Synthesis and structure-activity relationships of novel pyrimido[1,2-b]indazoles as potential anticancer agents against A-549 cell lines

Article abstract of DOI:10.1016/j.bmcl.2007.03.087

A series of novel pyrimido[1,2-b]indazoles 5, 7 have been prepared from 3-trifluoromethyl-5-phenyl-2,6-dicyano anilines 1 via novel indazole regioisomers 3 and 4 through a facile strategy. Specific examples were evaluated for anticancer activity in vitro and found to exhibit promising activity against A-549 cell lines and are more effective than Etoposide. QSAR models were developed and validated by cross-validation method. The results of the best QSAR model were further compared with the crystal structure of tubulin protein. The binding energies estimated were found to have a good correlation with the experimental inhibitory potencies.

Full text of DOI:10.1016/j.bmcl.2007.03.087

Bioorganic & Medicinal Chemistry Letters 17 (2007) 3445–3453
Synthesis and structure–activity relationships of
novel pyrimido[1,2-b]indazoles as potential anticancer
agents against A-549 cell linesq
T. Yakaiah,a B. P. V. Lingaiah,a B. Narsaiah,a,* B. Shireesha,b B. Ashok Kumar,c
S. Gururaj,d T. Parthasarathyc and B. Sridhare
aFluoroorganic Division, Indian Institute of Chemical Technology, Hyderabad 500 007, India
bCollege of Pharmaceutical Sciences, Kakatiya University, Warangal 506 302, India
cDepartment of Chemistry, Nizam College, Basheer Bagh, Hyderabad 500 001, India
dGlobal Institute of Biotechnology, Himayatnagar, Hyderabad 500 001, India
eLaboratory of X-ray Crystallography, Indian Institute of Chemical Technology, Hyderabad 500 007, India
Received 7 November 2006; revised 5 March 2007; accepted 27 March 2007
Available online 30 March 2007
Abstract—A series of novel pyrimido[1,2-b]indazoles 5, 7 have been prepared from 3-trifluoromethyl-5-phenyl-2,6-dicyano anilines 1
via novel indazole regioisomers 3 and 4 through a facile strategy. Specific examples were evaluated for anticancer activity in vitro
and found to exhibit promising activity against A-549 cell lines and are more effective than Etoposide. QSAR models were developed
and validated by cross-validation method. The results of the best QSAR model were further compared with the crystal structure of
tubulin protein. The binding energies estimated were found to have a good correlation with the experimental inhibitory potencies.
Ó 2007 Elsevier Ltd. All rights reserved.
The modern trend is directed towards discovery of new
organic molecules as potential anticancer agents by
adopting various synthetic approaches. In this process
some of the molecules considered to interfere effectively
with DNA either directly or inhibiting DNA–binding
enzymes led to identification of new promising antican-
cer agents.1 However, these molecules lack specificity to-
wards cytotoxicity and damage cell lines in the tissue as
a result no clear-cut drug is available. The principal
driving force for stacking and charge transfer interac-
tions is through hydrogen bonding and electrostatic
forces.2 In addition, strategically positioned suitable
substituents promote interference with cellular detoxifi-
cation pathways. Therefore, our attention was attracted
towards synthesis of a series of novel pyrimido[1,2-
b]indazoles with trifluoromethyl group at an appropriate
position in order to identify suitable lead compounds as
new pyrimido[1,2-b]indazoles as anticancer agents. Ear-
lier findings have been on indazole derivatives specifi-
cally known to be active as protein kinase inhibitors,
in cancer cell proliferative disorders, Alzheimer’s dis-
ease, viral infections, auto-immune diseases and neuro
degenerative disorders.3–6 Further, pyrimidine ring in
an organic molecule also shows prominent activity
against several diseases.7–9 Therefore interest is continu-
ously increasing on fusion of pyrimidine ring over indaz-
oles as a result pyrimido[1,2-b]indazoles are formed and
are considered to have promising activity against many
infections. Synthesis of pyrimido[1,2-b]indazoles is
mainly starting from 3-amino indazoles and their reac-
tion with 1,3-diketones,10–12 b-ketoester,13 propiolic acid
ester,14 DEEM15 or DMAD.16 In recent past, synthesis
based on microwave irradiation conditions17,18 is con-
sidered as a powerful synthetic tool due to its short reac-
tion times, operational simplicity with improved yields
and is of modern trend in organic synthesis. In continu-
ation of our efforts19–22 on the synthesis of novel ring
systems of biological interest, we wish to report the syn-
thesis of new indazole regioisomers 3, 4 and each isomer
is independently reacted with various substrates such as
1,3-diketones (symmetrical/unsymmetrical), diethyleth-
Keywords: Regioisomers; Indazoles; 1,3-Diketones; Pyrimido[1,2-b]in-
dazoles; Cyclisation; A-549 cell lines; QSAR modelling studies;
Tubuline.
q
IICT Communication No. 041004.
Corresponding author. Fax: +91 40 27160387x27193185; e-mail:
narsaiah@iict.res.in
*
0960-894X/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2007.03.087

3446
T. Yakaiah et al. / Bioorg. Med. Chem. Lett. 17 (2007) 3445–3453
oxymethylene malonoate (DEEM), ethoxymethylene
malononitrile (MMN), ethyl ethoxymethylene cyanoac-
etate (EMCA), diethyl N,N-dimethylaminomethylene
malonate (DAMM), b-ketoesters and dimethyl acetylen-
edicarboxylate (DMAD) under microwave irradiation
conditions obtaining pyrimido[1,2-b]indazoles 5, 7 in
high yields. Representative examples were screened for
anticancer activity against A-549 cell lines and found
to have more activity than Etoposide, a standard drug.
Molecular modelling studies further confirmed the
activity.
signals appeared in upfield compared to compound 4
and it is assumed to be the influence of CF3 and CN
groups. If CF3 and CN are para to each other, the
electron-withdrawing effect in ring nullifies as a result
the signals appeared in upfield as seen in compound
3. In case CF3 and CN are ortho to each, the cumula-
tive electron-withdrawing effect enhances and with that,
the signals for NH2 and C-5 proton in 1H NMR
appeared in downfield in compound 4. Further, phenyl
protons appeared as singlet due to magnetic equiva-
lence when CF3 and CN are para to each other and
multiplet if ortho to each other.
The 3-trifluoromethyl-5-phenyl-2,6-dicyano aniline 1
has three active functional groups ortho to each and
is subjected to diazotisation using NaNO2/HCl at
0 °C followed by potassium iodide obtaining corre-
sponding iodobenzene 2.21 It is further reacted with
hydrazine hydrate in refluxing ethanol resulting in
two 3-amino-4/6-trifluoromethyl-6/4-phenyl indazole
regioisomers 3, 4 in definite proportions. The reaction
is mainly nucleophilic substitution of iodine by hydra-
zine on benzene ring due to presence of powerful elec-
tron-withdrawing groups followed by selective attack
on one of the nitrile carbons at a time to result in
products 3, 4. The reaction is schematically drawn
below in Scheme 1.
The 13C NMR data of compounds 3 and 4 further sup-
port that the CF3 carbon in compounds 3 appeared as
quartet at d 121.31 and in compound 4 as quartet at d
122.76 with the same coupling constant 274 Hz. How-
ever, a characteristic difference of absorption of C-CF3
is observed. In compound 3 the C-CF3 appeared as
quartet at d 123.29 with coupling constant 31 Hz,
whereas in compound 4 the C-CF3 appeared as quartet
at d 130.16 with coupling constant 31 Hz. In addition ni-
trile carbon (CN) appeared at 126.99 for compound 3
and 113.16 for compound 4. Thus, the structure of each
isomer is determined. It is further confirmed by single
crystal X-ray analysis of compound 3 and presented in
Figure 1. CCDC 270847 contains supplementary crys-
tallographic data for the structure 3 (see Figs. 2 and 3).
Each of indazole regioisomers 3 and 4 is separated due
to their small difference in polarity and characterised
1
based on IR, H NMR, 13C NMR, 19F NMR (Table
Reaction of compounds 3 and 4 with 1,3-diketones. The
regioisomers 3 and 4 are independently reacted with
1,3-diketones (symmetrical and unsymmetrical) in a
sealed tube under microwave irradiation conditions
obtaining pyrimido[1,2-b]indazoles 5 in single step.
The sequence of reaction is initially nucleophilic attack
1), mass spectral data and X-ray analysis. Compound
3 is less polar than compound 4 based on TLC and
characteristic difference in chemical shifts of NH2
group and proton on C-5 carbon is clearly observed
in 1H NMR. In compound 3, NH2 and C-5 proton
CF3
CF3
CN
CF3
CN
CN
+
CN
I
KI
N2H4. H2O / EtOH
reflux, 2hrs
NaNO2 / HCl
CH3CN
0oC
_
N2Cl
Ph
NH2
Ph
Ph
CN
1
CN
2
Ph
CF3
CN
CF3
CF3
NH2
N
NH2
CN
..
NH2
CN
N
N
+
N
F3C
Ph
Ph
N
Ph
NH
CN NH
H
H
H
CN
CN
..
2
4 (45 %)
3 ( 55 %)
Scheme 1.
Table 1. 19F NMR (CDCl3) in ppm data for compounds 5d, 5f and 5k, 5m, 5n
3
4
5d
5f
5k
5m
5n
3-CF3
4-CF3
5-CF3
6-CF3
8-CF3
10-CF3
—
—
—
58.78
—
—
58.46
—
—
—
—
—
—
—
—
55.04
—
—
—
60.90
—
—
61.10
—
—
58.07
—
—
55.37
—
—
—
66.43
—
—
66.95
—
—
—
67.09
67.52
66.42

T. Yakaiah et al. / Bioorg. Med. Chem. Lett. 17 (2007) 3445–3453
3447
on one of the carbonyl carbons followed by cyclisation
onto other carbonyl carbon to form products 5. In case
of reaction of unsymmetrical 1,3-diketones like 1,1,1-tri-
fluoromethyl-2, 4-pentanedione, 1-phenyl-2, 4-butane-
dione, 4-phenyl-1, 1,1-trifluoro-2, 4-butanedione or
4-(4-methyl phenyl)-1,1,1-trifluoro-2, 4-butanedione
with compounds 3/4 it resulted exclusively in one isomer
5d, 5e, 5f or 5g/5k, 5l, 5m, or 5n, respectively. It is
known23,24 that the carbonyl attached to CF3 group in
1,3-diketones is prone to enolise as a result the carbonyl
remote from the CF3 preferentially reacts first, gives
exclusively one isomer and it is comparable with our ear-
lier experience.25 In addition, the mode of nucleophilic
attack of regioisomers 3 and 4 on unsymmetrical 1,3-
diketones is in two ways. In regioisomer 3 the primary
amine is more reactive than tertiary nitrogen, whereas
in regioisomer 4, primary amine is less reactive than ter-
tiary nitrogen. This is mainly due to their strategic posi-
tion with respect to trifluoromethyl group in benzene
ring. The reaction is schematically drawn below in
Scheme 2.
CF3
CN
δ 4.30
NH2)
N
(H
δ 7.18
N
Ph
H
Chemical shift (up field)
Figure 1.
Ph
δ 4.68
NH2)
N
δ 7.68 (H
N
F3C
H
CN
Chemical shift (down field)
Figure 2.
In order to characterise a specific product from the reac-
tion of unsymmetrical 1,3-diketones with regioisomers 3
and 4, 19F data are generated and tabulated in Table 1.
Reaction of compounds 3 and 4 with DEEM/EMMN/
EMCA/DAMM. The compounds 3 and 4 are indepen-
dently reacted with DEEM/EMMN/EMCA/DAMM in
sealed tube under microwave irradiation conditions with
450 W power and obtained pyrimido[1,2-b]indazoles 7.
In case of DEEM and DAMM an uncyclised intermedi-
ate 6a is formed from compound 3 and 6b from
Figure 3. X-ray crystal structure of 3. Displacement ellipsoids are
drawn at 30% probability level and H atoms are shown as small
spheres of arbitrary radii.
..
..
NH2
CF3
NH2
Ph
H
H
N
N
N
N
H
Ph
F3C
H
CN
CN
3
4
R
CF3
CF3
N
NH2
N
R
sealed tube
MW / 450W
O
O
N
+
R'
CH3COOH
90 Sec.
N
N
Ph
Ph
H
R'
CN
CN
5
3
5a) R = R' = CH3, 5b) R = R'= CF3, 5c) R =R'= C6H5, 5d) R = CH3, R' = CF3,
5e) R = CH3, R' = C6H5, 5f) R = C6H5, R' = CF3, 5g) R = p-CH3C6H5, R' = CF3.
R
Ph
Ph
NH2
N
N
R
sealed tube
MW / 450W
O
O
N
+
R'
CH3COOH
90 Sec.
N
N
F3C
F3C
H
R'
CN
CN
5
4
5h) R = R' = CH3, 5i) R = R'= CF3, 5j) R =R' = C6H5, 5k) R = CF3, R' = CH3,
5l) R = C6H5 ,R'= CH3, 5m) R = CF3, R' = C6H5, 5n) R = CF3, R' = p-CH3C6H5.
Scheme 2.

3448
T. Yakaiah et al. / Bioorg. Med. Chem. Lett. 17 (2007) 3445–3453
R
R
NH2
N
NH
COOEt
COOEt
sealed tube
MW / 450W
"R
H
COOEt
COOEt
N
+
N
N
CH3COOH
60 sec
'R
'R
H
H
CN
CN
R" = OEt, N(CH3)2
6a / 6b
3 / 4
6a) R = CF3, R' = Ph
3) R = CF3, R'= C6H5
6b) R = Ph, R' = CF3
4) R = C6H5, R' = CF3
R
R
NH
N
COOEt
COOEt
sealed tube
COOEt
MW / 500W
N
N
CH3COOH
120 sec
OH
N
N
'R
'R
H
CN
CN
6a / 6b
7a / 7b
6a) R = CF3, R' = Ph
7a) R = CF3, R' = Ph
6b) R = Ph, R' = CF3
7b) R = Ph, R' = CF3
R
R
NH2
EtO
N
R"
NH2
sealed tube
MW / 450W
CN
R"
N
+
N
N
H
N
CH3COOH
60 sec
'R
'R
H
CN
CN
7
3 / 4
R" = CN, COOEt
R"= CN, COOEt
3) R = CF3, R'= C6H5
4) R = C6H5, R'= CF3
7c) R = CF3, R'= C6H5, R" = CN, 7d) R = CF3,R'=C6H5, R"= COOEt, 7e) R = C6H5, R' = CF3 R" = CN,
7f) R = C6H5, R'= CF3 , R" = COOEt.
Scheme 3.
R"
R
R
NH2
N
HN
MeO2C
H3C
CO2Me /150 sec
OEt
N
+
O
N
N
'R
'R
H
/ 120 sec
CN
CN
O
O
3 / 4
7
MW / 450W,CH3COOH
sealed tube
3) R = CF3, R'= C6H5
4) R = C6H5, R' = CF3
7g) R = CF3, R'= C6H5, R"= CH3, 7h) R = CF3, R'= C6H5, R" = CH3,
7i) R = C6H5, R' = CF3, R" = CO2CH3, 7j) R = C6H5, R' = CF3, R" = CO2CH3,
Scheme 4.
compound 4 which are further cyclised to give products
7a and 7b under microwave irradiation conditions with
450 W power, respectively. It is attributed to the nucle-
ophilic nature of nitrile carbon (CN) versus ester car-
bonyl (CO2Et) and ester carbonyl is considered to be
sluggish. The 1H NMR of compounds 7d and 7f is quite
interesting in a way that the NH2 protons in each of
compounds appeared as two broad singlets with differ-
ent chemical shifts. This is due to extended conjugation
with ester carbonyl as a result proton became non-
equivalent. The reaction is schematically drawn below
in Scheme 3.
Reaction of compounds 3 and 4 with b-ketoesters/
DMAD. The compounds 3 and 4 are similarly reacted
with b-ketoesters/DMAD and respective products
obtained. In case of b-ketoester the reaction is initi-
ated by nucleophilic attack on carbonyl carbon fol-
lowed by cyclisation, whereas with DMAD, the
reaction is in addition followed by cyclisation. The
reaction is schematically drawn below in Scheme 4.
Representative examples of pyrimido[1,2-b]indazole
regioisomers with trifluoromethyl group at an appropri-
ate position were synthesized and representative exam-

T. Yakaiah et al. / Bioorg. Med. Chem. Lett. 17 (2007) 3445–3453
3449
Table 2. Anticancer activity profile
R3
6
5
R4
8
N
4
7
R2
9
3
N
2
R1
N
R5
10
11
CN
1
Compound
R1
R2
R3
R4
R5
% inhibitiona
7h
5a
5b
5l
@O
H
CH3
CH3
CF3
–Ph
–Ph
CF3
CF3
–Ph
CF3
CF3
–Ph
–Ph
–Ph
–Ph
–Ph
–Ph
–Ph
–Ph
CF3
–Ph
CF3
CF3
—
CF3
–Ph
–Ph
CF3
CH3
CF3
CF3
CF3
CF3
CF3
CF3
–Ph
CF3
–Ph
–Ph
—
0.44
3.74
CH3
CF3
CF3
–Ph
CF3
CH3
CF3
OH
CF3
CF3
NH2
NH2
CH3
CF3
—
H
H
4.54
4.80
H
5j
H
5.96
6.44
5k
5h
5n
7b
5m
5i
H
H
6.71
6.77
H
COOEt
p-CH3C6H4
H
8.24
8.90
H
–Ph
CF3
H
H
10.23
10.70
11.00
13.33
16.20
1.15
7d
7e
5e
5f
COOEt
COOEt
H
H
–Ph
–Ph
—
H
—
Std.
a All the pharmacological studies were performed in duplicate at 1 nM/ml (1 nM compound in 1 ml of 1% DMSO in PBS). % inhibition of cell growth
is measured based on absorbance at 550 nm by transferring plate to plate reader. Std. Etoposide.
ples were evaluated for anticancer activity against A-549
cell lines by MTT assay method (Table 2).26 The
compound 5f with trifluoromethyl group on C-3 carbon
showed highest activity, whereas compound 7h with
hydroxyl group on C-3 carbon showed lowest activity.
The compound inhibitory activity falls between 0.44%
(7h) and 16.2% (5f). Based on the activity profile, it is
broadly concluded that the presence of trifluoromethyl
group in pyrimidine ring in combination of phenyl or
additional trifluoromethyl group promotes high activity.
Alternatively, the presence of activating groups like
–OH, –NH2 may also increase the activity as shown in
compounds 7b, 7d and 7e. However, the stability of
the molecules is less due to the presence of ester group,
which may prone to metabolize into acid. The activity
data of these molecules are compared with those of Eto-
poside, a standard drug. Almost all the molecules
showed greater inhibition than the standard Etoposide.
The details of activity profile are tabulated in Table 2.
Cancer cell survival and viability in vitro were evaluated
using the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltet-
Table 3. Activity and physicochemical parameters of indazole derivatives
Compound
Activity
logP
HOMO In ev
LUMO In ev
MR
95.7375
5a
5b
5e
5f
1.7282
1.6440
1.1762
1.0915
1.4743
1.2912
1.5258
1.4921
1.6198
1.3516
1.4704
1.3851
1.2716
1.2596
2.6576
2.2403
4.4776
6.7002
6.3905
7.5018
5.5889
6.7002
7.8878
6.7002
6.3905
7.5018
7.9690
5.0915
4.5927
4.5927
5.1115
2.6690
ꢀ9.58
ꢀ8.49
ꢀ8.92
ꢀ8.69
ꢀ9.07
ꢀ8.47
ꢀ8.75
ꢀ8.83
ꢀ3.33
ꢀ8.7
ꢀ3.33
ꢀ4.58
ꢀ3.75
ꢀ3.79
ꢀ3.69
ꢀ3.80
ꢀ3.88
ꢀ3.77
ꢀ2.84
ꢀ3.72
ꢀ3.71
ꢀ4.22
ꢀ3.94
ꢀ4.01
ꢀ3.69
—
97.7579
115.9102
116.9204
96.7477
5h
5i
97.7579
135.1504
97.7579
5j
5k
5l
115.9102
116.9204
121.9616
103.5715
106.5778
106.5778
93.7935
5m
5n
7b
7d
7e
a7h
Std.
ꢀ8.69
ꢀ9.23
ꢀ9.20
ꢀ9.17
ꢀ10.25
55.1636
aConsidered as outlier.
Std. Etoposide.

3450
T. Yakaiah et al. / Bioorg. Med. Chem. Lett. 17 (2007) 3445–3453
Table 4. Model equations generated for the indazole analogues
Equation
1
Modelled Eq.
n
R
R2
SEE
F
Act = ꢀ0.013(0.005)MR + 2.898(0.531)
16
0.569
0.323
0.341
6.693
Q = 1.668; PRESS = 15.193; q2cv ¼ ꢀ0:089
2
3
4
5
6
Act = ꢀ0.123(0.064)logP + 2.278(0.397)
16
16
16
15
15
0.454
0.597
0.492
0.722
0.820
0.206
0.356
0.242
0.522
0.672
0.369
0.345
0.374
0.207
0.171
3.628
3.597
2.081
6.549
12.300
Q = 1.230; PRESS = 11.819; q2cv ¼ 0:152
Act = 0.000137(0.000)MR2 ꢀ 0.0392(0.033)MR + 4.103(1.575)
Q = 1.730; PRESS = 15.193; q2cv ¼ 0:326
Act = 0.013(0.040)logP2 ꢀ 0.475(0.419)logP + 3.200(1.230)
Q = 1.315; PRESS = 15.193; q2cv ¼ 0:343
Act = 0.0578(0.022)logP2 ꢀ 0.740(0.254)logP + 3.704(0.688)
Q = 3.487; PRESS = 0.578; q2cv ¼ 0:527
Act = 0.00024(0.000)MR2 ꢀ 0.0563(0.016)MR + 4.649(0.789)
Q = 4.795; PRESS = 0.358; q2cv ¼ 0:669
n, number of data points; SEE, standard error estimate; Q, quality factor; PRESS, predictive sum of squares; q2cv, cross-validated.
razolium bromide (MTT, Sigma Chemical Co., St.
Louis, MO) assay developed by Mosmann et al., with
some modifications.
With the aim of rationalizing the activity data obtained
for the Indazole derivatives, a molecular modelling
study was carried out in order to investigate the possi-
ble interaction of such compounds with tubulin pro-
tein. The stability of these molecules is derived from
logP,27 MR,28 HOMO, LUMO values. Further QSAR
studies were carried out on a series of indazole deriva-
tives in order to provide further insight into the key
structural features required to design potential drug
candidates of this class. Molecular docking methodolo-
gies ultimately seek to predict the best mode by which
a compound will fit into a binding site of a macromo-
lecular target. In addition to the synthetic work, an att-
emptive explore of the QSAR and docking studies of
indazole derivatives has been made to explain the
observed variance in biological activity as a function
of various molecular descriptors. This predicts the best
drug candidate providing an insight into substitutional
and configurational requirements for optimum receptor
pit, which leads to the development of best pharmaco-
phore activity.29–31
Test compounds are dissolved in 1 ml of DMSO. Cells
were seeded in 96-well plate (10,000 cell/well) and left
overnight in CO2 incubator (at 5%) to recover from han-
dling stress. Test compounds are added and incubated
for 12 h. A 5 mg/ml MTT solution was prepared by dis-
solving in PBS (Phosphate buffer solution) and filter-
sterilized. Added 20 ll of MTT solution from stock to
each well containing cells and incubated the plate for
another 5 h in a CO2 incubator at 37 °C. Media were
removed with needle and syringe. Added 200 ll of
DMSO to each well to dissolve separated crystals of
MTT. Kept the plate into incubator at 37 °C for
5 min. The % inhibition of cell growth is measured based
on absorbance at 550 nm by transferring the plate-to-
plate reader (This procedure is specific for adherent cells
in 96-well plates.)
Series1
Series2
Series3
Docking was carried out against tubulin protein (pdb
Code: 1SA1), which is the possible anticancer target of
indazole derivatives. Tubulin protein was chosen as a
target for docking (using Openeye software,32) because
of a relatively high tanimoto (2D structure finger print).
Similarly index33,34 of tubulin selected target protein was
retrieved from Protein data bank.35
2.5
2
1.5
1
0.5
0
The activity data log (1/IC50) and physicochemical prop-
erties logP, HOMO, LUMO and molar refractivity (MR)
of the indazole derivatives are presented in Table 3. These
data were subjected to regression analysis and the
corresponding model equations were generated using
SPSS 10.0 software36 and are shown in Table 4.
1
2
3
4
5
6
7
8
9 10 11 12 13 14 15
Data Points
Figure 4. Comaparison of observed activity versus predicted activities.
Series 1: observed activity; series 2: predicted activity (Eq. 5); series 3:
predicted activity (Eq. 6).
Table 5. Docking functions and QSAR descriptors for analysis32,36–38
Descriptor code
Description
SG
CG
Shapegauss (represents all the atoms as smooth Gaussian functions)-docking function score from Fred program
ChemGauss (combines Shapegaussian function with additional potentials between chemically complimentary
groups)-docking function score from Fred program
PLP
logP
MR
PLP (piecewise linear potential)-docking function score from Fred program
Logarithm of partitioning coefficient (calculated by clogP method)
Molar refractivity the measure of steric factor, bulkiness of the molecule

T. Yakaiah et al. / Bioorg. Med. Chem. Lett. 17 (2007) 3445–3453
3451
Figure 5. (a) Receptor cavity surface area. Amino acid in starmark. (b) Active site (amino acid) surface area.
Eqs. 1 and 2 show very lower correlation coefficients
in terms of R2 and Q values with respect to MR
and logP, respectively. Further improvement, a qua-
dratic term was included and Eqs. 3 and 4 were
obtained which is statistically valid but with lower R
and Q values (Table 4). Eqs. 5 and 6 (Table 4), which
were resulted by taking an outlier (compound 7h) on
the basis of their residual values, show a high statisti-
cal significance.
The cross-validation parameters also reveal that Eqs.
5 and 6 were the best regression models having the
higher value of Q, R2, qc2v and low SEE values. From
the graph (Fig. 4), a good agreement is observed by
comparing the observed and predicted activities
(derived from the best modelled equations) to the data
points.
Figure 7. Compounds 5j in purple and 5l in grey colour representation
in the active site region and their surface overlap.
Table 6. Scoring functions
Compound Chemgauss Clash PLP
Screenscore SHAPE
5a
5b
5e
5f
ꢀ61.1
26.92 ꢀ40.25 ꢀ121.67
ꢀ471.23
ꢀ458.93
ꢀ501.99
ꢀ523.52
ꢀ432.53
ꢀ432.7
ꢀ67.57
ꢀ65.48
ꢀ74.19
ꢀ62.82
ꢀ61.3
26.82 ꢀ28.05 ꢀ116.68
33.25 ꢀ26.42 ꢀ104.55
35.03 ꢀ32.12 ꢀ124.1
35.77 ꢀ28.51 ꢀ113.74
5h
5i
56.15
3.88 ꢀ57.54
5j
ꢀ67.84
ꢀ69.25
ꢀ74.56
ꢀ72.74
ꢀ75.27
ꢀ61.26
ꢀ71.86
ꢀ63.21
ꢀ63.64
ꢀ40.19
69.50 ꢀ17.56 ꢀ100.12
37.66 ꢀ17.2 ꢀ112.05
43.69 ꢀ33.84 ꢀ136.13
ꢀ476.65
ꢀ470.01
ꢀ524.37
ꢀ511.69
ꢀ519.31
ꢀ465.73
ꢀ537.06
ꢀ459.5
5k
5l
5m
5n
7b
7d
7e
7h
Std.
60.24
60.46
17.11 ꢀ55.41
2.16 ꢀ83.41
30.66 ꢀ32.88 ꢀ112.25
29.57 ꢀ48.03 ꢀ129.65
30.07 ꢀ30.66 ꢀ108.99
44.84 ꢀ41.95 ꢀ127.14
21.52 ꢀ32.64 ꢀ90.06
ꢀ448.56
ꢀ284.26
Figure 6. Active site region with amino acids.

3452
T. Yakaiah et al. / Bioorg. Med. Chem. Lett. 17 (2007) 3445–3453
˚
Figure 8. (a) 4 A grid model along with the best molecular surface area (5l) in pink colour, protein in ribbon model. (b) Ball and stick model of the
best molecule (5l) in the receptor pocket. Surface area of the receptor pocket is in red colour.
Since these are second-degree equations in logP and
MR, an optimum logP and MR that corresponds to
give maximum activity can be evaluated by finding the
maxima. Eqs. 5 and 6 were partially differentiated with
respect to logP and MR, respectively, the first derivative
was set to zero and a value of logPopt = 6.401 and
MRopt = 116.66 was obtained. Therefore, the analogue
with around 6.401 value of logP and around 116.66
value of MR is expected to show maximum activity.
thons for building up of additional ring onto indaz-
oles. Thus, it is demonstrated that pyrimidine ring
can be fused onto indazoles with a variety of sub-
strates and obtained pyrimido[1,2-b]indazoles under
microwave irradiation conditions. A set of novel inda-
zole derivatives have been synthesized and screened
for anticancer activity by MTT assay method against
A-549 cell lines. All the molecules showed higher
activity than the standard Etoposide. The optimum
lipophilic and bulkiness requirements are important
to fit in the active site of the target and to have more
hydrophobic affinity. The drug candidate with opti-
mum value of logP (6.401) and MR (116.66) shows
high activity. QSAR analysis and molecular docking
studies were compared for the investigated com-
pounds. The binding energies estimated by scoring
functions were found to have a good correlation with
the experimental inhibitory potencies. Based on the
binding conformations from molecular docking, it is
concluded that the presence of bulkier substituent
(Ph) at R3 position, in combination of CF3 group in
pyrimidine ring, promotes high activity.
Docking functions37–39 and QSAR descriptors for anal-
ysis are given in Table 5. The predicted binding energies
of Indazole analogues have shown good correlation with
the respective inhibitory activities, that is, 5l has lowest
scoring values and higher inhibitory activity than stan-
dard reference molecule (std) and other analogues.
To visualize the binding conformations of these ana-
logues within the active site of tubulin protein are dis-
˚
played in Figure 5. In the active site region (4 A) the
tubulin protein nine amino acids that can play an impor-
tant role are shown in Figure 4 (viz., Gln 11, Ala 12, Glu
71, Ser 140, Gly 142, Thr 145, Val 177, Asn 206 and Asn
228). As observed in Figure 5, 5l and 5j were docked in
the active site of protein with a significant different bind-
ing mode compared with standard molecule (see Figs. 6
and 7).
Acknowledgments
The authors are thankful to Dr. J.S. Yadav, Director,
IICT, and Shri S. Narayan Reddy, Head, Fluoroorganic
division, IICT, Hyderabad, for their constant encour-
agement and financial support from industry-sponsored
project. The author is thankful to the Head, Department
of Chemistry, Nizam College, for providing the facilities
for the work. We also thank Director Dr. N.V. Naidu
(Global Institute of Biotechnology, Himayatnagar,
Hyderabad) for providing the Software to complete
the work.
Furthermore, differences in the binding mode were ob-
served for 5l and 5j because of the presence of three phe-
nyl groups in 5j, increases in the bulkiness of the ligand
resulting in highest clash penalty with the active site of
the receptor amino acid (AA) (Table 6). Large substitu-
tion at R1 position of indazole analogues may lead to
collision with the residues Ser 178, Thr 179 and Ala
180 in the receptor site.
To further validate the binding mode best scored molecule
in the grid as well as surface form is given in Figure 8.
References and notes
In conclusion, the indazole regioisomers are prepared
by convenient route and are considered as active syn-
1. Rahman, A. U.; Malik, S. Tetrahedron Lett. 1985, 26,
2759.

T. Yakaiah et al. / Bioorg. Med. Chem. Lett. 17 (2007) 3445–3453
3453
2. Rahman, A. U.; Malik, S.; Sadiq, H. S.; Iqbal, C. M.;
Chao-zhou, N.; Jon, C.; Nigellidine, F. Tetrahedron Lett.
1995, 36, 1993.
3. Conception, N.; Patrick, M.; Thompson, F.; Fabienne, M.
T.; Eric, B.; Sylvie, W.; Cecile, C. PCT Int. Appl. WO 03
11, 833, 2003, 188 pp. (Fr) (C.A. 138: 153529x).
4. Amici, R. D.; Matteo A. K. M.; Barbara S.; Anna Vetti.
PCT Int. Appl. WO 03 28, 720, 2003, 132 pp. (Eng) (C.A.
138: 304278t).
5. Renhowe, P. A.; Shafer, C. M.; McBridge, C.; Silver, J.;
Pecchi, S.; Machajewski, T.; Mccrea, B.; Poon, D.;
Thomas, T. PCT Int. Appl. WO 03 4, 488, 2003, US
(C.A. 138: 106699t).
6. Molina, P.; Arques, A.; Victoria, M.; Vinader, V. Tetra-
hedron Lett. 1989, 30, 6237.
7. Griffin, R. J.; Lowe, P. R. J. Chem. Soc. Perkin Trans.
1992, 1, 1811.
8. Whilock, B. J.; Lipton, S. H.; Strong, F. M. J. Org. Chem.
1965, 115.
9. Schwan, T. J.; Howard Tieckemann, J. J. Med. Chem.
1965, 750.
10. Domergue, A.; Sureau, F.; Robert, F. M.; Georges, A.
(Ugine kuhlmann). Fr. 2, 192, 559, Appl. 72, 25, 693, 1974,
7 pp (C.A. 81: 152260V).
487; (d) Yadav, J. S.; Reddy, B. V. S. New J. Chem. 2000,
24, 489.
19. Venkat Reddy, G.; Rama Rao, V. V. V. N. S.; Maitraie,
D.; Ravikanth, S.; Yadla, R.; Reddy, S. N.; Narsaiah, B.;
Shanthan Rao, P. J. Fluorine Chem. 2003, 124, 203.
20. Rama Rao, V. V. V. N. S.; Venkat Reddy, G.; Maitraie,
D.; Ravikanth, S.; Yadla, R.; Narsaiah, B.; Shanthan
Rao, P. Tetrahedron 2004, 60, 12231.
21. Maitraie, D.; Venkat Reddy, G.; Rama Rao, V. V. V. N.
S.; Ravikanth, S.; Narsaiah, B.; Shanthan Rao, P.;
Ravikumar, K.; Sridhar, B. Tetrahedron 2005, 61, 3999.
22. Ravikanth, S.; Venkat Reddy, G.; Maitraie, D.; Rama
Rao, V. V. V. N. S.; Shanthan Rao, P.; Narsaiah, B. Syn.
Commun. 2004, 34, 4463.
23. Singh, S. P.; Kapoor, J. K.; Kumar, D.; Threadgil, M. D.
J. Fluorine Chem. 1997, 83, 73.
24. Joseph, P. K.; Shanmugasundaram, P.; Ramakrishnan, V.
T. Indian J. Chem. 1992, 31, 563.
25. Krishnaiah, A.; Narsaiah, B. J. Fluorine Chem. 2001, 109,
183.
26. Mosmann, T. J. Immunol. Methods 1983, 65, 55.
27. Bodor, N.; Buchwald, P. J. Phys. Chem. B 1997, 101, 3404.
28. Viswanadhan, V. N.; Ghose, A. K.; Revankar, G. R.;
Robins, R. K. J. Chem. Inf. Comput. Sci. 1989, 29, 163.
29. Rameshwar, N.; Krishna, K.; Ashok Kumar, B.; Partha-
sarathy, T. Bioorg. Med. Chem. 2006, 14, 319.
30. Ravi, K. A.; Ashok, K. B.; Parthasarathy, T.; Uma, V.
Bioorg. Med. Chem. 2005, 13, 1873.
11. Yu, V. M.; Chuiduk, V. A. Khim. Geterotsikl, Soedin 1974,
984 (CA: 81. 120560b).
12. Chuiguk, V. A.; Gladkaya, V. A. Ukr Khim. Zh. (Russ.
Ed). 1981, 47, 966 (C.A. 97: 23724f).
13. Salvatore, P. V. S.; Marino, L. M. J. Heteroatom Chem.
1973, 10, 261.
31. Sirisha, M. A.; Ashok, K. B.; Parthasarathy, T.; Uma, V.
Centl. Eur. J. Chem. 2004, 4, 696.
14. Hans, R.; Peiren, M. A.; Roert, M. Chem. Ber. 1972, 105,
794.
15. Dong, L. H. P.; Claude, C. Ger. Offen. 1978, 2, 822, 124.
(C.A. 90: 152181f).
16. Polanc, S.; Vestn, C.; Slov, C. Kem. Drug. 1984, 31, 115
(CA: 101. 191831h).
17. (a) Abramovitch, R. A. Org. Prep. Proced. Int. 1991, 23,
683; (b) Caddick, S. Tetrahedron 1995, 51, 10403; (c)
Loupy, A.; Petit, A.; Hamelin, J.; Texier-Boultet, F.;
Jacquault, P.; Mathe, D. Synthesis 1998, 1213; (d) Varma,
R. S. Green Chem. 1999, 1, 43.
18. (a) Yadav, J. S.; Reddy, B. V. S. Tetrahedron Lett. 2002,
43, 1905; (b) Yadav, J. S.; Reddy, B. V. S.; Madan, Ch.;
Kumar, G. M. Synlett 2001, 1781; (c) Yadav, J. S.; Reddy,
B. V. S.; Rasheed, M. A.; Kumar, H. M. S. Synlett 2000,
32. Fred 2.1, Omega 2.0, Openeye Scientific Software, Inc.,
3600 Cerrillos Road, Suite 1107, Santa Fe, NM 87507,
USA. www.eyesopen.com.
33. J. Chem., ChemAxon, ChemAxon Ltd, Maramaros koz 3/
a, Budapest 1037, Hungary.
34. Willett, P.; Barnard, J. M.; Downs, G. M. J. Chem. Inf.
Comput. Sci. 1998, 38, 983.
35. www.rcsb.orgnpdb.
36. McGann, M. R.; Almond, H. R.; Nicholls, A.; Grant, J.
A.; Brown, F. K. Biopolymers 2003, 68, 76.
37. Stahl, M.; Rarey, M. J. Med. Chem. 2001, 44, 1035.
38. Gehlhaar, D. K.; Verkhivker, G. M.; Rejto, P. A.;
Sherman, C. J.; Fogel, D. B.; Fogel, L. J.; Freer, S. T.
Chem. Biol. 1995, 2, 317.
39. SPSS software. Consult http://www.spss.com.