Synthesis, anticonvulsant activity and QSAR studies of some new pyrazolyl pyridines

Article abstract of DOI:10.1007/s00044-016-1597-8

Twenty-one new 3,5-bipyridinyl-1H-pyrazole derivatives (pyrazolyl pyridines) have been synthesized and evaluated for their anticonvulsant activity in animal models of epilepsy. The pyrazolyl pyridines, 7–27, were obtained through a general one-pot synthes

Full text of DOI:10.1007/s00044-016-1597-8

MEDICINAL
CHEMISTRY
RESEARCH
Med Chem Res
DOI 10.1007/s00044-016-1597-8
ORIGINAL RESEARCH
Synthesis, anticonvulsant activity and QSAR studies of some new
pyrazolyl pyridines
1
,2
3
Joohee Pradhan
•
Anju Goyal
Received: 22 December 2015 / Accepted: 10 May 2016
Ó Springer Science+Business Media New York 2016
Abstract Twenty-one new 3,5-bipyridinyl-1H-pyrazole
derivatives (pyrazolyl pyridines) have been synthesized
and evaluated for their anticonvulsant activity in animal
models of epilepsy. The pyrazolyl pyridines, 7–27, were
obtained through a general one-pot synthesis, from ketones
and acid chlorides via formation of 1,3-diketones in situ
carried out in hydrocarbon solvent and LiHMDS base. The
profile of anticonvulsant activity of final compounds was
established in the maximal electroshock (MES) and sub-
cutaneous pentylenetetrazole (sc PTZ) tests, after
intraperitoneal injection in rats and mice, respectively, at
doses of 30, 100, and 300 mg/kg. An observation was
carried out at two different time intervals—0.5 and 4 h.
Phenytoin was used as a standard antiepileptic against
MES convulsions and valproic acid against sc PTZ con-
vulsions. Furthermore, in addition to the primary anticon-
vulsant screening, the acute neurological toxicity was
determined in mice by the rotarod test and in rats by
positional sense test. The compounds showed anticonvul-
sant activity exclusively against MES convulsions. The
compounds were found especially active in 100 mg/kg
dose at both the time points, i.e., 0.5 and 4 h, depending
upon the lipophilicity of molecules as indicated by statis-
tically significant reduction in the time spent in tonic
extension phase (p \ 0.001). Further, the newly
synthesized compounds were subjected to two-dimensional
quantitative structure–activity relationship (2D QSAR)
analysis through multiple linear regression, principal com-
ponent regression, and partial least square regression anal-
ysis, and three-dimensional quantitative structure–activity
relationship (3D QSAR) analysis by k-nearest neighbor
molecular field analysis in conjunction with stepwise for-
ward–backward, genetic algorithm, and simulated anneal-
ing variable selection methods using the software VLife
MDS. The structure–activity relationship (SAR) as well as
quantitative structure–activity relationship (QSAR) studies
for anticonvulsant activity confirmed the crucial role of 3,5-
bipyridinyl-1H-pyrazole core fragment for anticonvulsant
activity.
Keywords Pyrazolyl pyridines Á Anticonvulsant Á
SAR Á 2D QSAR Á 3D QSAR Á KNN-MFA
Introduction
Epilepsy is the commonest neurological condition affecting
people of all ages, race, and social class. There are 50
million people living with epilepsy worldwide, and most of
them reside in developing countries. About 10 million
persons with epilepsy are there in India. Many people with
active epilepsy do not receive appropriate treatment for
their condition, leading to large treatment gap (Subbareddy
et al., 2014). Over the years, the field of epilepsy has
received a great deal of attention from research investiga-
tors in the hope of discovering new drugs that are more
effective and have minimal adverse effects. Though several
new anticonvulsants have been introduced, some types of
epilepsies are still not adequately controlled with the cur-
rent therapy. Adverse reactions and lack of efficacy for
&
Joohee Pradhan
juhipradhan123@gmail.com
1
2
3
Department of Pharmaceutical Sciences, Mohanlal Sukhadia
University, Udaipur, Rajasthan 313001, India
Pacific Academy of Higher Education and Research
University, Pacific Hills, Udaipur, Rajasthan 313001, India
B. N. Institute of Pharmaceutical Sciences, Sewashram Road,
Udaipur, Rajasthan 313001, India
123

Med Chem Res
certain types of epilepsies are some of the limitations of
existing medications (L o¨ scher and Schmidt 2002). Taking
into consideration the above limitations, many authors
conducted attempts to identify the structural features cru-
cial for anticonvulsant activity. On the basis of these
researches, several pharmacophoric models, enabling a
more rational design of new anticonvulsants, have been
described. Thus, one of the important core fragments of
anticonvulsants is defined by nitrogen heterocyclic system,
usually imide or lactam and phenyl or alkyl groups
attached to the heterocyclic system (Wong et al., 1986;
Bruno-Blanch et al., 2003; Malawska, 2005). This common
template is present in the structures of old, however well-
established antiepileptic drugs (AEDs), such as ethosux-
imide and phenytoin as well as among the newest drugs,
e.g., levetiracetam, brivaracetam, or seletracetam (Ro-
gawski and Porter, 1990; Bialer et al., 2007). Bearing in
mind the aforementioned, we have synthesized in our
laboratory a series of 3,5-bipyridinyl-1H-pyrazole deriva-
tives differently substituted at the 2-position of both the
pyridine rings, as candidates for new AEDs.
Quantitative structure–activity relationships (QSAR), as
a major factor in drug design, are mathematical equations
relating chemical structure to their biological activity
(Krogsgaard-Larsen et al., 2002). Anticonvulsant agents
have been the aim of many SAR and QSAR studies (Pal-
luotto et al., 1996; Marone et al., 1999; Hou and Xu, 2001;
Marder et al., 2001; Verli et al., 2002). Palludotto et al.
(1996) synthesized a series of 2-aryl-2,5-dihydropyri-
dazino[4,3-b]indol-3-one derivatives and tested them as
central benzodiazepine receptor ligands. These workers
used 2D and 3D QSAR on these molecules and observed
that the molar refractivity (MR) of the substituents was the
major factor controlling the binding of the ligands to their
receptors. A correlation between the theoretical descriptors
of the tricyclic neuroactive drugs and their biological mode
of action has been obtained by the theoretical studies of
Marone and coworkers (Marone et al., 1999). Three-di-
mensional QSAR analyses on the anticonvulsant activity of
a series of cinnamamides using comparative molecular field
analysis (CoMFA) and comparative molecular similarity
indices analysis (CoMSIA) approaches have been reported
by Hou and Xu (2001). These investigators found that the
interaction of these compounds with receptors is achieved
by electrostatic and hydrophobic forces. Marder et al.
(2001) have reported molecular modeling and QSAR anal-
ysis of flavone derivatives upon interaction with benzodi-
azepine binding site. The electronic properties of the ligands
were found to be the major factor affecting the ligand–re-
ceptor binding (Verli et al., 2002). VLife Molecular Design
Suite (VLife MDS; Supplied by VLife Science technolo-
gies, Pune, India.) is a medicinal chemist’s work bench for
Computer-Aided Drug Design and molecular discovery.
VLife MDS is completely scalable and customizable to
address the variegated problems in molecular design and
prediction of new molecules. VLife MDS facilitates com-
prehensive in silico approach to design, visualize, predict,
and analyze the small molecules as well as proteins and
study their interactions. Among its various applications,
molecular modeling, energy minimization, geometry opti-
mization, and 2D QSAR and 3D QSAR by k-nearest
neighbor molecular field analysis (KNN-MFA) studies were
used for the present study. KNN-MFA (Ajmani et al., 2006)
requires suitable alignment of given set of molecules. This is
Marketed AEDs predominantly target voltage-gated
?
cation channels (the a-subunits of voltage-gated Na
channels and also T-type voltage-gated Ca channels) or
2
?
influence GABA-mediated inhibition. Recently, a2 - d
2
?
voltage-gated Ca
channel subunits and the SV2A
synaptic vesicle protein have been recognized as likely
targets (Meldrum and Rogawski, 2007). Voltage-gated
?
Na channels are essential for action potentials, and their
mutations are the substrate for generalized epilepsy with
febrile seizures plus and benign familial neonatal infantile
?
seizures; Na channel inhibition is the primary mechanism
of carbamazepine, phenytoin, and lamotrigine and is a
probable mechanism for many other classic and novel
AEDs (Armijo et al., 2005). Most of the older anticon-
vulsants have close structural similarity. This is depicted in
Fig. 1. The structural similarity of newly synthesized
pyrazolyl pyridines to established anticonvulsant drugs,
especially Hydantoin derivatives (e.g., phenytoin) in hav-
ing five-membered heterocyclic systems with two nitrogen
?
atoms, gives rise to the possibility of Na channel inhibi-
tion as the most probable mechanism of action of these
compounds.
R1 R2
O
R1
O
R1
N
O
O
R2
R
2
N
R
2
HN
HN
HN
NH
NH
CH2
N
HN
R1
O
O
O
Barbiturates
Hydantoin
Succinimide
Synthesized Pyrazolyl pyridines
Fig. 1 Structural similarity of established anticonvulsant drugs with that of synthesized compounds
1
23

Med Chem Res
followed by generation of a common rectangular grid
around the molecules. The steric and electrostatic interac-
tion energies are computed at the lattice points of the grid
using a methyl probe of charge ?1. These interaction energy
values are considered for relationship generation and uti-
lized as descriptors to decide nearness between molecules.
The term descriptor is utilized in the following discussion to
indicate field values at the lattice points. The optimal
training and test sets were generated using the sphere
exclusion algorithm. This algorithm allows the construction
of training sets covering descriptor space occupied by rep-
resentative points. Once the training and test sets were
generated, KNN methodology was applied to the descriptors
generated over the grid.
Results and discussion
Chemistry
The title compounds pyrazolyl pyridines 7–27 (Table 1)
were synthesized from methyl pyridyl ketones (1a–6a) and
pyridyl acid chlorides (1b–6b) according to the
scheme (Fig. 2), where, as shown in Fig. 3, 1,3-diketones
were synthesized directly from ketones and acid chlorides
and were then converted in situ into pyrazoles by the
addition of hydrazine hydrate. This method was proved to
be extremely fast, general, and chemoselective. The reac-
tion was carried out under the atmosphere of nitrogen, in a
hydrocarbon solvent, so as to suppress the side reactions by
disfavoring a charged intermediate, only allowing the
enolate to react with the very electrophilic acid chloride.
LiHMDS was found to be the base of choice because as
reported previously (Heller and Natarajan, 2006), it does
In the present study, we report the synthesis, anticon-
vulsant evaluation, and QSAR studies of new pyrazolyl
pyridines to be considered as a template for the develop-
ment of new, potential anticonvulsant agents in future.
Table 1 List of synthesized compounds 7–27
N
N
R
2
N
HN
R1
Sl.
no.
Compound
ID
R
1
R
2
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
.
7.
–H
–H
.
8.
–NO
–NH
–Cl
2
2
–H
.
9.
–H
.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
–H
.
–C
6
H
H
5
4
–H
.
–C
6
CH
3
–H
.
–NO
–NH
–Cl
2
–NO
–NH
–Cl
2
.
2
2
.
0.
1.
2.
3.
4.
5.
6.
7.
8.
9.
0.
1.
–C
6
H
H
5
4
–C
6
H
H
5
4
–C
6
CH
3
–C
6
CH
3
–Cl
–NO
2
–Cl
–NH
2
–Cl
–C
6
H
H
5
4
–Cl
–C
6
CH
3
3
–NO
–NO
–NO
–NH
–NH
2
2
2
2
2
–NH
2
–C
6
–C
6
–C
6
–C
6
–C
6
H
H
H
H
H
5
4
5
4
4
CH
CH
CH
3
3
6 5
–C H
123

Med Chem Res
R2
N
Cl
O
O
R1
o
LiHMDS/ Toluene/ THF; 0 C
N
+
N H .H O
2
4
2
N
N
H
N
R2
N
R1
EtOH/THF/Toluene/AcOH; Reflux
1
a-6a
1b-6b
Compounds 7-27
Fig. 2 Scheme for the synthesis of title compounds 7–27
N
O
C
O
C
CH2
Cl
O
R1
R2
N H .H O N
R2
N
C
2
4
2
+
N
C
OLi
N
N
HN N
-27
R1
R1
R2
7
1
a -6a
1b-6b
a diketone intermediate
Fig. 3 Synthesis of title compounds through 1,3-diketone intermediate
-
1
not give an additional triketone intermediate as with the
use of other bases like NaHMDS or KHMDS, arising from
a metal diketone reacting with another equivalent of acid
chloride.
two bands at about 1000 and 1400 cm , respectively, due
to symmetrical N=O stretch which confirms the presence of
nitro group in these compounds. Compounds 9, 14, 19, 22,
25, and 26 showed two IR stretching vibrational bands at
-
1
Different ketones were prepared from 4-acetyl pyridine
or methyl pyridyl ketones by substitution at position 2-via
N-oxide intermediate as reported by us previously (Prad-
han and Goyal, 2015). Likewise, different acid chlorides
used were isonicotinoyl chloride and derivatives of isoni-
cotinoyl chloride prepared from ethyl isonicotinate via
N-oxide intermediate. Ethyl group acted as protecting
group which later on converted via alkaline hydrolysis to
parent acid after substitution at position 2. The acid so
obtained was then converted to acyl chloride by thionyl
chloride. The N-oxides in both the above cases were
reduced back to the base by palladium and carbon in ethyl
about 3405 and 3382 cm attributed to N–H str. in primary
aromatic amines. Further, a sharp singlet at d 1.43 ppm in
H NMR spectrum of these compounds showed the pres-
1
ence of two protons of NH group. Compound nos. 15, 18,
2
19, 20, and 21 containing chloro substitutions showed an IR
-
1
band at about 745–677 cm attributed to C–Cl str. in
aromatic halogen compounds. A proton NMR signal of the
same compounds between d 3.3–4.8 ppm indicates the
proton attached to the carbon-containing halogen atom.
Compound nos. 11, 16, 20, 23, 25, and 27 contained more
intense C–H stretching vibrational band at about
-
1
3040 cm , indicating the quantitative relationship between
number of aromatic C–H bonds and intensity of IR bands.
Compounds 12, 17, 21, 24, 26, and 27 containing tolyl
substituents on pyridine ring showed additional aliphatic
formate and methanol. The final compounds were fully
1
characterized by elemental analyses (C, H, N) and
H
1
3
NMR, C NMR, LC/MS spectra. The detailed physico-
chemical and analytical data are listed in the experimental
section. The structures of intermediates 1a–6a and 1b–6b
are given in Table 2.
3
-1
Sp C–H stretching vibration at 2872–3178 cm . Further,
1
H NMR spectrum of these compounds showed a signal at d
0.5–1.5 ppm which confirms the presence of simple Sp C–
3
1
3
IR spectrum of the compounds in general showed
heteroaromatic C–H stretching vibrations between 3000
H in the compounds. The C NMR spectrum of all the
compounds in general showed three peaks at about d
147.18, 146.18, and 103.52 ppm which correspond to car-
bon atoms of pyrazole ring. The carbon atom of the methyl
group at the tolyl fragment occurred as peak at d 21.13 ppm.
The peaks of carbon atoms of the aromatic rings were
detected at the expected chemical shifts within the range d
111–160 ppm. The molecular formulae of the synthesized
compounds 7–27 were confirmed by elemental analysis.
Elemental analysis was carried out for carbon, hydrogen,
and nitrogen. The results obtained were within ± 0.4 % of
-
and 3100 cm , =C–N str. for pyridines at about
1
-
1
1
3
250 cm ; 2° N–H str. in pyrazole between 3100 and
500 cm , C=N str. at about 1650 cm , aromatic C=C
-1 -1
-
1
str. between 1400 and 1700 cm and N–H bend at about
-1 1
200 cm . H NMR spectrum of the compounds in general
1
showed signal for aromatic protons between d 6.3 and
8
.5 ppm as multiplet. A sharp singlet formed between d
9
.85 and 10.62 ppm was attributed to the N–H proton. IR
spectrum of compounds 8, 13, 18, 22, 23, and 24 showed
1
23

Med Chem Res
Table 2 Intermediates for the synthesis of compounds 7–27
Intermediate code
Structure
Intermediate code
1b
Structure
Cl .HCl
1
a
O
O
N
N
1
1
-(pyridin-4-yl)ethanone
isonicotinoyl chloride
hydrochloride
2
a
O
2b
3b
O
Cl .HCl
N
NO2
N
NO
2
-(2-nitropyridin-4-yl)ethanone
2-nitropyridine-4-
carbonyl
hydrochloride
chloride
3
a
O
O
Cl
.HCl
N
NH2
N
NH
2
1
1
-(2-aminopyridin-4-yl)ethanone
2-aminopyridine-4-
carbonyl
chloride
hydrochloride
4
a
O
4b
5b
O
Cl .HCl
Cl
N
Cl
N
-(2-chloropyridin-4-yl)ethanone
2-chloropyridine-4-
carbonyl
hydrochloride
chloride
5
a
O
O
Cl .HCl
N
N
1
-(2-phenylpyridin-4-yl)ethanone
2-phenylpyridine-4-
carbonyl
chloride
hydrochloride
6
a
O
6b
O
Cl .HCl
N
N
1
-(2-p-tolylpyridin-4-yl)ethanone
2-p-tolylpyridine-4-
carbonyl
chloride
hydrochloride
the theoretical value. The molecular weight of the com-
pounds was confirmed by mass spectroscopy. The molec-
ular ion peaks obtained were in good agreement with the
molecular weight of the compounds. All the synthesized
compounds 7–27 exhibited satisfactory spectral data con-
sistent with their structure.
123

Med Chem Res
Anticonvulsant activity
compounds 13, 16, 17, 21, 22, and 27 showed no anti-
convulsant activity in all the tested doses at any time
interval. This indicated that bulky substitution (–C H CH )
The profile of anticonvulsant activity of final compounds
6
4
3
7
–27 was established in the MES and scPTZ tests, after
at R is detrimental for the said activity, but when com-
2
intraperitoneal (i.p.) injection in rats and mice, respec-
tively, at doses of 30, 100, and 300 mg/kg. An observation
was carried out at two different time intervals—0.5 and
bined with –NO or –NH , as in compounds 21 and 26,
2
2
respectively, increased potency as well as toxicity of the
compounds. The major group responsible for respiratory
depression in rats seemed to be –C H CH , coupled with
4
h. Furthermore, in addition to the primary anticonvulsant
6
4
3
screening, the acute neurological toxicity (NT) was deter-
mined in mice by the rotarod test and in rats by positional
sense test. The results of the preliminary pharmacological
studies are summarized in Table 3.
–H or NO as observed during pharmacological screening
2
of compounds 12 and 24. Compound 20 with –Cl and
–C H substitution at R and R , respectively, caused tre-
6
5
1
2
mors in rats at 300 mg/kg dose. Good anticonvulsant
activity of compound 7 at 100 mg/kg and no activity of
compound 27 at all three doses indicated that the nucleus
3,5-bipyridinyl-1H-pyrazole itself is crucial for anticon-
vulsant activity and the bulky substitution on the nucleus is
detrimental for the activity.
The compounds tested showed protection exclusively in
the MES seizures. Furthermore, in vivo data revealed that
anticonvulsant activity was closely connected with the type
of substituent at the pyridine moiety in the present pyra-
zolyl pyridines. Among all the synthesized compounds,
compound nos. 7, 8, 9, 10, 14, 18, and 19 having substi-
tution –H, –NO , –NH , or –Cl at R position were found to
The differences in times points in which the anticon-
vulsant protection was observed for compounds containing
pyrazolyl pyridine nucleus may result from the lipophilic
properties of these molecules. As it is presented in Table 3,
the more lipophilic molecules (higher clog P values)—11,
12, 15, 20, 23, 24, 25, and 26—showed activity mainly at
4 h, whereas less lipophilic compounds—7, 8, 9, 10, 14,
18, and 19—were effective only at time point of 0.5 h.
These observations may be connected with higher affinity
of more lipophilic compounds to the peripheral tissues that
causes slower distribution to the CNS. Further, inactivity of
compounds 16, 17, 20, and 27 against MES seizures sup-
ported the Lipinski’s rule of 5 which says that a drug
having log P value greater than 5 would not have good
absorption properties, so will not be a good choice as a
drug. Most CNS active drugs have log p less than 5 so it is
easily absorbed in blood–brain barrier where active diffu-
sion of drug is likely to have most lipophilic area like lipid
bilayer of membrane, and if the value is less than 0, it will
move toward hydrophilic compartment like blood serum
(Lien et al., 1973).
2
2
1
be the most active compounds (i.e., compounds effective in
a dose of 100 mg/kg at 0.5 h) which indicated that as
compared with other compounds in the series, a less bulky
substituent with variable electronic character favors the
anticonvulsant activity of these compounds against MES
convulsions. The aforementioned active compounds have
R substitution as either –H, –NH , or –NO . Here, it is
2
2
2
noteworthy that no compound with –Cl substitution at –R2
was active at 0.5 h in 100 mg/kg dose, but the criterion of
variable electronic character at R remains the same as with
2
R . Thus, by observing the SAR of active compounds, it is
1
imperative to say that bulk, but not the electronic character
of the substituent is important at both R and R position of
1
2
the compounds with the remark that –Cl substitution at R2
is not important for anticonvulsant activity. These obser-
vations indicate that receptor site may allow only less
bulky, smaller compound to interact with it, with little
electronic interactions; however, along with the bulk,
polarity as well as the electronegativity of the substituents
turned out to be important and decided the activity of the
compounds depending upon the position on the molecule
where they are attached. This has been proved through
QSAR study in the coming sections. Compounds 11, 12,
The overall SAR analysis indicated that pyrazolyl
pyridines with less bulky substituent of variable electronic
character at R were therapeutically active as well as less
1
toxic while bulkier substitution at R as in compounds 20,
2
1
5, 20, 23, 24, 25, and 26 showed protection for MES
24, and 26 caused neurotoxicity. Good anticonvulsant
activity of compound 7 indicated the crucial role of the
nucleus 3,5-bipydidinyl-1H-pyrazole for antiepileptic
seizures only 4 h after i.p. administration that means
delayed onset, however long duration of anticonvulsant
action of these compounds. As compared with the most
active compounds, these compounds had bulkier substitu-
tion either at R (comp. nos. 11 and 12) or at R (comp.
activity. -Chloro substituent, when attached to R , was
1
favorable for anticonvulsant activity as exemplified by
compounds 10, 15, 18, 19, and 20, but when attached to R2
(comp. 15), it caused delayed onset of action. The nucleus
itself was found to be a good anticonvulsant. Accordingly,
seven best compounds with anticonvulsant activity were 7,
8, 9, 10, 14, 18, and 19; all these compounds showed no
1
2
nos. 20, 23, 24, 25, and 26) position; an exception to this
was compound 15 having R = R = Cl which indicated
1
2
that although less bulky, -chloro substitution at both the
positions was responsible for delayed action. Additionally,
1
23

Med Chem Res
Table 3 Anticonvulsant activity of compounds 7–27 after i.p. administration in rats/mice
N
N
R
2
N
HN
R1
a
e
Treatment (comp. no.)
R
1
R2
Intraperitoneal administration in rats/mice
CLogP
1.7037
b
c
d
MES
.5 h
scPTZ
0.5 h
NT
0
4 h
4 h
0.5 h
h
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
.
–H
–H
100
100
100
100
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
300
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
300
–
–
–
–
–
–
–
–
–
–
.
–NO
–NH
–Cl
2
–H
–
1.62028
1.38979
2.4924
3.8017
4.3007
1.51137
1.06672
3.25683
5.8997
6.8977
2.38471
2.17849
4.5904
5.0662
1.30637
3.71828
4.19408
3.48779
3.98679
6.3987
2.605
.
2
–H
–
0.
1.
2.
3.
4.
5.
6.
7.
8.
9.
0.
1.
2.
3.
4.
5.
6.
7.
–H
–
–C
–C
6
H
H
5
4
–H
100
100
–
z
6
CH
3
–H
–
300
–NO
–NH
–Cl
2
–NO
–NH
–Cl
2
2
–
–
–
–
–
–
–
–
–
2
100
–
–
–
100
–
–
–C
6
H
H
5
4
–C
–C
6
H
5
4
–
–
–C
6
CH
3
6
H
CH
3
–
–
–
–Cl
–NO
–NH
2
100
100
–
–
–
–Cl
2
–
–
y
–Cl
–C
–C
6
H
5
4
100
–
300
–
–Cl
6
H
CH
3
3
–
–
–NO
–NO
–NO
–NH
–NH
2
2
2
2
2
–NH
2
–
–
–
–
–C
–C
–C
–C
–C
6
6
6
6
6
H
5
H
4
H
5
H
4
H
4
–
100
100
100
100
–
–
z
CH
300
–
300
–
–
–
CH
CH
3
3
300
–
300
–
–
–C
6
H
5
–
f
Phenytoin
Valproic acid
30
–
30
100
–
100
–
f
–
2.75
z
Response comments: respiratory depression, tremors
y
a
Doses of 30, 100, or 300 mg/kg were administered intraperitoneally in rats for MES test and in mice for scPTZ test. The data indicate the
minimum dose effective or neurotoxic in half or more animals tested. A dash indicates the absence of anticonvulsant activity or neurotoxicity at
the maximum dose administered
b
Maximal electroshock test
c
Subcutaneous pentylenetetrazole test
d
Neurotoxicity screening using rotarod test
e
clog P values calculated using a log P module of ChemDraw Ultra program, version 7.0.1 (Cambridge Soft Corporation, Cambridge, MA,
USA)
f
Phenytoin and valproic acid, reference antiepileptic drugs tested by use of ADD Program procedures in NIH/NINDS
neurotoxicity in the highest dose tested and can be sub-
jected to sodium channel binding assay for the determi-
nation of their mechanism of action. Thus, these
compounds may be useful for future development of anti-
convulsant drugs for grand mal epilepsy.
QSAR study
An intensive 2D QSAR study of 21 compounds (divided
into 7 test and 14 training sets) for anticonvulsant activity
against MES convulsions was carried out through multiple
123

Med Chem Res
linear regression (MLR), principal component regression
PCR), and partial least square (PLS) regression analysis
compounds gave similar results, most probably due to the
reason that present position of substituents was not
(
using VLife MDS. The most statistically significant model
obtained through PCR is presented below:
important. Requirement of lesser distance between R and
1
R2 further proves that smaller molecules show better
activity as observed with the results of anticonvulsant
screening.
À logð% inhibitionÞ ¼ þ0:1148ðÆ0:0271ÞT T N 3
À 0:0648ðÆ0:0204ÞSK MostHydrophobic
Hydrophilic Distance À 2:4069
Along with 2D QSAR analysis, the compounds were
also subjected to 3D QSAR analysis by k-nearest neighbor
molecular field analysis (KNN-MFA) in conjunction with
stepwise (SW) forward–backward, genetic algorithm (GA),
and simulated annealing (SA) variable selection methods
using the software VLife MDS. The best model selected
was having nearly the same observed and predicted activ-
ities of the compounds (Table 5, Fig. 7).
2
r _se = 0.2042
n = 14
2
q _se = 0.2579
2
pred_r = 0.5452
Degree_of_freedom = 11
2
r = 0.6384
2
q = 0.5227
2
pred_r se = 0.2987
F_test = 9.7083
Selected descriptors: S_442, S_164,E_272, E_442
Descriptor range:
S_442: -0.1376 to -0.1324,
S_164: 30.0000 to 30.0000,
2
The equation explains 63 % (r = 0.6384) of the total
E_272: -0.3004 to 0.8490,
2
variance in the training set. It also has an internal (q ) and
E_442: 5.2188 to 10.0000
2
external (pred_r ) predictive ability of *52 and *54 %,
k-nearest neighbor:
n:
2
respectively. The F-test = 9.70 shows the statistical sig-
nificance of 99.99 % of the model which means that
probability of failure of the model is 1 in 10,000. In
addition, the randomization test shows confidence of
14
Degree_of_freedom:
9
2
q :
2
0.7873
0.1440
0.3835
0.2898
q _se:
2
pred_r :
*
99.9 % that the generated model is not random, and
hence, it is chosen as the QSAR model. The plot of
observed versus predicted activity (Table 4 and Fig. 4)
provides an idea about how well the model was trained and
how well it predicts the activity of the external test set.
From the plot, it can be seen model is able to predict the
activity of training set quite well (all points are close to
regression line) as well as external test setup to *60 %
2
pred_r se:
The above KNN-MFA model obtained by simulated
annealing variable selection method shows that electro-
static interactions (E_272 and E_442 are electrostatic field
descriptors, former having a negative to positive range
while later having a positive range) as well as steric
interactions (S_442 and S_164 are steric field descriptors,
former having a negative range while later having a posi-
tive range) play major role in determining biological
activity. Negative range of electrostatic field descriptor
indicates that negative electrostatic potential is favorable
for increase in the activity, and hence, a more electroneg-
ative substituent group is preferred in that region, while
positive range indicates that positive electrostatic potential
is favorable for increase in the activity and hence a less
electronegative substituent group is preferred in that
region. A negative range of steric field descriptor indicates
that negative steric potential is favorable for increase in the
activity and hence less bulky substituent group is preferred
in that region while positive range indicates that positive
steric potential is favorable for increase in the activity and
hence more bulky substituent group is preferred in that
region.
(
only 1 point is relatively apart from the regression line)
providing confidence in predictive ability of the model.
The developed MLR model reveals that the descriptor
T_T_N_3 (i.e., any atom separated from nitrogen atom by
three bonds in a molecule) plays most important role
(
*60 %) in determining anticonvulsant activity (Fig. 5),
which mainly indicates the relationship with reference to
the introduction of a substituent at 3-position on the
pyrazole nucleus, or at 8- or 17-position on the pyridine
ring. The second descriptor SK MostHydrophobicHy-
drophilicDistance contributed negatively (*40 %) that
indicates that greater the distance between hydrophilic
(
e.g., –NH ) and hydrophobic (e.g., –C H ) groups in the
2 6 5
molecule, the lesser will be its activity. Taking into con-
sideration, both these descriptors, a pyrazolyl pyridine
having, for example, 3-NH and 8- or 17-C H , may be
2
6 5
more promising anticonvulsant (Fig. 6). Also, importance
of substituents at positions other than those present in the
designed compounds verified the results of pharmacologi-
cal screening that variety of substituent in the active
Statistically SA-KNN-MFA model was comparatively
better than the other two with respect to the internal
2
(q = 0.78) as well as the external (pred_r = 0.38) model
2
1
23

Med Chem Res
Table 4 Observed and predicted values of the compounds 7–27 for the best model obtained through 2D QSAR
Compound code
-log (% inhibition)
Observed value
Predicted value
Residual value
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
.
-1.965
-1.965
-1.98
-1.92355
-1.57528
-1.84207
-2.17589
-2.02769
-2.03513
-1.31367
-1.76902
-2.37278
-1.79467
-1.45282
-1.80739
-2.08528
-2.20621
-2.20291
-1.52125
-1.71268
-1.71287
-1.95584
-1.96255
-1.65765
-0.04145
-0.38972
-0.13793
0.201894
0.057691
0.107129
0.100665
-0.16798
0.425783
0.489673
0.038818
-0.16661
0.148277
0.223211
0.773908
0.423253
-0.25233
-0.25213
-0.03616
0.034553
0.419654
.
.
0.
1.
2.
3.
4.
5.
6.
7.
8.
9.
0.
1.
2.
3.
4.
5.
6.
7.
-1.974
-1.97
-1.928
-1.213
-1.937
-1.947
-1.305
-1.414
-1.974
-1.937
-1.983
-1.429
-1.098
-1.965
-1.965
-1.992
-1.928
-1.238
Fig. 4 Observed versus predicted anticonvulsant activity of test and
training sets obtained through 2D QSAR
Fig. 5 Contribution plot of statistically significant model obtained
through 2D QSAR
validation and correctly predicts activity *78 and *38 %
for the training and test set, respectively. It uses 2 elec-
trostatic and 2 steric field descriptors along with its
molecule. Plot of the KNN-MFA which shows the relative
position and ranges of the corresponding important elec-
trostatic/steric fields in the model provides the following
guidelines for design of new molecule. The 3D plot
2
k nearest neighbor (k = 2) to evaluate the activity of new
123

Med Chem Res
(
Fig. 8) shows the electrostatic and steric fields important
and 17) did not favor the activity; as a more bulky sub-
stitution may block the entrance to the binding pocket of
the receptor and if it is less bulky, it would permit the
entrance. However, the results suggested, as with 2D
QSAR analysis, a more bulky substitution at S_164, i.e., at
8- or 17-position of the parent molecule (Fig. 4) may be
allowed for the better activity. Thus, modification of these
fields according to the ranges obtained may result into
more potent compounds.
for interaction of molecules with space, i.e., with receptor,
thereby affecting the pharmacologic response. As was
observed with the pharmacological activity of present set
of compounds, the ones having substituents with similar
electronic character on both the pyridine rings were mostly
fond to be inactive (e.g., 13, 16, 17, and 27) while com-
pounds having substituents with different electronic char-
acter on either of the pyridine ring were active (e.g., 8, 9,
10, 18, and 19). A more bulky substitution at S_442 that
most of the inactive compounds had (e.g., comp. 16, 17, 21,
Experimental
Chemistry
Substituents
1
6
1
5
1
7
N
9
8
7
3
All the chemicals and solvents were purchased from
Sigma-Aldrich and were used without further purification.
Melting points were taken in open capillary tubes and are
uncorrected. The progress of the reaction as well as purity
of the compounds was confirmed by TLC using Merck
silica gel 60 F254-coated aluminum plates using chloro-
form–methanol (9:1) as solvent and detection by UV light.
IR spectra were recorded on a Bruker-alpha FTIR spec-
2
14 R2 19
1
3
1
0
N
6
4
N 1
1
1
N
1
2
H
R
1
8
5
1
The greater the distance,
lesser will be the activity
e.g. 3- NH and 8-or 17-C H may have good Anticonvulsant activity.
2
6 5
-
1
1
13
Fig. 6 Suggestions according to the result of 2D QSAR analysis for
design of improved anticonvulsant drugs
trometer (m max in cm ). H-NMR and C-NMR spectra
were recorded on Bruker NMR spectrometers (400 MHz
Table 5 Observed and predicted values of the compounds according to the best model obtained through 3D QSAR
Compound code
-log (% inhibition)
Observed value
Predicted value
Residual value
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
.
-1.965
-1.965
-1.98
-1.97891
-1.97677
-1.9688
-1.9765
-1.77761
-1.6895
-1.38682
-1.94002
-1.937
0.01391
0.011768
-0.0112
0.002501
-0.19239
-0.2385
0.173818
0.003021
-0.01
.
.
0.
1.
2.
3.
4.
5.
6.
7.
8.
9.
0.
1.
2.
3.
4.
5.
6.
7.
-1.974
-1.97
-1.928
-1.213
-1.937
-1.947
-1.305
-1.414
-1.974
-1.937
-1.983
-1.429
-1.098
-1.965
-1.965
-1.992
-1.928
-1.238
-1.57614
-1.513
0.271137
0.099
-1.6865
-1.94274
-1.96148
-1.96037
-1.38351
-1.95859
-1.96482
-1.92806
-1.99187
-1.57028
-0.2875
0.005741
-0.02152
0.531367
0.28551
-0.00641
-0.00018
-0.06394
0.063871
0.332281
1
23

Med Chem Res
septum, and the solution was cooled to 0 °C under nitro-
gen. LiHMDS (2.1 ml, 1.0 M in THF, 2.1 m mol) was
added quickly via syringe with stirring. After 1 min.,
substituted isonicotinoyl chloride hydrochloride (interme-
diates 1b–6b; 1 m mol) was added in one portion with
stirring. The flask was then removed from the ice bath and
allowed to stand for 1 min, and then 2 ml of AcOH was
added with stirring. EtOH (10 ml) and THF (5 ml) were
added to form a homogenous mixture, and then hydrazine
hydrate (2 ml, 1.1 g, and 34.43 m mol) was added. The
mixture was allowed to auto reflux and was held at that
temperature for 5 min or until the reaction completes. The
resulting solution was added to 1.0 M NaOH solution and
extracted with EtOAc. The organic fraction was then
washed with brine, dried over Na SO , and evaporated
2
4
under reduced pressure. The resulting residue was recrys-
tallized from 2-propanol/water to afford the product
(
compounds 7–27).
Fig. 7 Observed versus predicted anticonvulsant activity of test and
training sets obtained through 3D QSAR
4-(3-(Pyridin-4-yl)-1H-pyrazol-5-yl)pyridine (7) Yellow–
brown powder; Yield: 65 %; m.p.: 236–238 °C; Rf: 0.71;
FT-IR tmax (ATR cm ): 3388 (2°N–H str. in pyrazole),
3
-
1
036 (aromatic C–H str.), 1650 (C=N str.), 1407, 1486,
1
538 (C=C str. aromatic), 1239 (C–N str.), 1128 (N–H
1
bend.), H NMR (DMSO-d , 400 MHz,): d = 10.28 (s, 1H,
6
NH), 8.61–8.64 (d, 4H, Ar–H), 7.54–7.57 (d, 4H, Ar–H),
1
3
7
.05 (s, 1H, CH). C NMR (100 MHz, common NMR
solvents) d 150.7 (2C, C–Ar), 149.6 (2C, C–Ar), 147.5
C ), 146.8 (C ), 134. 4 (C–Ar), 132.8 (C–Ar), 120.7 (2C,
(
5
3
?
C–Ar), 119.8 (2C, C–Ar), 102.1 (C ). LCMS m/z [M]
4
222.09. Anal. Calcd. for C H N (222.25): C, 70.26; H,
13 10 4
4
.54; N, 25.21. Found: C, 70.16; H, 4.51; N, 25.11.
-Nitro-4-(3-(pyridin-4-yl)-1H-pyrazol-5-yl)pyridine (8) Yel-
low powder; Yield: 63 %; m.p.: 220–224 °C; R : 0.62; FT-
2
f
-
1
IR tmax (ATR cm ): 3174 (2°N–H str. in pyrazole), 3036
(aromatic C–H str.), 1650 (C=N str.), 1407,1486,1538
(
C=C str. aromatic), 1239 (C–N str.), 1127 (N–H bend.),
1
630, 1407 (N=O str.). H NMR (DMSO-d , 400 MHz,):
1
6
d = 10.13 (s, 1H, NH), 8.82–8.83 (d, 1H, Ar–H), 8.61 (s,
Fig. 8 3D plot of statistically significant model obtained through
KNN-MFA
1
H, Ar–H), 8.53–8.55 (d, 2H, Ar–H), 8.21–8.23 (d, 1H,
1
3
Ar–H), 7.50–7.51 (d, 2H, Ar–H), 6.87 (s, 1H, CH).
C
1
13
NMR (125 MHz, common NMR solvents) d 159.1 (C–Ar),
50.1 (2C, C–Ar), 149.85 (C–Ar), 147.45 (C ), 146.18
for H-NMR, and 100 and 125 MHz for C-NMR) using
TMS as the internal standard. MS spectra were recorded on
Water’s TOF Micro mass LC–MS spectrophotometer.
Microanalysis was performed on a PerkinElmer-240 CHN
elemental analyzer.
1
5
(C ), 134.44 (C–Ar), 134.24 (C–Ar), 123.99 (C–Ar),
3
1
19.80 (2C, C–Ar), 117.69 (C–Ar), 103.52 (C ). LCMS m/
4
?
z [M] 267.23. Anal. Calcd. for C H N O (267.24): C,
1
3 9 5 2
5
2
8.43; H, 3.39; N, 26.21. Found: C, 58.13; H, 3.29; N,
6.21.
General procedure for the synthesis of title compounds (7–
7)
2
4-(3-(Pyridin-4-yl)-1H-pyrazol-5-yl)pyridin-2-amine (9) Pale
yellow powder; Yield: 68 %; m.p.: 310–312 °C; R : 0.6;
f
-
1
Appropriate acetone (intermediates 1a–6a; 2 m mol) was
dissolved in 5 ml of dry toluene in a 100-ml RBF with a
FT-IR tmax (ATR cm ): 3405, 3382 (1°NH str.), 3038
2
(aromatic C–H str.), 1588 (C=N str.), 1415, 1492, 1512
123

Med Chem Res
1
C=C str. aromatic), 1588 (C–N str.). H NMR (DMSO-d ,
(
2H, Ar–H), 7.14–7.23 (m, 3H, Ar–H), 2.20 (s, 3H, CH ).
3
6
1
3
4
00 MHz,): d = 10.31 (s, 1H, NH), 8.64–8.66 (d, 2H, Ar–
H), 8.19–8.21 (d, 1H, Ar–H), 7.57–7.58 (d, 2H, Ar–H),
.08 (s, 1H, CH), 6.94–6.95 (d, 2H, Ar–H), 1.49 (s, 2H,
C NMR (125 MHz, common NMR solvents) d 159.54
(C–Ar), 150.58 (2C, C–Ar), 148.06 (C–Ar), 147.45 (C5),
146.18 (C3), 141.24 (C–Ar), 137.25 (C–Ar), 136.72 (C–
Ar), 134.22 (C–Ar), 130.25 (2C, C–Ar), 128.21 (2C, C–
Ar), 120.50 (C–Ar), 119.82 (2C, C–Ar), 117.40 (C–Ar),
7
1
3
NH2). C NMR (125 MHz, common NMR solvents) d
59.45 (C–Ar), 150.58 (2C, C–Ar), 149.15 (C–Ar), 147.45
C ), 146.24 (C ), 136.11 (C–Ar), 134.24 (C–Ar), 119.80
1
?
(
(
103.51 (C4), 21.11 (CH3). LCMS m/z [M] 312.13. Anal.
5
3
2C, C–Ar), 108.69 (C–Ar), 106.78 (C–Ar), 103.42 (C ).
4
Calcd. for C H N (312.37): C, 76.90; H, 5.16; N, 17.94.
20 16 4
?
LCMS m/z [M] 237.10. Anal. Calcd. for C H N
3 11 5
Found: C, 76.88; H, 5.26; N, 17.82.
-Nitro-4-(3-(2-nitropyridin-4-yl)-1H-pyrazol-5-yl)pyridine
13) Bright yellow crystals; Yield: 68 %; m.p.:
1
(
237.26): C, 65.81; H, 4.67; N, 29.52. Found: C, 64.13; H,
2
4
.29; N, 29.21.
(
-
1
2
-Chloro-4-(3-(pyridin-4-yl)-1H-pyrazol-5-yl)pyridine
320–322 °C; R : 0.56; FT-IR t
(ATR cm ): 3176
max
f
(
10) Pale white powder; Yield: 58 %; m.p.: 319–321 °C;
(2°N–H str. in pyrazole), 3106 (aromatic C–H str.), 1615
(C=N str.), 1597, 1343 (N=O str.), 1343 (C–N str.), 1211
-
1
Rf: 0.7; FT-IR tmax (ATR cm ): 3382 (2°N–H str. in
pyrazole), 3039 (aromatic C–H str.), 1590 (C=N str.), 1417,
1
(N–H bend.). H NMR (DMSO-d , 400 MHz,): d = 10.11
6
1
476, 1511 (C=C str. aromatic), 1239 (C–N str.). H NMR
1
(s, 1H, NH), 8.70–8.71 (d, 2H, Ar–H), 8.54 (s, 2H, Ar–H),
1
3
(
(
DMSO-d , 400 MHz,): d = 10.06 (s, 1H, NH), 8.52–8.54
d, 1H, Ar–H), 8.43–8.44 (d, 2H, Ar–H), 7.79 (s, 1H, Ar–
8.06–8.13 (m, 2H, Ar–H), 7.35 (s, 1H, CH). C NMR
(125 MHz, common NMR solvents) d 159.09 (C–Ar),
157.94 (C–Ar), 150.02 (C–Ar), 149.75 (C–Ar), 147.86
(C5), 146.16 (C3), 135.35 (C–Ar), 134.42 (C–Ar), 125.15
(C–Ar), 123.96 (C–Ar), 118.89 (C–Ar), 117.65 (C–Ar),
6
H), 7.62–7.64 (d, 1H, Ar–H), 7.39–7.41 (d, 2H, Ar–H),
.26 (s, 1H, CH). 13C NMR (125 MHz, common NMR
7
solvents) d 150.93 (C–Ar), 150.58 (2C, C–Ar), 149.12 (C–
Ar), 147.65 (C5), 146.15 (C3), 134.14 (C–Ar), 134.29 (C–
?
103.50 (C4). LCMS m/z [M] 312.48. Anal. Calcd. for
Ar), 121.45 (C–Ar), 120.28 (C–Ar), 119.80 (2C, C–Ar),
?
03.42 (C4). LCMS m/z [M] 256.21. Anal. Calcd. for
C H N O (312.24): C, 50.01; H, 2.58; N, 26.92. Found:
13 8 6 4
1
C, 50.11; H, 2.52; N, 26.86.
-(3-(2-Aminopyridin-4-yl)-1H-pyrazol-5-yl)pyridin-2-amine
14) Pale yellow powder; Yield: 62 %; m.p.:
C H ClN (256.69): C, 60.83; H, 3.53; Cl, 13.81; N,
1
3
9
4
4
2
1.83. Found: C, 60.86; H, 3.43; N, 21.66.
(
-
1
2
-Phenyl-4-(3-(pyridin-4-yl)-1H-pyrazol-5-yl)pyridine
350–352 °C; R : 0.62; FT-IR t
(ATR cm ): 3405,
max
f
(
11) Pale brown powder; Yield: 62 %; m.p.:
-
3382 (1°NH aromatic), 3038 (aromatic C–H str.), 1660
2
1
3
30–332 °C; R : 0.75; FT-IR t
(ATR cm ): 3374
max
(C=N str.), 1415, 1482, 1511 (C=C str. aromatic), 1242 (C–
1
f
(
(
2°N–H str. in pyrazole), 3030 (aromatic C–H str.), 1590
C=N str.), 1521, 1602, 1708 (C=C str. aromatic), 1345 (C–
N str.). H NMR (DMSO-d , 400 MHz,): d = 10.11 (s, 1H,
6
NH), 8.70–8.71 (d, 2H, Ar–H), 8.54 (s, 2H, Ar–H),
13
8.06–8.13 (m, 2H, Ar–H), 7.35 (s, 1H, CH). C NMR
1
N str.). H NMR (DMSO-d , 400 MHz,): d = 10.18 (s, 1H,
6
NH), 8.46–8.49 (d, 2H, Ar–H), 7.85 (s, 1H, Ar–H),
(125 MHz, common NMR solvents) d 161.05 (C–Ar),
159.45 (C–Ar), 149.73 (C–Ar), 149.15 (C–Ar), 147.88
(C5), 146.48 (C3), 139.16 (C–Ar), 135.83 (C–Ar),110.48
(C–Ar), 109.02 (C–Ar), 108.69 (C–Ar), 106.98 (C–Ar),
7
3
.74–7.76 (m, 2H, Ar–H), 7.50 (s, 1H, CH), 7.41–7.43 (m,
1
3
H, Ar–H), 7.11–7.23 (m, 3H, Ar–H).
C NMR
(
125 MHz, common NMR solvents) d 159.38 (C–Ar),
?
1
1
50.48 (C–Ar), 148.03 (C–Ar), 147.35 (C5), 146.18 (C3),
103.52 (C4). LCMS m/z [M] 252.11. Anal. Calcd. for
38.49 (C–Ar), 137.31 (C–Ar), 134.24 (C–Ar), 130.08 (C–
C H N (252.27): C, 61.89; H, 4.79; N, 33.31. Found: C,
13 12 6
Ar), 128.96 (2 C, C–Ar), 127.92 (2C, C–Ar), 120.42 (C–
Ar), 119.68 (2C, C–Ar), 117.38 (C–Ar), 103.22 (C4).
61.86; H, 4.68; N, 33.33.
-Chloro-4-(3-(2-chloropyridin-4-yl)-1H-pyrazol-5-yl)pyr-
idine (15) Pale white powder; Yield: 58 %; m.p.:
2
?
LCMS m/z [M] 298.12. Anal. Calcd. for C H N
1
9
14
4
(
298.34): C, 76.49; H, 4.73; N, 18.78. Found: C, 76.49; H,
-
1
3
80–281 °C; R : 0.65; FT-IR t
(ATR cm ): 3242
max
f
4
.80; N, 18.68.
(2°N–H str. in pyrazole), 2918 (aromatic C–H str.), 1663
4
-(3-(Pyridin-4-yl)-1H-pyrazol-5-yl)-2-p-tolylpyridine
(C=N str.), 1433, 1475, 1576 (C=C str. aromatic), 1238 (C–
1
N str.), 705 (C–Cl). H NMR (DMSO-d , 400 MHz,):
(
12) Pale yellow powder; Yield: 59 %; m.p.:
-
6
1
2
60–262 °C; R : 0.78; FT-IR t
(ATR cm ): 3374
max
d = 10.62 (s, 1H, NH), 8.52–8.54 (d, 2H, Ar–H), 7.78 (s,
2H, Ar–H), 7.61–7.66 (m, 2H, Ar–H), 6.87 (s, 1H, CH).
f
(
(
2°N–H str. in pyrazole), 3030 (aromatic C–H str.), 1708
C=N str.), 1521, 1538, 1602 (C=C str. aromatic), 1345 (C–
1
3
C NMR (125 MHz, common NMR solvents) d 153.19
1
N str.). H NMR (DMSO-d , 400 MHz,): d = 10.24 (s, 1H,
(C–Ar), 150.93 (C–Ar), 150.72, 149.42 (C–Ar), 147.88
(C5), 146.14 (C3), 136.68 (C–Ar), 134.29 (C–Ar), 121.42
(C–Ar), 120.45 (2C, C–Ar), 118.72 (C–Ar), 103.46 (C4).
6
NH), 8.53–8.55 (d, 3H, Ar–H), 7.91 (s, 1H, Ar–H),
7
.76–7.77 (m, 2H, Ar–H), 7.51 (s, 1H, CH), 7.47–7.50 (m,
1
23

Med Chem Res
?
LCMS m/z [M] 290.01. Anal. Calcd. for C H C N
?
z [M] 301.03. Anal. Calcd. for C H ClN O (301.69): C,
13 8 5 2
1
3
8
l2
4
(
291.14): C, 53.63; H, 2.77; N, 19.24. Found: C, 53.64; H,
51.76; H, 2.67; N, 23.21. Found: C, 51.66; H, 2.57; N,
23.18.
2
.74; Cl, 24.38; N, 19.18.
2
-Phenyl-4-(3-(2-phenylpyridin-4-yl)-1H-pyrazol-5-yl)pyr-
4-(3-(2-Chloropyridin-4-yl)-1H-pyrazol-5-yl)pyridin-2-amine
idine (16) Yellowish-white powder; Yield: 70 %; m.p.:
-
(
19) Yellow powder; Yield: 58 %; m.p.: 312–314 °C; R :
f
1
-
1
2
93–297 °C; R : 0.72; FT-IR t
(ATR cm ): 3310
max
f
0.62; FT-IR tmax (ATR cm ): 3405, 3382 (1°N–H str.),
3
(
(
2°N–H str. in pyrazole), 3047 (aromatic C–H str.), 1591
C=N str.), 1409, 1494, 1519 (C=C str. aromatic), 1269 (C–
039 (aromatic C–H str.), 1588 (C=N str.), 1415, 1492,
1
511 (C=C str. aromatic), 1588 (C–N str.),666 (C–Cl). H
1
1
N str.). H NMR (DMSO-d , 400 MHz,): d = 10.27 (s, 1H,
6
NMR (DMSO-d , 400 MHz,): d = 9.85 (s, 1H, NH), 8.75
6
NH), 8.56–8.55 (d, 2H, Ar–H), 7.93 (m, 4H, Ar–H),
(d, 1H, Ar–H), 8.14 (d, 1H, Ar–H), 7.75 (d, 1H, Ar–H),
7
6
.78–7.81 (m, 2H, Ar–H), 7.51 (s, 1H, CH), 7.16–7.30 (m,
7
.70 (s, 1H, Ar–H), 6.98 (s, 1H, CH), 6.85–6.86 (d, 2H,
1
3
H, Ar–H). C NMR (125 MHz, common NMR solvents)
1
3
Ar–H), 1.43 (s, 2H, NH2). C NMR (125 MHz, common
d 161.00 (C–Ar), 159.54 (C–Ar), 148.06 (2C, C–Ar),
47.88 (C5), 146.18 (C3), 138.59 (2C, C–Ar), 137.84 (C–
NMR solvents) d 157.08 (C–Ar), 153.17 (C–Ar), 150.75
1
(C–Ar), 147.85 (C5), 147.67 (C–Ar), 136.70 (C–Ar),
Ar), 137.35 (C–Ar), 130.08 (2C, C–Ar), 128.96 (4C, C–
Ar), 127.96 (4C, C–Ar), 120.93 (C–Ar), 120.52 (C–Ar),
1
35.74 (C–Ar), 134.30 (C3), 120.43 (C–Ar), 118.69 (C–
Ar), 113.77 (C–Ar), 113.39 (C–Ar), 101.30 (C4). LCMS m/
?
z [M] 271.16. Anal. Calcd. for C H ClN (271.7): C,
1
17.82 (C–Ar), 117.42 (C–Ar), 103.52 (C4). LCMS m/
1
3
10
5
?
z [M] 374.15. Anal. Calcd. for C H N (374.44): C,
2
5
18
4
57.47; H, 3.71; N, 25.78. Found: C, 57.44; H, 3.76; N,
8
1
0.19; H, 4.85; N, 14.96. Found: C, 80.16; H, 4.78; N,
4.88.
25.74.
4
-(3-(2-Chloropyridin-4-yl)-1H-pyrazol-5-yl)-2-phenylpyr-
2
-p-Tolyl-4-(3-(2-p-tolylpyridine-4-yl)-1H-pyrazol-5-yl)pyridine
idine (20) Pale white powder; Yield: 57 %; m.p.:
-1
(
17) Pale yellow brown powder; Yield: 60 %; m.p.:
-
245–247 °C; R : 0.69; FT-IR t
(ATR cm ): 3392
max
1
f
3
43–346 °C; R : 0.71; FT-IR t
(ATR cm ): 3310
max
f
(2°N–H str. in pyrazole), 3047 (aromatic C–H str.), 1590
(C=N str.), 1387, 1414, 1493 (C=C str. aromatic), 1242 (C–
(2°N–H str. in pyrazole), 3047 (aromatic C–H str.), 1591
(C=N str.), 1409,1477, 1575 (C=C str. aromatic), 1241 (C–
1
N str.). H NMR (DMSO-d , 400 MHz,): d = 10.23 (s, 1H,
NH), 8.61–8.63 (d, 1H, Ar–H), 8.53–8.54 (d, 1H, Ar–H),
1
6
N str.). H NMR (DMSO-d , 400 MHz,): d = 10.33 (s, 1H,
NH), 8.62–8.64 (d, 2H, Ar–H), 7.92–7.94 (d, 2H, Ar–H),
6
7
4
.77–7.90 (m, 5H, Ar–H), 7.45 (s, 1H, CH), 7.17–7.36 (m,
13
H, Ar–H). C NMR (125 MHz, common NMR solvents)
7
6
.75–7.78 (m, 4H, Ar–H), 7.42 (s, 1H, CH), 7.16–7.33 (m,
1
3
H, Ar–H), 2.23 (s, 6H, CH3). C NMR (125 MHz,
d 159.54 (C–Ar), 153.19 (C–Ar), 150.75 (C–Ar), 148.06
common NMR solvents) d 161.00 (C–Ar), 159.54 (C–Ar),
(C–Ar), 147.88 (C5), 146.18 (C3), 138.59 (C–Ar), 137.35
1
48.06 (C–Ar), 147.88 (C5), 146.18 (C3), 141.24 (C–Ar),
(C–Ar), 136.70 (C–Ar), 130.08 (C–Ar), 128.92 (2C, C–Ar),
1
37.84 (C–Ar), 137.35 (C–Ar), 136.72 (2C, C–Ar), 130.27
1
27.86 (2C, C–Ar), 120.52 (C–Ar), 120.44 (C–Ar), 118.72
?
(
4C, C–Ar), 128.23 (4C, C–Ar), 120.93 (C–Ar), 120.52 (C–
(C–Ar), 117.42 (C–Ar), 103.48 (C4). LCMS m/z [M]
Ar), 117.82 (C–Ar), 117.42 (C–Ar), 103.52 (C4), 21.13
?
2C, CH3). LCMS m/z [M] 402.18. Anal. Calcd. for
332.08. Anal. Calcd. for C H ClN (332.79): C, 68.57;
19 13
4
(
H, 3.94; N, 16.84. Found: C, 68.59; H, 3.98; N, 16.84.
C H N (402.49): C, 80.57; H, 5.51; N, 13.92. Found:C,
27 22 4
8
0.47; H, 5.70; N, 13.88.
4-(3-(2-Chloropyridin-4-yl)-1H-pyrazol-5-yl)-2-p-tolylpyr-
4
-(3-(2-Chloropyridin-4-yl)-1H-pyrazol-5-yl)-2-nitropyridine
idine (21) Pale yellow powder; Yield: 65 %; m.p.:
-
1
(
18) Pale yellow powder; Yield: 68 %; m.p.:
-
276–278 °C; R : 0.64; FT-IR t
(ATR cm ): 3391
max
f
1
2
94–296 °C; R : 0.68; FT-IR t
(ATR cm ): 3261
max
(2°N–H str. in pyrazole), 3047 (aromatic C–H str.), 1590
(C=N str.), 1240 (C–N str.). H NMR (DMSO-d6,
f
1
(2°N–H str. in pyrazole), 2919 (aromatic C–H str.), 1576
(C=N str.), 1371, 1475, 1539 (C=C str. aromatic), 1586,
400 MHz,): d = 10.20 (s, 1H, NH), 8.61 (d, 1H, Ar–H),
8.50 (d, 1H, Ar–H), 7.72–7.87 (m, 5H, Ar–H), 7.42 (s, 1H,
1
475 (N=O str.), 1378 (C–N str.), 694 (C–Cl). H NMR
1
1
3
(
(
DMSO-d , 400 MHz,): d = 9.90 (s, 1H, NH), 8.80–8.82
d, 1H, Ar–H), 8.61 (d, 1H, Ar–H), 8.26 (s, 1H, CH),
CH), 7.11–7.20 (m, 3H, Ar–H), 2.16 (s, 3H, CH3).
C
6
NMR (125 MHz, common NMR solvents) d 159.48 (C–
Ar), 153.19 (C–Ar), 150.65 (C–Ar), 148.06 (C–Ar), 147.82
(C ), 146.18 (C ), 141.24 (C–Ar), 137.35 (C–Ar), 136.70
8
1
.08–8.09 (d, 1H, Ar–H), 7.87 (s, 1H, Ar–H), 7.79–7.81 (d,
1
H, Ar–H), 6.86 (s, 1H, Ar–H). C NMR (125 MHz,
3
5
3
common NMR solvents) d 159.09 (C–Ar), 153.19 (C–Ar),
(C–Ar), 136.70 (C–Ar), 130.24 (C–Ar), 130.27 (C–Ar),
?
1
50.75 (C–Ar), 149.85 (C–Ar), 147.88 (C5), 146.18 (C3),
128.19 (C–Ar), 103.39 (C ), 21.11 (CH ). LCMS m/z [M]
4
3
1
36.70 (C–Ar), 134.44 (C–Ar), 123.99 (C–Ar), 120.45 (C–
346.09. Anal. Calcd. for C H ClN (346.81): C, 69.26;
2
0
15
4
Ar), 118.72 (C–Ar), 117.69 (C–Ar), 103.52 (C4). LCMS m/
H, 4.36; N, 16.15. Found: C, 69.16; H, 4.28; N, 16.21.
123

Med Chem Res
4
-(5-(2-Nitropyridin-4-yl)-1H-pyrazol-3-yl)pyridin-2-amine
C, 67.22; H, 4.23; N, 19.60. Found: C, 67.18; H, 4.25; N,
19.54.
(
22) Orange-Yellow powder; Yield: 58 %; m.p.:
-
1
3
24–326 °C; R : 0.55; FT-IR t
(ATR cm ): 3308,
max
f
4-(3-(2-Phenylpyridin-4-yl)-1H-pyrazol-5-yl)pyridin-2-amine
3383 (1° N–H str. in aromatic amines) 3177 (2°N–H str. in
pyrazole), 3039 (aromatic C–H str.), 1650 (C=N str.), 1599,
(
25) Yellow–brown powder; Yield: 61 %; m.p.: 352–
1
-
3
55 °C; R : 0.74; FT-IR t
(ATR cm ): 3405, 3382
max
f
1
1
4
411 (N=O str.) 1491, 1509, 1336 (C=C str. aromatic),
1
242 (C–N str.),1212 (N–H bend). H NMR (DMSO-d ,
(
(
1°N–H str. aromatic), 3040 (aromatic C–H str.), 1660
C=N str.), 1415, 1466, 1513 (C=C str. aromatic), 1242 (C–
6
00 MHz,): d = 10.28 (s, 1H, NH), 8.89–8.90 (d, 1H, Ar–
1
N str.). H NMR (DMSO-d , 400 MHz,): d = 10.33 (s, 1H,
6
H), 8.74 (s, 1H, Ar–H), 8.19–8.27 (m, 2H, Ar–H), 7.55 (s,
1
1
NH), 8.63 (d, 1H, Ar–H), 8.17–8.19 (d, 1H, Ar–H), 8.00 (s,
3
H, CH), 6.95–7.01 (m, 2H, Ar–H), 1.51 (s, 2H, NH2).
C
1
7
H, Ar–H), 7.89–7.91 (d, 2H, Ar–H), 7.50 (s, 1H, CH),
.25–7.37 (m, 4H, Ar–H), 6.92–6.93 (m, 2H, Ar–H), 1.54
NMR (125 MHz, common NMR solvents) d 161.05 (C–
Ar), 159.09 (C–Ar), 149.85 (C–Ar), 149.73 (C–Ar), 147.88
1
s, 2H, NH2). C NMR (125 MHz, common NMR sol-
3
(
(
(
C5), 146.18 (C3), 139.16 (C–Ar), 134.44 (C–Ar), 123.99
vents) d 159.48 (C–Ar), 159.25 (C–Ar), 149.15 (C–Ar),
C–Ar), 117.69 (C–Ar), 110.48 (C–Ar), 109.02 (C–Ar),
?
03.52 (C4). LCMS m/z [M] 282.08. Anal. Calcd. for
147.04 (C–Ar), 147.98 (C ), 146.15 (C ), 138.55 (C–Ar),
5 3
1
137.82 (C–Ar), 135.78 (C–Ar), 130.05 (C–Ar), 128.92 (2C,
C H N O (282.26): C, 55.32; H, 3.57; N, 29.77. Found:
13 10 6 2
C–Ar), 127.88 (2C, C–Ar), 120.94 (C–Ar), 117.72 (C–Ar),
C, 55.22; H, 3.69; N, 29.58.
-Nitro-4-(3-(2-phenylpyridin-4-yl)-1H-pyrazol-5-yl)pyridine
23) Yellow powder; Yield: 62 %; m.p.: 352–354 °C; R :
1
08.69 (C–Ar), 106.98 (C–Ar), 103.45 (C ). LCMS m/
4
?
2
z [M] 313.13. Anal. Calcd. for C H N (327.38): C,
20 17 5
(
73.37; H, 5.23; N, 21.39. Found: C, 73.68; H, 5.82; N,
21.38.
f
-
1
0
.63; FT-IR tmax (ATR cm ): 3106 (2°N–H str. in
pyrazole), 3062 (aromatic C–H str.), 1612 (C=N str.), 1344
1
C–N str.),1586, 1344 (N=O str.). H NMR (DMSO-d ,
(
4-(3-(2-p-tolylpyridine-4-yl)-1H-pyrazol-5-yl)pyridin-2-amine
6
400 MHz,): d = 10.30 (s, 1H, NH), 8.85–8.87 (d, 1H, Ar–
H), 8.71 (s, 1H, Ar–H), 8.60–8.62 (m, 1H, Ar–H),
(26) Pale yellow powder; Yield: 63 %; m.p.:
-
1
286–287 °C; R : 0.64; FT-IR tmax (ATR cm ): 3406,
f
8
7
4
.21–8.23 (d, 1H, Ar–H), 7.98–7.99 (s, 1H, Ar–H),
.87–7.89 (d, 2H, Ar–H), 7.59 (s, 1H, CH), 7.23–7.35 (m,
3383 (1° N–H str. aromatic), 3178 (C–H str. aliphatic)
3083 (2°N–H str. in pyrazole), 3041 (aromatic C–H str.),
1589 (C=N str.), 1415, 1492, 1551 (C=C str. aromatic),
1
H, Ar–H). C NMR (125 MHz, common NMR solvents)
3
1
d 161.00 (C–Ar), 159.09 (C–Ar), 149.85 (C–Ar), 148.04
1242 (C–N str.). H NMR (DMSO-d
6
, 400 MHz,):
(
C–Ar), 147.88 (C5), 146.18 (C3), 138.59 (C–Ar), 137.84
d = 10.24 (s, 1H, NH), 8.54 (d, 1H, Ar–H), 8.08–8.10 (d,
1H, Ar–H), 7.91 (s, 1H, Ar–H), 7.75–7.77 (d, 2H, Ar–H),
7.41 (s, 1H, CH), 7.27 (d, 1H, Ar–H), 7.13–7.14 (d, 2H,
(
C–Ar), 134.35 (C–Ar), 130.08 (C–Ar), 128.96 (2C, C–
Ar), 127.96 (2C, C–Ar), 122.99 (C–Ar), 120.93 (C–Ar),
17.81 (C–Ar), 117.69 (C–Ar), 103.50 (C4). LCMS m/
1
Ar–H), 6.83 (s, 2H, Ar–H), 2.18 (s, 3H, CH
), 1.46 (s, 2H,
3
?
z [M] 343.34. Anal. Calcd. for C H N O (343.34): C,
NH ). 13C NMR (125 MHz, common NMR solvents) d
1
9 13 5 2
2
6
2
6.47; H, 3.82; N, 20.40. Found: C, 66.45; H, 3.68; N,
0.20.
161.00 (C–Ar), 159.41 (C–Ar), 149.11 (C–Ar), 148.02 (C–
Ar), 147.85 (C5), 146.16 (C3), 141.14 (C–Ar), 137.77 (C–
Ar), 136.70 (C–Ar), 135.78 (C–Ar), 130.22 (2C, C–Ar),
2
-Nitro-4-(3-(2-p-tolylpyridine-4-yl)-1H-pyrazol-5-yl)pyr-
128.23 (2C, C–Ar), 120.83 (C–Ar), 117.79 (C–Ar), 108.58
idine (24) Pale yellow powder; Yield: 68 %; m.p.:
-
(
C–Ar), 106.87 (C–Ar), 102.82 (C4), 21.13 (CH3). LCMS
1
3
44–345 °C; R : 0.6; FT-IR t
(ATR cm ): 3106 (2°N–
H str. in pyrazole), 3047 (aromatic C–H str.), 2872 (ali-
?
m/z [M] 327.14. Anal. Calcd. for C H N (327.38): C,
f
max
2
0 17 5
7
2
3.37; H, 5.23; N, 21.39. Found: C, 73.32; H, 5.32; N,
1.43.
phatic C–H str.) 1597 (C=N str.), 1597, 1343 (N=O str.),
1
270 (C–N str.), 1210 (N–H bend). H NMR (DMSO-d ,
1
6
4
00 MHz,): d = 10.26 (s, 1H, NH), 8.82–8.83 (d, 1H, Ar–
H), 8.68 (s, 1H, Ar–H), 8.56 (d, 1H, Ar–H), 8.18–8.20 (d,
2-Phenyl-4-(3-(2-p-tolylpyridine-4-yl)-1H-pyrazol-5-yl)pyridine
(27) Pale brown powder; Yield: 57 %; m.p.: 347–349 °C;
-
1
1
7
H, Ar–H), 7.94 (s, 1H, Ar–H), 7.79–7.81 (d, 2H, Ar–H),
.56 (s, 1H, CH), 7.16–7.31 (m, 3H, Ar–H), 2.23 (s, 3H,
R
f
: 0.58; FT-IR tmax (ATR cm ): 3389 (2°N–H str. in
pyrazole), 3178 (C–H str. aliphatic) 3047 (aromatic C–H
1
3
CH3). C NMR (125 MHz, common NMR solvents) d
61.00 (C–Ar), 159.09 (C–Ar), 149.85 (C–Ar), 148.04 (C–
str.), 1660 (C=N str.), 1408, 1463, 1538 (C=C str. aro-
1
matic), 1269 (C–N str.). H NMR (DMSO-d
1
6
, 400 MHz,):
Ar), 147.88 (C5), 146.18 (C3), 141.24 (C–Ar), 137.84 (C–
Ar), 136.72 (C–Ar), 134.44 (C–Ar), 130.27 (2C, C–Ar),
d = 10.37 (s, 1H, NH), 8.65 (d, 2H, Ar–H), 8.01 (s, 2H,
Ar–H), 7.84–7.89 (m, 4H, Ar–H), 7.60 (s, 1H, CH),
1
28.23 (2C, C–Ar), 124.09 (C–Ar), 121.03 (C–Ar), 117.82
7.21–7.38 (d, 7H, Ar–H), 2.28 (s, 3H, CH ). 13C NMR
3
(
C–Ar), 117.66 (C–Ar), 102.99 (C4), 21.13 (CH3). LCMS
(125 MHz, common NMR solvents) d 161.05 (C–Ar),
159.54 (C–Ar), 148.04 (C–Ar), 148.04 (C–Ar), 147.68
?
m/z [M] 357.12. Anal. Calcd. for C H N O (357.37):
2
0 15 5 2
1
23

Med Chem Res
Table 6 Unicolumn statistics of selected training and test sets for 2D QSAR model development of compound for antiepileptic activity
Column name
Average
Max
Min
SD
Sum
Unicolumn statistics: training set
-
log (% inhibition)
Unicolumn statistics: test set
log (% inhibition)
-1.8050
-1.6910
-1.0980
-1.2380
-1.9920
-1.9830
0.3123
0.3480
-25.2700
-11.8370
-
(
(
C5), 146.18 (C3), 141.14 (C–Ar), 138.54 (C–Ar), 137.81
C–Ar), 137.25 (C–Ar), 136.69 (C–Ar), 130.25 (2C, C–Ar),
1
1
30.08 (C–Ar), 128.92 (2C, C–Ar), 128.21 (2C, C–Ar),
27.92 (2C, C–Ar), 121.05 (C–Ar), 120.42 (C–Ar), 117.82
(
C–Ar), 117.40 (C–Ar), 103.46 (C4), 21.15 (CH3). LCMS
?
m/z [M] 388.16. Anal. Calcd. for C H N (388.46): C,
2
6 20 4
8
1
0.39; H, 5.19; N, 14.42. Found: C, 80.88; H, 5.20; N,
4.42.
Pharmacology
The anticonvulsant studies of the synthesized compounds
were conducted as per National Institute of Neurological
and Communicative Disorders and Stroke (NINDS)
guidelines (Roger et al., 1984), at Animal house of Pacific
College of Pharmacy, PAHER University, Udaipur, having
institutional animal ethical committee with Registration no.
Fig. 10 Alignment of molecules for the development of 3D QSAR
models through KNN-MFA
1
622/PO/a/12/CPCSEA.
The profile of anticonvulsant activity of final com-
a period of at least 5 s in 97 % (CD ) of animals tested.
9
7
pounds 7–27 was established in the MES and scPTZ tests,
after intraperitoneal (i.p.) injection in mice at doses of 30,
PTZ is administered 0.5 and 4 h after injections of tested
compounds, and observation is carried out for 30 min. In
the control groups, the first episode of clonic convulsions is
observed between 6 and 15 min of observation. The
absence of clonic convulsions in the observed time period
of 0.5 and 4 h is interpreted as the compound’s ability to
protect against PTZ-induced seizures.
1
00, and 300 mg/kg. For each compound, the animals were
divided into three groups for three different doses, each
group consisting of six animals. An observation was car-
ried out at two different time intervals—0.5 and 4 h.
In the MES screen, an electrical stimulus of 0.2 s in
duration (150 mA in rats) is delivered via corneal elec-
trodes primed with an electrolyte solution containing an
anesthetic agent. The end point is the tonic extension of the
hind limbs. In the control (i.p. saline) groups, the proce-
dures cause immediate hind limb tonic extension. Rats not
displaying hind limb tonic extension were considered to be
protected from seizures. The % inhibition of duration of
tonic extensor phase was calculated for each animal for the
purpose of comparison of results.
Furthermore, in addition to the primary anticonvulsant
screening, the acute neurological toxicity (NT) was deter-
mined in mice by the rotarod test and in rats by positional
sense test.
To find out the statistical significance, the results of each
group of animals were compared with that of control with
the help of Student’s paired t test. The value of P \ 0.05
was considered to be statistically significant.
The scPTZ test utilizes a dose of pentylenetetrazole
QSAR
(
85 mg/kg in mice) that produces clonic seizures lasting for
2D QSAR
Fig. 9 3,5-Dipyridinyl-1H-
pyrazole nucleus as the template
for the alignment
N
All the synthesized compounds were subjected to 2D
QSAR analysis via three different regression methods, viz.,
multiple linear regression, principal component regression,
and partial least square regression methods to obtain the
statistically best model.
N
N
H
N
123

Med Chem Res
Table 7 Unicolumn statistics of selected training and test sets for 3D QSAR model development of compound for antiepileptic activity
Column name
Average
Max
Min
SD
Sum
Unicolumn statistics: training set
-
log (% inhibition)
Unicolumn statistics: test set
log (% inhibition)
-1.8050
-1.6910
-1.0980
-1.2380
-1.9920
-1.9830
0.3123
0.3480
-25.2700
-11.8370
-
The structures were constructed using the 2D draw
application and were converted to 3D structures. Energy
minimization and geometry optimization was conducted
using Merck molecular force field (MMFF) method with
root mean square (RMS) gradient set to 0.01 and iteration
limit to 10,000. The 2D descriptors (physicochemical and
alignment independent) were calculated for the optimized
compounds, and the invariable descriptors (the descriptors
that are constant for all the molecules) were removed as
they do not contribute to QSAR. The manual selection of
test and training set was done in the approximate ratio of 1:
3D QSAR
Alignment of all the 21 compounds was done using tem-
plate-based alignment, where a template structure (Fig. 9)
was defined and used as a basis for the alignment of a set of
molecules; the aligned structures (Fig. 10) were used for
3D QSAR study.
The electrostatic and steric descriptors were calculated
for given biological data (-log % inhibition) of the com-
pounds. For calculation of field descriptor values, both
electrostatic and steric field types with cutoffs 10.0 and
30.0 kcal/mol, respectively, were selected, and charge type
was selected as Gasteiger–Marsili. Dielectric constant was
set to 1.0 considering the distance-dependent dielectric
function. Probe setting was carbon atom with charge 1.0
and grid setting as follows:
3
according to Golbraikh and Tropsha (2002) which
resulted in models with statistically better predictive power
than by random selection method. Accordingly for each
biological activity, seven compounds were selected as test
set and remaining 14 compounds as training set. Calcula-
tion of unicolumn statistics (Table 6) further reflected the
right selection of test and training sets as shown in each
section of activity below.
From
To
Interval
2.0000
2.0000
2.0000
X
Y
Z
-10.115
0.306657
-4.0000
24.0225
14.2794
4.0000
The above showed that the test set is interpolative, i.e.,
derived within the minimum–maximum range of the
training set. The average and standard deviation of the
training and test sets provides insight into the relative
difference in mean and point density distribution (along
mean) of the two sets. In each of the cases below, mean in
the test set is higher than the training set shows the pres-
ence of relatively more active molecules as compared to
the inactive ones. Also, the similar standard deviation in
both the sets in each case indicates that the spread in both
the set with their respective mean is comparable.
This resulted in the calculation of 2080 field descriptors
1040 for each electrostatic and steric) for all the compounds
(
in separate columns. The invariable columns were removed
from the work sheet. The optimal test and training data sets
were generated using manual method as done for 2D QSAR
analysis. The method resulted in selection and seven com-
pounds as test set and remaining 14 as training set.
The set data was observed for activity distribution
through unicolumn statistics (Table 7) which revealed that
almost all the compounds in test set are within the mini-
mum–maximum limit of the training set.
All the 21 compounds were subjected to regression
analysis using MLR, PCR, and PLS as model-building
methods coupled with stepwise variable selection. QSAR
equations were generated by using -1og % inhibition
values as dependent variable and various descriptor values
as independent variables. Regression analysis was carried
out and the best model cross-validated. Cross-correlation
Building QSAR models by KNN-MFA Since these were
a large pool of descriptors available to build model, various
variable selection methods were used along with k-nearest
neighborhood (KNN) to find optimal subset of descriptors
for KNN-MFA model, viz., stepwise variable selection,
simulated annealing, and genetic algorithm.
limit was set as 0.5, number of variable in final equation as
2
, and term selection criteria as r , F-test ‘‘in’’ as 4 and F-
5
test ‘‘out’’ as 3.99. Variance cutoff was set to 0 and scaling
as autoscaling, number of random iteration was set to 10.
1
23

Med Chem Res
For stepwise forward–backward method, the cross-cor-
relation limit was set to 0.5 and term selection criteria as
must have a smaller substituent having electronic character
as suggested by 3D QSAR result. Thus, it can be concluded
that the nucleus 3,5-bipyridinyl-1H-pyrazole itself is the
pharmacophoric site of the molecule and by taking clues
from models obtained through 2D and 3D QSAR analysis,
new, better anticonvulsant agents can be developed.
2
q . F-test ‘‘in’’ was set to 4.0 and F-test ‘‘out’’ to 3.99. As
some additional parameters, variance cutoff was set as
˚
2
kcal/mol A and scaling and autoscaling, additionally the
K-nearest neighbor parameter setting was done within the
range of 2–5 and prediction method was selected as dis-
tance-based weighted average.
Acknowledgments Authors are highly thankful to sophisticated
analytical instrument facility (SAIF), Punjab University, Chandigarh,
India, for spectrophotometric and elemental analyses. We are also
grateful to Mr. Neeraj Kumar Kamra, Asst. Professor, Geetanjali
Institute of Pharmacy, Udaipur, Rajasthan, India, for his support
during the project.
For simulated annealing as variable selection method,
the cross-correlation limit was set as 0.5, terms in model as
4
0
, maximum temperature as 100, minimum temperature as
.01, iteration at given temperature as 5, decrease tem-
perature by as 10, seed as 0, perturbation limit as 1, and
2
term selection criteria as q .
Compliance with ethical standards
For the development of models through GA, the GA
parameter settings included cross-correlation limit as 0.5,
crossover probability as 0.95, mutation probability as 0.05,
Conflict of interest The authors declare no conflict of interest.
2
term selection criteria as q , population as 10, no. of gen-
References
erations 10,000, print after iterations 100, and seed as 0.
Convergence criteria 0.01, convergence ending criteria 5,
and chromosome length as 3. Additional parameter settings
were described in stepwise variable selection method.
Ajmani S, Jadhav K, Kulkarni SA (2006) Three-dimensional QSAR
using the K-nearest neighbor method and its interpretation.
J Chem Inf Model 46:24–31
Armijo JA, Shushtarian M, Valdizan EM, Cuadrado A, de las Cuevas
I, Ad ´ı n J (2005) Ion channels in epilepsy. Curr Pharm Des
1
1(15):1975–2003
Conclusion
Bialer M, Johannessen SI, Kupferberg HJ, Levy RH, Perucca E,
Tomson T (2007) Progress report on new antiepileptic drugs: a
summary of the eight Eilat conference (EILAT VIII). Epilepsy
Res 73(1):1–52
Bruno-Blanch L, Galvez J, Garcia-Domenach R (2003) Topological
virtual screening: a way to find new anticonvulsant drugs from
chemical diversity. Bioorg Med Chem Lett 13(16):2749–2754
Golbraikh A, Tropsha A (2002) Predictive QSAR modeling based on
diversity sampling of experimental datasets for the training and
test set selection. J Comput Aided Mol Des 16(5–6):357–369
Heller ST, Natarajan SR (2006) 1, 3-Diketones from acid chlorides
and ketones: a rapid and general one pot synthesis of pyrazoles.
Org Lett 8(13):2675–2678
Hou T, Xu X (2001) Three-dimensional quantitative structure activity
relationship analysis of some cinnamamides. Chemom Intell Lab
56:123–132
Krogsgaard-Larsen P, Liljefors T, Ulf Madsen (2002) Textbook of
drug design and discovery. Taylor and Francis, London, p 554
Lien EJ, Tong GL, Chou JT, Lien LL (1973) Structural requirements
for centrally acting drugs I. J Pharm Sci 62(2):246–250
In conclusion, the present project gave the library of 21
new 3,5-bipyridinyl-1H-pyrazole derivatives with seven
compounds having good antiepileptic potential against
maximum electroshock (MES) convulsions. The SAR and
QSAR studies revealed that the anticonvulsant activity of
pyrazolyl pyridines was related with the substituents on
pyridyl as well as pyrazole moiety. Two important con-
clusions from 2D QSAR analysis were as follows: First, the
electronic character of substituent in the present set of the
compound was less important than bulk of the same, and
second, a smaller molecule substituted at 3-position on
pyrazole and 8- and/or 17-position(s) of pyridine would
give rise to more effective anticonvulsant agents. Result of
3
D QSAR analysis was also in accordance with the results
Lo
¨
scher W, Schmidt D (2002) New horizons in the development of
antiepileptic drug. Epilepsy Res 50:3–16
of pharmacological screening as the range of steric and
electrostatic fields at the mentioned points suggested the
bulk and electronic character of the substituent at that
position. Two important conclusions drawn from 3D
QSAR analysis were as follows: Compounds having sub-
stitution on pyridine moieties with an atom or group having
the same electrostatic potential would be less active while
pyridyl substitutions with different electronic characters
would favor the activity. Second, bulk is important for
entrance into binding pocket of the receptor; compounds
with bulky substituents on both the pyridyl moieties were
found to be detrimental for the activity, for a compound to
be a better anticonvulsant, either of the pyridyl moieties
Malawska B (2005) New anticonvulsant agents. Curr Top Med Chem
(1):69–85
Marder M, Estiu G, Blanch LB, Viola H, Wasowski C, Medina JH
(2001) Molecular modelling and QSAR analysis of the interac-
tion of flavone derivatives with the benzodiazepine site of
5
A
GABA receptor complex. Bioorg Med Chem 9:323–335
Marone S, Rozas I, Weaver DF (1999) Theoretical structural analyses
of tricyclic neuroactive drugs: quantum pharmacologic descrip-
tors for clustering anticonvulsant, antidepressant, and antipsy-
chotic activities. J Mol Struct 467:25–30
Meldrum BS, Rogawski MA (2007) Molecular targets for antiepilep-
tic drug development. Neurotherapeutics 4(1):18–61
Palluotto F, Carotti A, Casini G, Campagna F, Genchi G, Rizzo M,
De Sarro GB (1996) Structure activity relationships of 2-aryl-
123

Med Chem Res
2
,5dihydropyridazino [4,3-b]indol-3(3H)-ones at the benzodi-
Subbareddy NS, Sinha S, Parthasarathy S (2014) Epilepsy: Indian
perspective. Ann Indian Acad Neurol 17(Suppl 1):S3–S11
Vlife MDS software package, version 4.4, supplied by Vlife science
technologies Pvt. Ltd, 1, Akshay 50, Anand park, Aundh, Pune,
India 411007
azepine receptor. Bioorg Med Chem 4(Pt. 12):2091–2104
Pradhan J, Goyal A (2015) Synthesis of some nuclear substitution
derivatives of 4-acetyl pyridine. Int J Pharm Chem Anal
2
(3):113–119
Rogawski MA, Porter RJ (1990) Antiepileptic drugs: pharmacological
mechanisms and clinical efficacy with consideration of promis-
ing developmental stage compounds. Pharmacol Rev
Verli H, Albuquerque MG, Bicca de Alencastro R, Barreiro EJ (2002)
Local intersection volume: a new 3D descriptor applied to
develop a 3D-QSAR pharmacophore model for benzodiazepine
receptor ligands. Eur J Med Chem 37(3):219–229
4
2(3):223–286
Roger JP, James JC, Gill DG, Bettie JH, Harvey JK, Barrett S, Billy
GW (1984) Antiepileptic drug development program. Cleve Clin
J Med 51(2):293–305
Wong MG, Defina JA, Andrews PR (1986) Conformational analysis
of clinically active anticonvulsant drugs.
29(4):562–572
J Med Chem
1
23