDA in 2013: Integrated Reactions, Kinetic Data,
Full data corresponding to Tables S8 and S9 (PDF)
MD simulation video of Pf-CTPS-compound 3a complex
(AVI)
MD simulation video of CK-compound 3a complex (AVI)
MD simulation video of GTS-compound 3a complex
(AVI)
MD simulation video of GTS complexed with alternative
pose of compound 3a (AVI)
AUTHOR INFORMATION
Corresponding Author
*Phone: +91-522-2772450. Fax: +91-522-2771941. E-mail:
yenpra@yahoo.com.
■
K
DOI: 10.1021/acs.jcim.5b00392
J. Chem. Inf. Model. XXXX, XXX, XXX−XXX

Journal of Chemical Information and Modeling
Article
Enzyme Function Data, Improved Disease Classification: New Options
and Contents in BRENDA. Nucleic Acids Res. 2013, 41, D764−D772.
(12) Irwin, J. J.; Shoichet, B. K. ZINC–a Free Database of
Commercially Available Compounds for Virtual Screening. J. Chem.
Inf. Model. 2005, 45, 177−182.
(13) SYBYL, version 7.3; Tripos Associates, St. Louis, MO, 2006.
(14) Morris, G. M.; Huey, R.; Lindstrom, W.; Sanner, M. F.; Belew, R.
K.; Goodsell, D. S.; Olson, A. J. AutoDock4 and AutoDockTools4:
Automated Docking with Selective Receptor Flexibility. J. Comput.
Chem. 2009, 30, 2785−2791.
(15) Phillips, J. C.; Braun, R.; Wang, W.; Gumbart, J.; Tajkhorshid, E.;
Villa, E.; Chipot, C.; Skeel, R. D.; Kale, L.; Schulten, K. Scalable
Molecular Dynamics with NAMD. J. Comput. Chem. 2005, 26, 1781−
1802.
(16) Wilkins, M. R.; Gasteiger, E.; Bairoch, A.; Sanchez, J. C.; Williams,
K. L.; Appel, R. D.; Hochstrasser, D. F. Protein Identification and
Analysis Tools in the ExPASy Server. Methods Mol. Biol. 1999, 112,
531−552.
(17) Muller, S. A.; Kohajda, T.; Findeiss, S.; Stadler, P. F.; Washietl, S.;
Kellis, M.; von Bergen, M.; Kalkhof, S. Optimization of Parameters for
Coverage of low Molecular Weight Proteins. Anal. Bioanal. Chem. 2010,
398, 2867−2881.
(18) Altschul, S. F.; Gish, W.; Miller, W.; Myers, E. W.; Lipman, D. J.
Basic Local Alignment Search Tool. J. Mol. Biol. 1990, 215, 403−410.
(19) Kelley, L. A.; Sternberg, M. J. Protein Structure Prediction on the
Web: a Case Study Using the Phyre Server. Nat. Protoc. 2009, 4, 363−
371.
(20) Cole, C.; Barber, J. D.; Barton, G. J. The Jpred 3 Secondary
Structure Prediction Server. Nucleic Acids Res. 2008, 36, W197−W201.
(21) Webb, B.; Sali, A. Comparative Protein Structure Modeling Using
MODELLER. Curr. Protoc. Bioinformatics 2014, 47, 5.6.1−5.6.32.
(22) Linding, R.; Jensen, L. J.; Diella, F.; Bork, P.; Gibson, T. J.; Russell,
R. B. Protein Disorder Prediction: Implications for Structural
Proteomics. Structure 2003, 11, 1453−1459.
(23) Marchler-Bauer, A.; Derbyshire, M. K.; Gonzales, N. R.; Lu, S.;
Chitsaz, F.; Geer, L. Y.; Geer, R. C.; He, J.; Gwadz, M.; Hurwitz, D. I.;
Lanczycki, C. J.; Lu, F.; Marchler, G. H.; Song, J. S.; Thanki, N.; Wang,
Z.; Yamashita, R. A.; Zhang, D.; Zheng, C.; Bryant, S. H. CDD: NCBI’s
Conserved Domain Database. Nucleic Acids Res. 2015, 43, D222−D226.
(24) Hall, T. A. BioEdit: a User-Friendly Biological Sequence
Alignment Editor and Analysis Program for Windows 95/98/NT.
Nucl. Acids. Symp. Ser. 1999, 41, 95−98.
Medium for Drug-Sensitivity Studies using SYBR Green I Assay. Exp.
Parasitol. 2011, 127, 318−3121.
(34) Sharma, M.; Chauhan, K.; Chauhan, S. S.; Kumar, A.; Singh, S. V.;
Saxena, J. K.; Agarwal, A.; Srivastava, K.; Kumar, S. R.; Puri, S. K.; Shah,
P.; Siddiqi, M. I.; Chauhan, P. M. S. Synthesis of Hybrid 4-
Anilinoquinoline Triazine as Potent Antimalarial Agents, their in Silico
Modeling and Bioevaluation as Plasmodium Falciparum Transketolase
and β-Hematin Inhibitors. Med. MedChemComm 2012, 3, 71−79.
(35) Lauritsen, I.; Willemoes, M.; Jensen, K. F.; Johansson, E.; Harris,
P. Structure of the Dimeric form of CTP Synthase from Sulfolobus
Solfataricus. Acta Crystallogr., Sect. F: Struct. Biol. Cryst. Commun. 2011,
67, 201−208.
(36) Goto, M.; Omi, R.; Nakagawa, N.; Miyahara, I.; Hirotsu, K.
Crystal Structures of CTP Synthetase Reveal ATP, UTP, and Glutamine
Binding Sites. Structure 2004, 12, 1413−23.
(37) Binkowski, T. A.; Naghibzadeh, S.; Liang, J. CASTp: Computed
Atlas of Surface Topography of Proteins. Nucleic Acids Res. 2003, 31,
3352−3355.
(38) MOE; The Molecular Operating Environment from Chemical
Computing Group Inc., Montreal, Quebec, Canada; http://www.
chemcomp.com.
(39) Wernimont, A. K.; Pizarro, J. C.; Artz, J. D.; Amaya, M. F.; Xiao,
T.; Lew, J.; Wasney, G.; Senesterra, G.; Kozieradzki, I.; Cossar, D.;
Vedadi, M.; Schapira, M.; Bochkarev, A.; Arrowsmith, C. H.; Bountra,
C.; Weigelt, J.; Edwards, A. M.; Hui, R.; Hills, T. Crystal Structure of
Choline Kinase from Plasmodium Falciparum, PF14_0020. PDB; DOI:
10.2210/pdb3fi8/pdb.
(40) Hiller, N.; Fritz-Wolf, K.; Deponte, M.; Wende, W.;
Zimmermann, H.; Becker, K. Plasmodium Falciparum Glutathione S-
Transferase–Structural and Mechanistic Studies on Ligand Binding and
Enzyme Inhibition. Protein Sci. 2006, 15, 281−289.
(41) Hofer, A.; Steverding, D.; Chabes, A.; Brun, R.; Thelander, L.
Trypanosoma Brucei CTP synthetase: a Target for the Treatment of
African Sleeping Sickness. Proc. Natl. Acad. Sci. U. S. A. 2001, 98, 6412−
6416.
(42) Torres-Rivera, A.; Landa, A. Glutathione Transferases from
Parasites: a Biochemical View. Acta Trop. 2008, 105, 99−112.
(43) Choubey, V.; Maity, P.; Guha, M.; Kumar, S.; Srivastava, K.; Puri,
S. K.; Bandyopadhyay, U. Inhibition of Plasmodium Falciparum Choline
Kinase by Hexadecyltrimethylammonium Bromide: a Possible Anti-
malarial Mechanism. Antimicrob. Agents Chemother. 2007, 51, 696−706.
(44) Alberge, B.; Gannoun-Zaki, L.; Bascunana, C.; Tran van Ba, C.;
Vial, H.; Cerdan, R. Comparison of the Cellular and Biochemical
Properties of Plasmodium falciparum Choline and Ethanolamine
Kinases. Biochem. J. 2010, 425, 149−158.
(25) Laskowski, R. A.; MacArthur, M. W.; Moss, D. S.; Thornton, J. M.
PROCHECK: a Program to Check the Stereochemical Quality of
Protein Structures. J. Appl. Crystallogr. 1993, 26, 283−291.
(26) Wiederstein, M.; Sippl, M. J. ProSA-Web: Interactive Web Service
for the Recognition of Errors in Three-Dimensional Structures of
Proteins. Nucleic Acids Res. 2007, 35, W407−10.
(27) Schuler, L. D.; Daura, X.; van Gunsteren, W. F. An Improved
GROMOS96 Force Field for Aliphatic Hydrocarbons in the Condensed
Phase. J. Comput. Chem. 2001, 22, 1205−1218.
(28) Guex, N.; Peitsch, M. C.; Schwede, T. Automated Comparative
Protein Structure Modeling with SWISS-MODEL and Swiss-
PdbViewer: a Historical Perspective. Electrophoresis 2009, 30, S162−
S173.
(29) Humphrey, W.; Dalke, A.; Schulten, K. VMD: Visual Molecular
Dynamics. J. Mol. Graphics 1996, 14, 33−38.
(45) Hansch, C.; Leo, A. In Substituent Constants for Correlation
Analysis in Chemistry and Biology; Wiley: New York, 1979; Chapter 6, pp
48−63.
(46) Shan, Y.; Kim, E. T.; Eastwood, M. P.; Dror, R. O.; Seeliger, M. A.;
Shaw, D. E. How does a Drug Molecule find its Target Binding Site? J.
Am. Chem. Soc. 2011, 133, 9181−9183.
(47) Stjernschantz, E.; Oostenbrink, C. Improved Ligand-Protein
Binding Affinity Predictions using Multiple Binding Modes. Biophys. J.
2010, 98, 2682−2691.
(48) Genheden, S.; Ryde, U. Improving the Efficiency of Protein−
Ligand Binding Free-Energy Calculations by System Truncation. J.
Chem. Theory Comput. 2012, 8, 1449−1458.
(30) Pedretti, A.; Villa, L.; Vistoli, G. VEGA–an Open Platform to
Develop Chemo-Bio-informatics Applications, Using Plug-in Architec-
ture and Script Programming. J. Comput.-Aided Mol. Des. 2004, 18, 167−
173.
(31) Video match. http://gromada.com/main/download.php (ac-
cessed June 16, 2015).
(32) Lin, C. C.; Hsieh, T. H.; Liao, P. Y.; Liao, Z. Y.; Chang, C. W.;
Shih, Y. C.; Yeh, W. H.; Chien, T. C. Practical Synthesis of N-substituted
Cyanamides via Tiemann Rearrangement of Amidoximes. Org. Lett.
2014, 16, 892−895.
(33) Singh, S.; Srivastava, R. K.; Srivastava, M.; Puri, S. K.; Srivastava, K.
In-Vitro Culture of Plasmodium Falciparum: Utility of Modified (RPNI)
L
DOI: 10.1021/acs.jcim.5b00392
J. Chem. Inf. Model. XXXX, XXX, XXX−XXX