Molecular modeling, structure activity relationship and immunomodulatory properties of some lupeol derivatives

Article abstract of DOI:10.1007/s00044-012-0183-y

Currently, scientists are focused on developing drug-like compounds as an alternative to the available immunosuppressive drugs with less side effects. Therefore, in the current study we tried to generate derivatives of lupeol and investigate their primary effects on the immune system. In the second part, a computational approach, integrating molecular docking, and molecular dynamics simulation was used to assess the likely mechanism of action and the preliminary analysis of the structure activity relationship (SAR). Our goal for this research was to develop an in-depth SAR for the design of future immunosuppressive lupeol analogs with the potential clinical use.

Full text of DOI:10.1007/s00044-012-0183-y

MEDICINAL
CHEMISTRY
RESEARCH
Med Chem Res (2013) 22:1795–1803
DOI 10.1007/s00044-012-0183-y
ORIGINAL RESEARCH
Molecular modeling, structure activity relationship
and immunomodulatory properties of some lupeol derivatives
•
Mohsen Shahlaei Syed Mustafa Ghanadian Abdul Majid Ayatollahi
•
•
•
•
•
M. Ahmed Mesaik Omer Mohamed Abdalla Suleiman Afsharypour
Mohammed Rabbani
Received: 2 April 2012 / Accepted: 25 July 2012 / Published online: 5 August 2012
Ó Springer Science+Business Media, LLC 2012
Abstract Currently, scientists are focused on developing
drug-like compounds as an alternative to the available
immunosuppressive drugs with less side effects. Therefore,
in the current study we tried to generate derivatives of
lupeol and investigate their primary effects on the immune
system. In the second part, a computational approach,
integrating molecular docking, and molecular dynamics
simulation was used to assess the likely mechanism of
action and the preliminary analysis of the structure activity
relationship (SAR). Our goal for this research was to
develop an in-depth SAR for the design of future
immunosuppressive lupeol analogs with the potential
clinical use.
Keywords Drug design ꢀ Biologic screening ꢀ
Natural product ꢀ Molecular modeling ꢀ
Structure activity relationship
Introduction
Currently, triterpenes are being investigated in the treat-
ment of autoimmune and non-autoimmune inflammatory
diseases like allergic asthma (Ayatollahi et al., 2011). In
such diseases, neutrophils, as a source of inflammatory
eicosanoids, cytokines, and tissue-degrading lysosomal
enzymes, appear shortly after an irritating or noxious
stimulus. Activation of primed neutrophils, either by
phagocytosis of opsonized bacteria or by frustrated
phagocytosis, generates the rapid production of ROS via
the action of NADPH oxidase. Oxygen radicals cause
oxidative damages to DNA, lipids, proteins, lipoproteins,
and may be implicated in mutations to immunoglobulins
leading toward the formation of rheumatoid factor (Helen
et al., 2010.). In such reactions, an increased level of cAMP
has a suppressive effect on the oxidative burst in neutro-
phils (Bani et al., 2006) where it is reported that cAMP-
dependent PKA phosphorylates a protein (RAP la)
belonging to the Ras superfamily which, in turn, blocks
assembly of NADPH oxidase components into the active
enzymatic complex (Ottonello et al., 1995). Similarly, in
peripheral blood lymphocytes, allergen-induced synthesis
of IL-2, IL-4, and IL-5 is inhibited by cAMP elevating
factors (Yeadon and Diamant, 2000). In this study, we tried
to investigate the phagocytic oxidative burst and the lym-
phocyte proliferation properties of some Lupeol derivatives
as well as their intermolecular interactions with adenyl
cyclase (AC) (Fig. 1) to introduce new anti-inflammatory
M. Shahlaei
Department of Medicinal Chemistry, Faculty of Pharmacy,
Kermanshah University of Medical Sciences, Kermanshah,
Islamic Republic of Iran
S. M. Ghanadian (&)
Isfahan Pharmaceutical Sciences Research Center,
Isfahan University of Medical Sciences, Isfahan,
Islamic Republic of Iran
e-mail: phytochemistry.center@gmail.com;
ghanadian@pharm.mui.ac.ir
A. M. Ayatollahi
Phytochemistry Research Center, Shahid Beheshti University
of Medical Sciences, Tehran, Islamic Republic of Iran
M. A. Mesaik ꢀ O. M. Abdalla
Dr. Panjwani Center for Molecular Medicine and Drug Research,
International Center for Chemical and Biological Sciences,
University of Karachi, Karachi 75270, Pakistan
S. Afsharypour ꢀ M. Rabbani
Faculty of Pharmacy and Pharmaceutical Sciences,
Isfahan University of Medical Sciences, Isfahan,
Islamic Republic of Iran
123

1796
Med Chem Res (2013) 22:1795–1803
100 %. Hexane/chloroform (20:80) fraction was further
purified using gradient mixtures of hexane/ethyl acetate
(0 ? 50) on silica to afford several fractions. Inferred by
1H-NMR spectra, Fr.2 and Fr.5 fractions containing tri-
terpenes were further purified on Sephadex (dichloro-
methane/methanol, 1:2) to yield subfraction rich in
triterpenes, visible as violet spots on thin layer chroma-
tography (TLC). Finally, Fr.2a and Fr.5a were subjected to
recycle HPLC Jaigel-2H&1H columns with chloroform as
mobile phase to obtain compounds 1, and 2 identified by
comparing their NMR, Mass, and spectroscopic data with
those reported in the literature (Macias et al., 1994;
´
Domınguez-Carmona et al., 2010; Viqar and Rahman,
1994).
Synthesis of derivatives
Ethyl betulinate (3). A mixture of betulinic acid (13.5 mg),
K2CO3 (200 mg), C2H5I (1 mL), and acetone (1 mL) was
stirred for 72 h at room temperature. When reached to the
end point, the reaction mixture was poured over distilled
water (13 mL) and extracted twice with ethyl acetate (4:1).
The organic layer was dried with anhydrous Na2SO4, fil-
tered, and the solvent was evaporated to obtain esterified
product. Then, the esterified compound was purified more
through preparative layer chromatography (hexane: ethyl
acetate, 7:3) and confirmed by H-NMR and mass spectra
Fig. 1 Chemical structures of lupeol (1), betulinic acid (2), ethyl
betulinate (3), and lupenyl acetate (4)
and immunosuppressive agents. Recently, pentacyclic tri-
terpenes have been widely reported to exhibit anti-inflam-
matory activities (Yogeeswari and Sriram, 2005;
Einzhammer and Xu, 2004). Based on these studies,
changes in the size of ring E, or the nucleophilicity, and the
strength of hydrogen bonding capability or acidity at C-19,
C-20, and C-28 positions are responsible for their differ-
ences in biologic effects. Therefore, making derivatives by
modification in these positions are assumed to improve
their immunomodulatory activities.
´
(Domınguez-Carmona et al., 2010).
Lupenyl acetate (4). 10 mg, separately with acetic anhy-
dride (1 mL) and pyridine (0.5 mL), were stirred at room
temperature for 3 days. At the final point of reaction, the
mixture was poured over ice and extracted twice with ethyl
acetate (2:1 v/v). The organic phase was washed with equal
volumesofHCl(5 %), NaOH(3 %), H2O,andNaClsaturated
followed by drying over anhydrous Na2SO4. After filtration,
the solvent evaporatedin vacuo to yield the acetylated product
which was then confirmed through H-NMR and Mass spectra
Materials and methods
Extraction and isolation of lupane-type triterpenes
Aerial flowering parts of Euphorbia aellenii Rech. F.
(Euphorbiaceae) were collected in August 2007 from
Northern Khorasan province (Iran) and a voucher specimen
(no. 2024) of the plant has been deposited in the herbarium
of the Pharmacognosy Department, School of Pharmacy
and Pharmaceutical sciences at the Isfahan University of
Medical Sciences (Iran). Air-dried plant was ground to fine
powder (7 kg) and macerated for 4 days with MeOH
(20 L 9 3) at room temperature. Filtration and in vacuo
evaporation resulted in a green gum (500 g) which was
partitioned between methanol and n-hexane. The defatted
extract was concentrated, dissolved in water, and succes-
sively extracted with chloroform, ethyl acetate, and
n-butanol. The chloroform fraction (240 g) was subjected
to column chromatography over the normal silica gel using
hexane/chloroform mixtures of increasing polarity up to
´
(Domınguez-Carmona et al., 2010).
Lupeol (1)
White powder, mp 215–217; IR (KBr) k max: 3440, 1648,
1605, 1280, 1025, 880 cm-1; 1H-NMR (CDCI3, 300 MHz)
dH, mult., J in Hz 4.67, 4.55 (each 1H, dd, J = 2.4, 1.5,
H-29), 3.17 (1H, dd, J = 10.8, 5.2, H-3), 2.36 (1H, dt,
J = 10.7,6.0, H-19), 1.71 (3H, s, H-30), 1.01, 0.96, 0.92
(each 3H, s, H-2, 27, 23), 0.81, 0.77, 0.74 (each 3H, s,
H-25, 28, 24); 13C-NMR (CDCI3, 100 MHz) dC 38.69 (C-
1), 27.4 (C-2), 78.9 (C-3), 38.9 (C-4), 55.3 (C-5), 18.3 (C-
6), 34.2 (C-7), 40.8 (C-8), 50.4 (C-9), 37.2 (C-10), 20.9
(C-11), 25.1 (C-12), 38.0 (C-13), 42.8 (C-14), 27.4 (C-15),
35.5 (C-16), 43.0 (C-17), 48.3 (C-18), 48.0 (C-19), 151.0
(C-20), 29.8 (C-21), 40.0 (C-22), 27.9 (C-23), 16.0 (C-24),
123

Med Chem Res (2013) 22:1795–1803
1797
18.0 (C-25), 16.1 (C-26), 14.6 (C-27), 15.4 (C-28), 109.3
(C-29), 19.3 (C-30); EI-HRMS of m/z 426.6998 (calc. for
C30H50O, 426.7194); EI-MS m/z 426, 424, 411, 316, 315,
257, 234, 229, 218, 205, 203, 189, 175, 149, 147, 133, 123,
121, 119, 105, 95, 55.
compound (0.5, 5, and 50 lg/mL) were prepared in 25 lL
of Hank’s Buffered Salt Solution containing calcium
chloride and magnesium sulfate (HBSS??) in 96-well flat-
bottomed plates for a final incubation volume of 100 lL.
Then, 25 lL of whole blood diluted 1:50 in suspension of
HBSS?? was added. Positive, negative, and blank control
wells were included in the assay. Whole blood phagocytes
and compounds were incubated for 30 min at 37 °C, then
25 lL luminol (Research Organics, Cleveland, OH, USA)
was added into each well, and 25 lL serum opsonized
zymosan (Fluka, Buchs, Switzerland) was finally added
except for negative and blank wells. Kinetic production of
the reactive oxygen species was monitored using the
luminometer (Labsystems luminoskan, Helsinki, Finland)
for 50 min in the repeated scan mode. Peak and total
integral chemiluminescence reading were expressed as
relative light unit (RLU).
Betulinic acid (2)
Needle crystalls, mp 294–296; IR (KBr) k max: 3446,
2924, 2862, 1685, 1645 1604, 1236, 1107, 1034; H-NMR
1
(CDCI3, 300 MHz) dH, mult., J in Hz: 4.72, 4.59 (each 1H,
br-s, H-29), 3.17 (1H, dd, J = 5.1, 10.8 Hz, H-3), 2.98
(1H, dt, J = 11.1, 3.0, H-19), 1.67 (3H, s, H-30), 0.95,
0.94, 0.91 (each 3H,s, H-26, 27, 23), 0.80, 0.73 (each 3H, s,
H-25, 24); 13C-NMR (CDCI3, 100 MHz) dc : 38.6 (C-1),
27.0 (C-2),78.8 (C-3), 38.7 (C-4), 55.3 (C-5), 18.2 (C-6),
34.2 (C-7), 40.6 (C-8), 50.4 (C-9), 37.0 (C-10), 20.8
(C-11), 25.4 (C-12), 38.2 (C-13), 42.3 (C-14), 30.5 (C-15),
32.1 (C-16), 56.1 (C-17), 49.1 (C-18), 46.9 (C-19), 150.6
(C-20), 37.0 (C-21), 29.6 (C-22), 27.5 (C-23), 15.2 (C-24),
15.8 (C-25), 16.0 (C-26), 14.5 (C-27), 179.1 (C-28), 19.2
(C-30), 109.4 (C-29). EI-HRMS of m/z 456.3606 (calc. for
C30H48O3, 456.3603); EI-MS m/z 456, 438, 423, 411, 395,
369,356, 302, 248, 220, 207, 203, 189, 175, 135, 95, 69, 55.
Lymphocyte proliferation assay
Peripheral human blood lymphocytes were incubated with
different concentrations of the test compounds (0.5, 5, and
50 lg/mL in duplicates) in supplemented (10 % Fetal
Bovine Serum) RPMI-1640 along with 5.0 lg/mL phyto-
hemagglutinin (Sigma Aldrich incorporation, USA) at
37 °C in CO2 environment for 72 h. Further incubation for
18 h after the addition of 0.5 lCi/well 3H-thymidine
(Amersham, Buckinghamshire, UK) was done and cells
were harvested using the cell harvester (Inotech Dottikon,
Switzerland). Finally, the proliferation level was deter-
mined by the radioactivity level as count per minute (CPM)
recorded from the b-scintillation counter (Beckman coul-
ter, LS 6500, Fullerton, CA, USA) (Nielsen et al., 1998).
Ethyl betulinate (3)
1H-NMR (CDCI3, 500 MHz) dH, mult., J in Hz: 4.71 (1H,
br-s, H-29a), 4.61 (1H, br-s, H-29b), 4.15 and 4.08 (each
1H, m, 28-COO-Et-H-10a & H-10b), 3.15 (1H, dd, J = 5.0,
11.5 Hz, H-3), 2.98 (1H, m, H-19), 1.66 (3H, s, H-30), 1.23
(3H, m, 28-COO-Et-H-2’), 0.94 (6H, br-s, H-26, 23), 0.89
(3H, s, H-27), 0.79 (3H, s, H-25), 0.73 (3H, s, H-24);
CI-MS m/z 485 (M ? 1), 467, 409, 393, 275, 206.
Molecular docking
Lupenyl acetate (4)
Using the crystal structure, AC5 protein retrieved from the
Protein Data Bank (PDB code: 1CJV), lupeol derivatives
(1–3) were selected for docking process to study the
molecular interactions between compounds of interest and
their site of action. The optimized structures by a semi
empirical method (AM1), implemented in HyperChem,
were used as input of Auto Dock Tools and the partial
charges of atoms were calculated by Gasteiger–Marsili
procedure (1980). After determining Kollman united atom
charges (Weiner et al., 1984), non-polar hydrogens were
merged to their corresponding carbons. As a final pro-
cessing step in protein preparation, desolvation parameters
were assigned to each atom. By means of auto grid tools,
the grid maps were generated adequately large to include
the active site of protein, as well as significant regions of
the surrounding surface. In all cases, a grid map of 60 grid
1H-NMR (CDCI3, 300 MHz) dH, mult., J in Hz 4.67, 4.56
(each 1 H, br-d, J = 2.5 Hz, H-29), 3.17 (dd, J = 11.0,
5.2, H-3), 2.04 (3H, s, COCH3), 1.71, 1.00, 0.96 (each 3H,
s, H-30, 26, 23), 0.92 (3H, s, H-27), 0.81, 0.77, 0.74 (each
3H, s, H-25, 28, 24); EI-MS m/z 468, 453, 357, 297, 276,
249, 218, 203, 189, 69, 55.
Biologic assays
Phagocyte chemiluminescence assay
As a parameter for evaluating phagocytosis in stimulated
polymorphonuclear cells, formation of the reactive oxi-
dants in whole blood was measured by the Luminol-
enhanced chemiluminescence assay procedure (Mischak
et al., 1993). In brief, three concentrations of each
˚
points and a grid spacing of 0.375 A were applied in each
123

1798
Med Chem Res (2013) 22:1795–1803
Cartesian direction. Considering the original ligand loca-
tion in the complex, the maps were centered on the ligand’s
binding site, searching favorable interactions with the
functional groups. Based on Lamarckian Genetic Algo-
rithm (Morris et al., 1998), by means of the pseudo-Solis
and Wets local search method (Solis and Wets, 1981), Auto
Dock Tools were employed to produce both grid and
data supported one double bond and therefore, five rings in
the molecule. Seven singlet methyls at dH 1.71, 1.01, 0.96,
0.92, 0.81, 0.77, and 0.74, along with a pair of olefinic
protons as a part of an exocyclic-methylene group at dH
4.67 and 4.55 together with a doublet of triplet of one
proton at dH 2.36 (J = 10.7, 6 Hz) characteristic for lupane
triterpenes (Ayatollahi et al., 2011) were observed in the
1H-NMR spectrum. The existence of one hydroxyl group
was also evident by NMR signals at dH 3.17. Taken toge-
ther, the spectral data of 1 was in complete accordance with
those previously reported for Lupeol (Viqar and Rahman,
1994). Compound 2 was also identified as betulinic acid
based on its similarities with compound 1 except for
additional acidic carbonyl group at dc 179.1 instead of
C-28 singlet methyl at dc 15.4 (Viqar and Rahman, 1994).
The structures of derivatives were also confirmed by
comparing their 1H-NMR and mass data with those
˚
docking parameter files. Applying 2.0 A clustering toler-
ance to construct clusters of the closest compounds, the
initial coordinates of the ligand were used as the reference
structure. Finally, docking results were investigated by
VMD1.8.6 (Humphrey et al., 1996).
Molecular dynamics simulation
Molecular dynamics (MD) simulation was carried out by
means of GROMOS96 force-field through the GROMACS
software (Ver. 4.5, www.gromacs.org). The structure of
adenyl cyclase 5 and ethyl betulinate ligand, as a repre-
sentative, obtained from the docking procedure were used
as a starting point for the MD simulation. Protein and
ligand were soaked in a cubic box of SPC (Single Point
Charge) water molecules (Berendsen et al., 1981) and
simulated using periodic boundary conditions. All the
protein atoms were at a distance equal or greater than
0.7 nm from the box edges. MD simulation was carried in
the isobaric-isothermal ensemble (NPT: 300 K and 1 bar).
By means of the productive MD, pressure was kept con-
stant at 1 bar by altering the box dimensions and the time
constant for pressure coupling was set to 1 ps. The LINCS
(Hess et al., 1997) algorithm was employed to constrain
bond lengths, permitting the application of 2 fs time step.
The Particle Mesh Ewald method (PME) was used in the
computation of the electrostatic forces (Darden et al.,
1993). Van der Waals and Coulomb interactions were
´
reported in the literature (Domınguez-Carmona et al.,
2010; Viqar and Rahman, 1994).
Biologic assays
Phagocyte chemiluminescence
Phagocytic cells, on activation, produce reactive oxygen-
free radicals which are then quantified by a luminol-
enhanced chemiluminescence assay. A comparison of the
phagocytic oxidative burst activity in the presence of
varying concentrations (0.5–50 lg/mL) of compounds
(1–4) is shown in Fig. 2. Lupeol (1) and betulinic acid (2)
showed a dose-dependent decrease in the oxidative burst in
neutrophils with IC50: 11.35 3.47, and 11.12 1.77,
respectively. Phagocyte chemiluminescence decreased
significantly (41.12 2.32 % at 50 lg/mL) by masking
the free 3-OH group in lupenyl acetate (4) and substitution
of C-28 hydrophilic acidic group in betulinic acid with
esteric group in ethyl betulinate (3) showed further sup-
pressive activity with IC50 value of 3.85 1.30 lg/mL.
˚
calculated within a cutoff of 10 A.
Statistical analysis
The IC50 values were calculated by means of a Excel-based
program and reported as mean SD of the mean. Signif-
icance was attributed to p-values (p \ 0.05), and the
probability values were obtained by the Student t test
between sample and control data.
T cell proliferation assay
The anti-proliferation effect of the test compounds was
determined by measuring the PHA-induced T cell prolif-
eration by determining radioactive thymidine incorpora-
tion. Lupeol (1) showed a dose-dependent decrease in
lymphocyte proliferation with IC50 of 13.16 0.9 lg/mL
in conformity with another study by Bani et al. (2006)
which showed suppressive action against cytotoxic CD8?
and helper CD4? T cells in balb/c mice. In comparison,
lupenyl acetate (4), with masked 3-OH group, showed a
significantly decreased inhibitory effect (38.92 9.39 %
at 50 lg/mL), while ethyl betulinate (3), with esteric
amphoteric group at C-28, presented an improved
Results and discussion
Chemistry
Compound 1 was identified as C30H50O (calculated:
426.7194) based on positive EI-HRMS of m/z 426.6998.
Six degrees of unsaturation, 13C-NMR, and DEPT spectral
123

Med Chem Res (2013) 22:1795–1803
1799
Fig. 2 Effect of compounds
1–4 on the neutrophils oxidative
burst. The luminol-dependent
chemiluminescence induced by
zymosan in the presence of
lupeol (1), betulinic acid (2),
ethyl betulinate (3), and lupenyl
acetate (4) compared to the
controls (?ve, -ve). Data were
presented as mean SD for
three measurements. One-way
ANOVA was used to calculate
p value against the positive
control (*p \ 0.05, **p \ 0.01,
and ***p \ 0.001). RLU
relative light units
inhibitory effect with IC50 value of 12.9 0.8 lg/mL.
Addition of betulinic acid (2) in the concentration ranges
0.5, 5 and 50 lg/mL resulted in stimulation of lymphocyte
proliferation at 50 lg/mL by 38.2 2.78 % agreed with
another report on increasing cellular and humoral immu-
nity by betulinic acid that stimulated the proliferation of
mice thymocytes, splenocytes, and human PBMC in a time
and dose-dependent manner (Yi et al., 2010).
Molecular docking
Fig. 3 Internal validation phase result; conformational overlaid of
5’-guanosine-diphosphate-monothioohosphate retrieved from crystal
structure (red structure) with that from Auto Dock model (blue
structure)
Through adenylyl cyclase (AC) pathway, cyclic AMP-
elevating agents downregulate the oxidative burst induced
by neutrophils (Bani et al., 2006). The activation of adenyl
cyclase converts adenosine triphosphate (ATP) into cyclic
adenosine monophosphate (cAMP) which starts a negative
signal pathway that reduces T cell activation process (Bani
et al., 2006) and cytokine production (Aandahl et al.,
2002). Among different known types of AC in mammals,
type 5 adenylyl cyclase (AC5) is known to be widely
distributed in the body with multiple regulatory functions
including immunomodulatory action (Ruppelt et al., 2007).
Clarification of ligand binding mechanisms is one of the
essential steps to introduce more selective and potent lead
compounds for a given target protein. To facilitate this,
computational methods, such as docking and MD simula-
tion, are among the best approaches for the estimation of
protein–ligand interactions. First, for the validation of
selected parameters and condition of docking procedure, an
internal validation phase was carried out. In this step of
geometries for each investigated ligand were separated into
clusters according to RMSD tolerance criterion. Besides
RMSD cluster analysis, Auto Dock also applies binding-
free energy assessment to assign the best binding confor-
mation. Energy items estimated by Auto Dock are described
by intermolecular energy (including van der Walls energy,
hydrogen bonding energy, desolvation energy, and electro-
static energy), internal energy, and torsional free energy
(Table 1). In addition, with respect to the docking results,
the procedure successfully placed the ligands into a binding
cavity among the Asn98, Leu298, Ala298, and Glu50
(Figs. 4, 5).
Molecular dynamics simulation and structure activity
relationship
docking,
50-Guanosine-diphosphate-monothioohosphate
As discussed above, we initially applied simple docking of
the ligand into the AC5 binding pocket. These results were
not conclusive because binding of ligand to a protein is a
dynamic process, and molecular docking alone can only
stand for instantaneous states and cannot provide complete
information on the best stable binding modes of ligand to
the receptor (Shahlaei et al., 2010a, b). Therefore,
was docked onto AC5 (the lowest energy pose for docking
is shown in Fig. 3). Superimposing the experimental and
predicted conformations, the RMSD was achieved as
˚
1.16 A which considered as a successfully docked
(Erickson et al., 2004). After internal validation phase,
docking process was performed on AC5 binding pocket
using tested compounds (1–4). The 200 docking
123

1800
Med Chem Res (2013) 22:1795–1803
Table 1 Values of various energies (in Kcal/mol) resulted from docking process for compound 1–4
Compound
Final intermolecular
energy
vdW ? Hbond ? desolv
energy
Electrostatic
energy
Internal energy
Torsional
free energy
Binding
energy
Docking
energy
1
2
3
4
-5.78
-4.84
-5.85
-6.18
-5.77
-5.23
-6.00
-6.02
-0.01
0.38
-0.46
-0.37
-0.68
-0.53
2.98
2.98
2.98
2.98
-2.8
-6.24
-5.21
-6.53
-6.71
-1.86
-2.87
-3.2
0.15
-0.16
Fig. 4 Docking simulation
result of ethyl betulinate
protein (s), is that such MD simulations give complete
details of the motions of protein and ligand (s) in the
system so the thermodynamic properties could be esti-
mated. Figure 6 shows the root mean-square deviation
(RMSD) of the Ca atoms of protein in the simulation with
respect to AC5 in the crystal structure. Within the first 2 ns
of MD, the RMSD with respect to the initial structure
˚
rapidly increases to a value near 2.7 A. After fluctuating
around this value for a further 2 ns, the RMSD was not
increased significantly. The RMSD value implies that this
protein structure has been affected by its docked ligand and
environment dramatically.
Fig. 5 2D scheme of interactions between ethyl betulinate and active
site
molecular dynamics simulation was carried out on the most
effective compound, ethyl betulinate (compound 3) to gain
further insight into the mechanism (s) that could explain
the experimentally observed activity and to cross check our
foundlings from docking. Worthy of attention, that what
we can expect from molecular dynamics simulations of
Fig. 6 Ca RMSD of adenyl cyclase 5 during the molecular dynamic
simulation
123

Med Chem Res (2013) 22:1795–1803
1801
It can be seen that the protein–ligand complex immersed
among water molecules stays in equilibrium throughout the
MD. Then, we deduce that the MD simulation has con-
structed an improved and more relaxed protein–ligand
complex, which can be analyzed for further studies. The
increase in the RMSD value at the beginning of the sim-
ulation shows that AC5 adopts a much more extended
conformation.
Comparison between the Ramachandran plots of protein
before and after MD cleared that there are some differences
in the backbone of dihedral angle values which slightly
affect the position of some residues in the Ramachandran
plots. As it can be seen in Fig. 7a, b, the presence of the
ligand in the active site of protein during MD simulation
leads to the displacement of these residues into the disal-
lowed regions. In other words, before the beginning of MD,
the protein has no residues in the generously allowed or
disallowed regions of the Ramachandran plot, but after the
MD, 1.6 % of the residues move to these unsuitable
regions.
Fig. 8 RMSD of ethyl betulinate during the 8 ns molecular dynamic
simulation
docking of ligands in the binding pocket. Explorative run
of MD on the complex between the protein and the
investigated ligand revealed that almost all residues of the
active site determined by docking were changed and some
new residues such as Arg197, Glu198, and Asp199 are
positioned in proximity of ligands and could participate in
the interaction (Fig. 9).
The RMSD of the ligand was calculated based on the
MD simulation of the system to obtain information on
position fluctuations and movements of ligand’s atoms.
Figure 8 showed that the RMSD of the ethyl betulinate
˚
increased from the initial conformation to 0.1 A after
˚
2.5 ns and dropped suddenly to 0.4 A. After fluctuating at
Arg140 assembles one hydrophilic interaction with ligand,
which accommodates the hydroxyl moiety of the ligand in
position3.Hence,ifwechangehydroxylgroupofthisposition
with hydrophobic groups, this interaction is vanished and
reduction of effect is observed. Alternatively, in position 28,
which has two major interactions including hydrophobic
interaction with Val195 and Met194 and hydrophilic inter-
action with Asn200 and Asp199 (Fig. 10), substitution with a
group with both hydrophilic and hydrophobic characteristics,
such as ethyl acetate, the maximum of interactions will occur.
Accordingly, phagocytic oxidative burst activity by betulinic
acid (2) with the acidic hydrophilic group on C-28 or lupeol
˚
about 3 ns, it rose up again to about 0.8 A. This indicated
that, after 8 ns simulation and preliminary fluctuations in
the magnitude of RMSD of ligand’s atoms, the ligand
obtained an equilibrium state characterized by the RMSD
profile.
At the end of MD, position and orientation of ligand in
the introduced binding site were changed, and this impor-
tant observation indicates useful application of MD after
Fig. 7 Ramachandran plot of adenyl cyclase type 5 protein; a before and, b after molecular dynamic simulation
123

1802
Med Chem Res (2013) 22:1795–1803
Fig. 9 3D representation of
ethyl betulinate in the binding
site of adenyl cyclase 5 after
molecular dynamic simulation
Fig. 10 2D representation of
interactions between ethyl
betulinate and adenyl cyclase 5
with the methyl hydrophobic group are weaker than ethyl
betulinate (3) with the amphoteric esteric group on C-28
(Fig. 1). It was in agreement with other reports on C-28 amide
derivatives of betulinic acid reported as more potent inhibitors
of tumor cell growth (Shintyapina et al., 2007). To date, there
existotherreportsalsointheliteraturedescribingthechemical
modification of lupane-type triterpenes to produce derivatives
with enhanced anti-HIV and cytotoxic activities (Baglin,
2003), but little is known about the importance of the various
functional groups in this skeleton for the expression of
immunomodulatory properties, especially on C-28; There-
fore, testing other derivatives by modification on this position
is suggested to improve their immunosuppressant effects.
Conclusion
By means of a computational approach, integrating
molecular docking and molecular dynamics simulation of
lupeol derivatives on adenyl cyclase 5 provided strong
123

Med Chem Res (2013) 22:1795–1803
1803
Lobstein A, Schoonjans K, Souchet M, Auwerx J, Saladin R, Wagner
A (2010) Structure-activity relationship study of betulinic acid, a
novel and selective TGR5 agonist, and its synthetic derivatives:
potential impact in diabetes. J Med Chem 53:178–190
Macias FA, Simonet AM, Esteban DM (1994) Potential allelopathic
lupane triterpenes from bioactive fractions of Melilotus messan-
ensis. Phytochemistry 36:1369–1379
evidence for these compounds as potential activators of
AC. These results are consistent with other studies on the
activation of AC and increasing cAMP level as possible
routes for the reduction of neutrophil oxidative burst or
regulation of T cell proliferation (Bani et al., 2006; Yeadon
and Diamant, 2000).
Mischak H, Goodnight JA, Kolch W, Martiny-Baron G, Schaechtle C,
Kazanietz MG, Blumberg PM, Pierce JH, Mushinski JF (1993)
Overexpression of protein kinase C-delta and epsilon in NIH 3T3
cells induces opposite effects on growth, morphology, anchorage
dependence, and tumorigenicity. J Biol Chem 268:6090–6096
Morris GM, Goodsell DS, Halliday RS, Huey R, Hart WE, Belew RK,
Olson AJ (1998) Automated docking using a lamarckian genetic
algorithm and an empirical binding free energy function.
J Comput Chem 19:1639–1662
Acknowledgments The authors are grateful to prof. Dr. Muham-
mad Iqbal Choudhary, and the H. E. J. Research Institute of Chem-
istry, International Center for Chemical and Biological Sciences,
University of Karachi, for their financial supports.
References
Nielsen MB, Gerwien J, Nielsen M, Geisler C, Ropke C, Svejgaard A
(1998) MHC class II ligation induces CD58 (LFA-3)-mediated
adhesion in human T cells. Exp Clin Immunogenet 15:61–68
Ottonello L, Morone MP, Dapino P, Dallegri F (1995) Cyclic AMP-
elevating agents down-regulate the oxidative burst induced by
granulocyte-macrophage colony-stimulating factor (GM-CSF) in
adherent neutrophils. Clin Exp Immunol 101:502–506
Aandahl EM, Moretto WJ, Haslett PA, Vang T, Bryn T, Tasken K,
Nixon DF (2002) Inhibition of antigen-specific T cell prolifer-
ation and cytokine production by protein kinase A type I.
J Immunol 169:802–808
Ayatollahi AM, Ghanadian M, Afsharypour S, Abdalla OM, Mirzai
M, Askari G (2011) Pentacyclic triterpenes in Euphorbia
microsciadia with their t-cell proliferation activity. Iran J Pharm
Res 10:287–294
Baglin I, Mitaine-Offer AC, Nour M, Tan K, Cave C, Lacaille-Dubois
MA (2003) A review of natural and modified betulinic acid,
ursolic and echinocystic acid derivatives as potential antitumor
and anti-HIV agents. Mini Rev Med Chem 3:525–539
Bani S, Kaul A, Khan B, Ahmad SF, Suri KA, Gupta BD, Satti NK,
Qazi GN (2006) Suppression of T lymphocyte activity by lupeol
isolated from Crataeva religiosa. Phytother Res 20:276–287
Berendsen H, Postma J, van Gusteren W, Hermans J (1981)
Interaction models for water in relation to protein hydratation.
Reidel, Dordrecht
¨
Ruppelt A, Mosenden R, Gronholm M, Aandahl EM, Tobin D,
´
´
Carlson CR, Abrahamsen H, Herberg FW, Carpen O, Tasken K
(2007) Inhibition of T cell activation by cyclic adenosine
5’-monophosphate requires lipid raft targeting of protein kinase
A type I by the A-kinase anchoring protein ezrin. J Immunol
179:5159–5168
Shahlaei M, Madadkar-Sobhani A, Fassihi A, Saghaie L (2011a)
Exploring a model of a chemokine receptor/ligand complex in an
explicitmembraneenvironmentbymoleculardynamicssimulation:
the human CCR1 receptor. J Chem Inf Model 51:2717–2730
Shahlaei M, Madadkar-Sobhani A, Mahnam K, Fassihi A, Saghaie L,
Mansourian M (2011b) Homology modeling of human CCR5
and analysis of its binding properties through molecular docking
and molecular dynamics simulation. Biochim Biophys Acta
Biomembr 1808:802–817
Darden T, York D, Pedersen L (1993) Particle mesh Ewald: an
Nꢀlog(N) method for Ewald sums in large systems. J Chem Phys
98:10089–10092
Shintyapina AB, Shults EE, Petrenko NI, Uzenkova NV, Tolstikov
CA, Pronkina NV, Kozhevnikov VS, Pokrovsky AG (2007)
Effect of nitrogen-containing derivatives of the plant triterpenes
betulin and glycyrrhetic acid on the growth of MT-4, MOLT-4,
CEM, and Hep G2 tumor cells Russ. J Bioorg Chem 33:579–583
Solis FJ, Wets JB (1981) Minimization by random search techniques.
Math Oper Res 6:19–30
Viqar UA, Rahman AU (1994) Handbook of natural products data,
vol 2. Elsevier Sciences, Amsterdam
Weiner SJ, Kollman PA, Case DA, Singh UC, Ghio C, Alagona G,
Profeta S, Weiner PK (1984) A new force-field for molecular
mechanical simulation of nucleic-acids and proteins. J Am Chem
Soc 6:765–784
Yeadon M, Diamant Z (2000) New and exploratory therapeutic agents
for asthma. Marcel Dekker, New York
Yi JE, Obminska-Mrukowicz B, Yuan LY, Yuan H (2010) Immu-
nomodulatory effects of betulinic acid from the bark of white
birch on mice. J Vet Sci 11:305–313
Yogeeswari P, Sriram D (2005) Betulinic acid and its derivatives:
a review on their biological properties. Curr Med Chem 12:
657–666
´
Domınguez-Carmona DB, Escalante-Erosa F, Garcıa-Sosa K,
Ruiz-Pinell G, Gutierrez-Yapu D, Chan-Bacab MJ, Gimenez-
´
´
Turba A, Pena-Rodriguez LM (2010) Antiprotozoal activity of
betulinic acid derivatives. Phytomedicine 17:379–382
Einzhammer DA, Xu ZQ (2004) Betulinic acid:
anticancer candidate. Idrugs 4:359–373
a promising
Erickson JA, Jalaie M, Robertson DH, Lewis RA, Vieth M (2004)
Lessons in molecular recognition: the effects of ligand and
protein flexibility on molecular docking accuracy. J Med Chem
47:45–55
Gasteiger J, Marsili M (1980) Iterative partial equalization of orbital
electro- negativity-a rapid access to atomic charges. Tetrahedron
36:3219–3228
Helen L, Robert J, Roger C, Steven W (2010) Neutrophil function in
inflammation and inflammatory diseases. Rheumatology
49:1618–1631
Hess B, Bekker H, Berendsen H, Fraahje J (1997) LINCS: a linear
constraint solver for molecular interactions. J Comp Chem
18:1463–1472
Humphrey W, Dalke A, Schulten K (1996) VMD: Visual molecular
dynamics. J Mol Graph 14:33–38
123