Synthesis and biological evaluation of new curcumin derivatives as antioxidant and antitumor agents

Article abstract of DOI:10.1007/s00044-012-0116-9

Twenty-four new compounds were prepared, taking curcumin as a lead, in order to explore their antioxidant and antitumor properties. The capacities of these derivatives to scavenge the 2,2′-azinobis(3-ethylbenzothiazoline-6- sulfonic acid) radical cation (ABTS^.+), and to protect human red blood cells (RBCs) from oxidative haemolysis were investigated. In addition, the percentage viability of different cell lines (Hep G2, WI38, VERO and MCF-7) was tested. The result of the antitumor testing was generally in accordance with those of the antioxidant assays. Compounds which bear o-methoxy substitution to the 4-hydroxy function in the phenyl ring (7g, 5g and 3g) exhibited significantly higher ABTS^.+-scavenging, antihaemolysis, and antitumor activities than other derivatives. In addition, molecular modelling studies were carried out for biologically active and inactive compounds, to study the structure-activity relationship, with the aim to elucidate which portions of the molecules are critical for the antioxidant and antitumor activity.

Full text of DOI:10.1007/s00044-012-0116-9

MEDICINAL
CHEMISTRY
RESEARCH
Med Chem Res (2013) 22:1147–1162
DOI 10.1007/s00044-012-0116-9
ORIGINAL RESEARCH
Synthesis and biological evaluation of new curcumin derivatives
as antioxidant and antitumor agents
•
Said M. Bayomi Hassan A. El-Kashef Mahmoud B. El-Ashmawy
•
•
•
Magda N. A. Nasr Magda A. El-Sherbeny Farid A. Badria
•
•
•
Laila A. Abou-zeid Mariam A. Ghaly Naglaa I. Abdel-Aziz
•
Received: 4 January 2012 / Accepted: 22 May 2012 / Published online: 7 June 2012
ꢀ Springer Science+Business Media, LLC 2012
Abstract Twenty-four new compounds were prepared,
takingcurcuminasa lead, inorder toexploretheirantioxidant
and antitumor properties. The capacities of these derivatives
to scavenge the 2,20-azinobis(3-ethylbenzothiazoline-6-sul-
fonic acid) radical cation (ABTS.?), and to protect human red
blood cells (RBCs) from oxidative haemolysis were investi-
gated. In addition, the percentage viability of different cell
lines (Hep G2, WI38, VERO and MCF-7) was tested. The
result of the antitumor testing was generally in accordance
with those of the antioxidant assays. Compounds which bear
o-methoxy substitution to the 4-hydroxy function in the
phenyl ring (7g, 5g and 3g) exhibited significantly higher
ABTS.?-scavenging, antihaemolysis, and antitumor activi-
ties than other derivatives. In addition, molecular modelling
studies were carried out for biologically active and inactive
compounds, to study the structure–activity relationship, with
the aim to elucidate which portions of the molecules are
critical for the antioxidant and antitumor activity.
Keywords Curcumin ꢀ Synthesis ꢀ
Free radical scavenging ꢀ Antihaemolytic effect ꢀ
Antitumor effect ꢀ Molecular modelling studies
Introduction
The area of free radical biology and medicine is developing
fast, since the discovery of the involvement of free radicals in
oxidative tissue injury and diseases. Free radicals and other
reactive oxygen species such as superoxide radical anion,
hydroxyl radical and hydrogen peroxide, are constantly gen-
erated through many biological processes and may be con-
sidered as a measure of biological inefficiency. The human
body uses an antioxidant system to neutralize the excessive
levels of reactive oxygen species that consists of enzymes
such as superoxide dismutases, catalases and glutathione
peroxidases, in addition to numerous non-enzymatic small
molecules that are widely distributed in the biological system
such as glutathione, a-tocopherol, ascorbic acid, b-carotene
and selenium (Memon and Sudheer, 2007). In general, the cell
is able to maintain an appropriate balance between oxidants
and antioxidants under normal conditions. The imbalance
between reactive oxygen species production and the available
antioxidant defence leads to a widely accepted phenomenon
called oxidative stress (Djordjevic, 2004).
S. M. Bayomi ꢀ M. B. El-Ashmawy ꢀ M. A. El-Sherbeny ꢀ
M. A. Ghaly ꢀ N. I. Abdel-Aziz (&)
Department of Medicinal Chemistry, Faculty of Pharmacy,
Mansoura University, Mansoura 35516, Egypt
e-mail: naglaabdalaziz2005@yahoo.com
Curcumin (1,7-bis(4-hydroxy-3-methoxyphenyl)-1,6-hept-
adiene-3,5-dione) is a naturally occurring phenolic com-
pound isolated as a yellow pigment from turmeric (dry
rhizomes of Curcuma longa, Zingiberaceae) which is
commonly used as a food colorant, spice and traditional
medicine in India and China (Geol et al., 2008; Sharma et al.,
2005). This compound has been the focus of many
recent biochemical investigations due to its various biolog-
ical and pharmacological properties, including antioxidant
H. A. El-Kashef
Department of Pharmacology, Faculty of Pharmacy,
Mansoura University, Mansoura 35516, Egypt
M. N. A. Nasr ꢀ L. A. Abou-zeid
Department of Pharmaceutical Organic Chemistry, Faculty
of Pharmacy, Mansoura University, Mansoura 35516, Egypt
F. A. Badria
Department of Pharmacognosy, Faculty of Pharmacy,
Mansoura University, Mansoura 35516, Egypt
123

1148
Med Chem Res (2013) 22:1147–1162
(Lee et al., 2009; Venkatesan and Rao, 2000) and anticancer
activities (Aggarwal et al., 2003). The cancer preventive
capability of curcumin is linked to its direct antioxidant
ability to eliminate free radicals and to reduce oxidative
stress (Dinkova-Kostova and Talalay, 2008). Consequently,
the past few years have witnessed tremendous interest in the
antioxidant activity of curcumin. The structural simplicity
and nontoxicity of curcumin, along with its defects such as
poor bioavailability (Anand et al., 2007), made this molecule
a promising lead compound for development of potential
antioxidant and cancer chemopreventive agents (Fadda
et al., 2010; Liang et al., 2009).
systems were built in compounds 6a–g to simulate coumarin
and flavonoid ring systems, which are well known for their
antioxidant and anticarcinogenic activities (Guthrie and
Carroll, 1998). Finally, certain fused pyrazoles 7e–g, 8a–
e and 8h–j, were included to formulate a more comprehen-
sive structure–activity relationship, if any.
Results and discussion
Chemistry
The synthetic routes of the designed compounds are illus-
trated in Schemes 1 and 2. Dimethylation of pentane-2,4-
dione using dimethyl sulphatein dimethylsulfoxide (DMSO)
and tetrahydrofuran (THF) as solvent, in the presence of
potassium carbonate, afforded 3,3-dimethylpentane-2,4-
dione 1, which was used directly in an aldol condensation
with various aromatic aldehydes 2a–g in ethanolic sodium
hydroxide solution to yield the corresponding 4,4-dimethyl-
1,7-bis((substituted)phenyl)hepta-1,6-diene-3,5-diones 3a–g
(Scheme 1). Refluxing curcumin with diethyl sulphate in
acetone, in the presence of anhydrous potassium carbonate,
afforded 1,7-bis(4-ethoxy-3-methoxyphenyl)-4,4-diethylhepta-
1,6-diene-3,5-dione 4 (Scheme 1).
O
O
H
3CO
H
OC
3
H
O
H
O
Curcumin
In this study, different series of curcumin derivatives were
synthesized (Schemes 1, 2) and their antioxidant activity
were assessed by ABTS.?-scavenging and antihaemolysis
experiments. To explore the substituent effect, type and
distribution pattern, and the role of central active methylene
hydrogens, compounds 3a–g and 4 were synthesized, where
the seven carbon spacer was retained. Compounds 5a–g and
5h–j were prepared to test the effect of decreasing length and
flexibility of the seven carbon spacer. Fused pyrano ring
On the other hand, aldol condensation of cyclohexanone
or cycloheptanone with different aromatic aldehydes
R1
R2
R3
CHO
R1
O
O
R1
R4
O
O
O
O
R2
R3
R2
R3
2a-g
(i)
H3C CH3
H3C
CH3
(ii)
R4
R4
1
3a-g
O
O
O
O
H3CO
HO
OCH3
OH
H3CO
C2H5O
OCH3
(iii)
C2H5 C2H5
OC2H5
Curcumin
4
a: R1 = R2 = R3 = R4 = H
b: R1 = R2 = R4 = H R3 = Cl
c: R1 = R2 = R4 = H R3 = OCH3
d: R1 = R4 = H, R2 = R3 = OCH3
e: R2 = R3 = H, R1 = R4= OCH3
f: R1 = R2 = R4 = H, R3 = OH
g: R1 = R4 = H, R2 = OCH3, R3 = OH
Scheme 1 Reaction protocol for the synthesis of 3a–g and 4: (i) (CH3)2SO4, K2CO3, DMSO/THF; (ii) NaOH, EtOH; (iii) (C2H5)2SO4, acetone,
K2CO3
123

Med Chem Res (2013) 22:1147–1162
1149
Scheme 2 Reaction protocol
for the synthesis of 5a–j, 6a–g,
7e–g, 8a–e and 8h–j: (i) NaOH,
EtOH; (ii) malononitril, butanol
or malononitril, DMF,
R1
O
R2
R3
CHO
+
R4
n
piperidine; (iii) hydrazine
hydrate, EtOH; (iv) 4-bromo
phenylhydrazine hydrochloride,
Na ethoxide, EtOH
2a-g
(i)
R1
O
R1
R2
R2
R3
R3
n
R4
R4
(ii)
(iv)
Br
5a-j
(iii)
NH2
CN
R1
N
N
R1
R1
O
R1
R2
R3
R2
R3
R2
R2
R1
N
NH
R1
R2
R3
R2
R3
R3
R3
n
R4
R4
R4
R4
6a-g
8a-e , 8h-j
R4
R4
7e-g
a: n= 1
n= 1
R
R
1 = R2 = R3 = R4 = H
1 = R2 = R4 = H R3 = Cl
b:
c: n= 1 R1 = R2 = R4 = H R3 = OCH3
d: n= 1 R1 = R4 = H, R2 = R3 = OCH3
R
2 = R3 = H, R1 = R4= OCH3
e:
n= 1
f: n= 1 R1 = R2 = R4 = H, R3 = OH
g:
n= 1 R1 = R4 = H, R2 = OCH3, R3 = OH
h: n= 2
1 = R2 = R4 = H R3 = OCH3
i: n= 2 R1 = R4 = H, R2 = R3 = OCH3
j:
R
n= 2
R
2 = R3 = H, R1 = R4= OCH3
(2a–e) in an ethanolic sodium hydroxide solution afforded
2,6-bis((substituted) benzylidene)cyclohexanone 5a–e and
2,7-bis(substituted benzylidene)cycloheptanones 5h–j, res-
pectively. The previously reported compounds 5f and 5g
were obtained via the interaction of cyclohexanone with
the hydroxybenzaldehydes 2f and 2g, respectively, in gla-
cial acetic acid and concentrated hydrochloric acid.
Condensation of 2,6-bis(4-methoxybenzylidene)cyclo-
hexanone 5c with malononitrile in n-butanol afforded 2-
amino-8-(4-methoxybenzylidene)-4-(4-methoxyphenyl)-5,
6,7,8-tetrahydro-4H-chromene-3-carbonitrile 6c. On the
other hand, the interaction of other 2,6-bis((substituted)
benzylidene)cyclohexanones 5a, b, d–g with malononitrile
was done in dimethylformamide (DMF) in the presence of
piperidine at room temperature, to get the correspond-
ing 5,6,7,8-tetrahydro-4H-chromene-3-carbonitriles, 6a, b,
d–g.
of sodium ethoxide yielded 2-(4-bromophenyl)-7-((substi-
tuted)benzylidene)-3-((substituted)phenyl)-3,3a,4,5,6,7-hexa-
hydro-2H-indazoles 8a–e and (E)-2-(4-bromophenyl)-8-
(substituted benzylidene)-3-(substituted phenyl)-2,3,3a,4,
5,6,7,8-cyclohepta[c]pyrazoles 8h–j, respectively (Scheme 2).
Biological evaluation
Assay for 2,20-azinobis(3-ethylbenzothiazoline-6-sulfonic
acid) radical (ABTS.?) scavenging activity
The antioxidant activity assay employed here is one of the
several assays that depend on measuring the consumption
of stable free radicals, to evaluate the free radical scav-
enging activity of the investigated compound. The meth-
odology assumes that consumption of the stable free
radical (Xꢀ) by hydrogen-donating antioxidants (YH) will
be determined by reaction as follows:
Finally, cyclocondensation of 5e–g using hydrazine
hydrate in refluxing ethanol afforded 7-(substituted ben-
zylidene)-3-(substituted phenyl)-3,3a,4,5,6,7-hexahydro-2H-
indazoles 7e–g. Cyclocondensation of 2,6-bis((substituted)
benzylidene)cyclohexanones 5a–e and 2,7-bis(substituted
benzylidene)cycloheptanones 5h–j with 4-bromophenyl-
hydrazine hydrochloride in absolute ethanol in the presence
Xꢀ þ YH ! XH þ Yꢀ:
The rate and/or the extent of the process measured in terms
of the decrease in Xꢀ concentration, would be related to the
ability of the added compounds (YH) to trap free radicals.
The decrease in colour intensity of the free radical solution
123

1150
Med Chem Res (2013) 22:1147–1162
due to scavenging of the free radical by the antioxidant
material is measured colourimetrically at a specific wave-
length. The assay employs the radical cation derived from
2,20-azinobis(3-ethylbenzothiazoline-6-sulfonic acid) (ABTS)
as stable free radical. The advantage of ABTS-derived free
radical method over other methods is that the produced col-
our remains stable for more than 1 h and the reaction is
stoichiometric (Chyong et al., 2011; Lissi et al., 1999). In this
study, all the synthesized compounds were subjected to
ABTS test (Table 1).
Table 1 continued
Compound Nos.
% inhibition
8h
8i
3.1
0.96
3.1
8j
90
80
70
60
50
40
30
20
10
Table 1 The percentage inhibition of the ABTS radical cation by the
tested compounds
Compound Nos.
% inhibition
Control of ABTS
0
0
Ascorbic acid
80.81
87.64
4.84
2.32
7.75
12.59
3.68
28.68
52.9
26.35
1.93
0.58
1.35
0
AA
C
3g
5g
6g
7f
7g
Curcumin
3a
3b
3c
Compounds
AA ascorbic acid, C curcumin
3d
3e
3f
From these results, it is concluded that compounds 7g,
5g, 6g, 7f and 3g exhibited more than 50 % inhibition of
the ABTS radical cation. The latter compounds, except
compound 7f, share an o-methoxy substituent to the
4-hydroxy function on the two aryl rings, though the core
of the molecules is completely different. Thus, the length
and the shape of the spacer between the two aryl rings
could be varied without affecting the scavenging activity.
However, compound 7f, characterized by having a pyra-
zole ring, appears amongst the most active compounds
indicating that the pyrazole NH might have a role in
trapping the ABTS free radical. The antioxidant activity of
the previously prepared compound 5g was also determined
and proved to be effective (Lee et al., 2009; Shang et al.,
2010; Ligeret et al., 2004; Weber et al., 2006).
3g
4
5a
5b
5c
5d
5e
1.16
9.49
74.22
1.74
0.77
5.03
0.19
1.55
0.38
16.86
0.38
16.66
57.75
25.6
54.45
76.74
1.74
2.9
5f
5g
5h
5i
5j
6a
6b
6c
Antihaemolysis assay
6d
6e
Human red blood cells (RBCs) are heterogeneous media,
and are particularly exposed to endogenous oxidative dam-
age because of their specific role as oxygen carriers. RBC
membrane is rich in polyunsaturated fatty acids which are
very susceptible to free radical-mediated peroxidation.
AAPH (2,20-azobis(2-amidinopropane) dihydrochloride)
has been extensively used as a free radical initiator for
biological studies because it is water-soluble and the rate of
free radical generation from AAPH can be easily controlled
and measured. Thermal decomposition of AAPH in the
aqueous dispersion of RBCs produces an initiating radical
6f
6g
7e
7f
7g
8a
8b
8c
5.03
0.19
0.77
8d
8e
123

Med Chem Res (2013) 22:1147–1162
1151
flavonoid ring systems, comprise four active members in
the antihaemolysis assay. This may focus on the impor-
tance of the pyrano-fused system in the in vitro assay.
While curcumin was the most active compound in the
ABTS assay, many compounds preceeded it in the anti-
haemolysis assay. These compounds vary in their struc-
tures from compounds retaining the seven carbon spacer
between the two aryl rings, to compounds having a 2H-
indazole or chromene ring embedded, to compounds hav-
ing a saturated cyclohexane ring core. This may reflect the
versatility of the active structures in the antihaemolysis
assay.
Table 2 The percentage reduction of RBCs haemolysis after prein-
cubation with these nine tested compounds at 2 lg/ml
Compound Nos.
% Reduction of RBCs haemolysis
Ascorbic acid
17.6
50.4
98.0
42.2
66.8
63.2
60.2
80.2
70.4
81.0
94.8
Curcumin
3g
5a
5g
6a
6b
6c
6e
7f
Antitumor testing
7g
The synthesized curcumin analogues were subjected to
antitumor testing by measuring their effect on the per-
centage viability of four different cell lines; human hepa-
tocellular liver carcinoma cell line (hep G2), human lung
fibroblast cell line (WI38), kidney epithelial cells of the
African green monkey (VERO), and human breast adeno-
100
80
60
40
20
´
carcinoma cell line (MCF-7) (Kalai et al., 2011). All the
synthesized compounds were subjected to antitumor test-
ing. Among the tested compounds, only compounds 3d, 3e,
3f, 3g, 4, 5f, 5g, 6d, 6f, 6g, 7e, 7f and 7g exhibited
promising activity against the different cell lines used,
while the rest of compounds showed no activity (Table 3).
In this study, the obtained data indicated that the most
active compounds against hep G2 cell line were in the
following order: curcumin & 6d [ 7g [ 5g [ 3g [ 3f.
On the other hand, compounds 6d, curcumin, 5g, 7g, 3d
and 5f proved to be effective against WI38 cell line.
Moreover, VERO cell line was found to be sensitive
toward compounds 7g, curcumin, 6d, 5g, 3e and 7f, while
compounds 7g, 6d, 5g, 3g, and 3f are particularly effective
against MCF-7 cell line. In all types of cell lines, antitumor
activity seems to be inherent to compounds 7g, 6d and 5g,
the first two exceeded curcumin in some cell lines.
The cyclic curcumin analogues 7g and 5g retain the
o-methoxyphenoxy moiety as a common structure. This
may reflect the importance of such moiety in antitumor
activity. However, the spacer between the two aryl rings
differs from that of curcumin. It is noteworthy that com-
pound 5g was previously tested for its cytotoxic activity
(Liang et al., 2009), and also proved effective. The notable
feature being the superiority of the chromene structure 6d,
even over curcumin, though it lacks the hydroxyl functions.
In conclusion, the results of the antitumor testing were
generally in accordance with those of the antioxidant
assays. Compounds which bear o-methoxy substitution to
the 4-hydroxy function (7g, 5g and 3g) exhibited signifi-
cantly higher ABTS.?-scavenging, antihaemolysis and
antitumor activities than other derivatives.
0
AA
C
3g 5a 5g 6a 6b 6c 6e 7f 7g
Compounds
AA ascorbic acid, C curcumin
(R.) which can attack the polyunsaturated lipids (LH) in
RBC membranes to induce lipid peroxidation. Since the
lipid peroxidation is a free radical chain reaction, the RBC
membrane is quickly damaged, leading to haemolysis. On
the other hand, whether antioxidants (YH) are present or
added to RBCs, they would react with the chain propagating
peroxyl radicals (LOO.) to stop the peroxidation, hence
inhibit haemolysis (Shang et al., 2010). Data obtained from
haemolysis assay indicated that the most of the tested
compounds produced 100 % haemolysis and hence exclu-
ded from the antihaemolysis assay. Hence, only nine com-
pounds, that did not induce haemolysis, were subjected to
the antihaemolysis assay (Table 2).
From these results, it is concluded that compounds
showing the highest protective activity towards AAPH-
induced oxidative RBCs haemolysis could be arranged as
follows: 3g [ 7g [ 7f [ 6c [ 6e [ 5g [ 6a [ 6b [ cur-
cumin. Compounds 3g and 7g are nearly equipotent. They
share the o-methoxy substituent to the 4-hydroxy function,
while the core of the two molecules is completely different.
Thus, the length and the shape of the spacer between the
two aryl rings could be varied without affecting activity.
Compounds 7, bearing resemblance to coumarin and
123

1152
Med Chem Res (2013) 22:1147–1162
Table 3 The results of the antitumor assay of curcumin and its tested analogues, together with 5-fluorouracil (5FU) as positive control, at 20 lg/ml,
on the percentage viability of the different cell lines
Compound Nos.
Hep G2
WI38
VERO
MCF-7
5FU
0
18
78
84
57
47
78
88
28
19
78
81
78
89
21
0
51
75
80
88
78
82
77
58
12
79
84
100
88
65
0
24
89
74
82
82
88
84
69
65
83
82
67
76
12
0
19
87
79
64
55
94
74
32
21
82
79
82
75
19
Curcumin
3d
3e
3f
3g
4
5f
5g
6d
6f
6g
7e
7f
7g
5FU 5-fluorouracil, C curcumin
Molecular modelling discussion
features responsible for the corresponding biological
activities.
Conformational analyses of most of the prepared com-
pounds were performed, then flexible alignment, and
electrostatic and hydrophobic mappings were applied for
certain selected compounds, based on the biological
results.
Compounds 3a–g
Compounds 3a–g with the seven carbon spacer bridge
between the two (substituted)phenyl terminal groups
showed that 3g, the most active analogue in this series,
exhibited a common alignment similar to the lead ligand
‘curcumin’. The 4,4-dimethyl substitution showed negli-
gible interference with the degree of flexibility of the
carbon bridge spacer. The three middle carbons of the
carbon bridge spacer got in narrow zigzag angle keeping
the trans configuration and the proper distance between the
two benzylidene rings.
Stochastic conformational analysis
Stochastic conformational search was performed using the
global minima conformers of the most active analogues,
the moderately active analogues, and the inactive analogue
of each series to confirm the proper conformational
123

Med Chem Res (2013) 22:1147–1162
1153
Compounds 7f and 7g
Compound 4
Both members of this series represent the most biologically
active analogues of the entire tested compounds. The
hexahydro-2H-indazole ring performed certain degree of
flexibility besides the angular conformational freedom of
the terminal benzylidene ring. The global minima con-
former kept the common structural features with the same
distance between the two terminal disubstituted phenyl
groups similar to the corresponding rings of the lowest
energy conformer of the lead analogue ‘curcumin’.
Compound 4, with the seven carbon spacer bridge between
the two substituted phenyl terminal groups, exhibited a
common alignment similar to the lead ligand ‘curcumin’.
The flexibility of the carbon spacer bridge with the con-
jugated double bonds and the diketones represented the
important structural characteristic. However, the 4,4-die-
thyl substitution played steric features with the degree of
flexibility of carbon bridge spacer where the angular con-
figuration of the carbon spacer got smaller, with certain
extent, keeping the two benzylidene rings in zigzag ori-
entation keeping the trans configuration around each dou-
ble bond.
Compounds 8a–e and 8h–j
All the compounds, 8a–e and 8h–j, showed limited confor-
mational changes due to occupying the five carbon spacer
bridge with the hexahydro-2H-indazole ring keeping the
chance of free conformational change restricted to the
angular position to the terminal benzylidene ring; however,
the hexahydro-2H-indazole ring still reserves certain elas-
ticity degree. N2 substitution with 4-bromophenyl ring
prompted high field of steric energy supporting the clarifi-
cation of the abolishing reactivity of these group analogues.
Compounds 5a–g and 5h–j
Compound 5g, with the bis(benzylidene)cyclohexanone,
showed the geometry of the lowest energy conformer
‘global minima’ with the trans configuration arranged in a
comparable manner to that of the lead ligand keeping the
proper distance between the two terminal benzylidene
rings.
Flexible alignment
Compounds 6a–g
Flexible alignment was performed to gain an insight into
the fingerprint features responsible for the different
Compounds 6a–g, with the five carbon spacer bridge
between the two (substituted)phenyl terminal groups,
showed that 6g, the most active analogue in this series,
exhibited limited flexibility due to the steric feature of
tetrahydro-4H-chromene ring keeping the chance of con-
formational change restricted to the angular position to the
terminal (4-hydroxy-3-methoxy)benzylidene ring and this
may explain the slightly impaired biological efficacy of the
members of this series.
Fig. 1 Flexible alignment of the most active compounds: 7g (in red)
and 5g (in blue) (Color figure online)
123

1154
Med Chem Res (2013) 22:1147–1162
Fig. 2 Flexible alignment of the moderately active compound 7f (in
red) and the most active compound 7g (in blue) (Color figure online)
Fig. 4 Flexible alignment of the least active compound 8e (in green)
and the most active compound 7g (in red) (Color figure online)
Fig. 3 Flexible alignment of
the moderately active
compound 3g (in green) and
curcumin (in red) (Color figure
online)
Fig. 5 Flexible alignment of the least active compound 6b (in blue)
and the most active compound 7g (in red) (Color figure online)
biological activity. The other features are totally accom-
modating the two structures with proper alignment score,
including the 4-phenoxy group with the spacer distance of
five carbons (Fig. 2).
biological activities. The structures of the most active
compounds 7g and 5g, with different structural cores, were
subjected to the flexible alignment techniques.
Compound 3g is curcumin analogue with the dimethyl
substitution at C4 of the seven carbon spacer bridge. The
introducing of 4,4-dimethyl groups caused moderate activity
as antitumor and antioxidant agent. On the other hand, the
presence of donating 4,4-dimethyl groups improved the an-
tihaemolysis activity to the maximal extent. The alignment
of 3g with its putative ligand was highly scored (Fig. 3).
Flexible alignment of the most active compounds;
7g and 5g
Mapping the MOE-generated alignment conformer (MOE,
2007.9 of Chemical Computing Group. Inc.,) indicated the
proper superposition of the two phenyl rings. The disub-
stituted functional groups; 4-hydroxy and 3-methoxy were
aligned with high scoring. Another fingerprint factor of
proper alignment is the five carbon spacer length present in
both the compounds. Moreover, the 2H-indazole of com-
pound 7g and the corresponding cyclohexyl ring of com-
pound 5g were deviated to certain extent showing the
restricted flexibility of the indazole ring in comparison with
the cyclohexyl moiety referring to the free rotation around
single double bond that are missing in compound 7g.
Compound 7g showed high electrostatic field due to the
N-atom rich indazole ring. Both 7g and 5g possessed better
free radical scavenging, antihaemolytic and antitumor
activities and that may be expressed as a comparative
relationship between the activity and the better alignment
represented by the lower score value of the configuration
(Fig. 1).
Flexible alignment of the inactive compounds; 8e and 6b
Examining the degree of similarity between the most active
analogue 7g and the least active compound 8e indicated
that the common features of alignment were the indazole
core; a flexible hydrophilic area, in addition to the ben-
zylidene fragment and the terminal phenyl group; the
hydrophobic area that performed proper alignment in all
the active analogues. The least active 8e has N2 of pyrazole
attached to p-bromophenyl moiety that introduced bulki-
ness to the total surface area and led to increment of the
steric energy. Moreover, changing the position on two
terminal phenyl groups with 2,5-dimethoxy substitution
instead of 3,4-disubstitution impaired all the given bio-
logical activities (Fig. 4).
Configurational analysis of the alignment data base
regarding the structural feature behind the abolishing of the
antioxidant, antitumor activities of compound 6b indicated
that the presence of 5,6,7,8-tetrahydro-4H-chromene ring
instead of the indazole ring played restrained effect to the
biological activity, in addition to the introduction of
4-chloro substitution instead of 4-hydroxy and 3-methoxy
disubstitution groups. Even the cyano and amino substitu-
tions at the 4H-chromene ring improved the hydrophilic
Flexible alignment of the moderately active compounds;
7f and 3g
Analysis of the moderately active analogues highlights the
common fingerprint features that are essential for elucida-
tion of the relative changing of the obtained activity.
Compound 7f showed identical alignment with the most
active analogue 7g except the absence of 3-methoxy group
that seemed to be essential functional feature for proper
123

Med Chem Res (2013) 22:1147–1162
1155
Fig. 6 Electrostatic map (left)
and hydrophobic map (right) for
the most active compound 7g
Fig. 7 Electrostatic map (left)
and hydrophobic map (right) for
the most active compound 5g
effect but the overall structural changes did not fulfil the
required alignment features (Fig. 5).
Electrostatic and hydrophobic mappings
Mapping the electrostatic, hydrophobic and the donor/
acceptor hydrogen bond characteristics were performed
using the lowest energy conformer obtained from the
geometrically constrained conformational search, in an
attempt to understand the variant degree of biological
reactivity of the tested compounds according to the elec-
trostatic feature mappings. The electrostatic feature map is
an application of the Poisson–Boltzmann equation (PBE)
to the prediction of electrostatically preferred locations of
hydrophobic, H-bond acceptor, and H-bond donor loca-
tions. Electrostatically positive regions (donor) are col-
oured blue, negative regions (acceptor) are coloured red,
and neutral regions are coloured white.
Fig. 8 Electrostatic map (left) and hydrophobic map (right) for the
least active compound 8e
Electrostatic mapping of compound 5g showed the
presence of 3-methoxy and the carbonyl oxygen groups
representing the acceptor regions that were coloured red.
Phenyl and cyclohexyl fragments were represented as
white neutral region. 4-Hydroxy groups showed donor
profile and were coloured blue. The hydrophobic distribu-
tions of the most active compound 5g had two phenyl and
cyclohexyl fragments that appeared green. On the other
hand, the hydroxyl and methoxy groups represented the
hydrophilic distributions and appeared purple (Fig. 7).
As a result, the most active compounds 7g and 5g can
adopt conformations with identical orientations and suffi-
cient agreement of the electrostatic and the hydrophobic
potential patterns.
Predicting electrostatically favourable regions
and hydrophobic regions of the most active
analogues 7g and 5g
Electrostatic mapping of 7g showed that the disubstituted
3-methoxy groups capped the functional groups repre-
senting the acceptor regions and coloured red. The pres-
ence of N1 and N2 of the indazole ring represented donor
regions (in blue). Hydrophilic predictions are coloured
purple and hydrophobic predictions are coloured green.
The hydrophobic distributions of the most active com-
pound 7g had two phenyl fragments and appeared green,
while the presence of the hydroxyl, methoxy and the
nitrogen indazole groups represented the hydrophilic dis-
tributions and appeared purple (Fig. 6).
Predicting electrostatically favourable regions
and hydrophobic regions of the least active
analogues 8e and 6b
Electrostatic mapping of 8e showed that the disubstituted
2,5-dimethoxy groups distributed the electrostatic features
turning the ortho, meta positions into an acceptor region
(red). The presence of the 4-bromophenyl ring performed
electrostatic donating behaviour turning the protruded area
into donor regions (blue). Varying the electrostatic
123

1156
Med Chem Res (2013) 22:1147–1162
Fig. 9 Electrostatic map (left)
and hydrophobic map (right) for
the least active compound 6b
Table 4 Calculated Lipinski’s rule of five for the most active compounds
Compound Nos.
ABTS assay
Antihaemolysis
Antitumor
Parameter
a-don
a-acc
M. Wt.
Log P
3g
6
3
5
2
1
1
6
3
2
9
4
2.0000
2.0000
3.0000
3.0000
2.0000
6.0000
5.0000
3.0000
5.0000
6.0000
396.439
366.413
319.384
379.436
368.385
5.094
4.177
3.349
3.331
3.718
5g
3
7f
Moderate
7g
1
1
Curcumin
a-acc number of H-bond acceptor, a-don number of H-bond donor, Log P calculated lipophilicity, M. Wt. molecular weight
distribution differently in comparison with the electrostatic
performance of the most active analogue 7g, designate the
significant elucidation concerning the abolishing of the
biological activity of compound 8e. The hydrophobic dis-
tributions (coloured green) of the least active compound 8e
are due to the presence of three phenyl fragments, while the
methoxy and the nitrogen indazole groups represent the
hydrophilic distributions (coloured purple), but located and
distributed differently in comparison with the most active
analogue 7g (Fig. 8).
chemists. Lipinski’s rule of five is a rule of thumb to
evaluate drug likeness, or determine if a chemical com-
pound with a certain pharmacological or biological activity
has properties that would make it a likely orally active drug
in humans. The rule was formulated by Lipinski et al. in
(1997), based on the observation that the most medication
drugs are relatively small and lipophilic molecules. The
rule is important for drug development where a pharma-
cologically active lead structure is optimized step-wise for
increased activity and selectivity, as well as drug-like
properties as described by Lipinski’s rule. The modification
of the molecular structure often leads to drugs with higher
molecular weight, more rings, more rotatable bonds, and a
higher lipophilicity (Oprea et al., 2001). Lipinski’s rule
says that, in general, an orally active drug has no more than
one violation of the following criteria:
Electrostatic mapping of 6b showed that the di-para-
chloro substituted groups distributed the electrostatic
features turning the para positions into a donor region
(blue). Presence of the rigid fused tetrahydro-4H-chro-
mene ring sited Van der Waal’s force that located heavily
in a different manner in comparison with the most active
analogue 7g. The cyano and amino groups assumingly
enrich the electrostatic interaction but located away from
the favourable considered area revealing to some extent
the reason of abolishing of the biological activity of
compound 6b. The hydrophobic distributions (coloured
green) of the least active compound 6b are highly
expressed due to the presence of firm tetrahydro-4H-
chromene ring and two phenyl fragments, while the cyano
and the amino groups represent the hydrophilic distribu-
tions (coloured purple), but located and distributed dif-
ferently in comparison with the most active analogue 7g
(Fig. 9).
•
•
Not more than 5 hydrogen bond donors (nitrogen or
oxygen atoms with one or more hydrogen atoms)
Not more than 10 hydrogen bond acceptors (nitrogen or
oxygen atoms)
•
•
A molecular weight under 500 Da
An octanol–water partition coefficient log P of less
than 5.
It is noteworthy that all numbers are multiples of five,
which is the origin of the rule’s name. Computational
studies were completed to evaluate the obedience of the
most active compounds to the Lipinski’s rule of five and
the results showed that all investigated compounds obey
these rules (Table 4) and these compounds should present
good passive oral absorption. The lipophilic characters of
the investigated compounds are in agreement with their
biological activities.
Lipinski’s rule of five
The introduction of Lipinski’s ‘rule of five’ has initiated a
profound shift in the thinking paradigm of medicinal
123

Med Chem Res (2013) 22:1147–1162
1157
Conclusion
residue obtained was purified on silica gel preparative TLC
plates using chloroform/methanol (9:1) as an eluent.
4,4-Dimethyl-1,7-bis(4-chlorophenyl)-hepta-1,6-diene-3,
Twenty-four new compounds have been synthesized and
subjected to antioxidant and antitumor screening. The
obtained data revealed that compounds which bear o-meth-
oxy substitution to the 4-hydroxy function (7g, 5g and 3g)
exhibited significantly higher ABTS.?-scavenging, anti-
haemolysis, and antitumor activities than other derivatives.
Conformational analyses, flexible alignment, and electro-
static and hydrophobic mappings were applied and were in
agreement with the experimental results.
5-diones (3b). Yield, 40 %; mp 155–157 ꢁC; IR (KBr) mmax
/
cm-1 1682 (C=O), 1599 (C=C). 1H NMR (CDCl3); d: 1.26
(s, 6H, 2CH3), 7.26–8.07 (m, 12H, 8Ar-H and 4 alkenyl
CH). MS m/z (%); 374, [M ? H]? (9.14), 44 (100.00).
Anal. Calcd for C21H18Cl2O2 (%):C, 67.57; H, 4.86. Found:
C, 67.90; H, 4.56.
4,4-Dimethyl-1,7-bis(4-methoxyphenyl)-hepta-1,6-diene-
3,5-diones (3c). Yield, 52 %; mp 126–128 ꢁC; IR (KBr)
m
max/cm-1 1677 (C=O), 1598 (C=C).1H NMR (CDCl3); d:
Experimental protocols
1.25 (s, 6H, 2CH3), 3.87 (s, 6H, 2OCH3), 6.95 (d, 2H, 2
alkenyl CH), 7.25–7.86 (m, 8H, 8ArH), 8.06 (d, 2H, 2
alkenyl CH). MS m/z (%); 364, M? (6.09), 121 (100.00).
Anal. Calcd for C23H24O4 (%):C, 75.80; H, 6.64. Found: C,
76.23; H, 6.87.
Synthetic procedures
Melting points (ꢁC) were recorded using a Fisher–Johns
melting point apparatus and are uncorrected. Microanaly-
ses were performed in the Microanalytical unit, Cairo
University. IR spectra (KBr) were recorded on Mattson
4,4-Dimethyl-1,7-bis(2,5-dimethoxyphenyl)-hepta-1,6-di
ene-3,5-diones (3e). Yield, 45 %; mp 67–69 ꢁC; IR (KBr)
m
max/cm-1 1685 (C=O), 1602 (C=C). 1H NMR (CDCl3); d:
5000 FT-IR spectrometer (m in cm-1). H NMR spectra
1.26 (s, 6H, 2CH3), 3.93 (s, 12H, 4OCH3), 6.89 (d, 2H, 2
alkenyl CH), 7.06–7.59 (m, 6H, 6Ar-H), 7.76 (d, 2H, 2
alkenyl CH). MS m/z (%); 425, M? (1.25), 55 (100.00).
Anal. Calcd for C25H28O6 (%):C, 70.74; H, 6.65. Found: C,
70.59; H, 6.34.
1
were obtained on FT-NMR spectrometer (200 MHz)
Gemini Varian using TMS as internal standard (chemical
shifts in ppm, d units). MS analyses were performed on
JEOL JMS-600H spectrometer. Thin layer chromatography
(TLC) was performed on silica gel G for TLC (Merck) and
spots were visualized by irradiation with ultraviolet
light (UV; 254 nm). Curcumin was purchased from Fluka.
The syntheses of compounds 3a (Weber et al., 2005),
3d (Roughley and Whiting, 1973), 5a (Singh et al., 2009),
5b (Singh et al., 2009), 5c (Wang et al., 2004), 5d (Singh
et al., 2009), 5f (Du et al., 2006), 5g (Du et al., 2006), 5h
(Singh et al., 2009), 5i (Singh et al., 2009), 6a (Zhou,
2003), and 6b (Zhou, 2003), have been previously reported.
4,4-Dimethyl-1,7-bis(4-hydroxyphenyl)-hepta-1,6-diene-
3,5-diones (3f). Yield, 40 %; mp 80–82 ꢁC; 1H NMR
(CDCl3); d: 1.22 (s, 6H, 2CH3), 6.83–7.74 (m, 12H, 8Ar-H
and 4 alkenyl CH), 9.76 (s, 2H, 2OH). MS m/z (%); 337,
[M ? H]? (0.86), 63 (100.00). Anal. Calcd for C21H20O4
(%):C, 74.98; H, 5.99. Found: C, 75.43; H, 5.57.
4,4-Dimethyl-1,7-bis(4-hydroxy-3-methoxyphenyl)-hepta-
1,6-diene-3,5-diones (3g). Yield, 45 %; mp 56–58 ꢁC; IR
(KBr) mmax/cm-1 1683 (C=O), 1600 (C=C). 1H NMR
(CDCl3); d: 1.30 (s, 6H, 2CH3), 3.82 (s, 6H, 2OCH3), 6.80
(d, 2H, 2 alkenyl CH), 7.10–7.79 (m, 8H, 6Ar-H and 2
alkenyl CH), 10.00 (s, 2H, 2OH). MS m/z (%); 396,
M?(13.96), 55 (100.00). Anal. Calcd for C23H24O6 (%): C,
69.68; H, 6.10. Found: C, 70.00; H, 6.41.
General method for synthesis of 4,4-dimethyl-1,7-bis
((substituted)phenyl)-hepta-1,6-diene-3,5-diones (3b, c, e–g)
To a cooled solution (10 ꢁC) of 2,4-pentanedione (1.33 g,
13.3 mmol) in DMSO (7 ml) and THF (4 ml), potassium
carbonate (4.60 g, 33.4 mmol) was added. Dimethyl sul-
phate (2.12 g, 16.8 mmol) was added dropwise while
stirring and the mixture was stirred at room temperature for
3 days. The expected 3,3-dimethylpentane-2,4-dione 1,
was used in the next step without isolation. To the stirred
reaction mixture, a solution of the appropriate aromatic
aldehyde 2b, c, e–g (26.6 mmol) in ethanol (20 ml) was
added. Ethanolic sodium hydroxide (50 ml, 10 %) was
added dropwise, and stirred overnight at room temperature.
After completion of the reaction (TLC), the mixture was
poured onto ice–water, then conc. HCl was added dropwise
until effervescence and precipitation ceased. The precipi-
tated solid was collected by filtration, air dried, and
recrystallized from ethyl acetate. For compounds 3f, g, the
General method for the synthesis of 1,7-bis(4-ethoxy-3-
methoxyphenyl)-4,4-diethylhepta-1,6-diene-3,5-dione (4)
Anhydrous potassium carbonate (2 g) was added to a stir-
red solution of curcumin (1.72 g, 5 mmol) in acetone
(100 ml), followed by diethyl sulphate (1.43 g, 9 mmol).
The mixture was heated at reflux temperature while stirring
for 22 h. Potassium carbonate was then filtered off, and the
solvent was evaporated. The residue was washed with
diethyl ether, and crystallized from ethanol to give com-
pound 4. Yield, 42 %; mp 125–127 ꢁC. IR (KBr) mmax
/
cm-1 1625 (C=O). 1H NMR (CDCl3); d: 1.42 (t, 6H,
2CH2CH3), 1.66 (t, 6H, 2OCH2CH3), 2.35 (q, 4H,
2CH2CH3), 4.10 (s, 6H, 2OCH3), 4.31 (q, 4H, 2OCH2CH3),
123

1158
Med Chem Res (2013) 22:1147–1162
6.67 (d, 2H, 2 alkenyl CH), 6.96–7.49 (m, 6H, 6Ar-H), 7.78
(d, 2H, 2 alkenyl CH). MS m/z (%); 481, M? (3.65), 69
(100.00). Anal. Calcd for C29H36O6 (%): C, 72.48; H, 7.55.
Found: C, 72.90; H, 8.02.
under reflux for 26 h. The precipitated product was filtered
and recrystallized from ethanol to yield compound 6c.
(E)-2-amino-8-(4-methoxybenzylidene)-4-(4-methoxy-
phenyl)-5,6,7,8-tetrahydro-4H-chromene-3-carbonitriles
(6c). Yield, 40 %; mp 211–213 ꢁC. IR (KBr) mmax/cm-1
3465 and 3367 (NH2), 2187 (C:N), 1670, 1631, 1600
General method for the synthesis of (2E, 6E)-bis(2,5-di-
methoxybenzylidene)cyclohexanone (5e) To a mixture of
cyclohexanone (0.98 g, 10 mmol) and 2,5-dimethoxy-
benzaldehyde 2e (3.32 g, 20 mmol), ethanolic sodium
hydroxide solution (50 ml, 10 %) was added dropwise. The
mixture was stirred at room temperature for 2 h, then
refrigerated overnight. The separated solid was collected
by filtration, washed with water, dried and recrystallized
from ethanol to give the title compound 5e. Yield, 91 %;
mp 130–131 ꢁC. IR (KBr) mmax/cm-1 1601 (C=O). 1H
NMR (CDCl3); d: 1.22 (m, 2H, CH2), 2.60 (t, 4H, 2CH2),
3.80 (s, 12H, 4OCH3), 6.80–7.59 (m, 8H, 6ArH and 2
ylidene CH). Anal. Calcd for C24H26O5 (%): C, 73.08; H,
6.64. Found: C, 73.07; H, 6.76.
1
(C=C), 1249 (C–N). H NMR (CDCl3); d: 1.11 (m, 2H,
CH2), 2.72 (m, 4H, 2CH2), 3.78,3.81 (s, 6H, 2OCH3),
4.00 (s, 1H, pyran H), 6.11 (br s, 2H, NH2, D2O
exchangeable), 6.45–7.10 (m, 9H, 8Ar-H and ylidene
CH). MS m/z (%); 399, M?-1 (29.4), 147 (100.00). Anal.
Calcd for C25H24N2O3 (%): C, 74.98; H, 6.04; N, 7.00.
Found: C, 75.00; H, 6.32; N, 6.74.
(E)-2-amino-8-(3,4-dimethoxybenzylidene)-4-(3,4-dime-
thoxyphenyl)-5,6,7,8-tetrahydro-4H-chromene-3-carbonit-
riles (6d). Yield, 55 %; mp 145–147 ꢁC. IR (KBr) mmax
/
cm-1 3389 and 3324 (NH2), 2185 (C:N), 1668, 1638,
1603 (C=C), 1249 (C–N).1H NMR (CDCl3); d: 1.20 (m,
2H, CH2), 2.72 (t, 4H, 2CH2), 3.75, 3.81 (s, 12H, 4OCH3),
3.91 (s, 1H, pyran H), 5.81 (br s, 2H, NH2, D2O
exchangeable), 6.58–7.20 (m, 7H, 6Ar-H and ylidene CH).
Anal. Calcd for C27H28N2O5 (%): C, 70.42; H, 6.13; N,
6.08. Found: C, 70.75; H, 6.35; N, 6.25.
General method for the synthesis of (2E, 7E)-bis(2,5-di-
methoxybenzylidene)cycloheptanone (5j) To a mixture of
cycloheptanone (1.12 g, 10 mmol) and 2,5-dimethoxybenz-
aldehyde 2e (3.32 g, 20 mmol), ethanolic sodium hydrox-
ide solution (50 ml, 10 %) was added dropwise. The
mixture was stirred at room temperature for 2 h, then
refrigerated overnight. The separated solid was collected by
filtration, washed with water, dried and recrystallized from
ethanol to give the title compound 5j. Yield, 53 %; mp
(E)-2-amino-8-(2,5-dimethoxybenzylidene)-4-(2,5-dime-
thoxyphenyl)-5,6,7,8-tetrahydro-4H-chromene-3-carbonit-
riles (6e). Yield, 64 %; mp 218–219 ꢁC. IR (KBr) mmax
/
cm-1 3408 and 3331 (NH2), 2191 (C:N), 1676, 1645,
1608 (C=C), and 1224 (C–N).1H NMR (CDCl3); d: 1.32
(m, 2H, CH2), 2.89 (t, 4H, 2CH2), 3.70; 3.75 (s, 12H,
4OCH3), 4.40 (s, 1H, pyran H), 6.56–7.02 (m, 7H, 6ArH
and ylidene CH), 8.32 (br s, 2H, NH2, D2O exchangeable).
Anal. Calcd for C27H28N2O5 (%): C, 70; H, 6.13; N, 6.08.
Found: C, 69.71; H, 6.35; N, 6.18.
1
122–124 ꢁC. IR (KBr) mmax/cm-1 1604 (C=O). H NMR
(CDCl3); d: 1.77 (m, 4H, 2CH2), 2.85 (t, 4H, 2CH2), 3.80
(s, 12H, 4OCH3), 6.86–6.89 (m, 4H, 4ArH), 7.27 (s, 2H,
2ArH), 7.94 (s, 2H, 2 ylidene CH). MS m/z (%); 408, M?
(17.90), 229 (100.00). Anal. Calcd for C25H28O5 (%): C,
73.51; H, 6.91. Found: C, 73.72; H, 6.04.
(E)-2-amino-8-(4-hydroxybenzylidene)-4-(4-hydroxy-
phenyl)-5,6,7,8-tetrahydro-4H-chromene-3-carbonitriles
(6f). Yield, 40 %; mp 128–130 ꢁC. IR (KBr) mmax/cm-1
3616 (OH), 3433 and 3342 (NH2), 2192 (C:N), 1670,
1635, 1595 (C=C), and 1263 (C–N).1H NMR (CDCl3); d:
1.20 (m, 2H, CH2), 3.19 (t, 4H, 2CH2), 3.52 (s, 2H, NH2,
exchangeable), 4.17 (s, 1H, pyran H), 6.66–7.84 (m, 9H,
8Ar-H and ylidene CH), 8.56 (s, 2H, 2OH, D2O
exchangeable). Anal. Calcd for C23H20N2O3 (%): C,
74.18; H, 5.41; N, 7.52. Found: C, 74.29; H, 5.39; N,
7.89.
General method for the synthesis of (E)-2-amino-8-
((substituted)benzylidene)-4-((substituted)phenyl)-5,6,7,8-
tetrahydro-4H-chromene-3-carbonitriles (6c–g) A mixture
of the appropriate 2,6-bis(substituted benzylidene)
cyclohexanone 5d–g (1 mmol) and malononitrile (0.28 g,
4 mmol) in DMF (7 ml) and piperidine (4 drops) was
stirred at room temperature for 24 h. The reaction mixture
was then concentrated. The precipitated solid was collected
by filtration and recrystallized from methanol (6e and g).
For compounds 6d and 6f, purification on silica gel pre-
parative thin layer chromatographic plates was done using
n-hexane/ethyl acetate 1:1 as an eluent, and recrystalliza-
tion was done from ethanol.
(E)-2-amino-8-(4-hydroxy-3-methoxybenzylidene)-4-(4-
hydroxy-3-methoxyphenyl)-5,6,7,8-tetrahydro-4H-chromene-
3-carbonitriles (6g). Yield, 40 %; mp 178–179 ꢁC. 1H
NMR (CDCl3); d: 1.42 (m, 2H, CH2), 2.89 (t, 4H, 2CH2),
3.74; 3.77 (s, 6H, 2OCH3), 3.84 (s, 1H, pyran H),
6.54–6.88 (m, 7H, 6ArH and ylidene H), 8.89 (s, 2H, NH2,
D2O exchangeable), 9.12 (s, 2H, 2OH, D2O exchangeable).
MS m/z (%); 432, M? (17.63), 55 (100.00). Anal. Calcd for
For compound 6c A mixture of 2,6-bis(4-methoxyben-
zylidene)cyclohexanone 5c (0.334 g, 1 mmol) and malon-
onitrile (0.07 g, 1 mmol) in n-butanol (10 ml) was heated
123

Med Chem Res (2013) 22:1147–1162
1159
C25H24N2O5 (%): C, 69.43; H, 5.59; N, 6.48. Found: C,
69.16; H, 5.72; N, 6.33.
(E)-2-(4-bromophenyl)-7-(benzylidene)-3-(phenyl)-3,3a,
4,5,6,7-hexahydro-2H-indazoles (8a). Yield, 50 %; mp
1
168–169 ꢁC. H NMR (CDCl3); d: 1.32–3.15 (m, 7H, ali-
General method for the synthesis of (E)-7-(substituted
benzylidene)-3-(substituted phenyl)-3,3a,4,5,6,7-hexahy-
dro-2H-indazoles (7e–g) A mixture of the 2,6-diarylide-
necyclohexanones 5e–g (1 mmol), and hydrazine hydrate
(1 ml, 20 mmol) in absolute ethanol (15 ml) was heated at
reflux while stirring for 21 h. The reaction mixture was
then evaporated, and the residue obtained was washed with
water, collected by filtration, dried, and recrystallized from
ethanol to yield the title compounds.
phatic), 4.55 (d, 1H, NCH), 6.86–7.46 (m, 15H, 14Ar-H
and ylidene CH). Anal. Calcd for C26H23BrN2 (%): C,
70.43; H, 5.23; N, 6.32. Found: C, 69.92; H, 5.28; N, 6.25.
(E)-2-(4-bromophenyl)-7-(4-chlorobenzylidene)-3-(4-
chlorophenyl)-3,3a,4,5,6,7-hexahydro-2H-indazoles (8b).
Yield, 55 %; mp 98–100 ꢁC. 1H NMR (CDCl3); d:
1.08–3.08 (m, 7H, aliphatic), 4.51 (d, 1H, NCH),
6.64–7.58 (m, 13H, 12Ar-H and ylidene CH). Anal.
Calcd for C26H21BrCl2N2 (%): C, 60.96; H, 4.13; N,
5.47. Found: C, 59.41; H, 4.38; N, 5.80.
(E)-7-(2,5-dimethoxybenzylidene)-3-(2,5-dimethoxyphen-
yl)-3,3a,4,5,6,7-hexahydro-2H-indazoles (7e). Yield, 78 %;
mp 70–71 ꢁC. IR (KBr) mmax/cm-1 3338 (N–H), 1593
(C=N), 1497 (C=C), 1221(C–N). 1H NMR (CDCl3); d:
1.55–3.42 (m, 7H, aliphatic), 4.25 (s, 12H, 4OCH3), 5.00 (br
s, 1H, NH, D2O exchangeable), 5.32 (dd, 1H, NCH),
7.26–7.72 (m, 7H, 6Ar-H and ylidene CH). Anal. Calcd for
C24H28N2O4 (%): C, 70.57; H, 6.91; N, 6.86. Found: C,
70.32; H, 6.85; N, 7.02.
(E)-2-(4-bromophenyl)-7-(4-methoxybenzylidene)-3-(4-
methoxyphenyl)-3,3a,4,5,6,7-hexahydro-2H-indazoles (8c).
Yield, 45 %; mp 70–72 ꢁC. IR (KBr) mmax/cm-1 1589
1
(C=N), 1499 (C=C), 1221 (C–N). H NMR (CDCl3); d:
1.12–3.13 (m, 7H, aliphatic), 3.73 (s, 3H, OCH3); 3.78 (s, 3
H, OCH3), 4.71 (d, 1H, NCH), 6.55–7.80 (m, 13H, 12Ar-H
and ylidene CH). MS m/z (%);503, M? (3.54), 135
(100.00). Anal. Calcd for C28H27BrN2O2 (%): C, 66.80; H,
5.41; N, 5.56. Found: C, 67.02; H, 5.33; N, 5.14.
(E)-7-(4-hydroxybenzylidene)-3-(4-hydroxyphenyl)-3,3a,
4,5,6,7-hexahydro-2H-indazoles (7f). Yield, 90 %; mp
166–168 ꢁC. IR (KBr) mmax/cm-1 3272 (N–H), 1606
(E)-2-(4-bromophenyl)-7-(3,4-dimethoxybenzylidene)-3-
(3,4-dimethoxyphenyl)-3,3a,4,5,6,7-hexahydro-2H-inda-
1
(C=N), 1510 (C=C), 1237 (C–N). H NMR (CDCl3); d:
zoles (8d). Yield, 45 %; mp 178–179 ꢁC. IR (KBr) mmax
/
1
1.41–3.41 (m, 7H, aliphatic), 5.32 (br s, 1H, NH, D2O
exchangeable), 5.52 (dd, 1H, NCH), 7.23–7.85 (m, 9H,
8Ar-H and ylidene CH), 9.8 (s, 2H, 2OH). MS m/z (%);
320, M? (3.67), 215 (100.00). Anal. Calcd for C20H20N2O2
(%): C, 74.98; H, 6.29; N, 8.74. Found: C, 74.65; H, 6.60;
N, 8.47.
cm-1 1595 (C=N), 1499 (C=C), 1227 (C–N).). H NMR
(CDCl3); d: 1.11–3.45 (m, 7H, aliphatic), 3.72; 3.77; 4.72
(s, 12 H, 4OCH3), 5.17 (d, 1H, NCH), 6.52–7.85 (m, 11H,
10Ar-H and ylidene CH). Anal. Calcd for C30H31BrN2O4
(%): C, 63.95; H, 5.55; N, 4.97. Found: C, 63.70; H, 5.27;
N, 4.40.
(E)-7-(4-hydroxy-3-methoxybenzylidene)-3-(4-hydroxy-3-
methoxyphenyl)-3,3a,4,5,6,7-hexahydro-2H-indazoles (7g).
Yield, 53 %; mp 179–180 ꢁC. IR (KBr) mmax/cm-1 3342
(E)-2-(4-bromophenyl)-7-(2,5-dimethoxybenzylidene)-3-
(2,5-dimethoxyphenyl)-3,3a,4,5,6,7-hexahydro-2H-inda-
zoles (8e). Yield, 60 %; mp 87–89 ꢁC. IR (KBr) mmax/cm-1
1588 (C=N), 1494 (C=C), 1219 (C–N).1H NMR (CDCl3);
d: 1.12–3.46 (m, 7H, aliphatic), 3.73; 3.78; 4.82 (s, 12H,
4OCH3), 5.07 (d, 1H, NCH), 6.42–7.80 (m, 11H, 10Ar-H
and ylidene CH). MS m/z (%); 563, M? (9.13), 212
(100.00). Anal. Calcd for C30H31BrN2O4 (%): C, 63.95; H,
5.55; N, 4.97. Found: C, 63.67; H, 5.46; N, 4.87.
1
(N–H), 1604 (C=N), 1514 (C=C), 1270 (C–N). H NMR
(CDCl3); d: 1.50–3.31 (m, 7H, aliphatic), 3.90 (s, 6H,
2OCH3), 5.10 (br s, 1H, NH, D2O exchangeable), 5.40 (dd,
1H, NCH), 7.22–7.82 (m, 7H, 6Ar-H and ylidene CH),
10.00 (s, 2H, 2OH). MS m/z (%); 380, M? (26.37), 378,
M?-2 (100.00). Anal. Calcd for C22H24N2O4 (%): C, 69.46;
H, 6.36; N, 7.36. Found: C, 69.21; H, 6.53; N, 7.71.
General method for the synthesis of (E)-2-(4-bromophe-
nyl)-8-((substituted) benzylidene)-3-((substituted) phenyl)-
2,3,3a,4,5,6,7,8-cyclohepta[c]pyrazoles (8h–j) A mixture
of the appropriate 2,7-bis(substituted benzylidene)cy-
cloheptanones 5h–j (1 mmol), 4-bromophenylhydrazine
hydrochloride (0.671 g, 3 mmol), and sodium ethoxide
(0.408 g, 6 mmol) in absolute ethanol (20 ml) was heated
under reflux for 20 h. The reaction mixture was then fil-
tered and the filtrate was concentrated. For compounds 8h
and 8i, purification on silica gel preparative TLC plates,
using n-hexane/ethyl acetate 1:1 as an eluent was done,
followed by crystallization from ethanol. For compound 8j,
General method for the synthesis of (E)-2-(4-bromophe-
nyl)-7-((substituted)benzylidene)-3-((substituted)phenyl)-3,
3a,4,5,6,7-hexahydro-2H-indazoles (8a–e) A mixture of
the appropriate 2,6-bis((substituted)benzylidene)cyclohexa-
nones 5a–e (1 mmol), 4-bromophenylhydrazine hydro-
chloride (0.671 g, 3 mmol), and sodium ethoxide (0.408 g,
6 mmol) in absolute ethanol (20 ml) was heated under
reflux for 20 h. The reaction mixture was then filtered and
the filtrate was concentrated. The separated solid was col-
lected by filtration, washed with water, and recrystallized
from ethanol.
123

1160
Med Chem Res (2013) 22:1147–1162
Antihaemolysis assay
the ethanolic filtrate was diluted with water, then extracted
with chloroform (3 9 10 ml). The residue after evapora-
tion was recrystallized from methanol.
Human RBCs were separated from heparinised blood that
was drawn from a healthy volunteer. The blood was cen-
trifuged at 2,000 rpm for 10 min to separate the RBCs
from plasma, then the RBCs were washed three times with
phosphate-buffered saline (PBS) at pH 7.4. During the last
washing, the cells were centrifuged at 2,000 rpm for
10 min to obtain a constantly packed cell volume. The 5 %
suspension of RBCs in PBS (pH 7.4) was incubated under
air atmosphere at 37 ꢁC for 5 min, into which the tested
compounds dissolved in DMSO were added to be exam-
ined for their direct haemolytic effect. The reaction mixture
was shaken gently while being incubated at 37 ꢁC. The
extent of haemolysis was determined spectrophotometri-
cally (Kuang et al., 1994); where aliquots of the reaction
mixture were taken out at appropriate time intervals,
diluted with 0.15 M NaCl, and centrifuged at 2,000 rpm
for 10 min to separate the RBCs. The percentage haem-
olysis was determined by measuring the absorbance of
haemoglobin in the supernatant at 540 nm and compared
with that of complete haemolysis by treating the same RBC
suspension with distilled water. The final concentration of
DMSO was 0.1 % (v/v) that did not interfere with the
determination. Every experiment was repeated five times.
Compounds which caused haemolysis were excluded from
the next step of the experiment. Other compounds were
evaluated for their ability to protect RBCs from AAPH-
induced oxidative damage (antihaemolysis assay). The
procedure was repeated, where the test compounds, dis-
solved in DMSO, were added and incubated for 30 min
before addition of a PBS solution of AAPH (50 mM)
(Sigma-Aldrich, St. Louis, MO, USA).
(E)-2-(4-bromophenyl)-8-(4-methoxybenzylidene)-3-(4-
methoxyphenyl)-2,3,3a,4,5,6,7,8-cyclohepta[c] pyrazoles
(8h). Yield, 50 %; mp 53–55 ꢁC. IR (KBr) mmax/cm-1
1600 (C=N), 1500 (C=C), 1247 (C–N). 1H NMR (CDCl3);
d: 1.21–3.16 (m, 9H, aliphatic), 3.84 (s, 6 H, 2OCH3), 4.19
(d, 1H, NCH), 6.48–7.76 (m, 13H, 12Ar-H and ylidene
CH). MS m/z (%); 517, M? (2.61), 398 (100.00). Anal.
Calcd for C29H29BrN2O2 (%): C, 67.31; H, 5.65; N, 5.41.
Found: C, 67.52; H, 5.34; N, 5.15.
(E)-2-(4-bromophenyl)-8-(3,4-dimethoxybenzylidene)-3-
(3,4-dimethoxyphenyl)-2,3,3a,4,5,6,7,8-cyclohepta [c]pyr-
azoles (8i). Yield, 50 %; mp 58–60 ꢁC. 1H NMR (CDCl3);
d: 1.20–3.18 (m, 9H, aliphatic), 3.71; 3.84; 3.87; 3.90 (s, 12
H, 4OCH3), 4.29 (d, 1H, NCH), 6.58–7.66 (m, 11H, 10Ar-
H and ylidene CH). MS m/z (%); 577, M? (9.45), 426
(100.00). Anal. Calcd for C31H33BrN2O4 (%): C, 64.47; H,
5.76; N, 4.85. Found: C, 64.23; H, 5.42; N, 5.00.
(E)-2-(4-bromophenyl)-8-(2,5-dimethoxybenzylidene)-3-
(2,5-dimethoxyphenyl)-2,3,3a,4,5,6,7,8-cyclohepta [c]pyr-
azoles (8j). Yield, 50 %; mp 77–79 ꢁC. IR (KBr) mmax
/
cm-1 1592 (C=N), 1497 (C=C), 1228 (C–N). H NMR
(CDCl3); d: 1.21–3.19 (m, 9H, aliphatic), 3.70; 3.84; 3.89;
3.91 (s, 12 H, 4OCH3), 4.30 (d, 1H, NCH), 6.62–7.86 (m,
11H, 10Ar-H and ylidene CH). MS m/z (%); 577, M?
(2.00), 377 (100.00). Anal. Calcd for C31H33BrN2O4
(%):C, 64.47; H, 5.76; N, 4.85. Found: C, 64.05; H, 5.80;
N, 4.61.
1
Biological testing
Antitumor testing
ABTS test
Thecellsweregrowninsuspensionculture, partlyfloatingand
partly attached, in Roswell Park Memorial Institute (RPMI)
1640 medium (Sigma chemical Co., St. Louis, USA), sup-
plemented with 10 % foetal bovine serum (GIBCO, UK).
They were maintained at 37 ꢁC in a humidified atmosphere
with 5 % CO2. The viability of the cells used in control
experiments (DMSO only without drug) exceeded 95 % as
determined with trypan blue. Test compounds were prepared
initially at concentration of 1 mg/ml DMSO.
A mixture of solution of ABTS (2 ml, 0.1 g/100 ml,
Sigma-Aldrich, St. Louis, MO, USA) and solution of MnO2
(3 ml, 25 mg/ml), prepared in phosphate buffer (pH 7,
0.1 M, BP 1998) was shaken, centrifuged, and decanted.
The absorbance (Acontrol) of the resulting green-blue solu-
tion (ABTS.? radical solution) was recorded at kmax
734 nm. The absorbance (Atest) was measured upon the
addition of 20 ll of 1 mg/ml solution of the test sample in
spectroscopic grade MeOH/buffer (1:1 v/v) to the ABTS
solution. The decrease in absorbance is expressed as %
inhibition which is calculated from the equation:
Molecular modelling methodology
Acontrol ꢁ Atest
% inhibition ¼
ꢂ 100:
Construction of the molecular structure of the lead ligand
and the synthesized analogues
Acontrol
Ascorbic acid 20 ll (2 mM) solution was used as
standard antioxidant (positive control). Blank sample was
run using solvent without ABTS.
Putative lead ligand ‘curcumin’ and the synthesized com-
pounds were constructed from fragment libraries using the
123

Med Chem Res (2013) 22:1147–1162
1161
Dinkova-Kostova AT, Talalay P (2008) Direct and indirect antiox-
idant properties of inducers of cytoprotective proteins. Mol Nutr
Food Res 52:5128–5138
Djordjevic VB (2004) Free radical in cell biology. Int Rev Cytol
237:57–89
Du Z, Liu R, Shao W, Mao X, Ma L, Gu L, Huang Z, Cham ASC
(2006) a-Glucosidase inhibition of natural curcuminoids and
curcumin analogs. Eur J Med Chem 41:213–218
El-Sherbeny MA, Abdel-Aziz AA-M, Ahmed MA (2010) Synthesis
and antitumor evaluation of novel diarylsulfonylurea derivatives:
molecular modeling applications. Eur J Med Chem 45:689–697
Fadda AA, Badria FA, El-Attar KM (2010) Synthesis and evaluation
of curcumin analogues as cytotoxic agents. Med Chem Res
19:413–430
Geol A, Kunnumakkara AB, Aggarwal BB (2008) Curcumin as
‘‘curecumin’’: from kitchen to clinic. Biochem Pharm 75:
787–809
Guthrie N, Carroll KK (1998) Inhibition of mammary cancer by citrus
flavonoids. In: Manthey JA, Buslig BS (eds) Flavonoids in the
living system. Plenum, New York, p 227
builder module of MOE. The partial atomic charges for
each analogue were assigned with the semiempirical
mechanical calculation method ‘AM1’ implemented in
MOE. Each structure was energy minimised and geomet-
rically optimized by using the MMFF94 force field with
gradient 0.05.
Conformational search and the derived structural
global minima
The partially charged and energy minimised structures of
the tested compounds were subjected to conformational
search using the stochastic conformational search module
implemented in MOE (El-Sherbeny et al., 2010, Al-Omary
et al., 2010). Such search was performed without chiral
conversion and with using chiral constraints around the
˚
double bonds using the energy cutoff 10 A, the dihedral
˚
minimization RMS gradient 100 kcal/mol A and the
˚
Cartesian minimization RMS gradient 0.001 kcal/mol A.
Halgren TAJ (1996) Merck molecular force field. I. Basis, form,
scope, parameterization, and performance of MMFF94. J Comput
Chem 17:490–519
´
Kalai T, Kuppusamy ML, Balog M, Selvendiran K, Rivera BK,
The iteration limit was allowed to 1000 iterations where the
˚
root mean square (RMS) tolerance fixed to 0.1 kcal/mol A.
Kuppusamy P, Hideg K (2011) Synthesis of N-substituted 3,5-
bis(arylidene)-4-piperidones with high antitumor and antioxidant
activity. J Med Chem 54:5414–5421
The conformers with the lowest energy were selected as the
global minima for further modelling studies.
Kuang ZH, Wang PF, Zheng RL, Liu Z, Liu Y (1994) Making
vitamin C liposoluble enhances its protective effect against
radical induced hemolysis of erythrocytes. Chem Phys Lipids
91:95–97
Flexible alignment search
Labute P, Williams C, Feher M, Sourial E, Schmidt JM (2001)
Flexible alignment of small molecules. J Med Chem 44:1483–
1490
The fingerprint key structural features of the global minima
conformers of the selected analogues were determined
using the implemented flexible alignment search module
(Labute et al., 2001, Halgren, 1996). The iteration limit
was set to 200 iterations for each compound, turning on the
option of preserving the geometrical configuration of the
double bond centers using the MMFF94 force field opti-
mization energy to get proper alignment feature analysis.
Lee K, Ab Aziz FH, Syahida A, Abas F, Shaari K, Israf DA, Lajis NH
(2009) Synthesis and biological evaluation of curcumin-like
diarylpentanoid analogues for anti-inflammatory, antioxidant
and antityrosinase activities. Eur J Med Chem 44:3195–3200
Liang G, Shao L, Wang Y, Zhao C, Chu Y, Xiao J, Zhao Y, Li X,
Yang S (2009) Exploration and synthesis of curcumin analogues
with improved structural stability both in vitro and in vivo as
cytotoxic agents. Bioorg Med Chem 17:2623–2631
Ligeret H, Barthelemy S, Zini R, Tillement J, Labidalle S, Morin D
(2004) Effects of curcumin and curcumin derivatives on
mitochondrial permeability transition pore. Free Radic Biol
Med 36:919–929
Acknowledgments The authors extend their appreciation to
Mansoura University research center for funding this study.
Lipinski CA, Lombardo F, Dominy BW, Feeney PJ (1997) Exper-
imental and computational approaches to estimate solubility and
permeability in drug discovery and development settings. Adv
Drug Del Rev 23:3–25
References
Lissi E, Modak B, Torres R, Escobar J, Urzua A (1999) Total
antioxidant potential of resinous exudates from Heliotropium
species, and a comparison of ABTS and DPPH methods. Free
Radic Res 30:471–477
Aggarwal BB, Kumar A, Bharti AC (2003) Anticancer potential of
curcumin: preclinical and clinical studies. Anticancer Res
23:363–398
Al-Omary FAM, Abou-zeid LA, Nagi MN, Habib EE, Abdel-Aziz
AA-M, El-Azab AS, Abdel-Hamide SG, Al-Omar MA,
Al-Obaid AM, El-Subbagh HI (2010) Non-classical antifolates.
Part 2. Synthesis, biological evaluation, and molecular modeling
study of some new 2,6-substituted quinazolin-4-ones. Bioorg
Med Chem 18:2849–2863
Memon VP, Sudheer AR (2007) Antioxidant and anti-inflammatory
properties of curcumin. In: Aggarwal BB, Surh Y, Shishodia S
(eds) The molecular targets and therapeutic uses of curcumin in
health and disease. Springer, New York, pp 106–107
MOE 2007.9 of Chemical Computing Group. Inc.
Oprea TI, Davis AM, Teague SJ, Leeson PD (2001) Is there a
difference between leads and drugs? A historical perspective.
J Chem Inf Comput Sci 41:1308–1315
Roughley PJ, Whiting DA (1973) Experiments in the biosynthesis of
curcumin. J Chem Soc Perkin Trans 1(20):2379–2388
Shang Y, Jin X, Shang X, Tang J, Liu G, Dai F, Qian Y, Fan G, Liu Q,
Zhou B (2010) Antioxidant capacity of curcumin-directed
Anand P, Kunnumakkara AB, Newman RB, Aggarwal BB (2007)
Bioavailability of curcumin: problems and promises. Mol Pharm
4:807–818
.?
´
Chyong FH, Hui P, Cedric B, Jadranka T-S, Paul A (2011) ABTS
scavenging activity of polypyrrole, polyaniline and poly(3,4-
ethylenedioxythiophene). Polym Int 60:69–77
123

1162
Med Chem Res (2013) 22:1147–1162
analogues: structure–activity relationship and influence of
microenvironment. Food Chem 119:1435–1442
catalyzed by Yb(OTf)3 under solvent-free conditions. Synthesis
18:3060–3064
Sharma RA, Gescher AJ, Verma WP (2005) Curcumin: the story so
far. Eur J Cancer 41:1958–1968
Weber WM, Hunsaker LA, Abcouwer SF, Deck LM, Vander Jagt DL
(2005) Anti-oxidant activities of curcumin and related enones.
Bioorg Med Chem 13:3811–3820
Weber WM, Hunsaker LA, Gonzales AM, Heynekamp JJ, Orlando
RA, Deck LM, Vander Jagt DL (2006) TPA-induced up-
regulation of activator protein-1 can be inhibited or enhanced by
analogs of the natural product curcumin. Biochem Pharmacol
72:928–940
Singh N, Pandey J, Yadav A, Chaturvedi V, Bhatnagar S, Gaikwad
AN, Sinha SK, Kumar A, Shukla PK, Tripathi RT (2009) A
facile synthesis of a,a’-(EE)-bis(benzylidene)-cycloalkanones
and their antitubercular evaluations. Eur J Med Chem 44:
1705–1709
Venkatesan P, Rao MN (2000) Structure–activity relationships for the
inhibition of lipid peroxidation and the scavenging of free
radicals by synthetic symmetrical curcumin analogues. J Pharm
Pharmacol 52:1123–1128
Zhou J (2003) One-step synthesis of pyridine and 4H-pyran deriv-
atives from bisarylidenecyclohexanone and malononitrile under
microwave irradiation. Syn Commun 33:99–103
Wang L, Sheng J, Tian H, Han J, Fan Z, Qian C (2004) A convenient
synthesis of a,a’-bis(substituted benzylidene)cycloalkanones
123