Antinociceptive and antiinflammatory activities and QSAR studies on 2-substituted-4,5-diphenyl-1H-imidazoles

Article abstract of DOI:10.1016/j.bmc.2006.10.025

This paper describes the pharmacological evaluation pertaining to antinociceptive (hot plate and tail flick) and antiinflammatory (based on Carrageenan-induced paw oedema) activities, and QSAR studies on 2-substituted-4,5-diphenyl-1H-imidazoles. Compounds with phenyl substitution with -F, -Cl, -NH₂, -N(CH₃)₂, -OH and -OCH₃ at the p-position showed higher activity than the other substitutions in all the three studies. QSARs developed for the 60 and 120 s hot plate data indicate that the models for both the cases not only fit the data very well (R² > 0.9, R_adj² > 0.86), but also have very good predictive capability (q² > 0.81). The descriptors used in the model relate to surface area, volume, dipole moment and ADME properties of the molecule. Good QSARs for the 60 and 120 s tail flick data are developed. The models fit the data well (R² > 0.8, R_adj² > 0.74), and in addition have good predictive capability (q² > 0.66). Surface area, specifically polar surface area, HOMO and molecular connectivity index appear in the models. Very good QSAR model is developed for the antiinflammatory data (R² = 0.86, R_adj² = 0.822 and q² = 0.64) with aqueous solubility, number of hydrogen bond donor groups, surface area and principal moment of inertia as the molecular descriptors.

Full text of DOI:10.1016/j.bmc.2006.10.025

Bioorganic & Medicinal Chemistry 15 (2007) 1083–1090
Antinociceptive and antiinflammatory activities and QSAR
studies on 2-substituted-4,5-diphenyl-1H-imidazoles
A. Puratchikodya and Mukesh Dobleb,*
aDepartment of Pharmaceutical Engineering and Technology, Bharathidasan Institute of Technology,
Bharathidasan University, Tiruchirappalli 620024, India
bDepartment of Biotechnology, Indian Institute of Technology Madras, Chennai 600036, India
Received 23 August 2006; revised 6 October 2006; accepted 11 October 2006
Available online 18 October 2006
Abstract—This paper describes the pharmacological evaluation pertaining to antinociceptive (hot plate and tail flick) and antiin-
flammatory (based on Carrageenan-induced paw oedema) activities, and QSAR studies on 2-substituted-4,5-diphenyl-1H-imidaz-
oles. Compounds with phenyl substitution with –F, –Cl, –NH2, –N(CH3)2, –OH and –OCH3 at the p-position showed higher
activity than the other substitutions in all the three studies. QSARs developed for the 60 and 120 s hot plate data indicate that
the models for both the cases not only fit the data very well (R2 > 0.9, Ra2dj > 0:86), but also have very good predictive capability
(q2 > 0.81). The descriptors used in the model relate to surface area, volume, dipole moment and ADME properties of the molecule.
Good QSARs for the 60 and 120 s tail flick data are developed. The models fit the data well (R2 > 0.8, Ra2dj > 0:74), and in addition
have good predictive capability (q2 > 0.66). Surface area, specifically polar surface area, HOMO and molecular connectivity index
appear in the models. Very good QSAR model is developed for the antiinflammatory data (R2 = 0.86, Ra2dj ¼ 0:822 and q2 = 0.64)
with aqueous solubility, number of hydrogen bond donor groups, surface area and principal moment of inertia as the molecular
descriptors.
Ó 2006 Elsevier Ltd. All rights reserved.
1. Introduction
Several papers report QSAR studies on benzimidazole
derivatives. The in vitro activity of benzimidazole
chloroaryloxyalkyl derivatives against Salmonella typhi
O-901 and Staphylococcus aureus A 15091 is found to
depend on three molecular descriptors namely, HOMO
(Highest occupied molecular orbital) energy, hydration
energy and number of primary carbon atoms of the
molecule.16 2-Aminobenzimidazole moiety connected
to the 4 (5) position of an imidazole ring through di-
or tri-methylene chains exhibits activity on rat brain
membranes and the activity is found to correlate with
logP through a parabolic relationship.17 Activities of
benzimidazoles against Bacillus subtilis also indicate a
similar relationship. Sener et al. have developed the
QSAR of activity of benzimidazoles against Klebsiella
pneumoniae18 and concluded that the more potent
compound possesses an oxazolo pyridine ring system
substituted with an electron-withdrawing group at
position 5 and benzyl moiety at position 2.19 In addition,
they found that the activity of these compounds against
Candida albicans highly correlated with LUMO (lowest
unoccupied occupied molecular orbital), molecular
weight, resonance effect and HOMO.20 Oxazolo
pyridine ring system with the substitution of a benzyl
moiety at position 2 exhibits good activity against K.
Imidazole derivatives have occupied a unique place in
the field of medicinal chemistry. They have wide range
of biological activities. They are well-known analgesics,1
antiinflammmatory,2 antiparasitic,3 platelet aggregation
inhibitors,4 and antiepileptic agents.5 Imidazole can be
found in many other drugs such as dacarbazine,6 metro-
nidazole,7 cimetidine,8 flumazenil,9 thyroliberin,10 meth-
imazole,11 pilocarpine,12 and etomidate13 which are used
as antineoplastic, antibiotic, antiulcerative, benzodiaze-
pine antagonist, prohormone, antihyperthyroid, musca-
rinic receptor antagonist and hypnotic agents,
respectively. In view of such reports and in continuation
to our work on the imidazole derivatives,14,15 we report
here the in vivo pharmacological evaluation and
QSAR studies on new 2-substituted-4,5-diphenyl-1H-
imidazoles.
Keywords: 2-Substituted-4,5-diphenyl-1H-imidazoles; Molecular desc-
riptors; Surface area; HOMO; Genetic function algorithm.
*
Corresponding author. Tel.: +91 44 22374107; fax: +91 44
22374102; e-mail: mukeshd@iitm.ac.in
0968-0896/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2006.10.025

1084
A. Puratchikody, M. Doble / Bioorg. Med. Chem. 15 (2007) 1083–1090
pneumoniae and groups which possessed hydrogen-ac-
cepting capability improved the activity.21
O
O
C6H5
C6H5
O
2NH4OCOCH3
R
H
2. Results and discussion
Compounds listed in Table 1 were obtained by the meth-
od outlined in Figure 1. The synthesis of 1–25 was car-
ried out by condensation of benzil with ammonium
acetate and appropriate aldehyde(s). Substituted benzal-
dehyde(s) is used to give compounds 1–21. Aldehyde
containing alkyl, alkenyl or styryl unit is used to give
compounds 22–25. The yields of 1–25 fall in the range
of 65–82%. Most of them are colourless crystalline sol-
ids. The preparation procedure and properties of these
imidazole derivatives have been discussed somewhere
else in more detail.14,15
glacial acetic acid
stir, room temperature
1-2 hr
C6H5
C6H5
3H2O
CH3COOH
N
R
HN
Table 1 shows the result of antinociceptive and antiin-
flammatory studies. In the hot plate and tail flick mod-
els, there was no significant difference in the mean
predrug reaction time among the different groups. Thir-
ty minutes after the administration of the drug, reaction
time increased significantly for the test and standard
groups when compared to the predrug reaction time.
The compounds 12, 13, 17, 18, 19 and 21 showed their
potency as antinociceptive agents by their greater in-
crease in reaction time. The compounds 2, 4, 6, 7, 10
Figure 1. Synthesis scheme.
and 11 increased the pain threshold significantly after
30 min. The groups such as –F, –Cl, –NH2, –N(CH3)2,
–OH and –OCH3 at the p-position of the phenyl substi-
tuted imidazoles (compounds 12, 13, 17, 18, 19 and 21,
respectively) produced significant antinociceptive activi-
Table 1. Compounds synthesised and their corresponding biological activity
C6H5
C6H5
HN
N
R
Compound
R
Hot plate method. Experiments Tail flick method. Experiments % reduction in paw volume
performed after (in s) performed after (in s)
0
15
30
60
120
0
15 30 60 120
4.12 4.13 4.67 5.12 5.78 30.24
4.14 5.78 7.27
1
2
Phenyl
4.63 4.58 5.28 6.07
4.67 4.62 6.25 8.2
6.1
2-Chlorophenyl
2-Nitrophenyl
2-Hydroxyphenyl
2-Methylphenyl
2-Methoxyphenyl
3-Chlorophenyl
3-Nitrophenyl
3-Methylphenyl
3-Methoxyphenyl
9.01 4.2
4.71 4.09 4.1
9.16 4.2
8.85 62.8
4.39 30.24
9.12 60.47
5.52 18.61
9.23 65.12
8.18 55.82
3
4.5
4.89 4.93 6.32 8.59
4.52 4.62 4.67
4.38 4.38
4.18 5.82 7.53
4
5
4.5
4.9
4.55 4.67 4.79
4.94 6.89 8.65
5.79 4.11 4.12 4.52 4.98
9.25 4.21 4.21 5.98 7.64
8.91 4.12 4.12 4.98 6.89
5.62 4.09 4.09 4.36 5.74
6
7
4.68 5.14 6.18 7.78
4.55 4.56 4.6 4.62
4.56 4.6 4.63 4.67
4.67 5.02 6.15 7.42
7.92
8
5.82
5.93
4.66
9.31
9
5.72 4.1
4.1
4.48 5.89
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
8.79 4.12 4.12 4.56 6.75
8.99 4.16 4.12 5.57 7.18
7.99 37.21
8.34 58.14
4-Hydroxy-3-methoxyphenyl 4.68 5.23 6.2
4-Fluorophenyl
4-Chlorophenyl
4-Bromophenyl
4-Iodophenyl
4-Nitrophenyl
4-Aminophenyl
4-Dimethylaminophenyl
4-Hydroxyphenyl
4-Methylphenyl
4-Methoxyphenyl
H
4.78 4.93 7.5
4.66 4.9
9.33 10.83 4.55 4.87 7.14 8.92 10.67 79.07
7.43 8.98 10.35 4.52 4.67 6.92 8.37 10.5
6.38 4.2
81.4
6.83 74.42
4.45 4.65 6.37 6.38
4.45 4.64 5.33 6.18
4.45 5.33 6.5
6.18 4.16 4.26 5.21 5.89
5. 83 4.13 4.14 5.09 5.12
6.5 27.91
5.13 25.59
4.37 4.6
4.66 5.82
4.48 4.69 7.01 8.25 10.08 4.22 4.27 6.18 7.77 10.05 81.4
4.5 4.76 7.16 8.33 10.3 4.26 4.45 6.32 7.93 10.18 69.77
4.55 4.81 7.21 8.63 10.33 4.4 4.48 6.79 8.18 10.33 83.73
4.35 4.5 4.62 4.83 4.82 4.14 4.14 4.82 4.82 4.89 11.63
4.87 5.02 7.89 9.67 10.87 4.58 4.98 7.45 9.17 10.83 86.05
4.66 4.83 4.83 5.05
4.65 5.1 5.33 5.47
5.2
3.9
3.92 4.01 4.01
4.02 20.94
4.35 25.59
4.29 32.56
4.75 55.82
Methyl
2-Propenyl
2-Styryl
5.52 4.12 4.17 4.28 4.35
5.47 3.98 4.02 4.23 4.27
4.66 5.08 5.12 5.33
4.67 5.17 5.58 5.94
6.12 4.16 4.28
0
4.68

A. Puratchikody, M. Doble / Bioorg. Med. Chem. 15 (2007) 1083–1090
1085
ty (P < 0.001). Presence of groups like –Cl, –OH and –
1.2
1
OCH3 at the o-position of the phenyl substituted imi-
dazoles (compounds 2, 4 and 6, respectively) showed
moderately significant activity (P < 0.01). Substitution
of groups such as –Cl, –OCH3 and –4-OH–3-OCH3 at
the m-position of the phenyl substituted imidazoles
(compounds 7, 10 and 11, respectively) exhibited signif-
icantly less activity (P < 0.02).
0.8
0.6
0.4
0.2
0
In the acute inflammation model, the compounds 12, 13,
18 and 19 showed maximum inhibition of carrageenan-
induced rat paw oedema at the end of 3 h. Moreover,
oedema suppressant effect of the compounds 2, 4, 6, 7,
11, 14, 17, 21 and 25 was found to be moderately signif-
icant as compared to control (P < 0.01). Carrageenan-
induced hind paw oedema is the standard experimental
model of acute inflammation. Carrageenan is the phlo-
gistic agent of choice for testing antiinflammatory drugs
as it is not known to be antigenic and is devoid of appar-
ent systemic effects. Moreover, the experimental model
exhibits a high degree of reproducibility.22 Carrageen-
an-induced oedema is a biphasic response. The first
phase is mediated through the release of histamine, sero-
tonin and kinins, whereas the second phase is related to
the release of prostaglandin and other slow-reacting
substances which peak at 3 h.23 The compound 21
exhibited maximum inhibition of 86.0% and, 25 showed
a minimum inhibition of 55.8%, while the standard drug
indomethacin showed an inhibition of 95.3% in the car-
rageenan-induced rat paw oedema (acute) model. The
inhibition of the test drugs was however less than that
of the standard drug indomethacin.
1
8 9 5 16 20 25 3 23 24 22 14 15 2 4 6 7 11 10 12 21 13 19 17 18
Objects
Figure 3. Cluster analysis of tail flick data (120 s data).
1.4
1.2
1
0.8
0.6
0.4
0.2
0
Figures 2–4 show the cluster analysis of the activity data
for the hot plate, tail flick and antiinflammatory
studies, respectively. Compounds 12 (4-fluorophenyl),
13 (4-chlorophenyl), 17 (4-aminophenyl), 18 (4-dimethyl-
amino- phenyl), 19 (4-hydroxyphenyl) and 21 (4-meth-
oxyphenyl) exhibit highest activity in all the three in vivo
studies and are located in a single cluster. This is
possibly because these compounds have electron-donat-
ing groups. In addition compounds having N or O are
more hydrophilic than the other groups. Compound 1
1
3 24 15 16 23 10 5 22 8 9 20 2 6 4 7 25 11 12 13 17 19 21 14 18
Objects
Figure 4. Cluster analysis of antiinflammatory data.
(phenyl) exhibits the lowest activity in all the three in vivo
studies, probably because of its hydrophobic nature.
Table 2 shows the results of the ANOVA of the data
from hot plate studies. Statistically significant difference
is seen between the response time data collected at var-
ious time intervals (F = 23.5, P < 0.001), with a steady
increase in the means of 25 data values. The mean of
the hot plate response time increases from 4.6 to 7.6 s
when the data collection is carried out at 120 instead
of 0 s. Table 3 shows the results of the ANOVA of the
data from tail flick studies. Statistically significant differ-
ence is seen between the response time data collected at
various time intervals (F = 24.38, P < 0.001), with a
steady increase in the means of the 25 data values. The
mean of the tail-flick response time is 4.2–7.3 s when
the data collection is carried out at 0 and 120 s,
respectively.
1.6
1.4
1.2
1
0.8
0.6
0.4
0.2
0
Table 4 lists the best QSAR models for the 60 and 120 s
hot plate data. The results indicate that the models
developed for both the cases not only fit the data well
(R2 > 0.9), but also have very good predictive capability
(q2 > 0.81). Description of the descriptors used in the
1
25 15 14
3
20
5
16
9
8
23 24 22
2
11
7
10
4
6 12 21 13 19 18 17
Objects
Figure 2. Cluster analysis of hot plate data (120 s data).

1086
A. Puratchikody, M. Doble / Bioorg. Med. Chem. 15 (2007) 1083–1090
Table 2. One-way ANOVA of the data from hot plate method
Experiments performed after (s)
0
15
30
5.91
60
6.86
120
Mean of 25 data (s)
Variance
4.61
0.022
4.81
0.053
7.61
4.762
1.1518
2.908
F (cal)
23.445
Factor
Sum of squares
Degrees of freedom
Mean SS
Between groups
Within groups
166.89
213.54
4
120
41.72
1.779
(*P 6 0.001)
Total
380.43
124
Table 3. One-way ANOVA of the data from tail flick studies
Experiments performed after (s)
0
15
30
60
120
Mean of 25 data (s)
Variance
4.19
0.026
4.26
0.066
5.37
1.03
6.37
2.50
7.3
5.64
Factor
Sum of squares
Degrees of freedom
Mean SS
F (cal)
Between groups
Within groups
181.42
221.34
4
119
45.36
1.86
24.38
(*P 6 0.001)
Total
402.77
123
Table 4. Best QSAR model for the data from hot plate method
Regression model
R2
R2adj
q2
F
MSSE
Experiments performed after 60 s
Activity =
0.696 + 0.413 Jurs-TASA ꢀ 2.447
0.896
0.862
0.806
25.97
0.289
*
ADME_Absorption-T2-2D ꢀ 0.0397
*
Jurs-SASA + 6.233 ADME_BBB_2D ꢀ 0.213
*
*
*
Jurs-PNSA-3 + 0.00514 Weiner
*
Experiments performed after 120 s
Activity = ꢀ10.363 + 0.382
0.922
0.902
0.868
45.1
0.35
Jurs-RNCS ꢀ 1.776
*
ADME_Absorption_T2_2D + 0.0992
Vm ꢀ 0.0955 ADME_PSA_2D + 0.146 Dipole-mag
*
*
*
*
Table 5. Details of the descriptors used in the QSAR models listed in Table 4
Descriptors name
Jurs-TASA
Explanation
Total hydrophobic surface area: sum of solvent-accessible surface areas of atoms with absolute value of partial
charges less than 0.2.
Total molecular solvent-accessible surface area.
Jurs-SASA
Jurs-PNSA-3
Atomic charge weighted negative surface area: sum of the product of solvent-accessible surface area X partial
charge for all negatively charged atoms.
Vm
Wiener
Molecular volume inside the contact surface. Molecular volume is related to binding and transport.
The sum of the chemical bonds existing between all pairs of heavy atoms in the molecule.
Indicates the strength and orientation behaviour of a molecule in an electrostatic field.
Blood brain barrier permeation-depends on H-bonding capacity, local hydrophobicity, molecular size,
lipophilicity and flexibility.
Dipole-mag
ADME_BBB_2D
ADME_PSA_2D
ADME polar surface area.
ADME_Absorption_T2_2D Multivariate distance among nitrogen, oxygen atoms and attached hydrogen atoms from van Der Waals polar
surface area.
models is listed in Table 5. All the descriptors relate to
surface area, volume, dipole moment and ADME
(Absorption, Distribution, Metabolism and Excretion)
properties of the molecule. Dipole moment is caused
due to the asymmetry in the 3-D structure of the mole-
cule. Molecular volume, shape and hydrophobicity of
the substituent group are found to affect fungicidal
activity of benzimidazoles.24 Figure 5 compares the
hot plate data and model predictions in the form of a
parity plot for the experiments carried out at 120 s.
Table 6 lists the best QSAR models for the 60 and 120 s
tail flick data. The results indicate that the models devel-
oped for both the cases not only fit the data well
(R2 > 0.8), but also have good predictive capability
(q2 > 0.66). In addition to surface area, specifically polar

A. Puratchikody, M. Doble / Bioorg. Med. Chem. 15 (2007) 1083–1090
1087
12
11
10
9
12
11
10
9
8
8
7
7
6
6
5
5
4
4
4
5
6
7
8
9
10
11
12
4
5
6
7
8
9
10
11
12
Activity (data)
Activity (data)
Figure 5. Parity plot of hot plate data (experiments carried out at
120 s).
Figure 6. Parity plot of tail flick data (experiments carried out at
120 s).
surface area, HOMO, the electronic descriptor and
molecular connectivity index also appear in the model
(see Table 7). HOMO has been identified as a molecular
descriptor by others to describe in vitro activity of benz-
imidazole chloroaryloxyalkyl derivatives against S. typhi
O-901 and S. aureus A 15091.16,20 The polar surface area
is an indication of the hydrophilic nature of the mole-
cule. Relative lyphophilic character, as defined by octa-
nol/water partition coefficient (namely logP), is a very
important property that determines the biological activ-
ity of substituted imidazoles.25 Figure 6 shows the parity
plot of the tail flick experiments carried out at 120 s and
the model predictions.
of the hydrophilic nature of the molecule. Descriptor
that characterises the hydrogen bond donor/acceptor
capability is found to be essential to describe the anti-
bacterial activity of substituted benzimidazoles against
K. pneumoniae.21 Similar descriptors have been identi-
fied by other researchers as well for different imidaz-
oles.20 Figure 7 compares the antiinflammatory data
and model predictions.
3. Conclusions
Imidazole derivatives have a wide-ranging biological
activity. In this paper, the determination of antinocicep-
tive and antiinflammatory activities of 2-substituted-4,5-
diphenyl-1H-imidazoles is reported. The antinociceptive
activity is based on hot plate and tail flick methods and
the antiinflammatory studies are based on carageenan-
induced paw oedema. Compounds with phenyl substitu-
tion with –F, –Cl, –NH2, –N(CH3)2, –OH and –OCH3 at
p-position showed higher activity than all other substitu-
tions in all the three studies. Electron-donating groups
Table 8 lists the best QSAR model for the antiinflamma-
tory data. The model fits the data very well (R2 = 0.86)
and also has a good predictive capability (q2 = 0.64). A
description of the descriptors used in the models is listed
in Table 9. Aqueous solubility, number of hydrogen
bond donor groups, surface area and principal moment
of inertia are the descriptors that appear in the model
for the activity data. Aqueous solubility is an indication
Table 6. Best QSAR model for data from tail flick method
Regression model
R2
R2adj
q2
F
MSSE
Experiments
performed after 60 s
Activity = ꢀ12.254 ꢀ 1.9950 ADME_Absorption_level_2D
0.80
0.742 0.665 14.85 0.489
*
+ 0.3704 Jurs-RNCS + 15.224 Jurs-FPSA-1 + 2.4993
* *
CHI-V-3_P + 0.2030 HOMO
*
*
Experiments
performed after 120 s
Activity = ꢀ10.002 + 3.242 ADME_BBB_2D ꢀ 1.9495
0.863 0.836 0.789 31.5
0.74
*
ADME_Absorption_T2_2D + 0.4206 Jurs-RNCS + 5.7414
*
CHI-V-3_P
*
*
Table 7. Details of the descriptors used in the QSAR models listed in Table 6
Descriptors name Explanation
ADME_Absorption_level_2D The Absorption level is calculated based on the position of each molecule in the polar surface area versus
AlogP98 plot.
HOMO
Highest energy level in the molecule that contains electrons molecules with high HOMOs are more able to donate
their electrons and are hence relatively reactivity of the molecule.
Molecular connectivity index.
CHI-V-3_P
Jurs-RNCS
Jurs-FPSA-1
Relative negative charge surface area: solvent-accessible surface area of most negative atom divided by descriptor.
Fractional charged partial surface areas.

1088
A. Puratchikody, M. Doble / Bioorg. Med. Chem. 15 (2007) 1083–1090
Table 8. Best QSAR model for data from antiinflammatory study (% reduction in paw volume)
Regression model
R2
R2adj
q2
F
MSSE
93.88
Activity = 264.84 + 57.73 ADME_Solubility ꢀ 177.54 Jurs-RPSA
0.859
0.822
0.642
231.8
*
*
ꢀ 9.87 Jurs-WNSA-3 + 31.43 Hbond donor + 0.03 PMI-mag
*
*
*
Table 9. Details of the descriptors used in the QSAR model listed in Table 8
Descriptors name
Explanation
ADME_Solubility
Jurs-RPSA
Prediction of aqueous solubility.
Total polar surface area divided by the total molecular solvent-accessible surface area.
Surface-weighted charged partial surface area.
Number of hydrogen bond donor groups.
Jurs-WNSA-3
Hbond donor
PMI-mag
Principal moment of inertia calculated about the principal axes of a molecule.
104
94
84
74
64
54
44
34
24
14
4
4.2. Pharmacology
The test compounds were evaluated for antinociceptive
and antiinflammatory activities. Student’s t test27 was
performed for all the activities to ascertain the signifi-
cance of the exhibited activities. The test compounds
and the standard drugs were administered in the form
of suspension (0.2% Tween suspension as vehicle) in
the same route of administration.
4.3. Animal experimentation
Biological evaluation of 1–25 was carried out in the
Department of Animal Biotechnology, Bharathidasan
University, Tiruchirappalli. Animal facility of this insti-
tute is approved by CPCSEA (Reg.No.418/a/01/CPC-
4
24
44
64
84
104
Activity (data)
Figure 7. Parity plot of antiinflammatory data (% reduction in paw
volume).
SEA).
The
experimental
protocols
for
the
antinociceptive and antiinflammatory activities have
been approved by the Institutional Animal Ethics Com-
mittee and conducted according to the guidelines of
Indian National Science Academy for the use and care
of experimental animals. The animals were maintained
at a well-ventilated, temperature-controlled (30 1 °C)
animal room for 7 days prior to the experimental period
and provided with food and water ad libitum. The ani-
mals were acclimatized to laboratory conditions before
the test.
and hydrophilicity play an important role in the biolog-
ical activity. Lowering of activity was observed with
hydrophobic groups. Quantitative structure–activity
relationships are developed correlating the observed bio-
logical activity with structural descriptors. For all the
cases the models developed are able to fit the data very
well (R2 > 0.8, Ra2dj > 0:74) with good predictive capabil-
ity (q2 > 0.64). The molecular descriptors used in the
models relate to surface area, volume, dipole moment,
ADME, HOMO, molecular connectivity index, aqueous
solubility, number of hydrogen bond donor group, and
principal moment of inertia. A genetic function algo-
rithm technique is used to select the descriptors from a
large pool of calculated molecular properties. This tech-
nique is useful in identifying the best descriptors when
their number is much larger when compared to the
experimental data.
4.4. Antinociceptive activity
Hot plate and tail flick methods28,29 were employed to
determine antinociceptive activity. The hot-plate test
was assessed on eleven groups of six mice. The Swiss albi-
no mice (20–25 g) were used in the test. One group served
as standard (received pentazocine 10 mg/kg ip), the other
group served as control (received 0.2% Tween suspension
5 ml/kg, ip). The remaining 25 groups received between
9.5 and 12.58 mg/kg of the compounds 1–25 ip. The bas-
al reaction time of the animal was noted by placing the
animals in the hot plate at 55 0.5 °C before and 15,
30, 60, and 120 min after the administration of test drugs.
Cut-off time was kept at 15 s in order to avoid injury to
the tail. Tail flick to radiant heat (Tail-flick apparatus
type 812, Hugo Sachs Elektronik, Germany) was used
to measure acute nociceptive responses in mice. The
Swiss albino mice (20–25 g) were used in this study and
worked up as mentioned in the above method. The inten-
4. Experimental
4.1. Acute toxicity study
Albino mice of both sexes were divided into 125 groups.
Each group had 10 animals. They were treated with test
drugs (ip) with the dose range from 40 to 220 mg/kg.
The animals were observed for mortality until 72 h
and the LD50 was calculated using graphical method.26

A. Puratchikody, M. Doble / Bioorg. Med. Chem. 15 (2007) 1083–1090
1089
sity of the thermal stimulus was adjusted to produce 3–
4 s latency in tail-flick response. The latency was mea-
sured just before, 15, 30, 60 and 120 min after injections.
The trial was automatically terminated at 12 s if a re-
sponse did not occur (cut-off time). The prolongation
of the latency times compared with the values of the con-
trol was used for statistical comparison.
evolving random initial models using a genetic algo-
rithm. GFA can build models using not only linear poly-
nomials but also higher-order polynomials, splines and
other nonlinear functions. The goodness of the regres-
sion fits were estimated using statistical parameters such
as,
R2
(=1 ꢀ SSE/TSS),
R2adjð¼ 1 ꢀ ðn ꢀ 1Þ
ð1 ꢀ R2Þ=ðn ꢀ p ꢀ 1ÞÞ, q2 (=1 ꢀ PRESS/TSS) and F ratio
(=(n ꢀ 2)R2/(1 ꢀ R2)) and MSSE (=mean sum of square
of error = SSE/n), where TSS = total sum of squares and
SSE = sum of squares, PRESS = predictive sum of
squares based on leave-one-out method. R2 and Ra2dj
are indication of the model fit, while q2 is an indication
of the predictive capability of the model. A large F indi-
cates that the model fit is not a chance occurrence. R2,
R2adj and q2 above a value of 0.6 indicate good model fit.
4.5. Antiinflammatory activity
Antiinflammatory activity was determined by carrageen-
an-induced rat paw oedema method.30 Wistar rats of
either sex (120–150 g) were used in this study and
worked up as mentioned in the above method. The stan-
dard drug used here was indomethacin (10 mg/kg ip).
The control group received 0.2% Tween suspension
5 ml/kg ip. All the remaining 25 groups received between
9.5 and 12.58 mg/kg of the compounds 1–25 ip. After an
hour of test drug administration the animals were inject-
ed with 0.05 ml suspension of carrageenan (1.0% w/v in
0.9% NaCl) in the left hind paw planter aponeurosis.
The volume of paw was measured using the mercury dis-
placement technique with the help of a plethysmometer
(UGO-Basile, Italy). It was measured both in control as
well as in standard animals including the test animals at
an interval of 1, 2 and 3 h after carrageenan injection.
The initial volume of paw was measured immediately
before the injection. The increase in paw volume after
3 h was calculated. The percentage inhibition of inflam-
mation after 3 h was calculated by using Eq. 1.
References and notes
1. Ucucu, D.; Karaburn, N. G.; Isikdag, I. Farmaco 2001, 56,
285.
2. Maeda, S.; Suzuki, M.; Iwasaki, T.; Matsumoto, K.;
Iwasawa, Y. Chem. Pharm. Bull. 1984, 32, 2536.
3. Quattara, L.; Debaert, M.; Cavier, R. Farmaco (Sci.)
1987, 42, 449.
4. Seko, N.; Yoshino, K.; Yokota, K.; Tsukamoto, G. Chem.
Pharm. Bull. 1991, 39, 651.
5. Puratchikody, A.; Yasodha, A.; Ruckmani, K.; Sarkun-
am, K.; Nallu, M. Indian J. Heterocycl. Chem. 2003, 14,
79.
6. Shealy, Y. F.; Krauth, C. A.; Montgomery, J. A. J. Org.
Chem. 1962, 27, 2150.
7. Brogden, R. N.; Heel, R. C.; Speigt, T. M. Drugs 1978, 16,
387.
8. Brimblecombe, R. W.; Duncan, W. A. M.; Durant, G. J.;
Emmet, C.; Gamellin, C. R.; Parsons, M. E. J. Int. Med.
Res. 1975, 3, 86.
Percentage inhibition ¼ 100½1 ꢀ ðV tÞ=ðV cÞꢁ
ð1Þ
where Vt = Mean relative changes in the paw volume of
each of the rats after the administration of carrageenan
and test or standard compound.
9. Hunkeler, W.; Mohler, H.; Pieri, L.; Polc, P.; Bonetti, E.
P.; Cumin, R.; Schaffner, R.; Haefely, W. Nature 1981,
290, 514.
10. Fluoret, G. J. Med. Chem. 1970, 13, 843.
11. Cooper, D. S. N. Engl. J. Med. 1984, 311, 1353.
12. Mayorga, A. J.; Cousins, M. S.; Trevitt, J. T.; Coulan, A.;
Gianutsos, G.; Salamone, J. D. Eur. J. Pharmacol. 1999,
364, 7.
13. Godefroi, E. F.; Janssen, P. A. J.; Van der Eycken, C. A.
M.; Van Heertum, A. H. M. T.; Niemegeers, C. J. E. J.
Med. Chem. 1965, 8, 220.
14. Puratchikody, A.; Gopalakrishnan, S.; Nallu, M. Indian J.
Heterocycl. Chem. 2004, 14, 149.
15. Puratchikody, A.; Gopalakrishnan, S.; Nallu, M. Indian J.
Pharm. Sci. 2005, 67, 47.
16. Khalafi-Nezhad, A.; Soltani Rad, M. N.; Mohabatkar,
H.; Asrari, Z.; Hemmateeneja, B. Bioorg. Med. Chem.
2005, 13, 1931.
17. Mor, M.; Bordi, F.; Silva, C.; Rivara, S.; Zuliani, V.;
Vacondio, F.; Rivara, M.; Barocelli, E.; Bertoni, S.;
Ballabeni, V.; Magnanini, F.; Impicciatore, M.; Plazzi,
P. V. Bioorg. Med. Chem. 2004, 12, 663.
18. Yalcin, I.; Aki-Sener, E. Int. J. Pharm. 1993, 98, l.
19. Sener, E.; Yalcin, I.; Sungur, E. Quant. Struct-Act. Relat.
1991, 10, 223.
Vc = Mean relative changes in the paw volume of each
of the rats after the administration of carrageenan injec-
tion in the control group.
4.6. Modelling methodology
ANOVA, cluster analysis and principal component
analysis were performed using a standard statistical soft-
ware package, KyplotÒ. The structure of the various
molecules as given in Table 1 was drawn and the energy
was minimized by using Universal force field available in
Cerius2 softwareÒ (Acceryls Inc, USA). Two hundred
and twenty-nine descriptors that include topological,
charge, geometrical, aromaticity indices, constitutive
properties, quantum mechanics and thermodynamics
were evaluated for all the molecules. Several literature
reports give a very detailed description of these descrip-
tors.31–33 A genetic function algorithm (GFA) technique
was used to select the descriptors from this large pool
for using in the QSAR model.
The GFA uses a genetic algorithm to perform a search
over the space of possible QSAR models using the lack
of fit (LOF) score to estimate the fitness of each model.
Such evolution of a population of randomly constructed
models leads to the discovery of highly predictive
QSAR. The populations of the models are created by
20. Turker, L.; Sener, E.; Yalcin, I.; Akbulut, U.; Kayalidere,
I. Sci. Pharm. 1990, 58, 107.
21. Yalcin, I.; Sener, E. Int. J. Pharm. 1993, 98, 1.
22. Winter, C. A.; Risley, E. A.; Nuss, G. W. Proc. Soc. Exp.
Biol. Med. 1962, 111, 544.

1090
A. Puratchikody, M. Doble / Bioorg. Med. Chem. 15 (2007) 1083–1090
23. Vinegar, R.; Schreiber, W.; Hugo, R. J. Pharmacol. Exp.
Ther. 1969, 166, 96.
28. Hosseinzadeh, H.; Ramezani, M.; Salmani, G.-A. J.
Ethnopharmacol. 2000, 73, 379.
24. Takahashi, J.; Kirino, O.; Takayama, C.; Nakamura, S.;
Noguchi, H.; kato, T.; Kamoshita, K. Pestic. Biochem.
Phys. 1988, 30, 262.
25. Ertepinar, H.; Gok, Y.; Geban, O.; Ozden, S. Eur. J. Med.
Chem. 1995, 30, 171.
26. Miller, L. C.; Tainter, M. L. Pro. Soc. Exp. Biol. Med.
1994, 57, 261.
27. Armitage, P. Statistical Methods in Medical Research, 1st
ed.; Blackwell Scientific Publications: London, 1971, pp
217.
29. D’Amour, F. G.; Smith, D. L. A. J. Pharmacol. Exp. Ther.
1941, 72, 74.
30. Vetrichelvan, T.; Jegadeesan, M. Indian J. Pharmacol.
2003, 35, 388.
31. Todeschini, R.; Lasagni, M.; Marengo, E. J. Chemomet.
1994, 8, 263.
32. Todeschini, R.; Consonni, V. Handbook of Molecular
Descriptors; Wiley-VCH: Weinheim, 2000.
33. Karelson, M. Molecular Descriptors in QSAR/QSPR;
Wiley Interscience: NewYork, 2000.