Possible anticancer agents: synthesis, pharmacological activity, and molecular modeling studies on some 5-N -Substituted-2-N-(substituted benzenesulphonyl)-L(+)Glutamines

Article abstract of DOI:10.1007/s00044-017-1858-1

On the basis of our earlier work, fortyone 5-N-substituted-2N-(substituted benzenesulphonyl)-L(+)glutamines were synthesized and screened for cancer cell inhibitory activity. The best active compounds showed 91% tumor cell inhibition, whereas other three compounds showed more than 80% inhibition. Two-dimensional quantitative structure–activity relationship modeling and three-dimensional quantitative structure–activity relationship k-nearest neighbor molecular field analysis studies were done to get an insight into structural requirements toward further improved anticancer activity. Considering the fact that these compounds are competitive inhibitors of glutaminase, a molecular docking study followed by molecular dynamic simulation analysis were performed. The work may help to develop new anticancer agents.

Full text of DOI:10.1007/s00044-017-1858-1

MEDICINAL
CHEMISTRY
RESEARCH
Med Chem Res
DOI 10.1007/s00044-017-1858-1
ORIGINAL RESEARCH
Possible anticancer agents: synthesis, pharmacological activity,
and molecular modeling studies on some 5-N-Substituted-2-N-
(substituted benzenesulphonyl)-L(+)Glutamines
Tarun Jha1 Soumya Basu1 Amit Kumar Halder1 Nilanjan Adhikari1
Soma Samanta1
●
●
●
●
Received: 3 August 2015 / Accepted: 2 March 2017
© Springer Science+Business Media New York 2017
Abstract On the basis of our earlier work, fortyone 5-N-
substituted-2N-(substituted benzenesulphonyl)-L(+)gluta-
mines were synthesized and screened for cancer cell inhi-
bitory activity. The best active compounds showed 91%
tumor cell inhibition, whereas other three compounds
showed more than 80% inhibition. Two-dimensional
quantitative structure–activity relationship modeling and
three-dimensional quantitative structure–activity relation-
ship k-nearest neighbor molecular field analysis studies
were done to get an insight into structural requirements
toward further improved anticancer activity. Considering
the fact that these compounds are competitive inhibitors of
glutaminase, a molecular docking study followed by
molecular dynamic simulation analysis were performed.
The work may help to develop new anticancer agents.
Introduction
Due to the utilization in malignant cells both for an
anaplerotic compound and for its reductive power
(DeBerardinis et al. 2008a, b), glutamine is found to be a
crucial nutrient for the rapidly growing cells (Newsholme
et al. 2003; Turner and McGivan 2003; Martin et al. 1998;
Wasa et al. 1996; Matsuno and Hirai 1989; Bhattacharya
and Maity 2000). Glutamine plays an important role in
tumor cell growth by supplying its amide nitrogen atom.
It is observed in the biosynthesis of other amino acids,
purine, pyrimidine, amino sugar and coenzymes (Zalkin and
Smith 1998) with versatile mechanisms (Massiere and
Badet-Denisot 1998). Glutamine provides multiple con-
tributions by participating in protein, purine, and pyrimidine
metabolisms. Almost one third of the circulating amino
acids and nitrogen are transported by glutamine. It was also
found to be a principal carrier of nitrogen from the skeletal
muscles to the visceral organs (Klimberg and McClellan
1996). The levels of glutamine are undetectable in malig-
nant strains of Earlich Ascite tumor cells (EATC) and ascite
fluids (Marquez et al. 1989). Nevertheless, several reports
revealed that glutamine transporter ASTC2 is overexpressed
in gliomas, colorectal carcinoma, hepatoma, and neurobal-
stoma, and hence, a high glutamine uptake is found in
different cancer cells (Huang et al. 2014; Xie et al. 2014;
Hassanein et al. 2013; Ogura et al. 2011; Dolinska et al.
2003; Wasa et al. 2002; Witte et al. 2002). Moreover,
glutamine is also found to modulate gene expression on
different tissues. It triggers antiapoptotic Bcl-2 protein and
helps to inhibit proapoptotic CD95 which, in turn, pro-
tecting these cells from apoptosis and thus, aids in malig-
nancy (Chang et al. 2002). Furthermore, the higher level of
glutamine catabolism by glutaminase enzyme has been
proved to be a crucial feature of tumors in different origins.
●
●
Keywords Anticancer agent Glutamine analog 2D-
●
●
●
QSAR 3D-QSAR Docking MD simulation
Electronic supplementary material The online version of this article
(doi:10.1007/s00044-017-1858-1) contains supplementary material,
which is available to authorized users.
* Tarun Jha
tjupharm@yahoo.com
1
Natural Science Laboratory, Division of Medicinal and
Pharmaceutical Chemistry, Department of Pharmaceutical
Technology, Jadavpur University, P.O. Box-17020, Kolkata
700032, India

Med Chem Res
During the exponential phase of cellular growth in several
neoplasms, phosphate dependent kidney type glutaminase
(KGA) activity was found to be the highest (Lu et al. 2010;
Szeliga and Obara-Michlewska 2009). In rapidly pro-
liferating cells of rat and human hepatomas, EATCs, human
breast cancer cells and in human leukemias, KGA isoform
is found to be highly overexpressed. Not only that, inhibi-
tion of KGA in EATCs activated apoptosis and sensitized
these cells to hydrogen peroxide and methotrexate toxicity.
Thus, inhibition of enzyme glutaminase is a potential target
for anticancer therapy (Szeliga and Obara-Michlewska
2009). Therefore, it may be hypothesized that compounds
which are able to reduce the uptake and utilization of glu-
tamine and also may block or inhibit glutaminase in
malignant cells, may elicit considerable anticancer poten-
cies. In this article, we report some new 5-N-substituted-2-
N-(substituted benzenesulphonyl)-L(+) glutamines as pos-
sible anticancer agents and also to explore theoretically the
effect of these compounds on glutaminase enzyme. The
general structure of 5-N-substituted-2-N-(substituted ben-
zenesulphonyl)-L(+) glutamines is shown in Fig. 1.
This work is done as a part of our composite program of
rational drug design and development (Adhikari et al. 2015;
Kumar et al. 2015; Mondal et al. 2015; Das et al. 2015a, b;
Hazra et al. 2015; Chatterjee et al. 2015; Chakraborty et al.
2015; Halder et al. 2015; Adhikari et al. 2014; Mondal et al.
2014; Chakraborty et al. 2014; Kumar et al. 2014; Chat-
terjee et al. 2014; Das et al. 2014; Adhikari et al. 2013a, b;
Halder et al. 2013a, b; Mondal et al. 2013; Adhikari et al.
2012). The newly synthesized compounds were character-
ized and biologically evaluated by in vivo method con-
sidering the percent tumor cell inhibition (%TCI) as
the biological activity parameter. To understand crucial
structural and physicochemical factors responsible for
the antitumor activities, two-dimensional quantitative
structure–activity relationship (2D-QSAR) and three-
dimensional quantitative structure–activity relationship
(3D-QSAR) studies were performed. It was assumed that
glutaminase is one of the targets of this series of com-
pounds. Therefore, molecular docking and molecular
dynamic simulation-based theoretical approaches were
taken to understand the possible interactions of the best
active compounds with the enzyme. In this attempt, the
X-ray crystallographic structure of human glutaminase was
docked with the best active compound and the ligand
receptor interactions were reported. The stability of the
docked ligand–receptor complex was analyzed by mole-
cular dynamics (MD) simulation.
Materials and methods
Chemistry
The preparations of the title compounds were carried out in
four steps. The work started with the chlorosulphonylation
of benzene derivatives (1–7) to obtain corresponding sub-
stituted benzenesulphonyl chlorides (8–14) (Huntress and
Carten 1940). These benzenesulphonyl chlorides (8–14)
were separately condensed with L(+) glutamic acid to get
2-N-(substituted benzenesulphonyl)–L(+) glutamic acids
(15–21). In this condensation, the reaction medium was
maintained alkaline by 2 N NaOH solution so that the
hydrochloric acid formed during the reaction was removed.
Cyclization of the resulting diacids (15–21) with acetyl
chloride produced 1-N-(substituted benzenesulphonyl)-5-
oxopyrrolidine-2-carboxylic acids (22–28). Aminolysis of
these monoacids (22–28) with different amines resulted in
the corresponding glutamine analogs (29–69). The route of
synthesis is presented in Fig. 2.
Chemical characterization of all these compounds
(29–69) was done by both analytical and spectrometric
methods. Melting points of these compounds (29–69) were
measured and verified by Mel-Temp Electrothermal
apparatus-a capillary melting point apparatus and CTRO-
NICS—a digital melting point apparatus. Structures of these
synthesized final compounds were confirmed by infrared
(IR)-, nuclear magnetic resonance (NMR)-, and mass
spectroscopy as well as by the elemental analysis. All final
compounds as well as L(+)-glutamic acid showed optical
activity when observed in a polarimeter. IR spectra of these
synthesized compounds (29–69) were recorded on SHI-
MADZU FTIR—8400 S Model by using KBr pellets. 1H
NMR spectra were recorded at 25 °C in the pulsed Fourier
Transformation mode on Bruker DRX 300 MHz NMR
spectrometer. Chemical shifts are recorded in δ ppm (parts
per million) in comparison with tetramethylsilane (Me4Si)
as the internal standard for solutions in deuterated dime-
thylsulfoxide (DMSO-d6). Splitting patterns are represented
as s (singlet), d (doublet) and m (multiplet). Positions of
hydrogen described in 1H NMR interpretation are as per
the general structure (Fig. 1) and substitutions at the R4
position are designated as the superscript “″” (double dash).
All these reaction steps were monitored by the analytical
O
R1
O
6'
H
N
2
4
2'
3
5
S
NHR4
1'
3'
1
O
O
OH
4'
R2
5'
R3
Fig. 1 General structure 5-N-substituted-2-N-(substituted benzene-
sulphonyl)-L(+) glutamines

Med Chem Res
O
R1
R1
O
O
R1
Cl
H2N
O
O
H
N
OH
S
S
OH
OHSO2Cl
0-5 0C
OH
2(N) NaOH
O
O
O
OH
R2
R2
R2
(15-21)
R3
R3
CH3COCl
R3
(8-14)
(1-7)
Cyclization
O
O
R1
O
H
N
R1
O
N
1. R4NH2
Amination
S
NHR4
S
O
O
OH
O
OH
2. Deamination
with 6(N) HCl
O
R2
R2
R3
(29-69)
(22-28)
R3
Fig. 2 General synthetic procedures of 5-N-substituted-2-N-(substituted benzenesulphonyl)-L(+) glutamines (29–69)
thin layer chromatography done on silica gel G plates. The
spots were identified by keeping the thin layer chromto-
graphic plates in iodine chamber.
Molecular modeling studies
The synthesized compounds were subjected to molecular
modeling study by different methods such as 2D QSAR,
k-nearest neighborhood-molecular field analysis (kNN-
MFA) 3D QSAR, molecular docking and MD simulation
analyses.
Pharmacological activity
Biological evaluation of all these synthesized final com-
pounds (29–69) was performed for their anticancer activ-
ities against Erlich ascites carcinoma (EAC) cells in Swiss
Albino mice (Samanta et al. 2004). Female Swiss Albino
mice of 10 weeks old with an average body weight of
18–20 g. were used for in vivo biological evaluation. Two
groups of Swiss Albino mice each containing six healthy
female mice (one test and the other control) were considered
for the biological activity. All mice were kept in ambient
room temp (20–27 °C) with basal metabolic diet and water
ad libitum during these experiments. These final compounds
(29–69) were dissolved separately into sterile phosphate
buffer saline for the biological screening. Intraperitonially,
these compounds (29–69) were administered individually
after 24 h incubation of 2 × 106 EAC cells to each mouse.
The total number of tumor cells was counted under a
microscope with the help of a haemocytometer (Marienfeld,
Germany). The mean of Ascites cells/ml was considered for
evaluation of anticancer activity. The %TCI of these com-
pounds (29–69) was determined individually. Mitomycin C,
Azaserin, and DON in sterile phosphate buffers (pH 7.2)
were utilized as standard drugs which showed 100% inhi-
bitions. Mitomycin C was considered as the universal
standard and azaserin as well as DON were chosen as the
specific standard drugs to compare the activity of the test
compounds.
2D-QSAR study
2D-QSAR was performed on 41 synthesized compounds
(29–69) to understand their structural requirements for the
higher anticancer activity.
Biological activity and descriptors
As far as the 2D-QSAR analysis is concerned, the percen-
tage tumor cell inhibition (%TCI) of these fortyone (41)
synthesized compounds (29–69) was considered. These %
TCI values were subsequently converted into the logarith-
mic scale [Log (%TCI)] and used as the dependent para-
meter. For a reliable and predictive 2D-QSAR modeling,
selection of descriptors is the crucial criteria. As the bio-
logical activity and structural variations of these compounds
(29–69) were comparatively less, keeping it in mind, atom-
based descriptors as well as simple whole molecular
descriptors were given importance for understanding the
structural requirements for potential anticancer activity. The
calculation of the atom-based quantum chemical descriptors
was performed on (Chem 3D Pro version 5.0 and Chem
Draw Ultra version 5.0 are programs of Cambridge Soft
Corporation, USA) Chem 3D Pro package (CambridgeSoft

Med Chem Res
18
O
2004) and k-means cluster analysis (Tropsha 2003; Snede-
cor and Cochran 1967) techniques were used for this pur-
pose. In the Y-based ranking method, molecules were first
ranked from the higher to the lower values of the observed
biological activity. Then molecules in the 4th, 8th, 12th, and
so on rows were collected and treated as the test set. The k-
means cluster analysis separated and arranged different
objects into groups according to their Euclidian distances in
multidimensional space depending on the biological activity
as well as those descriptors. From these clusters, the test set
compounds were selected randomly.
8
R1
O
H
N
19
11
16
1
15
17
10
S
7
NHR4
6
2
12
O
9
O
13
OH
14
5
3
R2
4
R3
Fig. 3 Arbitrary numbering of 5-N-substituted-2-N-(substituted ben-
zenesulphonyl)-L(+)glutamines (29–69)
Corporation, U.S.A.) and (Hyperchem Professional Release
7.0 is a computer program developed by Hypercube Inc.,
Gainesville, Florida) Hyperchem Release 7.0 Pro Package
(Hypercube Inc., Gainesville, Florida). Molecules were
numbered arbitrarily (shown in Fig. 3) keeping the serial
number of atoms same in all molecules for atom-based
descriptor calculations.
Multiple regression analysis and variable selection
Multiple stepwise regression analysis was used for the
selection of descriptors (Adhikari et al. 2015; Mondal et al.
2015; Mondal et al. 2014; Adhikari et al. 2014, 2013a, b;
Mondal et al. 2013; Adhikari et al. 2012) by utilizing F
value as the stepping criteria (F = 3.0 for inclusion, F = 2.9
for exclusion). Intercorrelated independent parameters
(correlation coefficient > 0.50) were not considered and
discarded during model development through stepwise
regression analysis. Regression equations were evaluated by
The energy minimization of these structures was done
individually using molecular mechanical (MM+) force
fields available in both of these softwares without cut-off for
non-bonded interactions, solvation, and constrains. In Chem
3D Pro. Package software, further minimizations and
charge calculations were performed under MOPAC module
according to semi empirical Austin Model 1 (AM1) method
(Leach 2001) using restricted Hartee-Fock (RHF): closed
shell wave function whereas in Hyperchem software, these
energy minimized structures were optimized geometrically
by semi empirical AM1 method (Leach 2001) using the
Polak-Ribiere algorithm with a RMS gradient of 0.1 kcal/Ǻ
mol. Wang-Ford charges of these atoms were calculated
from the energy minimized geometry of these compounds
by Chem 3D Pro package software. Frontier electron den-
sities of these atoms were calculated by Hyperchem Release
7.0 Pro Package software. Simple whole molecular
descriptors such as constitutional, atom-based fragments,
geometrical, empirical descriptors were calculated by
(DRAGON web version 2.1 is a QSAR software developed
by Milano Chemometrics and QSAR Research Group,
Dipartimento di Scienze dell’Ambiente e del Territorio
Universitàdegli Studi di Milano—Bicocca) Dragon Soft-
ware (Dragon 2.1 web version, Milano Bicocca). The
Verloop STERIMOL parameters like L, B1 and B5 were
collected from the literature (Krogsgaard–Larsen et al.
2002).
correlation coefficient (R), adjusted R2 (R2 ), variance ratio
a
(F) at specified degrees of freedom (df), probability factor
related to F ratio (p), standard error of estimate (SEE).
Significant level of regression coefficient and intercepts of
all equations were determined by using t-statistics and p-
values of the corresponding parameters.
Validation of the QSAR models
As far as the validation of the developed model was con-
cerned, leave-one-out (LOO) cross-validation method
(Tetko et al. 2001) was used to validate models. Regarding
the internal predictability parameters of the developed
model, those were justified by predicted residual sum of
squares (PRESS), cross-validated R2 (R2CV), standard
deviation error of prediction (SDEP) and standard error of
PRESS (SPRESS) whereas the external predictabilities of
these models were judged by R2
values (Golbraikh and
Pred
Tropsha 2002). The R2CV and the R2Pred value are calculated
according to the following equations:
ꢀ
ꢁ
P
2
Y
ObsðTrainingÞ ꢀ YPredðTrainingÞ
YObsðTrainingÞ ꢀ Ymean
R2CV ¼ 1 ꢀ
ð1Þ
ð2Þ
ꢀ
ꢁ
P
2
ꢀ
ꢁ
P
2
Division into training and test sets
Y
ObsðTestÞ ꢀ YPredðTrainingÞ
YObsðTestÞ ꢀ Ymean
R2Pred ¼ 1 ꢀ
ꢀ
ꢁ
P
2
Proper splitting of dataset into a test set and a training set is
an essential criterion for unbiased predictive model devel-
opment. In the current 2D-QSAR approach, two splitting
methods, the Y-based ranking method (Hemmateenejad
In the Eq. (1), the Ymean is the mean activity of the
training set while the YObs(training) and the Ypred(training) are
the experimental and predicted activity values of the

Med Chem Res
training set compounds respectively. A high R2
value
Alignment of molecules
CV
(R2CV > 0.50) indicates a good internal predictive ability of
a model. In eq. (2) YPred(Test) and YObs(Test) are the predicted
and the observed activity values of the test set compounds
respectively and the Ymean is the mean activity of the
The 3D structures of these compounds were drawn using
the “draw molecule” function in VLife QSAR Plus 1.0
(VLife Science and Technologies, Pune, India). The
MMFF94 force field and Gasteiger-Marsili charges fol-
lowed by AM1 (Leach 2001) Hamiltonian method were
used for the energy minimization of individual molecules as
well as for the entire series with the convergence criterion of
0.001 kcal/mol Ǻ. The location of each atom is vital for
kNN-MFA study as the descriptor calculation is based on
the 3D-space grid. Therefore, the mode to find out the
conformation of each molecule and the method to align
molecules together are two sensitive conditions to make a
reasonable model. The lowest energy of the most active
molecule of the dataset (compound 59) was selected as the
reference molecule to fit the training set and the test set
compounds by using the align molecules function available
with the software. The resulting aligned molecules were
imported for building kNN-MFA 3D-QSAR models.
training set. The recommended value of R2
should be
Pred
greater than 0.50 for successful predictability (Golbraikh
and Tropsha 2002). The root mean square (rm2) value (Roy
and Roy 2008) were also taken into consideration as an
2
external predictability parameter. The recommended rm
value should be more than 0.50. The rm2 is calculated as per
the following equation:
h
i
ꢀ
ꢁ
1=2
rm ¼ r2 ꢁ 1 ꢀ r2 ꢀ r02
ð3Þ
2
In the Eq. (3), r2 is the correlation coefficient between the
2
observed and the predicted activity whereas r0 is the cor-
relation coefficient between the observed and the predicted
activity passing through the origin. Applicability domain of
a model is determined by the extent of extrapolation that
considers the leverage values of compounds both in the test
set and the training set in the developed model. Prediction is
considered unreliable for compounds of leverage value
greater than 3p/n where p is the number of model variables
plus one and n is the number of the objects used to calculate
the model (Tropsha et al. 2003). As stepwise regression
procedure is susceptible to chance correlation, randomiza-
tion test is performed where values of observed variance are
scrambled randomly and repeatedly to generate the 2D-
QSAR models. Resulting scores are compared with that
of original 2D-QSAR models (generated from non-
randomized activity values). If the original 2D-QSAR
models are statistically significant, their regression coeffi-
cient (R) should be significantly greater than that of per-
muted data (Deswal and Roy 2006). In the current study, for
95% confidence interval, 19 trials were made (Nargotra
et al. 2009) and average scrambled values were named as
R2SCR. The best equation was selected by good internal
predictability (justified by Q2) and external predictability
Dataset and sphere exclusion principle
Since a dataset of 41 compounds is a small dataset for kNN-
MFA 3D-QSAR study, earlier reported 53 compounds
(Samanta et al. 2004) are included in the dataset along
with this 41 compounds to get a meaningful 3D-QSAR
study. Selection of the training and the test set for kNN-
MFA 3D-QSAR model was done by considering the fact
that the test set compounds should express diversity in
structures and a range of biological activities similar to the
training set. The whole dataset was divided into the training
set and the test set using sphere exclusion algorithm. This
algorithm allows constructing the training sets covering all
descriptor space areas occupied by representative points
based on dissimilarity value. By setting dissimilarity value
of 6.5, total 16 compounds were isolated from the dataset
and treated as the test set. The remaining 78 compounds
were used as the training set on which the model was
developed.
(justified by R2
and rm2) for the test set-training set
Pred
combinations. An average Q2 value of 0.60 and R2
and
Pred
2
rm value of 0.50 were set as the selection criteria. The
recommended ratio of the number of predictor parameters to
number of data point of 1:5 was maintained in this work
(Walker et al. 2003; Eriksson et al. 2003).
Descriptor generation
To develop the kNN-MFA descriptor fields, a 3D cubic
lattice with grid spacing of 2 Å in x, y, and z dimensions
were formed to encompass these aligned molecules. The
kNN-MFA descriptors were calculated to generate steric
energies and electrostatic fields with default cut off energy
of 10 and 30 kcal/mol respectively with an sp3 carbon
probe atom having a Van der waals radius of 1.52 Ǻ and a
charge of +1.0. Gasteiger-Marsili charge was set for
steric, electrostatic and hydrophobic energy calculations.
The generated kNN-MFA steric and electrostatic fields were
kNN-MFA 3D-QSAR study
The kNN-MFA is a popular 3D-QSAR technique and it was
utilized for current 3D-QSAR molecular modeling study.
The kNN-MFA 3D-QSAR analysis was done by using
VLife QSAR Plus 1.0 (VLife Sciences and Technologies,
India).

Med Chem Res
scaled by the standard method in the software VLife
QSAR Plus 1.0 (VLife Sciences and Technologies, Pune,
India).
hydrogen bonding with the carboxyl group of the bound
glutamic acid. Glutamic acid forms hydrophobic interac-
tions with Val484, Ser 286, and Tyr249.
Evaluation of models
Docking procedure
The statistical analysis of kNN field was done by the
distance-based weighted average and the predictive value of
the model was evaluated by standard LOO cross validation
method (Tetko et al. 2001). The cross-validated correlation
coefficient served as a measure of the quality of the model.
The predictive r2 calculation was based on molecules in the
test set and was used to estimate validated r2 by leave-one-
out method (q2), cross-validated standard error (q2_se),
predicted r2 for external test set (pred_r2) and standard error
for predicted r2 (pred_r2se).
GOLD v5.0.1 (Astex Technology, Chembridge, UK, 2011)
was used to predict the binding of flexible molecules
to the protein binding site. Generic algorithm incorporated
into GOLD v5.0.1 software was used for the ligand-protein
docking that allows full ligand and partial protein flexibility
(Jones et al. 1997). As far as the docking is concerned,
the ligand binding site was defined as a 8 Ǻ radius
from bound ligand and used as an input for GOLD calcu-
lations. Default GOLD fitness functions and default evo-
lutionary parameters such as population size = 100;
selection pressure = 1.1, operations = 1,00,000, island = 2,
niche size = 2, migration = 10, mutation = 95, crossover =
95 were used during docking calculations. Ten docking runs
were performed per structure unless three of the 10 poses
were within 1.5 Ǻ RMSD of each other. For understanding
the interaction ability of these compounds, Goldscore was
considered. The more the GOLD fitness score, the better is
the binding ability of the ligand.
Molecular docking study
As these compounds are analogs of glutamine, it is assumed
that this series of compounds may probably act through the
competitive inhibition of glutaminase enzyme. Based on the
assumption, the best active compound (compound 59) was
docked against the human kidney type glutaminase enzyme
(KGA). Molecular docking study was performed in GOLD
molecular docking tools (Astex Technology, Chembridge,
UK, 2011; Jones et al. 1995).
MD simulation
As far as the MD simulation was concerned, based on the
docking interactions, it was performed on the apo, glutamic
acid bound and compound-59 bound complexes of gluta-
minase (KGA). The GROMACS 4.0.4 program (Hess et al.
2008; Van Der Spoel et al. 2005) with GROMOS96 43al
force field at constant temperature and pressure ensemble
was used for performing all these simulation analyses. In
the MD simulations, all these protein atoms were sur-
rounded by a cubic water box pf SPC3 water molecules that
extended 10 Å away from the protein and periodic boundary
conditions were applied in all directions. These systems
were neutralized with Na+ and Cl− counter ions replacing
the solvent molecules. Energy minimization was performed
using the steepest descent algorithm for 10000 steps. A 100
ps position restrained MD simulation was performed for
every system followed by 500 ps production of MD simu-
lations with a time step of 2 fs a constant pressure (1 atm)
and temperature (300 K). The electrostatic interaction was
calculated by the Particle-Mesh Ewald (PME) method. All
bonds were constrained using LINCS algorithm. The
GROMACS topology for the ligands was obtained from
PRODRG web server. The stability of the protein com-
plexes in the MD simulation was also monitored using the
root mean square deviation (RMSD) calculation with
respect to their initial structure. The RMSD of glutaminase-
glutamic acid substrate complex showed the lowest value
Protein structure
The X-ray crystallographic structure of human phosphate
dependent kidney type glutaminase (KGA) was docked
with the best active compound (compound 59). The PDB
structure of human glutaminase in complex with L-glutamic
acid (PDB ID 3CZD) was downloaded from the Research
Collaborator for Structural Bioinformatics web site (www.
rscb.org). The X-ray crystal structure had the resolution of
2.40 Ǻ. Glutaminase forms a homodimer and each mono-
mer is bifurcated fold with the active site between the two
domains. The first domain is comprised of five stranded
anti-parallel β-sheet surrounded by α-helices and loops
(residues 222–281 and 424–531). The second domain is
completely α-helical (residues 282–423). In the inter-
domain cleft, L-glutamic acid was added to the protein
during crystallization. It made hydrogen bonds with
Gln285, Ser286, Asn335, Glu381, Asn388, Tyr414,
Tyr466, and Val484. The inhibition of human kidney type
glutaminase (KGA) by glutamic acid was reported earlier
(www.rscb.org.;
description/3CZD/).
It was also reported (www.thesgc.org/structures/
structure_description/3CZ) that although Lys289 and
Ser286 are present in close proximity, Ser286 makes
www.thesgc.org/structures/structure_

Med Chem Res
Table 1 Physical data of the intermediate compounds (8–28)
compared to all other systems. The complex of glutaminase
and the best active compound (compound 59) showed
considerably lower RMSD value which indicates the
stability of this complex. All complexes in these simulations
except the apoform of the enzyme converged at around
2.0 Å.
Cpda R1
R2
R3
Mp(°C) % Yield Molecular
formula
b
8
H
H
H
H
Liquid
C6H5SO2Cl
C7H7SO3Cl
C8H9SO4Cl
C8H9SO2Cl
C10H13SO2Cl
C6H4SO2ClF
C8H9SO2Cl
9
H
OCH3
H
40–42
71.32
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
a
OCH3
CH3
H
OCH3 105–108 91.56
CH3
i-C4H9
F
H
Liquid
Liquid
Liquid
Liquid
79.57
81.26
80.44
97.56
H
H
H
Results and discussion
Chemistry
CH3
H
H
CH3
H
H
125–127 84.56
C11H13O6NS
H
OCH3
H
H
75–78
37.75
50.52
C12H15O7NS
C13H17O8NS
C13H17O6NS
C15H21O6NS
C11H12O6NSF
C13H17O6NS
C11H11NO5S
C12H13NO6S
C13H15 NO7S
C13H15 NO5S
C15H19NO5S
OCH3
CH3
H
OCH3 87–89
Substituted benzenesulphonyl chlorides (8–14) were pre-
pared from the corresponding substituted benzenes (1–7)
through chlorosulphonylation (Huntress and Carten 1940).
To obtain 2-N-(substituted benzenesulphonyl)–L(+) glu-
tamic acids (15–21), these sulphonyl chlorides (8–14) were
condensed separately with L(+) glutamic acid. During the
condensation process, alkaline medium (2 N NaOH) was
maintained that helped to remove hydrochloric acid
formed. The resulting diacids (15–21) were subjected to
cyclization reaction with acetyl chloride to get 1-N-(sub-
stituted benzenesulphonyl)-5-oxopyrrolidine-2-carboxylic
acids (21–28). Aminolysis of these monoacids (21–28)
separately with various amines produced the corresponding
glutamine analogs (29–69) (Fig. 1).
CH3
i-C4H9
F
H
135–141 49.92
132–134 85.53
125–127 78.93
102–104 82.46
145–147 56.66
H
H
H
CH3
H
H
CH3
H
H
H
OCH3
H
H
95–97
81.53
OCH3
CH3
H
OCH3 124–126 82.26
CH3
i-C4H9
F
H
173–175 85.55
112–114 86.12
H
H
H
78–80
82–84
72.13
90.25
C11H10FNO5S
CH3
H
CH3
C13H15NO5S
Compound number;
b
Directly purchased from Spectrochem, Mumbai
The route of synthesis of these glutamines (29–69) was
presented schematically in Fig. 2. The structures and phy-
sical data of the intermediate compounds are shown in
Table 1.
Compound 30
The structures and physical data of the final compounds
(29–69) are shown in Table 2.
All title compounds (29–69) were obtained as crystalline
solids with appreciable yields except two (compounds 35
and 36). Spectral and elemental analysis data of all the final
compounds (29–69) are shown below:
Yield 81.48%, M. P. 178–180 °C, MS (FAB):M+H+ peak
at m/z 372. IR (KBr, cm−1): 3467, 3110 (N–H str. of
CONH), 2977 (Ar C–H str.), 2839 (ali C–H str.), 1701
(C=O str.), 1552, 1444 (ali C–H def.), 1307 (S=O str. of
SO2NH,asymmetric), 1255 (asymmetric C–O–C str), 1151
(S=O str. of SO2NH, symmetric), 1099 (symmetric C–O–C
str),991,829,804 (Ar–C–H def.). 1H NMR (300 MHz,
CDCl3): δ 12.51 (s, 1H, COOH), δ 7.69 (d, 1H, H-3′, H-5′),
δ 7.67 (m, 1H, CONH-5), δ 7.58 (s, 1H, SO2NH), δ 7.47 (d,
2H, H-2′, H-6′), δ 3.93 (m, 1H, H-2), 1.47-1.26 (m, 9H,
three CH3 of t-But), 2.23 (m, 2H, H-4), 2.07–1.92 (m, 2H,
HA-3, HB-3). Anal. (C, H, N) calcd: 51.61, 6.45, 7.53;
found: 51.75, 6.58, 7.42
Compound 29
Yield 44.74%, M. P. 174–176 °C, MS (FAB):M+H+ peak
at m/z 407. IR (KBr, cm−1): 3103 (N–H str. of CONH),
2925 (ali C–H str.), 1755(C=O str.), 1697, 1450 (ali C–H
def.), 1481, 1367 (NO2 str.), 1340(S=O str. of SO2NH,
asymmetric), 1176 (S=O str. of SO2NH, symmetric), 1087,
1024, 958, 756 (Ar–C–H def.). 1H NMR (300 MHz,
CDCl3): δ 12.65 (s, 1H, COOH), δ 8.23-7.9 (m, 4H,
nitrobenzyl protons), 7.87 (dd, J = 13.3, 2H, H-2′, H-6′),
7.72 (d, 1H, H-4′), δ 7.42(dd, 2H, J = 5.7, H-3′, H-5′), δ
3.78 (m, 1H, H-2), δ 2.19 (m, 2H, H-4), δ 2.11–1.99
(m, 2H, HA-3, HB-3). Anal. (C, H, N) calcd: 50.12, 4.17,
10.32; found: 50.31, 4.67, 10.39
Compound 31
Yield 86.09%, M. P. 90-92 °C, MS (FAB):M+H+ peak at
m/z 437. IR (KBr, cm−1): 3533, 3110 (N–H str. of CONH),
2975 (Ar C–H str.), 2839 (ali C–H str.), 1733 (C=O str.),
1531, 1461 (ali C–H def.), 1350 (S=O str. of SO2NH,
asymmetric), 1269 (asymmetric C-O-C str), 1161 (S=O str.

Med Chem Res
Table 2 Structures and the physicochemical data of 5-N-substituted-2-N-(substituted benzenesulphonyl)-L(+) glutamines (29–69)
Cpda
R1
R2
R3
R4
Mp(°C)
% Yield
%TCIe
Log(%TCI)f
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
Mitob
Azac
DNd
H
H
H
C6H4NO2
t-C4H9
C6H4NO2
C2H4C6H5
H
174–176
178–180
90–92
44.74
81.48
86.09
89.13
57.14
44.75
16.11
14.83
53.80
87.42
67.21
56.68
73.46
92.67
75.63
47.26
62.13
54.26
64.14
89.64
78.35
69.53
84.81
72.40
86.65
89.05
96.99
76.27
77.55
77.55
78.74
74.69
91.12
98.18
48.34
52.86
79.52
83.59
94.61
48.36
89.43
–
5.00
57.57
46.81
21.58
39.83
47.48
53.60
67.23
66.96
74.11
51.84
61.13
51.41
31.30
85.26
53.90
80.38
38.82
61.64
62.37
63.56
56.85
79.26
56.60
56.51
75.76
28.19
65.04
79.58
88.38
90.92
28.50
62.87
63.17
67.46
37.96
16.46
68.20
79.75
63.26
39.46
100.00
100.00
100.00
0.699
1.760
1.670
1.334
1.600
1.676
1.729
1.827
1.826
1.870
1.715
1.786
1.711
1.495
1.931
1.840
1.905
1.589
1.790
1.795
1.803
1.755
1.899
1.753
1.752
1.879
1.450
1.813
1.847
1.946
1.959
1.455
1.798
1.800
1.829
1.579
1.216
1.834
1.902
1.801
1.596
–
H
OCH3
OCH3
OCH3
H
H
H
H
H
H
131–133
145–147
122–124
73–75
OCH3
OCH3
OCH3
OCH3
OCH3
OCH3
OCH3
OCH3
OCH3
OCH3
CH3
CH3
CH3
CH3
CH3
CH3
CH3
CH3
CH3
CH3
CH3
H
OCH3
OCH3
OCH3
OCH3
OCH3
OCH3
OCH3
OCH3
OCH3
OCH3
H
H
CH3
H
C2H5
H
n-C3H7
n-C4H9
t-C4H9
n-C6H13
c-C6H11
CH2C6H5
C2H4C6H5
H
65–67
H
111–113
171–173
52–54
H
H
H
148–150
162–164
160–162
150–152
137–139
130132
137–139
148–150
150–152
143–145
152–154
200–203
173–175
147–149
141–143
173–175
155–157
173–175
173–175
165–167
179–181
202–204
102–104
133–135
193–195
166–168
119–121
171–173
99–101
130–132
–
H
H
CH3
CH3
CH3
CH3
CH3
CH3
CH3
CH3
CH3
CH3
CH3
i-C4H9
i-C4H9
i-C4H9
i-C4H9
i-C4H9
i-C4H9
i-C4H9
i-C4H9
i-C4H9
F
H
CH3
H
C2H5
H
n-C3H7
i-C3H7
i-C4H9
n-C5H11
n-C6H13
C6H5
H
H
H
H
H
H
C6H4NO2
C2H4C6H5
H
H
H
H
H
CH3
H
H
i-C3H7
n-C4H9
i-C4H9
n-C5H11
c-C6H11
C6H5
H
H
H
H
H
H
H
H
H
H
H
H
CH2C6H5
CH3
H
H
H
F
H
t-C4H9
C6H4NO2
C2H4C6H5
t-C4H9
C6H4NO2
C2H4C6H5
–
H
F
H
H
F
H
CH3
CH3
CH3
–
H
CH3
CH3
CH3
–
H
H
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
a
Compound number;
Mitomycin,
b
c
d
e
f
Azaserine;
DON;
%Tumor cell inhibition;
Logarithm of tumor cell inhibition
of SO2NH, symmetric), 1089 (symmetric C-O-C str), 960,
842, 742 (Ar–C–H def.). H NMR (300 MHz, CDCl3): δ
12.56 (s, 1H, COOH), δ 7.83-7.66 (m, 5H, phenyl protons),
δ 7.11 (dd, 2H, J = 7.2, H-3′, H-5′), δ 7.08 (d, 2H, H-2′, H-
6′), δ 7.45 (s, 1H, SO2NH), δ 3.87 (s, 3H, OCH3), δ 3.78 (m,
1H, H-2), δ 2.30 (m, 2H, H-4), δ 1.99–1.92 (m, 2H, HA-3,
HB-3). Anal. (C, H, N) calcd: 49.43, 4.35, 9.61; found:
49.76, 4.16, 9.84
1

Med Chem Res
1
Compound 32
(symmetric C-O-C str),983, 827(Ar–C–H def.). H NMR
(300 MHz, CDCl3): δ 12.67 (s, 1H, COOH), 7.62 (d, J =
4.3, 1H, H-3′), 7.31 (d, J = 1.8, 1H, H-4′), 7.05 (s, 1H, H-
6′), 6.78 (m, 2H, CONH2), 3.87 (s, 3H, 5′ OCH3), 3.80
(m, 2H, N-CH2-1″),3.76 (s, 3H, 2′ OCH3), 3.68 (m, 1H,
H-2), δ 2.60 (m, 3H, N-CH3-1‴), 2.16 (m, 2H, H-4), δ 1.22
(m, 3H, CH3-2″). Anal. (C, H, N) calcd: 48.13, 5.88, 7.49;
found: 48.24, 5.81, 7.54
Yield 89.13%, M. P. 131–133 °C, MS (FAB):M+H+ peak
at m/z 420. IR (KBr, cm−1): 3396, 3259 (N–H str. of
CONH), 3028 (ArC–H str.), 2837 (ali C–H str.), 1712(C=O
str.), 1542, 1454 (ali C–H def.), 1301 (S=O str. of SO2NH,
asymmetric), 1257 (asymmetric C–O–C str), 1159 (S=O
str. of SO2NH, symmetric), 1091 (symmetric C–O–C str),
975, 831, 746 (Ar–C–H def.). 1H NMR (300 MHz, CDCl3):
δ 12.58 (s, 1H, COOH), 7.11 (d, 2H, H-3′, H-5′), 7.08 (dd,
J = 12.8, 2H, H-2′, H-6′), δ 7.45 (s, 1H, SO2NH), δ 3.87 (s,
3H, OCH3), δ 3.69 (m, 1H, H-2), 2.30 (m, 2H, H-4), δ
1.99–1.92 (m, 2H, HA-3, HB-3). Anal. (C, H, N) calcd:
57.14, 5.71, 6.67; found: 57.43, 5.68, 6.73
Compound 36
MS (FAB):M+H+ peak at m/z 388. IR (KBr, cm−1): 3357,
3286 (N–H str. of CONH), 3097 (Ar C–H str.), 2937
(ali C–H str.), 1706 (C=O str.), 1554, 1440(ali C–H def.),
1299 (S=O str. of SO2NH, asymmetric), 1226 (asymmetric
C–O–C str), 1159 (S=O str. of SO2NH, symmetric), 1039
(symmetric C–O–C str),983, 825,746 (Ar–C–H def.). 1H
NMR (300 MHz, CDCl3): δ 12.56 (s, 1H, COOH), δ 7.62 (d,
J = 3.5, 1H, H-3′), δ 7.34 (d, J = 1.9, 1H, H-4′), δ 6.80 (m,
2H, CONH2), δ 3.88 (s, 3H, 5′ OCH3), δ 3.74 (s, 3H, 2′
OCH3), δ 3.63 (m, 1H, H-2), δ 3.34 (m, 2H, N-CH2-1″), δ
2.66 (m, 3H, N-CH3-1‴), δ 2.16 (m, 2H, H-4), δ 1.22 (m, 3H,
CH3-2”), δ 1.65 (m, 2H, CH2-2‴), δ 0.97 (m, 3H, CH3-3‴).
Anal. (C, H, N) calcd: 49.48, 6.18, 7.22; found: 49.64,
6.17, 7.24
Compound 33
MS (FAB):M+H+ peak at m/z 334. IR (KBr, cm−1): 3465,
3334 (N–H str. of CONH2), 3083 (ArC–H str.), 2840 (ali
C–H str.), 1716 (C=O str.), 1579, 1456 (ali C–H def.), 1307
(S=O str. of SO2NH, asymmetric), 1278 (asymmetric
C–O–C str), 1164 (S=O str. of SO2NH, symmetric), 1043
1
(symmetric C–O–C str), 945, 790, 740 (Ar–C–H def.). H
NMR (300 MHz, CDCl3): δ 12.62 (s, 1H, COOH), δ 7.64
(d, 1H, J = 6.8, H-3′), δ 7.27 (d, 1H, H-4′), δ 7.05 (s, 1H, H-
6′), δ 6.75 (m, 2H, CONH2), δ 3.82 (s, 3H, 5′ OCH3), δ 3.74
(s, 3H, 2′ OCH3), δ 3.68 (m, 1H, H-2), δ 2.14 (m, 2H, H-4),
δ 1.85-1.65(m, 2H, H-3). Anal. (C, H, N) calcd: 43.11, 5.39,
8.38; found: 43.24, 5.41, 8.54
Compound 37
MS (FAB):M+H+ peak at m/z 402. IR (KBr, cm−1): 3299
(N–H str. of CONH),3083 (Ar C–H str.), 2958 (ali C–H str.),
1720 (C=O str.), 1546, 1442 (ali C–H def.), 1328 (S=O str.
Of SO2NH, asymmetric), 1247 (asymmetric C–O–C str),
1157 (S=O str. of SO2NH, symmetric), 1045(symmetric
Compound 34
MS (FAB):M+H+ peak at m/z 360. IR (KBr, cm−1): 3296,
3112 (N–H str. of CONH), 3002 (ArC–H str.), 2925, 2840
(ali C–H str.), 1714 (C=O str.), 1531, 1438 (ali C–H def.),
1307 (S=O str. of SO2NH, asymmetric), 1240 (asymmetric
C–O–C str), 1166 (S=O str. of SO2NH, symmetric), 1047
1
C–O–C str), 989, 806, 740 (Ar–C–H def.). H NMR (300
MHz, CDCl3): δ 12.52 (s, 1H, COOH), 7.62 (d,J = 5.6, 1H,
H-3′), 7.31 (d, J = 2.1, 1H, H-4′), 7.05 (s, 1H, H-6′), 6.78 (m,
2H, CONH2), 3.87 (s, 3H, 5′ OCH3), 3.76 (s, 3H, 2′ OCH3),
3.68 (m, 1H, H-2), δ 3.29 (m, 2H, N-CH2-1″), 2.16 (m, 2H,
H-4), δ 1.22 (m, 3H, CH3-2″), 1.49-1.33 (m, 4H, CH2-2″,
CH2-3″), 0.99–0.93 (m, 3H, CH3-4″). Anal. (C, H, N) calcd:
50.75, 6.47, 6.96; found: 50.87, 6.51, 7.06
1
(symmetric C–O–C str),977, 823, 790 (Ar–C–H def.). H
NMR (300 MHz, CDCl3): δ 12.61 (s, 1H, COOH), δ 7.62
(d, J = 4.6, 1H, H-3′), δ 7.31 (d, J = 2.4, 1H, H-4′), δ 7.05
(s, 1H, H-6′), δ 6.78 (m, 2H, CONH2), δ 3.87 (s, 3H, 5′
OCH3), δ 3.76 (s, 3H, 2′ OCH3), δ 3.68 (m, 1H, H-2), δ 2.60
(m, 3H, N-CH3-1′′), δ 2.16 (m, 2H, H-4), δ 1.85-1.65(m,
2H, H-3). Anal. (C, H, N) calcd: 46.67, 5.55, 7.78; found:
46.84, 5.46, 7.78
Compound 38
MS (FAB):M+H+ peak at m/z 402. IR (KBr, cm−1): 3203,
3101 (N–H str. of CONH), 2977 (Ar C–H str.), 2918
(ali C–H str.), 1701 (C=O str.), 1537, 1498 (ali C–H def.),
1319 (S=O str. of SO2NH, asymmetric), 1224 (asymmetric
C-O-C str), 1155 (S=O str. of SO2NH, symmetric), 1037
Compound 35
MS (FAB):M+H+ peak at m/z 374. IR (KBr, cm−1): 3361,
3124 (N–H str. of CONH), 2974 (Ar C–H str.), 2925 (ali
C–H str.), 1706 (C=O str.), 1550, 1442 (ali C–H def.), 1309
(S=O str. of SO2NH, asymmetric), 1228 (asymmetric
C–O–C str), 1161 (S=O str. of SO2NH, symmetric), 1041
1
(symmetric C-O-C str), 990, 829, 804 (Ar–C–H def.). H
NMR (300 MHz, CDCl3): δ 12.52 (s, 1H, COOH), 7.52-
7.10 (m, 4H, Phenyl protons), 6.78 (m, 2H, CONH2), 3.87
(s, 3H, 5′ OCH3), 3.76 (s, 3H, 2′ OCH3), 3.68 (m, 1H, H-2),

Med Chem Res
2.16 (m, 2H, H-4), 1.46-1.27 (m, 9H, three methyl groups
of t-But). Anal. (C, H, N) calcd: 50.75, 6.47, 6.96; found:
50.87, 6.45, 7.12
Compound 42
MS (FAB):M+H+ peak at m/z 450. IR (KBr, cm−1): 3303
(N–H str. of CONH), 3026 (Ar C–H str.), 2939 (ali C–H
str.), 1542, 1440 (ali C–H def.), 1336 (S=O str. of SO2NH,
asymmetric), 1234 (asymmetric C–O–C str), 1153 (S=O
str. of SO2NH, symmetric), 1033 (symmetric C–O–C str),
817, 746 (Ar–C–H def.). 1H NMR (300 MHz, CDCl3):
δ 12.64 (s, 1H, COOH), 7.64 (d, J = 6.2, 1H, H-3′), 7.27 (d,
1H, H-4′), 7.05 (s, 1H, H-6′), 6.75 (m, 2H, CONH2), 3.82
(s, 3H, 5′ OCH3), 3.74 (s, 3H, 2′ OCH3), 3.69 (s, 2H, N-
CH2-1″), 2.77 (m, 2H, CH2-2″) 2.14 (m, 2H, H-4), 1.85-
1.65 (m, 2H, H-3). Anal. (C, H, N) calcd: 56.00, 5.78, 6.22;
found: 56.15, 5.87, 6.45
Compound 39
MS (FAB):M+H+ peak at m/z 430. IR (KBr, cm−1): 3305,
(N–H str. of CONH), 3082 (Ar C–H str.), 2931 (ali C–H
str.), 1649 (C=O str.), 1550, 1442 (ali C–H def.), 1326
(S=O str. of SO2NH, asymmetric), 1224 (asymmetric
C–O–C str), 1159 (S=O str. of SO2NH, symmetric), 1045
1
(symmetric C–O–C str),929, 806, 700 (Ar–C–H def.). H
NMR (300 MHz, CDCl3): δ 12.52 (s, 1H, COOH),
7.52–7.10 (m, 4H, Phenyl protons), 6.78 (m, 2H, CONH2),
3.87 (s, 3H, 5′ OCH3), 3.76 (s, 3H, 2′ OCH3), 3.68 (m, 1H,
H-2), δ 3.21 (m, 2H, NH-CH2-1″), 2.16 (m, 2H, H-4), δ
1.22 (m, 3H, CH3-2″), δ 0.83–1.63 (m, 13H, CH2-2″, CH2-
3″, CH2-4″, CH2-5″, CH2-6″,). Anal. (C, H, N) calcd: 53.02,
6.98, 6.51; found: 53.17, 6.99, 7.65
Compound 43
MS (FAB):M+H+ peak at m/z 302. IR (KBr, cm−1): 3346,
(N–H str. of CONH), 3056 (Ar C–H str.), 2860 (ali C–H
str.), 1716 (C=O str.), 1595, 1448 (ali C–H def.), 1344
(S=O str. of SO2NH, asymmetric), 1149 (S=O str. of
SO2NH, symmetric), 973, 781, 705 (Ar–C–H def.). 1H
NMR (300 MHz, CDCl3): δ 12.56 (s, 1H, COOH), 7.28
(d, J = 7.2, 1H, H-3′), 7.20 (d, J = 2.3, 1H, H-4′), 7.48
(s, 1H, SO2NH), 6.75 (m, 2H, CONH2), 3.66 (m, 1H, H-2),
2.27 (m, 2H, H-4), 2.02-2.07 (m, 2H, H-3),1.70- 0.99 (m,
6H, six methyl protons of xylene). Anal. (C, H, N) calcd:
47.68, 5.96, 9.27; found: 47.77, 5.99, 9.23
Compound 40
MS (FAB):M+H+ peak at m/z 428. IR (KBr, cm−1): 3379,
3286 (N–H str. of CONH), 2991 (Ar C–H str.), 2935 (ali
C–H str.), 1733 (C=O str.), 1553, 1463 (ali C–Hdef.), 1332
(S=O str. of SO2NH, asymmetric), 1222 (asymmetric C-O-
C str), 1159 (S=O str. of SO2NH, symmetric), 1049
1
(symmetric C–O–C str),977, 796, 744 (Ar–C–H def.). H
Compound 44
NMR (300 MHz, CDCl3): δ 12.62 (s, 1H, COOH), 7.64 (d,
J = 5.8, 1H, H-3′), 7.27 (d, J = 1.6, 1H, H-4′), 7.05 (s, 1H,
H-6′), 6.75 (m, 2H, CONH2), 3.82 (s, 3H, 5′ OCH3), 3.74
(s, 3H, 2′ OCH3), 3.68 (m, 1H, H-2), 2.14 (m, 2H, H-4),
1.85–1.65(m, 2H, H-3), δ 1.01-1.90 (m, 22H, Cyclohexyl
protons). Anal. (C, H, N) calcd: 53.27, 6.54, 6.54; found:
53.45, 6.52, 6.69
MS (FAB):M+H+ peak at m/z 328. IR (KBr, cm−1): 3339,
3253 (N–H str. of CONH), 3062 (Ar C–H str.), 2931 (ali
C–H str.), 1714 (C=O str.), 1537, 1494 (ali C–H def.), 1323
(S=O str. of SO2NH, asymmetric), 1155 (S=O str. of
SO2NH, symmetric), 972,771, 713 (Ar–C–H def.). 1H
NMR (300 MHz, CDCl3): δ 12.63 (s, 1H, COOH), 7.28 (d,
J = 6.5, 1H, H-3′), 7.24 (d, J = 1.4, 1H, H-4′), 7.48 (s, 1H,
SO2NH), 6.72 (m, 2H, CONH2), 3.64 (m, 1H, H-2), δ 2.67
(m, 3H, N-CH3-1‴), 2.24 (m, 2H, H-4), 2.02-2.06 (m, 2H,
H-3),1.63- 0.97 (m, 6H, six methyl protons of xylene).
Anal. (C, H, N) calcd: 51.22, 6.10, 8.54; found: 51.43,
5.97, 8.45
Compound 41
MS (FAB):M+H+ peak at m/z 436. IR (KBr, cm−1): 3316,
3106 (N–H str. of CONH), 3012 (Ar C–H str.), 2876 (ali
C–H str.), 1700 (C=O str.), 1556, 1440 (ali C–H def.), 1332
(S=O str. of SO2NH, asymmetric), 1247 (asymmetric
C–O–C str), 1160 (S=O str. of SO2NH, symmetric), 1045
Compound 45
1
(symmetric C–O–C str), 977, 798, 748 (Ar–C–H def.). H
MS (FAB):M+H+ peak at m/z 342. IR (KBr, cm−1): 3373,
3251 (N–H str. of CONH), 2974 (Ar C–H str.), 2931
(ali C–H str.), 1718 (C=O str.), 1564, 1452 (ali C–H def.),
1323 (S=O str. of SO2NH, asymmetric), 1157 (S=O str. of
SO2NH, symmetric), 975, 798, 750 (Ar–C–H def.).
1H NMR (300 MHz, CDCl3): δ 12.63 (s, 1H, COOH), δ
7.28 (d, 1H, H-3′), δ 7.24 (d, 1H, H-4′), δ 7.48 (s, 1H,
NMR (300 MHz, CDCl3): δ 12.63 (s, 1H, COOH), 7.66 (d,
J = 5.8, 1H, H-3′), 7.25 (d, 1H, J = 2.4, H-4′), 7.03 (s, 1H,
H-6′), 6.72 (m, 2H, CONH2), 3.80 (s, 3H, 5′ OCH3), 3.72
(s, 3H, 2′ OCH3), 3.60 (s, 2H, N-CH2-1″), 2.18 (m, 2H,
H-4), 1.86-1.65(m, 2H, H-3). Anal. (C, H, N) calcd: 55.04,
5.50, 6.42; found: 55.16, 5.56, 6.56

Med Chem Res
SO2NH), δ 7.23 (m, 1H, CONH), δ 3.64 (m, 1H, H-2), δ
2.67 (m, 3H, N-CH3-1‴), 2.24 (m, 2H, H-4), 2.02-2.06 (m,
2H, H-3),1.63–0.97 (m, 6H, six methyl protons of xylene).
Anal. (C, H, N) calcd: 52.63, 6.43, 8.19; found: 52.95,
6.55, 8.32
C–H str.), 1716 (C=O str.), 1564, 1454 (ali C–H def.), 1323
(S=O str. of SO2NH, asymmetric), 1155 (S=O str. of
SO2NH, symmetric), 973, 771, 711 (Ar–C–H def.). 1H
NMR (300 MHz, CDCl3): δ 12.57 (s, 1H, COOH), 7.34 (s,
1H, H-3′), 7.23 (s, 1H, H-4′), 7.41 (s, 1H, SO2NH), δ 2.67
(m, 2H, N-CH2-1‴), 2.28 (m, 2H, H-4), 2.00–2.07 (m, 2H,
H-3), 1.65–0.98 (m, 6H, six methyl protons of xylene),
0.92–0.84 (m, 9H, CH2-2″, CH2-3″, CH2-4″, CH3-5″). Anal.
(C, H, N) calcd: 56.25, 7.29, 7.29; found: 56.39, 7.22, 7.65
Compound 46
MS (FAB):M+H+ peak at m/z 356. IR (KBr, cm−1): 3377,
3251 (N–H str. of CONH), 3060 (Ar C–H str.), 2875 (ali
C–H str.), 1716 (C=O str.), 1564, 1454 (ali C–H def.), 1323
(S=O str. of SO2NH, asymmetric), 1155 (S=O str. of
SO2NH, symmetric), 972, 773, 709 (Ar–C–H def.). 1H
NMR (300 MHz, CDCl3): δ 12.66 (s, 1H, COOH), δ 7.28
(d, 1H, H-3′), δ 7.24 (d, J = 1.9, 1H, H-4′), δ 7.48 (s, 1H,
SO2NH), δ 7.23 (m, 1H, CONH), δ 3.64 (m, 1H, H-2), δ
2.67 (m, 3H, N-CH3-1‴), 2.24 (m, 2H, H-4), δ 2.02–2.06
(m, 2H, H-3),δ 1.63–0.97 (m, 6H, six methyl protons of
xylene), δ 1.22 (m, 3H, CH3-2″). Anal. (C, H, N) calcd:
53.93, 6.74, 7.86; found: 54.13, 6.77, 7.94
Compound 50
MS (FAB):M+H+ peak at m/z 398. IR (KBr, cm−1): 3375,
3249 (N–H str. of CONH), 3060 (Ar C–H str.), 2856 (ali
C–H str.), 1716 (C=O str.), 1562, 1454 (ali C–H def.), 1323
(S=O str. of SO2NH, asymmetric), 1155 (S=O str. of
SO2NH, symmetric), 975, 773, 713 (Ar–C–H def.). 1H
NMR (300 MHz, CDCl3): δ 12.55 (s, 1H, COOH), 7.31 (s,
1H, H-3′), 7.26 (s, 1H, H-4′), 7.43 (s, 1H, SO2NH), δ 2.65
(m, 2H, N-CH2-1‴), 2.28 (m, 2H, H-4), 1.99–2.03 (m, 2H,
H-3), 1.65–0.98 (m, 6H, six methyl protons of xylene),
1.39-1.31 (m, 8H, CH2-2″, CH2-3″, CH2-4″, CH2-5″), δ
0.96-0.90 (m, 3H, CH3-6″). Anal. (C, H, N) calcd: 57.29,
7.53, 7.03; found: 57.43, 7.52, 6.85
Compound 47
MS (FAB):M+H+ peak at m/z 356. IR (KBr, cm−1): 3377,
3272 (N–H str. of CONH), 2972 (Ar C–H str.), 2875 (ali
C–H str.), 1718 (C=O str.), 1558, 1450 (ali C–H def.), 1340
(S=O str. of SO2NH, asymmetric), 1157 (S=O str. of
SO2NH, symmetric), 983, 831, 703 (Ar–C–H def.). 1H
NMR (300 MHz, CDCl3): δ 12. 45 (s, 1H, COOH), 7.30 (s,
1H, H-3′), 7.21 (s, 1H, H-4′), 7.43 (s, 1H, SO2NH), δ 3.55
(m, 1H, N-CH-1,‴), 2.27 (m, 2H, H-4), 2.02-2.07 (m, 2H,
H-3), 1.63–0.97 (m, 6H, six methyl protons of xylene), δ
1.22-1.18 (m, 6H, CH3-2″, CH3-3‴). Anal. (C, H, N) calcd:
53.93, 6.74, 7.86; found: 54.11, 6.76, 7.55
Compound 51
MS (FAB):M+H+ peak at m/z 390. IR (KBr, cm−1): 3352,
3249 (N–H str. of CONH), 3056 (Ar C–H str.), 2866 (ali
C–H str.), 1726 (C=O str.),1552, 1444 (ali C–H def.), 1317
(S=O str. of SO2NH, asymmetric), 1153 (S=O str. of
SO2NH, symmetric), 975, 754, 709 (Ar–C–H def.). 1H
NMR (300 MHz, CDCl3): δ 12.55 (s, 1H, COOH), 7.31 (s,
1H, H-3′), 7.26 (s, 1H, H-4′), 7.43 (s, 1H, SO2NH), 2.28 (m,
2H, H-4), 1.99-2.03 (m, 2H, H-3), 1.65–0.98 (m, 6H, six
methyl protons of xylene), 7.21-6.99 (m, 5H, phenyl pro-
ton). Anal. (C, H, N) calcd: 58.46, 5.64, 7.18; found: 58.63,
5.62, 6.97
Compound 48
MS (FAB):M+H+ peak at m/z 370. IR (KBr, cm−1): 3377,
3271 (N–H str. of CONH), 2952 (Ar C–H str.), 2866 (ali
C–H str.), 1718 (C=O str.), 1571, 1450 (ali C–H def.), 1323
(S=O str. of SO2NH, asymmetric), 1157 (S=O str. of
SO2NH, symmetric), 983, 767, 727 (Ar–C–H def.). 1H
NMR (300 MHz, CDCl3): δ 12.54 (s, 1H, COOH), 7.34 (s,
1H, H-3′), 7.23 (s, 1H, H-4′), 7.41 (s, 1H, SO2NH), δ 2.67
(m, 2H, N-CH2-1‴), 2.28 (m, 2H, H-4), 2.00–2.07 (m, 2H,
H-3), 1.65–0.98 (m, 6H, six methyl protons of xylene),
0.92–0.84 (m, 7H, CH-2″, CH3-3″, CH3-4″). Anal. (C, H,
N) calcd: 55.13, 7.03, 7.57; found: 55.42, 7.02, 7.85
Compound 52
MS (FAB):M+H+ peak at m/z 435. IR (KBr, cm−1): 3452
(N–H str. of CONH), 3055 (Ar C–H str.), 2925 (ali C–H
str.), 1733 (C=O str.), 1602, 1452 (ali C–H def.), 1353
(S=O str. of SO2NH, asymmetric), 1172 (S=O str. of
SO2NH, symmetric), 956, 798, 707 (Ar–C–H def.). 1H
NMR (300 MHz, CDCl3): δ 12.65 (s, 1H, COOH), 8.23-7.9
(m, 4H, nitrobenzyl protons), 7.31 (s, 1H, H-3′), 7.26
(s, 1H, H-4′), 7.43 (s, 1H, SO2NH), 2.28 (m, 2H, H-4),
1.99-2.03 (m, 2H, H-3), 1.65–0.98 (m, 6H, six methyl
protons of xylene). Anal. (C, H, N) calcd: 52.41, 4.83, 9.65;
found: 52.63, 4.82, 9.67
Compound 49
MS (FAB):M+H+ peak at m/z 384. IR (KBr, cm−1): 3377,
3253 (N–H str. of CONH), 3060 (Ar C–H str.), 2867 (ali

Med Chem Res
Compound 53
CONH), δ 3.55 (m, 1H, N-CH-1‴), δ 2.48 (m, 2H, H2-4), δ
1.31 (m, 2H, H2-3), δ 1.22–1.18 (m, 6H, CH3-2″, CH3-3‴).
Anal. (C, H, N) calcd: 56.25, 7.29, 7.29; found: 56.61,
7.35, 7.07
MS (FAB):M+H+ peak at m/z 418. IR (KBr, cm−1): 3326,
3249 (N–H str. of CONH), 3028 (Ar C–H str.), 2862 (ali
C–H str.), 1589, 1456 (ali C–H def.), 1398 (S=O str. of
SO2NH, asymmetric), 1168 (S=O str. of SO2NH, sym-
metric), 987, 779, 748 (Ar–C–H def.). 1H NMR (300 MHz,
CDCl3): δ 12.67 (s, 1H, COOH), 7.31 (s, 1H, H-3′), 7.26 (s,
1H, H-4′), 7.43 (s, 1H, SO2NH), 2.28 (m, 2H, H-4), 1.99-
2.03 (m, 2H, H-3), 1.65–0.98 (m, 6H, six methyl protons of
xylene), 3.69 (s, 2H, N-CH2-1″), 2.77 (m, 2H, CH2-2″).
Anal. (C, H, N) calcd: 60.29, 6.22, 6.70; found: 60.45,
6.26, 6.67
Compound 57
MS (FAB):M+H+ peak at m/z 398. IR (KBr, cm−1): 3369,
3269 (N–H str. of CONH), 2960 (Ar C–H str.), 2867 (ali
C–H str.), 1720 (C=O str.), 1558, 1454 (ali C–H def.), 1336
(S=O str. of SO2NH, asymmetric), 1163 (S=O str. of
SO2NH, symmetric), 981, 756, 730 (Ar–C–H def.). 1H
NMR (300 MHz, CDCl3): δ 12.56 (s, 1H, COOH), δ 7.88
(d, 2H, H-2′, H-6′), δ 7.64 (d, J = 9.5, 1H, H-3′, H-5′), δ
7.61 (s, 1H, SO2NH), δ 6.55 (m, 1H, CONH2), δ 3.51 (m,
1H, H-2), δ 2.46 (m, 2H, H2-4), δ 1.31 (m, 2H, H2-3), 1.37-
0.89 (m, 9H, i-But protons). Anal. (C, H, N) calcd: 57.28,
7.54, 7.03; found: 57.31, 7.61, 6.91
Compound 54
MS (FAB):M+H+ peak at m/z 342. IR (KBr, cm−1): 3431,
3269 (N–H str. of CONH), 2954 Ar C–H str.), 2867 (ali
C–H str.), 1708 (C=O str.), 1562, 1452 (ali C–H def.), 1344
(S=O str. of SO2NH, asymmetric), 1163 (S=O str. of
Compound 58
1
SO2NH, symmetric), 968, 792 (Ar–C–H def.). H NMR
(300 MHz, CDCl3): δ 12.67 (s, 1H, COOH), 7.31 (s, 1H, H-
3′), 7.26 (s, 1H, H-4′), 7.43 (s, 1H, SO2NH), 2.28 (m, 2H,
H-4), 1.99-2.03 (m, 2H, H-3), 1.65–0.98 (m, 6H, six methyl
protons of xylene), 3.69 (s, 2H, N-CH2-1″), 2.77 (m, 2H,
CH2-2″). Anal. (C, H, N) calcd: 52.63, 6.43, 8.19; found:
52.89, 6.55, 7.97
MS (FAB):M+H+ peak at m/z 398. IR (KBr, cm−1): 3375,
3265 (N–H str. of CONH), 2954 (Ar C–H str.), 2867 (ali
C–H str.), 1718 (C=O str.), 1562, 1461 (ali C–H def.), 1326
(S=O str. of SO2NH, asymmetric), 1163 (S=O str. of
SO2NH, symmetric), 977, 792, 730 (Ar–C–H def.). 1H
NMR (300 MHz, CDCl3): δ 12.56 (s, 1H, COOH), δ 7.88
(d, 2H, H-2′, H-6′), δ 7.64 (d, 1H, H-3′, H-5′), δ 7.61 (s, 1H,
SO2NH), δ 6.55 (m, 1H, CONH2), δ 3.51 (m, 1H, H-2), δ
2.46 (m, 2H, H2-4), δ 1.31 (m, 2H, H2-3), 1.37-0.89 (m, 9H,
i-But protons). Anal. (C, H, N) calcd: 57.28, 7.54, 7.03;
found: 57.43, 7.52, 6.97
Compound 55
MS (FAB):M+H+ peak at m/z 356. IR (KBr, cm−1): 3375,
3257 (N–H str. of CONH), 3066 (Ar C–H str.), 2869
(ali C–H str.), 1712 (C=O str.), 1564, 1458 (ali C–H def.),
1325 (S=O str. of SO2NH, asymmetric), 1163 (S=O str. of
SO2NH, symmetric), 972, 792, 736 (Ar–C–H def.).
1H NMR (300 MHz, CDCl3): δ 12.60 (s, 1H, COOH), δ
7.88 (d, 2H, H-2′, H-6′), δ 7.64 (d, 1H, H-3′, H-5′), δ 7.61
(s, 1H, SO2NH), δ 7.55 (m, 1H, CONH), δ 2.67 (m, 3H,
N-CH3-1‴), δ 3.51 (m, 1H, H-2), δ 2.48 (m, 2H, H2-4),
δ 1.31 (m, 2H, H2-3), 1.37-0.89 (m, 9H, i-But protons).
Anal. (C, H, N) calcd: 53.93, 6.74, 7.86; found: 54.21,
6.75, 7.77
Compound 59
MS (FAB):M+H+ peak at m/z 412. IR (KBr, cm−1): 3373,
3255 (N–H str. of CONH), 2952 (Ar C–H str.), 2867 (ali
C–H str.), 1712 (C=O str.), 1566, 1463 (ali C–H def.), 1325
(S=O str. of SO2NH, asymmetric), 1163 (S=O str. of
SO2NH, symmetric), 972, 794, 736 (Ar–C–H def.). 1H
NMR (300 MHz, CDCl3): δ 12.64 (s, 1H, COOH), δ 7.88
(d, 2H, H-2′, H-6′), δ 7.64 (d, 1H, H-3′, H-5′), δ 7.61 (s, 1H,
SO2NH), δ 7.55 (m, 1H, CONH-5), δ 3.51 (m, 1H, H-2), δ
2.72 (m, 2H, N-CH2-1‴), δ 2.46 (m, 2H, H2-4), δ 1.31 (m,
2H, H2-3), 1.19-0.89 (m, 14H, CH-2″, CH3-3″, CH3-4″,
CH3-5″). Anal. (C, H, N) calcd: 58.25, 7.77, 6.79; found:
58.41, 7.73, 6.84
Compound 56
MS (FAB):M+H+ peak at m/z 384. IR (KBr, cm−1): 3377,
3257 (N–H str. of CONH), 2958 (Ar C–H str.), 2867 (ali
C–H str.), 1712 (C=O str.), 1598 (ali C–H def.), 1330
(S=O str. of SO2NH, asymmetric), 1161 (S=O str. of
SO2NH, symmetric), 983, 910, 730 (Ar–C–H def.). 1H
NMR (300 MHz, CDCl3): δ 12.62 (s, 1H, COOH),
δ 7.88 (dd, J = 14.5, 2H, H-2′, H-6′), δ 7.64 (dd, J = 8.9,
2H, H-3′, H-5′), δ 7.61 (s, 1H, SO2NH), δ 7.55 (m, 1H,
Compound 60
MS (FAB):M+H+ peak at m/z 424. IR (KBr, cm−1): 3371,
3263 (N–H str. of CONH), 3053 (Ar C–H str.), 2852 (ali
C–H str.), 1720 (C=O str.), 1558, 1450 (ali C–H def.), 1336

Med Chem Res
(S=O str. of SO2NH, asymmetric), 1163 (S=O str. of
SO2NH, symmetric), 977, 794, 732 (Ar–C–H def.). 1H
NMR (300 MHz, CDCl3): δ 12.61 (s, 1H, COOH), δ 7.88
(dd, J = 12.4, 2H, H-2′, H-6′), δ 7.64 (dd, J = 8.9, 1H, H-3′,
H-5′), δ 7.61 (s, 1H, SO2NH), δ 7.55 (m, 1H, CONH-5), δ
3.51 (m, 1H, H-2), δ 2.72 (m, 2H, N-CH2-1‴), δ 2.46
(m, 2H, H2-4), δ 1.31 (m, 2H, H2-3), 1.01–1.90 (m, 11H,
Cyclohexyl protons). Anal. (C, H, N) calcd: 59.43, 7.55,
6.80; found: 59.64, 7.55, 6.57
Compound 64
MS (FAB):M+H+ peak at m/z 360. IR (KBr, cm−1): 3429,
3105 (N–H str. of CONH), 3068 (Ar C–H str.), 2839 (ali
C–H str.), 1704 (C=O str.), 1552, 1492 (ali C–H def.), 1336
(S=O str. of SO2NH, asymmetric), 1151 (S=O str. of
SO2NH, symmetric), 1097 (C-F str.), 989,788, 771
1
(Ar–C–H def.). H NMR (300 MHz, CDCl3): δ 12.43(s,
1H, COOH), 7.46-7.13 (m, 4H, phenyl protons), 3.69 (m,
1H, H-2), 3.15 (m, 3H, N-CH3-1″), δ 2.47 (m, 2H, H2-4),
1.68 (m, 2H, H2-3), 1.46-1.27 (m, 9H, three methyl groups
of t-But). Anal. (C, H, N) calcd: 50.00, 5.83, 7.78; found:
50.34, 5.76, 7.83
Compound 61
MS (FAB):M+H+ peak at m/z 418. IR (KBr, cm−1): 3355,
3253 (N–H str. of CONH), 3062 (Ar C–H str.), 2867
(ali C–H str.), 1720 (C=O str.), 1550, 1444 (ali C–H def.),
1319 (S=O str. of SO2NH, asymmetric), 1161 (S=O str. of
SO2NH, symmetric), 973, 794, 752 (Ar–C–H def.). 1H
NMR (300 MHz, CDCl3): δ 12.60 (s, 1H, COOH), δ 7.88
(d, 2H, H-2′, H-6′), δ 7.64 (d, 1H, H-3′, H-5′), δ 7.61 (s, 1H,
SO2NH), δ 7.55 (m, 1H, CONH-5), δ 3.51 (m, 1H, H-2), δ
2.72 (m, 2H, N-CH2-1‴), δ 2.46 (m, 2H, H2-4), δ 1.31 (m,
2H, H2-3). Anal. (C, H, N) calcd: 60.29, 6.22, 6.70; found:
60.47, 6.32, 6.49
Compound 65
MS (FAB):M+H+ peak at m/z 425. IR (KBr, cm−1): 3481,
3178 (N–H str. of CONH), 3068 (Ar C–H str.), 2839 (ali
C–H str.), 1704 (C=O str.), 1593, 1492 (ali C–H def.), 1342
(S=O str. of SO2NH, asymmetric), 1180 (S=O str. of
SO2NH, symmetric), 1087 (C-F str.), 962,786, 771
1
(Ar–C–H def.). H NMR (300 MHz, CDCl3): δ 12.49(s,
1H, COOH), 8.23-7.9 (m, 4H, nitrobenzyl protons), 7.46-
7.13 (m, 4H, phenyl protons), 3.69 (m, 1H, H-2), 3.15 (m,
3H, N-CH3-1″), δ 2.47 (m, 2H, H2-4), 1.68 (m, 2H, H2-3),
1.46-1.27 (m, 9H, three methyl groups of t-But). Anal.
(C, H, N) calcd: 48.00, 4.94, 9.88; found: 48.24, 4.87, 9.82
Compound 62
MS (FAB):M+H+ peak at m/z 432. IR (KBr, cm−1): 3330,
3249 (N–H str. of CONH), 3029 (Ar C–H str.), 2850 (ali
C–H str.), 1525, 1458 (ali C–H def.), 1317 (S=O str. of
SO2NH, asymmetric), 1157 (S=O str. of SO2NH, sym-
Compound 66
MS (FAB):M+H+ peak at m/z 408. IR (KBr, cm−1): 3379,
3257 (N–H str. of CONH), 3066 (Ar C–H str.), 2866 (ali
C–H str.), 1710 (C=O str.), 1566, 1454 (ali C–H def.), 1338
(S=O str. of SO2NH, asymmetric), 1159 (S=O str. of
SO2NH, symmetric), 1095 (C-F str.), 973, 746 (Ar–C–H
def.). 1H NMR (300 MHz, CDCl3): δ 12.49(s, 1H, COOH),
8.23-7.9 (m, 4H, nitrobenzyl protons), 7.46-7.13 (m, 4H,
phenyl protons), 3.69 (m, 1H, H-2), 3.15 (m, 3H, N-CH3-
1″), δ 2.47 (m, 2H, H2-4), 1.68 (m, 2H, H2-3), 1.46-1.27 (m,
9H, three methyl groups of t-But). Anal. (C, H, N) calcd:
55.88, 5.15, 6.86; found: 55.76, 5.33, 6.79
1
metric), 987,792, 746 (Ar–C–H def.). H NMR (300 MHz,
CDCl3): δ 12.59 (s, 1H, COOH), δ 7.88 (d, 2H, H-2′, H-6′),
δ 7.64 (d, 1H, H-3′, H-5′), δ 7.61 (s, 1H, SO2NH), δ 6.55
(m, 1H, CONH2), 3.66 (s, 2H, N-CH2-1″), δ 3.51 (m, 1H,
H-2), δ 2.46 (m, 2H, H2-4), δ 1.31 (m, 2H, H2-3), 1.37–0.89
(m, 9H, i-But protons). Anal. (C, H, N) calcd: 61.11, 6.48,
6.48; found: 61.34, 6.43, 6.49
Compound 63
Compound 67
MS (FAB):M+H+ peak at m/z 308. IR (KBr, cm−1): 3454
(N–H str. of CONH), 3068 (Ar C–H str.), 2839 (ali C–H
str.), 1704 (C=O str.), 1549, 1492 (ali C–H def.), 1326
(S=O str. of SO2NH, asymmetric), 1159 (S=O str. of
SO2NH, symmetric), 1101 (C-F str), 962, 786, 771
MS (FAB):M+H+ peak at m/z 370. IR (KBr, cm−1): 3413,
3141 (N–H str. of CONH), 2977 (Ar C–H str.), 2837 (ali
C–H str.), 1718 (C=O str.), 1562, 1454 (ali C–H def.), 1321
(S=O str. of SO2NH, asymmetric), 1153 (S=O str. of
1
1
(Ar–C–H def.). H NMR (300 MHz, CDCl3): δ 12.43(s,
SO2NH, symmetric), 985, 783 (Ar–C–H def.). H NMR
1H, COOH), 7.46-7.13 (m, 4H, phenyl protons), 3.69 (m,
1H, H-2), 3.15 (m, 3H, N-CH3-1″), δ 2.47 (m, 2H, H2-4),
1.68 (m, 2H, H2-3). Anal. (C, H, N) calcd: 46.75, 4.87,
9.09; found: 46.98, 4.66, 9.01
(300 MHz, CDCl3): δ 12.56 (s, 1H, COOH), 7.28 (d, 1H,
H-3′), 7.20 (d, J = 2.1, 1H, H-4′), 7.48 (s, 1H, SO2NH),
6.75 (m, 2H, CONH2), 3.66 (m, 1H, H-2), 2.27 (m, 2H, H-
4), 2.02-2.07 (m, 2H, H-3), 1.47- 1.26 (m, 6H, six methyl

Med Chem Res
protons of xylene). Anal. (C, H, N) calcd: 55.13, 7.03, 7.57;
found: 55.54, 7.34, 7.81
descriptors were taken into consideration. Highly correlated
descriptors (correlation coefficient > 0.60) were grouped
together and descriptor with the highest correlation with
biological activity was selected from the group. From the
remaining 234 descriptors, equations were developed by
stepwise forward (F = 3.0 for inclusion) and stepwise
backward (F = 2.9 for exclusion) multiple linear regression
analysis using F value as the stepping criteria. The corre-
lation matrix among the selected descriptors used to develop
QSAR model was shown in the supplementary materials
(Table S1). Calculated values of important descriptors used
in QSAR models are shown in the Supplementary materials
(Table S2).
Before dividing the dataset into the training and the test
sets, regression analysis was performed with the total
dataset. Three compounds were found as outliers (Cpd. 55,
61, and 66) as these compounds showed standard residuals
more than two units for most of the developed models. At
first, the dataset was divided into the test set and the training
set by the Y-based ranking method. Molecules were first
arranged from the highest to the lowest values based on
their biological activities and then molecules in the 4th, 8th,
12th, and so on rows were collected. This way eight com-
pounds (32, 34, 36, 45, 46, 53, 54, and 68) were selected for
the test set compounds (test set I) whereas the rest 30 was
treated as the training set molecules (training set I). Step-
wise regression on the training set generated different
equations but the number of descriptors was restricted up to
five following recommended ratio of 1: 5 for the number of
predictor parameters to number of data point (Topliss and
Edwards 1979; Eriksson et al. 2003). The best equation was
selected on the basis of statistical qualities and predictability
as well as overall fitting. It is shown below:
Compound 68
MS (FAB):M+H+ peak at m/z 435. IR (KBr, cm−1): 3463
(N–H str. of CONH), 2975 (Ar C–H str.), 2862 (ali C–H
str.), 1720 (C=O str.), 1562, 1460 (ali C–H def.), 1348
(S=O str. of SO2NH, asymmetric), 1164 (S=O str. of
SO2NH, symmetric), 958, 796, 744 (Ar–C–H def.). 1H
NMR (300 MHz, CDCl3): δ 12.65 (s, 1H, COOH), 8.23-7.9
(m, 4H, nitrobenzyl protons), 7.31 (s, 1H, H-3′), 7.26 (s,
1H, H-4′), 7.43 (s, 1H, SO2NH), 2.28 (m, 2H, H-4), 1.99-
2.03 (m, 2H, H-3), 1.65- 0.98 (m, 6H, six methyl protons of
xylene). Anal. (C, H, N) calcd: 52.41, 4.83, 9.65; found:
52.63, 4.57, 9.82
Compound 69
MS (FAB):M+H+ peak at m/z 418. IR (KBr, cm−1): 3326,
3249 (N–H str. of CONH), 3029 (Ar C–H str.), 2862 (ali
C–H str.), 1720 (C=O str.), 1583, 1454 (ali C–H def.), 1396
(S=O str. of SO2NH, asymmetric), 1147 (S=O str. of
SO2NH, symmetric), 989,748 (Ar–C–H def.). 1H NMR
(300 MHz, CDCl3): δ 12.66 (s, 1H, COOH), 7.31 (s, 1H, H-
3′), 7.26 (s, 1H, H-4′), 7.43 (s, 1H, SO2NH), 2.28 (m, 2H,
H-4), 1.99-2.03 (m, 2H, H-3), 1.65–0.98 (m, 6H, six methyl
protons of xylene), 3.69 (s, 2H, N-CH2-1″), 2.77 (m, 2H,
CH2-2″). Anal. (C, H, N) calcd: 60.29, 6.22, 6.70; found:
60.44, 6.43, 6.59
Anticancer activity
All title compounds (29–69) were biologically evaluated for
their antitumor activities against EAC cells in Swiss Albino
mice. The %TCI of these compounds against the control
was determined individually. Mitomycin C was chosen as
the universal standard and azaserin and DON were used as
the specific standard drugs to compare the activity of the
test compounds and these standard drugs showed 100%
inhibition. The %TCI of all title compounds (29–69) and
their logarithmic values (log %TCI) are shown in Table 2.
Logð%TCIÞ ¼ 2:247ð 0:362Þ ꢀ 0:045ð 0:015ÞnCaH
þ0:00006ð 0:000Þf ðNÞ5 ꢀ 1:069ð 0:214Þq11
þ2:149ð 0:621Þq13 þ þ0:300ð 0:113ÞB1R1
ð4Þ
n = 30; R = 0.865; R2 = 0.750; R2A = 0.698; F (5, 24) =
14.375; p < 0.00000; SEE = 0.137; R2CV = 0.515; PRESS
= 0.874; SSY = 1.803; SDEP = 0.174; SPRESS = 0.196;
R2SCR = 0.193; n’ = 8; R2Pred = 0.689; rm2 = 0.535
Molecular modeling study
The 2D-QSAR, kNN-MFA 3D-QSAR and molecular
docking studies as well as MD simulation analysis results
are as follows:
Where n and n′ are number of compounds in the training
and the test sets respectively. The Eq. (4) explains 69.80%
and predicts 51.51% of variances of the biological activity.
2D-QSAR study
The external predictability of the model is justified by R2
and rm values which are 0.689 and 0.535, respectively.
Pred
2
Total 347 atom-based and whole molecular descriptors were
calculated for the 2D-QSAR study. Intercorrelated
Three atom based descriptors q11, q13 and f(N)5 were
found to be important. The q11 and q13 are Wang-Ford

Med Chem Res
Table 3 Pharmacological activities of 5-N-substituted-2-N-
(substituted benzenesulphonyl)-L(+) glutamines (70–122)
charges of atom numbers 11 and 13 respectively (Fig. 3).
The negative coefficient of q11 shows that the lower value
of the charge (the negative charge) at the atom number 11
may increase the biological activity. The positive coefficient
associated with q13 suggests that with the increase in the
value of Wang-Ford charge at the atom number 13, the
value of the biological activity may be improved, i.e., the
negative charge of the atom should be decreased for the
increased anticancer activity. As these two atoms are pre-
sent in the aliphatic chain and are attached to the electro-
negative atoms/groups like the nitrogen, the carbonyl, etc, it
may be concluded that the aliphatic part of the glutamine
moiety of these compounds are important for the binding
interactions. The f(N)5 is the frontier electron density of the
atom number 5 present in the aromatic ring of the general
structure (Fig. 3). The positive coefficient of f(N)5 indicates
that the increased nucleophilic attack at the atom number 5
may be favorable for the antitumor activity. Two whole
molecular descriptors—nCaH (number of unsubstituted
aromatic rings) and B1R1 (Verloop strerimol parameter for
R1 substitution) were also found to be important. The
negative coefficient associated with nCaH suggests that the
unsubstituted aromatic rings may not be favorable for
anticancer activity, i.e., substitutions at the aromatic rings
may play important roles for the better binding in the
receptor site. The positive coefficient of B1R1 suggests that
the higher value of this parameter may be conducive to the
biological activity.
Cpda
R1′
R2′
R3′
R4′
R5′
%TCI
R1
2'
3
3'
4'
2
4
SO2NH
R2
H
1'
COOH CON
6'
5'
R4
1
5
R3
70
H
NO2
H
CH3
H
n-C4H9
n-C4H9
n-C3H7
CH3
80.74
89.36
83.098
21.74
31.34
10.47
41.38
32.75
45.45
48.86
31.69
61.97
40.00
56.42
41.76
48.71
20.88
30.76
39.56
12.68
19.13
20.8
71
CH3
Cl
H
NO2
CH3
H
72
H
H
73
H
NO2
NO2
NO2
NO2
NO2
NO2
NO2
H
H
74
H
H
H
C2H5
75
H
H
H
n-C3H7
i-C3H7
C6H5
76
H
H
H
77
H
H
H
78
H
H
H
n-C4H9
n-C6H13
i-C3H7
n-C4H9
i-C4H9
C6H5CH2
n-C4H9
n-C6H13
n-C3H7
i-C4H9
C6H5
79
H
H
H
80
Cl
H
CH3
CH3
CH3
CH3
H
81
Cl
H
H
82
Cl
H
H
83
Cl
H
H
84
H
H
Br
Br
Br
Br
Br
H
85
H
H
H
86
H
H
H
87
H
H
H
88
H
H
H
As only one test set-training set combination may pro-
duce biased model, the dataset was also divided by k-means
cluster analysis (k-MCA) technique. Cluster analysis was
allowed to split the dataset into four clusters (10, 3, 12 and
13 compounds in those four clusters) and 7 compounds (35,
48, 49, 53, 59, 64, and 67) were selected randomly from
each cluster to form the test set (test set II). As the number
of compounds in the training set raised to 31, six descriptors
were allowed to generate models from the training set
(training set II). The best found equation is:
89
H
H
H
i-C4H9
i-C3H7
i-C4H9
H
90
H
H
CH3
CH3
H
H
91
H
H
H
92
CH3
CH3
CH3
CH3
CH3
CH3
CH3
CH3
CH3
CH3
CH3
H
H
NO2
NO2
NO2
NO2
NO2
NO2
NO2
NO2
NO2
NO2
NO2
H
19.35
46.77
66.19
45.00
42.20
76.29
71.80
33.30
70.15
78.89
67.38
40.32
56.06
31.29
32.28
61.80
65.14
57.97
62.85
93
H
H
CH3
94
H
H
C2H5
95
H
H
n-C3H7
i-C3H7
i-C4H9
C6H11
C6H5
96
H
H
97
H
H
98
H
H
99
H
H
Logð%TCIÞ¼ 2:194ð 0:327Þꢀ0:044ð 0:013ÞnCaH
þ0:0002ð 0:000Þf ðEÞ8
100
101
102
103
104
105
106
107
108
109
110
H
H
C6H5CH2
n-C5H11
n-C6H13
H
H
H
H
H
NO2
NO2
NO2
NO2
NO2
NO2
NO2
NO2
CH3
CH3
CH3
CH3
CH3
CH3
CH3
CH3
þ0:00007ð 0:000Þf ðNÞ5 ꢀ 1:002ð 0:185Þq11
H
H
CH3
H
H
C2H5
H
H
n-C3H7
n-C5H11
n-C6H13
i-C3H7
i-C4H9
þ2:287ð 0:529Þq13 þ 0:323ð 0:105ÞB1R1
ð5Þ
H
H
H
H
n = 31; R = 0.900; R2 = 0.810; RA2 = 0.762; F (6, 24) =
16.998; p < 0.00001; SEE = 0.123; R2CV = 0.527; PRESS
= 0.903; SSY = 1.911; SDEP = 0.158; SPRESS = 0.179;
H
H
H
H
R2SCR = 0.179; n′ = 7; R2
= 0.693; rm2 = 0.638.
Pred

Med Chem Res
R5′ %TCI
The Eq. (5) explains 76.20% and predicts 52.70% of
Table 3 continued
variances of the dependent parameter. The equation has
Cpda
R1′
R2′
R3′
R4′
sufficient predictability as justified by R2
and rm
2
Pred
111
112
113
114
115
116
117
118
119
120
121
122
a
H
H
H
H
H
H
H
H
H
H
H
H
NO2
NO2
NO2
H
CH3
H
H
H
H
H
H
H
H
H
H
H
H
C6H11
38.75
28.22
47.88
34.65
41.20
24.73
50.68
34.06
70.47
54.07
65.23
52.36
values of 0.693 and 0.638 respectively. However, the best
found equation for this test set-training set combination
fails to give new descriptor other than f(E)8. This is an
atom based descriptor that stands for the frontier electron
density for electrophilic attack at the atom number 8 which
according to the general structure is one of the two oxy-
gens of the sulphonyl group. Oxygen, having two lone
pairs of electrons is very much susceptible to electrophilic
attack. From the equation, it may be hypothesized that the
electrophilic attack on this atom should be increased for
the higher activity of these molecules. All regression
coefficients used in current 2D-QSAR study are significant
at more than 95% confidence interval as shown by p- and t-
values. The p-and t-values of regression equations were
shown in the supplementary materials (Table S3). The
observed, calculated, residual, predicted residual and pre-
dicted values of these two models (Eq. 4 and Eq. 5) are
shown in the supplementary materials (Table S4 and Table
S5, respectively).
CH3
C6H5CH2
C6H5
CH3
C2H5
C2H5
C2H5
C2H5
C2H5
C2H5
C2H5
C2H5
C2H5
CH3
H
C2H5
H
n-C3H7
n-C4H9
n-C5H11
n-C6H13
i-C3H7
C6H5CH2
C6H5
H
H
H
H
H
H
Compound Number
O
H
H
NH2
HN
3D-QSAR kNN-MFA study
H
O
H
The 2D-QSAR model may not be able to show all
necessary features for the higher activity. Therefore, a 3D-
QSAR study is an essential tool for the prediction of
necessary features of these molecules of interest. With a
hope to get a realistic 3D-QSAR model, 41 compounds of
the current dataset (Table 2) and 53 compounds of earlier
published dataset (compound 70–122) (Samanta et al.
2004) (shown in Table 3) were combined to a set of 94
compounds and these were subjected to KNN-MFA
modeling. As the structure of bioactive conformer is
unknown, the lowest energy conformation of the opti-
mized structure of the best active compound (compound
59) was used for alignment of all molecules. The template
structure is shown in Fig. 4.
After alignment of these molecules, steric, electrostatic,
and hydrophobic charges were calculated. The %TCI
activities were directly used as the dependent parameter.
The dataset was divided into the training set and the test set
by sphere exclusion principle using dissimilarity value 6.5.
The test set comprised of 16 molecules (compounds 36, 37,
44, 54, 57, 63, 68, 70, 76, 78, 81, 84, 94, 106, 115, and
116). The kNN-MFA forward-backward regression (F =
3.0 to enter, F = 2.9 to remove) provided the best model
that is shown below:
S
O
O
X
O
H
Fig. 4 Template structure for the alignment of structures in kNN-MFA
analysis
Statistics:
kNN = 3
Degree of freedom = 73
q2 = 0.4923, (q2 = 0.529, Deleted compounds = 105,
121)
q2_se = 14.6641
Predr2 = 0.1001 (Predr2 = 0.697, Deleted compounds =
36, 97)
pred_r2se = 17.4348
Descriptor Range:
H_1533 0.2574 0.2695
S_1222 -0.0210 -0.0145
S_539 -0.0139 -0.0075
S_1471 -0.0031 -0.0024
The kNN-MFA 3D-QSAR model showed the contribu-
tion of descriptors H_1533, S_1222, S_539 and S_1471,
i.e., hydrophobic interaction field at the lattice point 1533
and the steric interaction fields at lattice points 1222, 539
and 1471 respectively. The model shows q2 value of 0.492
and Predr2 = 0.100; Deletion of two compounds (105, 121)
kNN method:
Training set size = 78, Test set size = 16
Selected Descriptors:
H_1533, S_1222, S_539, S_1471

Med Chem Res
from the training set raised the value of q2 up to 0.529.
Deletion of another two compounds (36, 97) from the test
set yielded predr2 0.697. A comparison between the
observed and the predicted values in the model showed that
the most of the lower active molecules showed the higher
values of the deleted residuals. The dataset was directly
divided into the higher active (%TCI > 0.50) and lower
active (%TCI < 0.50). Compounds were deleted when a
higher active compounds was predicted as a lower active
compound and vice versa as well as the deleted residual was
more than 25 unit. The aligned structures of these molecules
as well as descriptors are shown in Fig. 5.
The kNN-MFA 3D-QSAR study showed importance of
hydrophobic and steric requirements of these glutamine
analogs on the three dimensional space. It is observed that
two steric interaction fields were located near to the sub-
stituted phenyl ring (Fig. 3) and another steric interaction
field was located near the R4 position (Fig. 3). The negative
value of all steric interactions at the lattice point 1222,
1471, and 539, respectively (Fig. 3) suggested that the
decrease of steric interactions at these positions may be
conducive to the higher anticancer activity of these gluta-
mine analogs. Another hydrophobic interaction field was
found to be important at the R4 position (Fig. 3). The
positive value of the hydrophobic interaction at lattice point
1533 suggested that the hydrophobic interaction at this
position may be beneficial for the higher anticancer activity.
Therefore, it may be assumed that the lesser bulky sub-
stitution with the higher hydrophobicity at the
R4 substitution (Fig. 3) may be conducive to the higher
anticancer activity of these glutamine analogs.
Molecular docking study
Molecular docking study (Jones et al. 1997) was performed
with GOLD software (Astex Technology, Chembridge, UK,
2001). First of all, the original ligand, L-gluatmic acid was
docked to the binding site of the same in 3CZD protein to
understand how well the docking experiment was per-
formed. At first, the docking was performed setting all
amino acids in the binding site rigid. The ligand glutamic
acid showed only two hydrogen bonding interactions which
did not match with the reported binding interactions of
L-glutamic acid at the binding site. Therefore, amino acids
in the binding site were rendered flexible and docking was
performed again. This time L-glutamic acid showed Gold-
score fitness 42.88. The interactions are shown in Fig. 6.
In the docked conformation, the ligand formed hydrogen
bonds with Gln285, Ser286, Asn335, Glu381, Asn388 and
Tyr249 and forms short contacts with Gln285, Ser286,
Asn335, Glu381, Asn388, Tyr414, and Tyr466. No addi-
tional interactions were found. Although the binding site
and interactions of L-glutamic acid were reported (Jones
et al. 1995; Venkatachalam et al. 2003) the same for L-
glutamine was not found in the literature. After validating
the original ligand, L-glutamine was also docked in the
same binding site keeping all parameters same. The best
pose showed Goldscore fitness of 40.85. The lower fitness
value was due to the less number of hydrogen bond with the
binding site amino acids. It showed that hydrogen bond
interactions with Glu381, Tyr 249, Tyr414 and Tyr 466 and
short interactions with Gln285, Ser286, Asn335, Glu381,
Fig. 5 Graphical representation
of the kNN-MFA forward
backward regression model

Med Chem Res
Fig. 6 Best docking pose as
well as interactions of
L-glutamic acid
Fig. 7 Best docking pose as
well as interactions of
L-glutamine
Asn388, Tyr414, and Tyr466. The binding interactions are
given in Fig. 7.
The compound 59 was found to form hydrogen bonds
with Tyr466, Val 484, Ser286, Tyr414, Asn388, and
Glu281. The better Goldscore fitness was due to the higher
hydrogen bonding interactions with other amino acids like
Gln285, Ser286, Asn335, Glu381, Asn388, Tyr414,
Tyr466, and Val484. The lipophilic parts of the compounds
like the phenyl ring and its aliphatic substitution like the
The best active compound (compound 59) was docked
similarly. The best pose had Goldscore fitness of 65.82,
which was much higher than that of two natural ligands of
glutaminase. The best docked conformation is shown in
Fig. 8.

Med Chem Res
Fig. 8 Best docking pose as
well as interactions of Cpd. 59
0.30
0.25
0.20
0.15
0.10
0.05
0.00
-0.05
isobutyl group were found to form short contacts with
amino acid residues of the active site. Hence, it may be
concluded that the best active compound may interact with
the binding site of L-glutamic acid because of much
structural similarity with the ligand and important moieties
like the phenyl ring and the sulphonyl group render the
better stability in the binding site of the protein.
1
2
3
MD simulation
1
2
3
Apoenzyme
Enzyme+Compound
Enzyme + Glutamic acid (substrate)
The geometries of the compound 59 was fully optimized
using the density functional theory method with the help of
Becke’s three parameter hybrid density functional, B3LYP/
6–31 G (d, p) using GAUSSIAN3 program (Frisch et al.
2003). This structure was subsequently docked at the glu-
tamic acid binding pocket of 3UNW. The best docked pose
showed Goldscore 64.57 which was much higher than that
of L-glutamic acid (40.94). The best docked pose (Fig. 8)
was found to form hydrogen bond interactions with Tyr249,
Ser286, Glu 381, Asn335, and Tyr466. Almost all polar
residues showed hydrogen bond acceptor or donor char-
acteristics. The sulfonyl group binds with Tyr466 whereas
the adjacent amido group interacts with Asn335. The car-
boxylic group was found to form hydrogen bond interaction
with Ser 286 and Lys289. The amido group attached to the
n-pentyl residue formed hydrogen bond interaction with
Tyr249 and Asn388.
1
51
101 151 201 251 301 351 401 451 501
Time (ps)
Fig. 9 RMSD plot to investigate the stability of the system
Based on these docking interactions, MD simulations
were done on the apo, glutamic acid bound and the com-
pound 59 bound complexes of glutaminase (KGA). These
simulations were performed using the GROMACS 4.0.4
program with GROMOS96 43al force field at constant
temperature and pressure ensemble. In the MD simulations,
all protein atoms were entrapped by a cubic water box of
SPC3 water molecules that extended 10 Å from the protein
and periodic boundary conditions were utilized in all
directions. These systems were balanced with Na+and Cl−
counter ions replacing the solvent molecules. The energy
minimization was performed using the steepest descent

Med Chem Res
Fig. 10 Structural requirements
of glutamine derivatives for
better anticancer activity
H-bond interaction
with Asn388
H-bond interaction
with Ser286
H-bond interaction
with Val484
Decreasing charge
favors activity
18
O
Steric effect
favorable
8
O
H
N
10
16
19
11
15
1
17
S
N
6
H
2
7
12
O
9
O
13
OH
14
5
3
4
H-bond interaction
with Glu381
H-bond interaction
with Tyr414
Nuclecophilic
attack favorable
Substitution of
the aromatic ring
favors activity
Increasing charge
favors activity
algorithm for 10000 steps. A 100 ps position restrained MD
simulations was performed for every system. It was fol-
lowed by 500 ps production of MD simulations with a time
step of 2 fs constant pressure (1 atm) and temperature (300
K). The electrostatic interactions were calculated by the
PME method. All bonds were constrained using LINCS
algorithm. The GROMACS topologies for ligands were
obtained from PRODRG web server.
The stability of the protein complexes in the MD simu-
lation was also monitored using RMSD calculation with
respect to their initial structure. The RMSD vs. time plot is
shown in Fig. 9.
The RMSD of glutaminase-L-glutamic acid substrate
complex has shown the lowest value compared to all other
systems. However, the complex of glutaminase and the best
active compound (compound 59) has shown RMSD value
higher than 2.0 Å initially but after 60 ps it maintained
consistently below 2.0 Å.
as molecular docking studies and MD simulation were
adopted here separately. The 2D-QSAR study reveals the
importance of the sulphonyl group, substitutions of the
phenyl ring and charges of the alkyl part of glutamine
moiety. The 3D-QSAR study by kNN-MFA mainly focused
in the steric and hydrophobic interactions that are close to
the substituted phenyl residue and aliphatic/aromatic groups
attached to the amido moiety of glutamine residue. The
docking study performed with the human kidney type glu-
taminase (KGA) enzyme depicts that the best active com-
pound 59 may bind to the L-glutamic acid binding site of
the protein. Apart from hydrogen bonding interactions
formed by electronegative elements like the oxygen and the
nitrogen atoms, close contacts formed by steric Van der
waals interaction render the better stability in the binding
site. The molecular docking study followed by molecular
dynamic simulation showed that these compounds may be
acting as glutaminase inhibitors through L-glutamic acid
binding site of glutaminase enzyme. Atoms and substituents
important for the better anticancer activity obtained in
molecular modeling studies of these glutamine analogs are
presented schematically on the best active compound
(compound 59) in Fig. 10. The current study may help in
further tailoring of this series of compounds for obtaining
compounds with the better anticancer activity.
Conclusion
The study carried out here give some important information
about the required structural features and possible interac-
tions of 5-N-substituted-2-N-(substituted benzenesulpho-
nyl)-L(+) glutamines. The highest antitumor activity was
found for the compound 59. Although there are consider-
able biological activity variations among these synthesized
compounds, lack of structural variations may limit finding
of necessary information from a single molecular modeling
approach. Therefore, a multiple ligand-based designing
methodology by 2D-QSAR and 3D-QSAR analyses as well
Acknowledgements Authors are thankful to the All India Council
for Technical Education (AICTE), New Delhi, Council for Scientific
and Industrial Research (CSIR), New Delhi and University Grants
Commission (UGC), New Delhi for providing financial support. Two
authors (AKH and NA) thank Council for Scientific and Industrial
Research (CSIR), New Delhi for providing a Senior Research Fel-
lowship (SRF). We are also thankful to the authority of Jadavpur
University for providing us the facility required for the work.

Med Chem Res
Compliance with ethical standards
DeBerardinis R, Sayed N, Ditsworth D, Thompson CB (2008b) Brick
by brick: metabolism and tumor cell growth. Curr Opin Genet
Dev 18:54–61
Conflict of interest The authors declare that they have no competing
Deswal S, Roy N (2006) Quantitative structure activity relationship
studies of aryl heterocycle-based thrombin inhibitors. Eur J Med
Chem 41:1339–1346
interests.
Dolinska M, Dybel A, Zablocka B, Albrecht J (2003) Glutamine
transport in C6 glioma cells shows ASCT2 system characteristics.
Neurochem Int 43:501–507
References
Eriksson L, Jaworska J, Worth AP, Cronin MT, McDowell RM, Gra-
matica P (2003) Methods for reliability and uncertainty assessment
and for applicability evaluations of classification- and regression-
based QSARs. Environ Health Perspect 111:1361–1375
Frisch MJ, Trucks GW, Schlegel HB, Scuseria GE, Millam MA,
Daniels AD, Kudin KN, Strain MC, Farkas O, Tomasi BJV,
Cossi M, Cammi R, Mennucci B, Pomelli C, Adamo C, Clifford
S, Ochterski J, Petersson GA, Ayala PY, Cui Q, Morokuma K,
Salvador P, Dannenberg JJ, Malick DK, Rabuck AD, Ragha-
vachari K, Foresman JB, Cioslowski J, Ortiz JV, Baboul AG,
Sefanov BB, Liu G, Liashenko A, Piskorz P, Komaromi I,
Gomperts R, Martin RL, Fox DJ, Keith T, A-Laham MA, Peng
CY, Nanayakkara A, Challacombe M, Gill PMW, Johnson B,
Chen W, Wong MW, Andes JL, Gonzalez C, Head-Gordon M,
Replogle ES, Pople JA (2003) Gaussian 03-Revision B.03.
Gaussian Inc, Pittsburgh
Adhikari N, Halder AK, Mondal C, Jha T (2013a) Exploring structural
requirements of aurone derivatives as antimalarials by validated
DFT-based QSAR, HQSAR, and COMFA-COMSIA approach.
Med Chem Res 22:6029–6045
Adhikari N, Halder AK, Mondal C, Jha T (2013b) Ligand based
validated comparative chemometric modeling and pharmaco-
phore mapping of aurone derivatives as antimalarial agents. Curr
Comput Aid Drug Des 9:417–432
Adhikari N, Halder AK, Mondal C, Jha T (2014) Structural findings of
quinolone carboxylic acids in cytotoxic, antiviral, and anti-HIV-1
integrase activity through validated comparative molecular
modeling studies. Med Chem Res 23:3096–3127
Adhikari N, Halder AK, Saha A, Saha KD, Jha T (2015) Structural
findings of phenylindoles as cytotoxic antimitotic agents in
human breast cancer cell lines through multiple validated QSAR
studies. Toxicol in Vit 29:1392–1404
Adhikari N, Jana D, Halder AK, Mondal C, Maiti MK, Jha T (2012)
Chemometric modeling of 5-Phenylthiophenecarboxylic acid
derivatives as anti-rheumatic agents. Curr Comput Aid Drug Des
8:182–195
Bhattacharya P, Maity P (2000) Localization of phosphate dependent
glutaminase in ascites fluid of ovarian cancer patient. Pathol
Oncol Res 6:217–223
Chakraborty D, Maity A, Jain CK, Hazra A, Bharitkar YP, Jha T,
Majumder HK, Roychoudhury S, Mondal NB (2015) Cytotoxic
potential of dispirooxindolo/acenaphthoquino andrographolide
derivatives against MCF-7 cell line. Med Chem Commun
6:702–707
Chakraborty D, Maity A, Jha T, Mondal NB (2014) Spermicidal and
contraceptive potential of desgalactotigonin: a prospective alter-
native of nonoxynol-9. Plos One 9:e107164
Chang WK, Yang KD, Chuang H, Jan JT, Shaio MF (2002) Glutamine
protects activated human T cells from apoptosis by up-regulating
glutathione and Bcl-2 levels. Clin Immunol 104:151–160
Chatterjee N, Das S, Bose D, Banerjee S, Jha T, Saha KD (2014)
Leishmanial lipid suppresses the bacterial endotoxin-induced
inflammatory response with attenuation of tissue injury in sepsis.
J Leuk Biol. 96:325–336
Chatterjee N, Das S, Bose D, Banerjee S, Jha T, Saha KD (2015)
Leishmanial lipid affords protection against oxidative stress
induced hepatic injury by regulating inflammatory mediators and
confining apoptosis progress. Toxicol Lett 232:499–512
Das S, Chatterjee N, Bose D, Banerjee S, Jha T, Saha KD (2015a)
Leishmanial sphingolipid induces apoptosis in Sarcoma 180
cancer cells through regulation of tumour growth via angiogenic
switchover. Tumor Biol 36:3109–3118
Das S, Chatterjee N, Bose D, Banerjee S, Jha T, Saha KD (2015b)
Antineoplastic impact of leishmanial sphingolipid in tumour
growth with regulation of angiogenic event and inflammatory
response. Apoptosis 20:869–882
Das S, Chatterjee N, Bose D, Banerjee S, Pal P, Jha T, Saha KD (2014)
Lipid isolated from a Leishmania donovani strain reduces Escher-
ichia coli induced sepsis in mice through inhibition of inflammatory
responses. Mediat Inflamm 2014, Article number 409694
DeBerardinis R, Lum JJ, Hatzivassiliou G, Thompson CB (2008a) The
biology of cancer: metabolic reprogramming fuels cell growth
and proliferation. Cell Metabol 7:11–20
Golbraikh A, Tropsha A (2002) Beware of q2! J Mol Graph Model
20:269–276
Gold Ver. 5.0.1, Astex technology, Cambridge, 2011. https://www.
ccdc.cam.ac.uk/solutions/csd-discovery/components/gold/
Halder AK, Saha A, Jha T (2013a) Exploring QSAR and pharmaco-
phore mapping of structurally diverse selective matrix
metalloproteinase-2 inhibitors. J Pharm Pharmacol 65:1541–1554
Halder AK, Saha A, Jha T (2013b) Exploration of structural and
physicochemical requirements and search of virtual hits for
aminopeptidase N inhibitors. Mol Divers 17:123–137
Halder AK, Saha A, Saha KD, Jha T (2015) Stepwise development of
structure–activity relationship of diverse PARP-1 inhibitors through
comparative and validated in silico modeling techniques and mole-
cular dynamics simulation. J Biomol Struct Dynam 33:1756–1779
Hassanein M, Hoeksema MD, Shiota M, Qian J, Harris BK, Chen H,
Clark JE, Alborn WE, Eisenberg R, Massion PP (2013) SLC1A5
mediates glutamine transport required for lung cancer cell growth
and survival. Clin Cancer Res 19:560–570
Hazra A, Mondal C, Chakraborty D, Halder AK, Bharitkar YP,
Mondal SK, Banerjee S, Jha T, Mondal NB (2015) Towards the
development of anticancer drugs from andrographolide: semi-
synthesis, bioevaluation, QSAR analysis and pharmacokinetic
studies. Curr Top Med Chem 15:1013–1026
Hemmateenejad B (2004) Optimal QSAR analysis of the carcinogenic
activity of drugs by correlation ranking and genetic algorithm-
based PCR. J Chemometr 18:475–485
Hess B, Kutzner C, van der Spoel D, Lindahl E (2008) GRGMACS 4:
Algorithms for highly efficient, load balanced, and scalable
molecular simulation. J Chem Theo Comput 4:435–447
Huang F, Zhao Y, Zhao J, Wu S, Jiang Y, Ma H, Zhang T (2014)
Upregulated SLC1A5 promotes cell growth and survival in col-
orectal cancer. Int J Clin Exp Pathol 7:6006–6014
Huntress EH, Carten FH (1940) Identification of organic compounds.
I. Chlorosulfonic acid as a reagent for the identification of Aryl
Halides. J Am Chem Soc 62:511–514
Jones G, Willet P, Glen RC (1995) Molecular recognition of receptor
sites using a genetic algorithm with a description of desolvation. J
Mol Biol 245:43–53
Jones G, Willet P, Glen RC, Leach AR, Taylor R (1997) Development
and validation of a genetic algorithm for flexible docking. J Mol
Biol 267:727–748

Med Chem Res
Klimberg VS, McClellan JL (1996) Claude H. Organ, Jr. Honorary
Lectureship. Glutamine, cancer, and its therapy. Am J Surg
172:418–424
Krogsgaard-Larsen P, Liljefors T, Madsen ULF (2002) Text book of
drug design & discovery. Taylor & Francis, London
Kumar A, Chowdhury SR, Chakrabarti T, Majumder HK, Jha T,
Mukhopadhyay S (2014) A new ellagic acid glycoside and DNA
topoisomerase IB inhibitory activity of saponins from Putranjiva
roxburghii. Nat. Prod. Commun 9:675–677
Roy PP, Roy K (2008) On some aspects of variable selection for partial
least squares regression models. QSAR Comb Sci 27:302–313
Samanta S, Srikanth K, Banerjee S, Debnath B, Gayen S, Jha T (2004)
5-N-Substituted-2-(substituted benzenesulphonyl) glutamines as
antitumor agents. Part II: synthesis, biological activity and QSAR
study. Bioorg Med Chem 12:1413–1423
Snedecor GW, Cochran WG (1967) Statistical methods. Oxford &
IBH, New Delhi
Van Der Spoel D, Lindahl B, Hess B, Groenhof G, Mark AE (2005)
Kumar A, Chowdhury SR, Jatte KK, Chakrabarti T, Majumder HK,
Jha T, Mukhopadhyay S (2015) Anthocephaline, a new indole
alkaloid and cadambine, a potent inhibitor of DNA topoisomerase
IB of Leishmania donovani (LdTOP1LS), isolated from Antho-
cephalus cadamba. Nat Prod Commun 10:297–299
GROMACS: Fast, flexible, and free.
26:1701–1718
Szeliga M, Obara-Michlewska M (2009) Glutamine in neoplastic cells:
focus on the expression and roles of glutaminases. Neurochem Int
55:71–75
J Comput Chem
Leach, AR (2001) Molecular modeling principles and applications,
2nd edn. Prentice Hall, England
Lu W, Pelicano H, Huang P (2010) Cancer metabolism: is glutamine
sweeter than glucose? Cancer Cell 18:199–200
Tetko IV, Tanchuk VY, Villa AE (2001) Prediction of n-octanol/water
partition coefficients from PHYSPROP database using artificial
neural networks and E-state indices. J Chem Inf Comput Sci
41:1407–1421
Marquez J, Sanchez-Jimenez FM, Medina A, Quesada AR, Nunez de
Castro I (1989) Nitrogen metabolism in tumor bearing mice. Arch
Biochem Biophys 268:667–675
Martin M, Beauvoit B, Voisin PJ, Canoini P, Guerin B, Rigoulet M
(1998) Energetic and morphological plasticity of C6 glioma cells
grown on 3-D support; effect of transient glutamine deprivation. J
Bioenerg Biomembr 30:565–578
Massiere F, Badet-Denisot MA (1998) The mechanism of
glutamine-dependent amidotransferases. Cell Mol Life Sci
54:205–222
Matsuno T, Hirai H (1989) Glutamine synthetase and glutaminase
activities in various hepatoma cells. Biochem Int 19:219–225
Mondal C, Halder AK, Adhikari N, Jha T (2013) Cholesteryl ester
transfer protein inhibitors in coronary heart disease: Validated
comparative QSAR modeling of N, N-disubstituted trifluoro-3-
amino-2-propanols. Comp Biol Med 43:1545–1555
Mondal C, Halder AK, Adhikari N, Jha T (2014) Structural findings of
cinnolines as anti-schizophrenic PDE10A inhibitors through
comparative chemometric modeling. Mol Divers 18:655–671
Mondal C, Halder AK, Adhikari N, Saha A, Saha KD, Gayen S,
Jha T (2015) Comparative validated molecular modeling of
p53-HDM2 inhibitors as antiproliferative agents. Eur J Med
Chem 90:860–875
Nargotra A, Koul S, Sharma S, Khan IA, Kumar A, Thota N, Koul JL,
Taneja SC, Qazi GN (2009) Quantitative structure-activity rela-
tionship (QSAR) of aryl alkenyl amides/imines for bacterial
efflux pump inhibitors. Eur J Med Chem 44:229–238
Newsholme P, Procopio J, Ramos Lima MS, Pithon-Curi TC, Curi R
(2003) Glutamine and glutamate--their central role in cell meta-
bolism and function. Cell Biochem Funct 21:1–9
Topliss RP, Edwards P (1979) Chance factors in studies of quantitative
structure-activity relationships. J Med Chem 22:1238–1244
Tropsha A (2003) Recent trends in quantitative structure-activity
relationships. In: Abraham DJ (ed) Burger’s medicinal chemistry
and drug discovery, vol 1. 6th edn. Wiley, New York, p 49–75
Tropsha A, Gramatica P, Gomber VK (2003) The Importance of Being
Earnest: Validation is the Absolute Essential for Successful
Application and Interpretation of QSPR Models. QSAR Comb
Sci 22:69–77
Turner A, McGivan JD (2003) Glutaminase isoform expression in cell
lines derived from human colorectal adenomas and carcinomas.
Biochem J 370:403–408
Venkatachalam CM, Jiang X, Oldfield T, Waldman M (2003)
LigandFit: a novel method for the shape-directed rapid docking of
ligands to protein active sites. J Mol Graph Model 21:289–307
VLife (2011) QSAR Plus 1.0 is a molecular modelling software of VLife
Sciences and Technologies, Pune, India. www.vlifesciences.com
Walker JD, Jaworska J, Comber MH, Schultz TW, Dearden JC (2003)
Guidelines for developing and using quantitative structure-
activity relationships. Environ Toxicol Chem 22:1653–1665
Wasa M, Bode BP, Abcouwer SF, Collins CL, Tanabe KK, Souba
WW (1996) Glutamine as a regulator of DNA and protein bio-
synthesis in human solid tumor cell lines. Ann Surg 224:189–197
Wasa M, Wang HS, Okada A (2002) Characterization of L-glutamine
transport by a human neuroblastoma cell line. Am J Physiol Cell
Physiol 282:C1246–C1253
Witte D, Ali N, Carlson N, Younes M (2002) Overexpression of the
neutral amino acid transporter ASCT2 in human colorectal ade-
nocarcinoma. Anticancer Res 22:2555–2557
Xie J, Li P, Gao HF, Qian JX, Yuan LY, Wang JJ (2014) Over-
expression of SLC38A1 is associated with poorer prognosis in
Chinese patients with gastric cancer. BMC Gastroenterol. 14:70
Zalkin H, Smith JL (1998) Enzymes utilizing glutamine as an amide
donor. Adv Enzymol Relat Areas Mol Biol 72:87–144
Ogura M, Takarada T, Nakamichi N, Kawagoe H, Sako A, Nakazato
R, Yoneda Y (2011) Exacerbated vulnerability to oxidative stress
in astrocytic C6 glioma cells with stable overexpression of the
glutamine transporter slc38a1. Neurochem Int 58:504–511