Design, synthesis and 3D-QSAR analysis of novel 2-hydrazinyl-4- morpholinothieno[3,2-d]pyrimidine derivatives as potential antitumor agents

Article abstract of DOI:10.1016/j.ejmech.2012.09.002

A series of 2-hydrazinyl-4-morpholinothieno[3,2-d]pyrimidine derivatives were synthesized and evaluated for their cytotoxic activities against five cancer cell lines. Most of them exhibited moderate to significant cytotoxic activities and high-selectivity against one or more cell lines, and nearly all of them had higher potency than positive controls against MDA-MB-231 cell line. The most promising compound 15f showed strong cytotoxic activities against H460, HT-29 and MDA-MB-231 cell lines, which were 1.7- to 66.5-folds more active than 2-(1H-Indazol-4-yl)-6-((4-(methylsulfonyl)-1-piperazinyl)methyl)-4-(4- morpholinyl)thieno[3,2-d]pyrimidine(GDC-0941). To investigate the SARs of thieno[3,2-d]pyrimidine derivatives in more details, CoMFA (q² = 0.436, r² = 0.937) and CoMSIA (q² = 0.706, r² = 0.947) models on H460 cell line were established. The generated 3D-QSAR models can be used for further rational design of novel thienopyrimidines as highly potent and selective cytotoxic agents.

Full text of DOI:10.1016/j.ejmech.2012.09.002

European Journal of Medicinal Chemistry 57 (2012) 162e175
Contents lists available at SciVerse ScienceDirect
European Journal of Medicinal Chemistry
journal homepage: http://www.elsevier.com/locate/ejmech
Original article
Design, synthesis and 3D-QSAR analysis of novel 2-hydrazinyl-4-
morpholinothieno[3,2-d]pyrimidine derivatives as potential antitumor agents
,
Wufu Zhu a b, Yajing Liu a, Xin Zhai a, Xiao Wang a, Yan Zhu a, Di Wu a, Hongyu Zhou a, Ping Gong a,
,
Yanfang Zhao a
*
a Key Laboratory of Original New Drugs Design and Discovery of Ministry of Education, Shenyang Pharmaceutical University, 103 Wenhua Road, Shenhe District, Shenyang 110016,
PR China
b School of Pharmacy, Jiangxi Science and Technology Normal University, Nanchang, Jiangxi 330013, China
a r t i c l e i n f o
a b s t r a c t
Article history:
A series of 2-hydrazinyl-4-morpholinothieno[3,2-d]pyrimidine derivatives were synthesized and eval-
uated for their cytotoxic activities against five cancer cell lines. Most of them exhibited moderate to
significant cytotoxic activities and high-selectivity against one or more cell lines, and nearly all of them
had higher potency than positive controls against MDA-MB-231 cell line. The most promising compound
15f showed strong cytotoxic activities against H460, HT-29 and MDA-MB-231 cell lines, which were 1.7-
to 66.5-folds more active than 2-(1H-Indazol-4-yl)-6-((4-(methylsulfonyl)-1-piperazinyl)methyl)-4-(4-
morpholinyl)thieno[3,2-d]pyrimidine(GDC-0941). To investigate the SARs of thieno[3,2-d]pyrimidine
derivatives in more details, CoMFA (q2 ¼ 0.436, r2 ¼ 0.937) and CoMSIA (q2 ¼ 0.706, r2 ¼ 0.947) models
on H460 cell line were established. The generated 3D-QSAR models can be used for further rational
design of novel thienopyrimidines as highly potent and selective cytotoxic agents.
Received 30 March 2012
Received in revised form
31 August 2012
Accepted 1 September 2012
Available online 10 September 2012
Keywords:
2-Hydrazinyl-4-morpholinothieno[3,2-d]
pyrimidine
Synthesis
3D-QSAR
Ó 2012 Elsevier Masson SAS. All rights reserved.
CoMFA
CoMSIA
Cytotoxic activity
1. Introduction
tumor cell lines and is currently undergoing Phase I clinical trials
[8e10]. Further optimization of GDC-0941 led to several backup
Despite efforts to prevent it, cancer remains a global killer,
taking the lives of over 7 million people a year. What’s more, deaths
from cancer worldwide are projected to continue rising, with an
estimated of 13.1 million in 2030 [1]. Although a range of antitumor
drugs and therapies are available, drug resistance and adverse side
effects are still serious problems. Therefore, the search for highly
efficient and safe chemotherapeutic agents for treating cancer
remains desirable.
Thienopyrimidine derivatives have attracted great attention
over many years due to their broad bioactivities, including anti-
tumor [2e5], antimicrobial [6], anti-inflammatory [7]. Of these
active compounds, 4-morpholinothieno[3,2-d]pyrimidines have
been reported to show marked antitumor activities with different
biotargets and mechanisms [4,5,8]. For example, GDC-0941(1,
Fig. 1), which is a potent, selective, orally bioavailable inhibitor of
PI3K, exerted antiprolifeative effects against an array of human
clinical candidates with similar structures, such as GNE-477 (2,
Fig. 1) [4], GNE-493 (3a, Fig. 1) [5], and GNE-490 (3b, Fig. 1) [5]. The
structureeactivity relationships (SARs) studies of these (pre)clin-
ical compounds revealed that the morpholino substituted thieno
[3,2-d]pyrimidine scaffolds play a key role in their anti-tumor
activities [3e5,8].
However, nearly all of these reported 4-morpholinothieno[3,2-
d]pyrimidines were obtained by introducing heterocyclic groups
to the C-2 position of thienopyrimidine core through Suzuki
Coupling reaction. The potential drawbacks of these derivatives
are their strong structure rigidity, which may lead to a poor
pharmacokinetic profile. Prompted by these observations, we
speculated that compounds with flexible substituent at this
position might possess equal even enhanced anti-tumor activity
as well as an improved pharmacokinetic profile. It’s worthwhile to
search and discover some flexible molecules or functional groups.
The flexible hydrazinyl group, which was widely used in the
design of anticancer agents (4 [11], 5 [12], 6 [13,14], Fig. 1),
aroused our attention and was applied to the design of new
compounds.
* Corresponding author. Tel./fax: þ86 24 2398 6429.
E-mail address: yanfangzhao@126.com (Y. Zhao).
0223-5234/$ e see front matter Ó 2012 Elsevier Masson SAS. All rights reserved.
http://dx.doi.org/10.1016/j.ejmech.2012.09.002

W. Zhu et al. / European Journal of Medicinal Chemistry 57 (2012) 162e175
163
Fig. 1. Structures of 4-morpholinothieno[3,2-d]pyrimidines (GDC-0941,GNE-477,GNE-490,GNE-493), hydrazinyl group-containing compounds (4,PAC-1,6) and substituted
heterocyclic compounds (7,8).
In view of above, we focused on the synthesis of novel 4-
morpholinothieno[3,2-d]pyrimidines bearing flexible hydrazinyl
groups at C2-position. Especially, we thought to introduce some N-
benzyl heterocycles to the hydrazinyl groups. The N-benzyl indole
moieties, which are frequently found in medicinal chemistry
(Oncrasin-1 [15], bis(indolyl) hydrazide-hydrazones [16,17], 8 [18],
Fig. 1),were introduced to afford compounds 15ae15o (Fig. 2). And
according to the theory of isostere principle, the indole ring was
then replaced by benzimidazole moiety (17ae17f, Fig. 2).
However, these benzimidazole-containing compounds (17ae17f)
showed disappointing cytotoxic activities and thus we shifted
our focus back to the N-benzyl indole series. To investigate the
influences of C-6 position substituent to this series of derivatives,
a hydroxy-isopropyl group that existed both in GNE-490 and GNE-
493 was introduced to yield compounds 20ce20k and 20p (Fig. 2).
These modifications might increase the bioavailability of the new
compounds.
Moreover, 3D-QSAR analysis employed by CoMFA and CoMSIA
models on H460 cell line will be presented in this paper as well.
2. Chemistry
The preparation of target compounds 15ae15o, 17ae17f, 20ce
20k and 20p was described in Scheme 1 [8]. The commercially
available methyl 3-amino-2-thiophenecarboxylate 9 was condensed
with urea at 190 ꢀC for 2.5 h to give thienopyrimidinedione 10.
Chlorination of 10 with phosphorus oxychloride proceeded
smoothly to 11 as a pale yellow solid. Regioselective nucleophilic
displacement of the 4-chloride with morpholine, followed by
hydrazinolysis gave access to the key intermediate 13. Subsequently,
13 condensed with substituted benzyl-1H-indole-3-carbaldehydes
14ae14o or 1-benzyl-1H-benzo[d]imidazole-2-carbaldehydes
16ae16f to yield compounds 15ae15o and 17ae17f, respectively.
On the other hand, compound 12 treated with n-BuLi and subse-
quently trapped with acetone to furnish a tertiary alcohol which was
then used in a substitution reaction with NH2NH2$H2O to generate
another key intermediate 19. Similarly, intermediate 19 condensed
Herein we disclose the design, synthesis and cytotoxic activity
of 2-hydrazinyl-4-morpholinothieno[3,2-d]pyrimidines against
H460, HT-29, MDA-MB-231, U87MG and H1975 cancer cell lines.
Fig. 2. Design strategy for 2-hydrazinyl-4-morpholinothieno[3,2-d]pyrimidine derivatives.

164
W. Zhu et al. / European Journal of Medicinal Chemistry 57 (2012) 162e175
Scheme 1. Reagents and conditions: (a) 5eq urea, 190 ꢀC, 2.5 h; (b) POCl3, DMF(cat.), reflux, 8 h; (c) 2.1eq morpholine, MeOH, 0 ꢀC, 30min, rt, 1e2 h; (d) 80% NH2NH2$H2O, 80 ꢀC, 4e
6 h; (e) substituted aldehyde, reflux, 4e8 h; (f) acetone, n-BuLi, THF, ꢁ78 ꢀC to rt, 5 h; (g) ArCH2Cl, K2CO3, DMF,80 ꢀC; (h) DMF, n-BuLi, THF, ꢁ78 ꢀC to rt, 6e8 h.
with the corresponding benzyl-1H-indole-3-carbaldehydes 14ce
14k and 14p to afford the target compounds 20ce20k and
20p,respectively.
Scheme 1 led selectively to the E isomer, which was confirmed by
2D NOESY NMR spectroscopic approach with the aid of confor-
mational analysis with Tripos Force Field by Sybyl 6.91 software
package (TRIPOS Associates Inc.) (Figs. 3 and 4) [19,20].
The substituted benzyl-1H-indole-3-carbaldehydes 14ae14p
were prepared by nucleophilic substitution of the various benzyl
chloride with 1H-indole-3-carbaldehyde 21 in the presence of
K2CO3 in N,N-dimethylformamide (DMF). The 1-benzyl-1H-benzo
[d]imidazole-2-carbaldehydes 16ae16f were synthesized via
a two-step reaction. Reaction of benzimidazole with the corre-
sponding benzyl chloride in the presence of K2CO3 in DMF yielded
23aꢁ23f. Subsequently, lithiation of 23aꢁ23f, followed by the
addition of DMF, gave the aldehydes 16ae16f in 75% yield.
The structures of target compounds were confirmed by IR, 1H
NMR, 13C NMR, 2D NOESY NMR and MS spectra. All the target
compounds can exist in the E or Z isomeric forms due to the
carbonenitrogen double bonds formed in condensation reaction.
The presence/absence of bulky groups and the differences in
stability and energy are generally accepted as the main reasons for
the preferential formation of one isomer over the other. It is
interesting to note that the experimental procedure shown in
For the representative compound 17a or 20p, the NOESY effect
was observed between the proton of NeH and the proton in imine
in E isomer (E-17a and E-20p) (see supplementary information),
which should not be observed in the putative Z isomer (Z-17a and
Z-20p) due to the larger intra-molecular H-H distances (Fig. 3). For
compound 17a, it is revealed experimentally by the NOESY spec-
trum, in which an evident cross-peak system from the dipolar
interactions between N]CH (8.30, s) and NeH (11.43, s) is clear
evidence of the presence of the E-17a isomer. Similarly, for
compound 20p, the NOE interactions between N]CeH (8.26, s)
and NeH (10.48, s) confirmed the E isomer. Moreover, variable
temperature experiments performed on E-17a and E-20p in the
range 220e300 K do not produce any changes in the NMR spectra,
revealing that Z and E isomers are not easily interconverted. On the
basis of the above analysis, the relative stereochemistry of 17a and
20p was established unambiguously.

W. Zhu et al. / European Journal of Medicinal Chemistry 57 (2012) 162e175
165
had higher potency than positive control GDC-0941 against MDA-
MB-231 cancer cell line.
The SARs based on IC50 values (Table 1) showed that variations
in substitutions of the benzyl group had a marked impact on the
cytotoxicity. For 15ae15o and 20ce20k, 20p, substitution of the
benzyl group at the 3-position or 4-position was well tolerated and
3-F substitution (15f and 20f) produced the best potency. In most
cases, compounds with substituents at the 3-position (15c, 15f, 15h,
20c, 20f, 20h) were 2.6- to 12-folds more active than those without
substituents at the same position (15a). Generally, electron-
donating groups (EDGs) and weak electron-withdrawing groups
(EWGs) at C-4 position were preferred. It was noticeable that strong
EWGs at C-4 position weakened the activity dramatically, such as
compounds 15i and 20i with cyano group at C-4 position were 1.9-
to 38.7-folds weak relative to the positive control GDC-0941.
Furthermore, small fluorine atom at 2-position of the benzyl
group was more favorable, as compound 15e was more potent than
15b against all the tested cancer cell lines. However, introduction of
two substituents (such as chlorine in this study) in the benzyl group
resulted in a significantly decrease in activity, which might be due
to its steric hindrance. It was obviously that compounds substituted
by two chlorine groups (15je15l and 20je20l) were more active
than those with one chlorine group (15be15d and 20ce20d).
Further investigations were carried out in details to study the
effect of hydroxy-isopropyl substituent at 6-position of the thie-
nopyrimidine scaffold on the activity. It can be noted from Table 1
that the introduction of hydroxy-isopropyl group led to a remark-
ably increase in the activity against HT-29 and MDA-MB-231 cell
lines and a slightly decrease against H460 cell line (15ce15j vs.20ce
20k). This dramatic boost in the HT-29 and MDA-MB-231 potency
and selectivity by the hydroxy-isopropyl substituent might be
ascribed to the increased hydrogen bonds between the hydroxyl
group and receptors necessary for activity.
Fig. 3. Key NOESY of representative compounds 17a and 20p.
3. Results and discussion
3.1. Biological evaluation
The targetcompounds and their precursors 13,19 were evaluated
for their cytotoxic activities against H460 (human lung cancer), HT-
29 (human colon cancer), MDA-MB-231 (human breast cancer),
U87MG (human glioblastoma) and H1975 (human lung cancer) cell
lines in vitro together with reference compounds 4, PAC-1 and GDC-
0941 by MTT assay. In addition, the most promising compound 15f
was further evaluated against four cancer cell lines A549 (human
lung cancer), H226 (human lung cancer), HepG2 (human liver
cancer) and SGC-7901 (human stomach cancer). The results
expressed as IC50 were summarized in Tables 1 and 2 and the IC50
values are the average of at least three independent experiments.
As illustrated in Table 1, most of the synthesized compounds
showed moderate to significant cytotoxic activities and twenty-two
of them were more potent than GDC-0941 against one or more cell
lines. Nearly all the compounds were more active than their
precursors 13, 19, suggesting that the presence of N-benzyl indole
or N-benzyl benzimidazole hydrazinyl moiety enhanced their anti-
tumor activities dramatically. The most promising compound 15f
showed strong cytotoxic activities against H460, HT-29 and MDA-
Turning to the compounds 17ae17f, interestingly, replacement
of N-benzyl indole with its bioisostere N-benzyl benzimidazole
moiety remarkably enhanced the activity and selectivity against
HT-29 cancer cell line but failed to improve the activity against the
other tested cell lines. In our opinion, these differences in phar-
macological activity may be closely related to the electron densities
of five-membered heterocycles and the specificity of cancer cells.
The cytotoxic activities of the most promising compound 15f as
well as GDC-0941 toward four additional cancer cell lines were
shown in Table 2. Obviously, compound 15f were more active than
GDC-0941 against H226, SGC-7901 cell lines, which further
confirmed its strong antitumor activityagainst different cancer lines.
MB-231 cell lines with IC50 values of 0.23
1.11 M, which were 3.8,1.7 and 66.5 times more active than GDC-
0941 (0.87 M, 0.66 M, 73.82 M), respectively.
mM, 0.39 mM and
m
m
m
m
In general, the target compounds were more potent against
H460 and HT-29 cell lines than against MDA-MB-231, U87MG and
H1975 cell lines. These results suggested that this series of
compounds possessed selectivity for H460 and HT-29 cancer cell
lines. It’s worth mentioning that almost all the target compounds
3.2. 3D-QSAR study
3.2.1. 3D-QSAR model
All structures of thieno[3,2-d]pyrimidine derivatives were built
and aligned onto the common substructure 2-hydrazinyl-4-
Fig. 4. Global optimal conformations for compounds 17a (left) and 20p(right) generated by Sybyl 6.91 software package.

166
W. Zhu et al. / European Journal of Medicinal Chemistry 57 (2012) 162e175
Table 1
Structures and cytotoxic activities of compounds 13, 19, 15ae15o, 17ae17f, 20ce20k and 20p against H460, HT-29, MDA-MB-231, U87MG and H1975 cancer cell lines in vitro.
M)a
pIC50
b
Compd.
R
IC50
(m
H460
HT29
MDA-MB-231
U87MG
H1975
13
19
e
42.21 ꢂ 3.01
29.66 ꢂ 2.58
1.41 ꢂ 0.13
3.98 ꢂ 0.11
0.54 ꢂ 0.09
0.36 ꢂ 0.085
0.86 ꢂ 0.22
0.23 ꢂ 0.095
0.41 ꢂ 0.04
0.34 ꢂ 0.13
1.62 ꢂ 0.21
5.77 ꢂ 0.33
10.42 ꢂ 1.41
0.74 ꢂ 0.21
0.83 ꢂ 0.14
0.38 ꢂ 0.098
0.52 ꢂ 0.070
25.56 ꢂ 1.41
79.36 ꢂ 3.52
4.56 ꢂ 0.12
2.60 ꢂ 0.31
3.62 ꢂ 0.27
37.14 ꢂ 4.21
0.84 ꢂ 0.069
0.55 ꢂ 0.095
1.10 ꢂ 0.064
0.42 ꢂ 0.07
1.10 ꢂ 0.57
0.36 ꢂ 0.097
4.53 ꢂ 1.63
8.90 ꢂ 0.81
4.03 ꢂ 0.72
7.22 ꢂ 1.21
9.2 ꢂ 1.35
23.42 ꢂ 3.13
56.21 ꢂ 5.02
4.70 ꢂ 0.72
>100
63.75 ꢂ 4.01
>100
ND
ND
>100
ND
5.96 ꢂ 0.73
15.64 ꢂ 1.54
22.58 ꢂ 0.59
12.59 ꢂ 0.46
8.83 ꢂ 1.03
9.68 ꢂ 0.21
ND
ND
ND
>100
4.97 ꢂ 0.19
11.23 ꢂ 0.63
4.84 ꢂ 0.37
ND
ND
ND
ND
ND
e
e
e
15a
15b
15c
15d
15e
15f
15g
15h
15i
H
12.39 ꢂ 1.49
42.74 ꢂ 2.24
5.85
5.40
6.27
6.45
6.06
6.65
6.39
6.46
5.79
5.24
4.98
6.13
6.08
6.42
6.28
4.59
4.10
5.34
5.58
5.44
4.43
6.08
6.26
5.96
6.37
5.96
6.44
5.34
5.05
5.39
5.14
e
2-chloro
3-chloro
4-chloro
2-fluoro
3-fluoro
4-fluoro
3-cyano
4-cyano
2,4-dichloro
2,6-dichloro
3,4-dichloro
4-methyl
4-tert-butyl
3-trifluoromethyl
H
2-chloro
3-chloro
4-chloro
4-tert-butyl
2,3-dichloro
3-chloro
4-chloro
2-fluoro
3-fluoro
4-fluoro
3-cyano
4-cyano
2,4-dichloro
2,6-dichloro
2,3-dichloro
e
>100
ND
2.98 ꢂ 0.45
5.76 ꢂ 0.95
3.28 ꢂ 0.48
0.39 ꢂ 0.056
1.79 ꢂ 0.12
4.86 ꢂ 1.14
25.51 ꢂ 2.21
46.52 ꢂ 3.17
35.36 ꢂ 3.07
16.20 ꢂ 2.46
3.31 ꢂ 0.69
1.69 ꢂ 0.35
1.68 ꢂ 0.23
1.08 ꢂ 0.27
0.40 ꢂ 0.091
1.01 ꢂ 0.12
0.77 ꢂ 0.054
4.18 ꢂ 1.39
6.46 ꢂ 1.01
0.87 ꢂ 0.028
0.87 ꢂ 0.042
1.05 ꢂ 0.127
0.38 ꢂ 0.058
1.01 ꢂ 0.097
1.39 ꢂ 0.089
2.54 ꢂ 0.147
5.38 ꢂ 0.91
3.36 ꢂ 0.65
3.70 ꢂ 0.32
20.2 ꢂ 2.09
0.97 ꢂ 0.053
0.66 ꢂ 0.081
9.74 ꢂ 1.41
2.7 ꢂ 0.56
9.24 ꢂ 1.74
1.11 ꢂ 0.061
12.32 ꢂ 1.67
2.0 ꢂ 0.043
>100
>100
ND
>100
19.01 ꢂ 0.78
>100
ND
ND
ND
ND
15j
15k
15l
>100
20.47 ꢂ 2.15
>100
>100
15m
15n
15o
17a
17b
17c
17d
17e
17f
20c
20d
20e
20f
9.73 ꢂ 2.08
1.43 ꢂ 0.075
83.77 ꢂ 5.01
21.26 ꢂ 1.87
>100
37.68 ꢂ 3.21
13.88 ꢂ 0.64
8.17 ꢂ 0.51
>100
3.22 ꢂ 0.42
5.14 ꢂ 0.76
1.66 ꢂ 0.082
7.20 ꢂ 0.36
1.53 ꢂ 0.073
5.61 ꢂ 0.31
6.31 ꢂ 1.07
>100
33.12 ꢂ 1.69
8.00 ꢂ 1.01
2.79 ꢂ 0.52
ND
ND
ND
ND
ND
13.37 ꢂ 2.11
ND
ND
ND
8.91 ꢂ 1.08
7.43 ꢂ 1.04
3.92 ꢂ 0.86
7.82 ꢂ 1.12
ND
23.75 ꢂ 0.25
33.78 ꢂ 1.32
15.56 ꢂ 0.72
19.66 ꢂ 0.32
20g
20h
20i
ND
6.64 ꢂ 0.87
ND
ND
ND
ND
ND
ND
ND
ND
20j
ND
ND
ND
ND
ND
7.77 ꢂ 0.89
20k
20p
4c
62.12 ꢂ 1.21
>100
>100
PAC-1c
GDC-0941c
e
3.57 ꢂ 1.62
0.87 ꢂ 0.20
6.11 ꢂ 0.75
e
e
73.82 ꢂ 3.06
0.27 ꢂ 0.021
e
Bold values show IC50 of < 1 mM. ND: Not determined.
a
IC50: concentration of the compound (mM) producing 50% cell growth inhibition after 72 h of drug exposure, as determined by the MTT assay. Each experiment was run at
least three times.
b
pIC50 values of H460 human tumor cell lines.
Used as a positive control.
c
morpholinothieno[3,2-d]pyrimidine nucleus (Fig. 5a, shown in
bold), with the potent compound 15f (Fig. 5a) as template. The
structure alignment was illustrated in Fig. 5b. To further investigate
the SARs of thieno[3,2-d]pyrimidine derivatives, the cytotoxic data
of H460 cell line were used to build the 3D-QSAR models. 31
compounds with determinate IC50 values were employed and their
pIC50 values were calculated based on the original IC50 data. The
detailed operation steps were similar to the methods in literature
and were shown in the experimental section [21e23].
Table 2
Cytotoxic activities of compound 15f and GDC-0941 against A549, H226, HepG2 and
SGC-7901 cancer cell lines in vitro.
3.2.2. CoMFA and CoMSIA analysis
The PLS statistic results of CoMFA and CoMSIA models were
summarized in Table 3. The predicted versus the experiment pIC50
values are listed in Table 4 and are depicted graphically in Fig. 6 for
CoMFA and CoMSIA models, respectively. The cross-validation
correlation coefficient (q2) which marked the predictive capacity
and the conventional correlation coefficient (r2) which marked self-
consistence were two key parameters to evaluate the qualities of
PLS analysis. A q2 value over 0.3 is considered significant for the
chance of significant correlation being <95%. For CoMFA model, the
Compd.
IC50
(
m
M)a
A549
H226
HepG2
SGC-7901
15f
8.7 ꢂ 0.92
6.9 ꢂ 0.45
0.91 ꢂ 0.14
2.9 ꢂ 0.64
1.1 ꢂ 0.21
1.0 ꢂ 0.23
0.8 ꢂ 0.11
1.8 ꢂ 0.69
GDC-0941b
a
IC50: concentration of the compound (mM) producing 50% cell growth inhibition
after 72 h of drug exposure, as determined by the MTT assay. Each experiment was
run at least three times.
b
Used as a positive control.

W. Zhu et al. / European Journal of Medicinal Chemistry 57 (2012) 162e175
167
Fig. 5. (a) Structure of the template compound 15f, common substructure is in bold, (b) 3D-QSAR structure alignment and superposition of 31 compounds using compound 15f as
the template.
q
2 and r2 was 0.436 and 0.937, while for CoMSIA model was 0.706
favorable regions for sterically bulkier groups and the yellow
contours are indicative of regions that are sterically less favorable.
There is one big green contour covering the 3- and 4-position of the
benzyl group as shown in Fig. 7(A), which may explain why
compounds 15a (R ¼ H) is less potent than compounds 15m (R ¼ 4-
methyl), 15n (4-tert-butyl) and 15o (3-trifluoromethyl). Conversely,
the sterically unfavorable regions indicated by yellow contours
spotted around the 2- and 6-position of the benzyl group, as
observed for the activity order of those compounds, 15e (2-fluoro)>
15b (2-chloro), 15j (2,4-dichloro) > 15k (2,6-dichloro), and 20e (2-
fluoro) > 20j (2,4-dichloro), 20k (2,6-dichloro), 20p (2,3-dichloro).
The contour map of the electrostatic field of CoMFA model is
shown in Fig. 7 (B). The red contours represent regions that lead to
the enhancement of activity with electron rich groups, and contrary
to that the blue regions represent electron deficient regions and can
lead to an increase in the activity of compounds. Several small red
regions are found around the 3-(5-) or 4-position of the benzyl
group, which can explain the fact that compound 15i (R ¼ 4-cyano)
is less active than compounds 15d (2-chloro), 15g (4-fluoro), 15m
(R ¼ 4-methyl) and 15n (4-tert-butyl), and compound 20i (R ¼ 4-
cyano) is less active than compounds 20d (2-chloro) and 20g (4-
fluoro). On the other hand, the blue contour near the 2-position
and 6-position of benzene ring sheds light on the fact that the
activity of compound 15e (R ¼ 2-fluoro) is higher than those of the
compounds 15b (R ¼ 2-chloro), 15j (R ¼ 2, 4-dichloro) and 15k
(R ¼ 2,6-dichloro).
and 0.947. The statistic results indicated good predictive ability of
CoMFA and CoMSIA models. The optimal numbers of components
using to generate CoMFA and CoMSIA models is 6 and 5, respec-
tively, which were reasonable according to the number of
compounds used to derive the models. The standard errors of
estimate of CoMFA and CoMSIA models were also reasonably low
amounting to 0.186 and 0.167. Fisher test results were 59.326 and
89.399 for CoMFA and CoMSIA models, respectively. As shown in
Table 3, for CoMFA model, the contributions of steric, electrostatic
fields were 50.4% and 49.6%, while the contributions of steric,
electrostatic, hydrophobic, hydrogen bond donor and acceptor
fields of CoMSIA model were 8.9%, 30.8%, 44.8%, 2.1%, and 13.4%,
respectively. The CoMFA and CoMSIA models established in the
present study are quite reliable and sufficiently efficient in guiding
further plausible modification in the molecules for obtaining better
activity.
3.2.3. Contour analysis
To view the field effect on the target property, CoMFA and
CoMSIA contour maps were generated. The contour maps can
identify the important regions where any change in the steric,
electrostatic, hydrophobic, hydrogen bond donor and hydrogen
bond acceptor fields may affect the biological activity. The visuali-
zation of the results of the CoMFA and CoMSIA models were per-
formed using the StDev*Coeff mapping option contoured by
contribution. All of the contours represented the default 80 and 20%
level contributions for favored and disfavored regions. 3D-QSAR
contour maps of both CoMFA and CoMSIA models were depicted in
Figs. 7 and 8, compounds 15f with the highest pIC50 value and 17d
were chosen to superimpose into the contour maps to better
illustrate the SAR information.
3.2.3.2. CoMSIA contour analysis. The CoMSIA contour maps,
derived from using steric, electrostatic, hydrophobic, hydrogen
bond donor and hydrogen bond acceptor fields, are represented in
Fig. 8. CoMSIA steric and electrostatic contours (Fig. 8(A) and (B))
are similar to those of CoMFA. However, some differences are noted
as well. In case of CoMSIA, there is a large yellow contour over-
lapping the benzene ring, revealing that bulkier substituents will
not be preferred in this region.
3.2.3.1. CoMFA contour analysis. The CoMFA contour map of steric
field is shown in Fig. 7(A). The green contours are indicative of
Table 3
PLS statistics of CoMFA and CoMSIA 3D-QSAR models.
q2a
Nb
r2c
SEd
Fe
Fractionf
Steric
Electrostatic
Hydrophobic
Donor
Acceptor
CoMFA
CoMSIA
0.436
0.706
6
5
0.937
0.947
0.186
0.167
59.326
89.399
0.504
0.089
0.496
0.308
e
e
e
0.448
0.021
0.134
a
Cross-validated correlation coefficient.
b
c
d
e
f
Optimum number of components obtained from cross-validated PLS analysis and same used in final non-cross-validated analysis.
Non-cross-validated correlation coefficient.
Standard error of estimate.
F-test value.
Field contributions.

168
W. Zhu et al. / European Journal of Medicinal Chemistry 57 (2012) 162e175
Table 4
Experimental and predicted activities against H460 cell line (pIC50) of the target compounds for CoMFA and CoMSIA models.
Compd.
CoMFA
CoMSIA
pIC50 (experimental)
pIC50 (predicted)
Residuals
pIC50 (experimental)
pIC50 (predicted)
Residuals
15a
15b
15c
15d
15e
15f
15g
15h
15i
5.85
5.40
6.27
6.45
6.06
6.65
6.39
6.46
5.79
5.24
4.98
6.13
6.08
6.42
6.28
4.59
4.10
5.34
5.58
5.44
4.43
6.08
6.26
5.96
6.37
5.96
6.44
5.34
5.05
5.39
5.14
6.027
5.813
6.237
6.255
5.961
6.369
6.254
6.323
5.939
5.163
4.916
6.237
6.231
6.449
6.28
5.033
4.158
5.136
5.367
5.459
4.348
6.146
6.189
5.954
6.486
5.980
6.392
5.481
4.862
5.267
5.210
ꢁ0.1772
ꢁ0.4126
0.0329
0.1955
0.0989
0.2814
0.1360
0.1371
5.85
5.40
6.27
6.45
6.06
6.65
6.39
6.46
5.79
5.24
4.98
6.13
6.08
6.42
6.28
4.59
4.10
5.34
5.58
5.44
4.43
6.08
6.26
5.96
6.37
5.96
6.44
5.34
5.05
5.39
5.14
6.056
5.709
6.291
6.264
6.069
6.242
6.23
6.591
5.580
5.361
4.985
6.154
6.185
6.366
6.35
4.942
4.098
5.251
5.374
5.393
4.427
6.205
6.318
5.952
6.464
6.001
6.397
5.285
4.960
5.267
5.150
ꢁ0.2063
ꢁ0.3091
ꢁ0.0210
0.1862
ꢁ0.0087
0.4078
0.1600
ꢁ0.1315
ꢁ0.1487
0.2105
15j
15k
15l
0.0765
ꢁ0.1212
ꢁ0.0053
ꢁ0.0242
ꢁ0.1049
0.0536
ꢁ0.0698
ꢁ0.3524
0.0022
0.0889
0.2061
0.0466
0.0031
ꢁ0.1254
ꢁ0.0577
0.0080
ꢁ0.0945
ꢁ0.0412
0.0433
0.0550
0.0895
0.1226
ꢁ0.0104
0.0640
ꢁ0.1075
ꢁ0.1507
ꢁ0.0291
0.0005
ꢁ0.4430
ꢁ0.0580
0.2043
15m
15n
15o
17a
17b
17c
17d
17e
17f
20c
20d
20e
20f
20g
20h
20i
20j
20k
20p
0.2131
ꢁ0.0189
0.0819
ꢁ0.0659
0.0707
0.0059
ꢁ0.1162
ꢁ0.0198
0.0482
ꢁ0.1408
0.1882
0.1234
ꢁ0.0704
The hydrophobic contour map of CoMSIA model is displayed in
Fig. 8(C). In the CoMSIA hydrophobic field, yellow regions and
white regions indicate the areas where hydrophobic and hydro-
philic properties are preferred, respectively. Two big yellow
contours near the 3- and 4- positions of benzene ring (both of 15f
and 17d) indicate the necessity of the hydrophobic group attached
to the benzene ring for higher activity. This is a possible reason why
compounds 15m (R ¼ 4-methyl) and 15o (R ¼ 4-tert-butyl) have
higher potency than compound 15a (R ¼ H), and why compounds
17c (R ¼ 4-chloro), 17e (R ¼ 4-tert-butyl) are more active than 17a
(R ¼ H).There is one white contour near the C-2 position of benzene
ring, suggesting that hydrophilic groups in this region are beneficial
to the activity.
The hydrogen bond donor contour map of CoMSIA model is
displayed in Fig. 8(D). In the CoMSIA hydrogen bond donor field,
hydrogen bond donor favored regions are represented by cyan
contours and unfavorable regions by purple contours. The big cyan
contour near eNHe functional group specifies that hydrogen bond
donor is favored there. The little blue contour near thiophene ring is
consistent with the fact that almost all the compounds with
hydroxy-isopropyl groups (15ce15i) are less potent than those
without substituents at the same position (20ce20i). This suggests
that the hydroxy-isopropyl group at this position is not favorable
for the cytotoxic activity against H460 cell line.
to the nitrogen atom of imidazole ring is in agreement with the fact
that compounds containing benzimidazole moieties (17ae17f) are
generally less potent than those with indole moieties (15ae15o,
20ce20k, 20p). Another red contour around 4-position of
benzene ring could explain why compound 15i with cyano group in
this position is less potent than those without hydrogen bond
acceptor group (15d, 15g, 15me15n), and why 20i (R ¼ 4-cyano) is
less potent than compounds 20d (R ¼ 4-chloro) and 20g (R ¼ 4-
fluoro).
4. Conclusions
In summary, thirty-one thieno[3,2-d]pyrimidines bearing flex-
ible substituted hydrazinyl groups at C2-position were designed
and synthesized. Twenty-two of them were more potent than GDC-
0941 and compound 15f showed strongest cytotoxic activities
against H460, HT-29 and MDA-MB-231 cell lines which were 3.8,1.7
and 66.5 times more active than GDC-0941, respectively. The initial
SARs showed that the flexible group benzyl-indole or benzyl
-benzimidazole hydrazinyl moieties were necessary for these
compounds to possess potent cytotoxic activities and the former
were much better. Variations in substitutions of the benzyl group
had a significant impact on the cytotoxicity. Substitution of the
benzyl group at the 3-position or 4-position was well tolerated and
3-F substitution produced the best potency. Moreover, the intro-
duction of hydroxy-isopropyl group was benefit to the selectivity
for HT-29 cell line. Both CoMFA (q2 ¼ 0.436, r2 ¼ 0.937) and CoMSIA
(q2 ¼ 0.706, r2 ¼ 0.947) models gave good statistical results in terms
of q2 and r2 values, and provided significant insights that could be
used in the further design of novel, potent and selective cytotoxic
agents. Studies on the mechanism of action of these compounds are
in progress and will be reported in the future.
The hydrogen bond acceptor contour map of CoMSIA model is
displayed in Fig. 8(E). In the CoMSIA hydrogen acceptor field,
magenta contours represent regions where hydrogen bond
acceptor substituents on ligands can be favored and the red ones
represent areas where such substituents on compounds may be
disfavored. It can be found that there is one big magenta region
around the imino group, which proved the importance of thi-
spharmacophore for all the target compounds. The red region close

W. Zhu et al. / European Journal of Medicinal Chemistry 57 (2012) 162e175
169
5. Experimental
5.1. Chemistry
All melting points were obtained on a Büchi Melting Point B-540
apparatus (Büchi Labortechnik, Flawil, Switzerland) and were
uncorrected. The IR spectra were recorded by means of the KBr
pellet technique on a Bruker FTS 135 spectrometer. 1H NMR and 13
C
NMR spectra were recorded on Bruker ARX-300, 300 MHz or
Bruker ARX-600, 600 MHz spectrometers (Bruker Bioscience, Bill-
erica, MA, USA) with TMS as an internal standard. Mass spectra
(MS) were taken in ESI mode on Agilent 1100 LCeMS (Agilent, Palo
Alto, CA, USA). Elemental analysis was determined on a Carlo-Erba
1106 Elemental analysis instrument (Carlo Erba, Milan, Italy). TLC
analysis was carried out on silica gel plates GF254 (Qindao Haiyang
Chemical, China). All the materials were obtained from commercial
suppliers and used without purification, unless otherwise specified.
Yields were not optimized.
5.2. Preparation of 1H-thieno[3,2-d]pyrimidine-2,4-dione (10)
A mixture of methyl 3-amino-2-thiophenecarboxylate (100 g,
0.64 mol) and urea (191 g, 3.2 mol) was heated at 190 ꢀC for 2.5 h.
Upon cooling to about 120 ꢀC, the reaction mixture was poured into
sodium hydroxide (2000 mL, 1N) solution and any insoluble
material removed by filtration. The mixture was then acidified (HCl,
2N) to yield 1H-thieno[3,2-d]pyrimidine-2,4-dione as a white
precipitate, which was collected by filtrationꢁand dried (105 g, 98%).
m.p. >300 ꢀC. ESI-MS m/z: 167.1 [M ꢁ H] . 1H NMR (300 MHz,
DMSO-d6)
1H), 11.61 (s, 1H).
d
: 6.93 (d, J ¼ 5.2 Hz, 1H), 8.04 (d, J ¼ 5.2 Hz, 1H), 11.19 (s,
5.3. Preparation of 2, 4-dichloro-thieno[3,2-d]pyrimidine (11)
A
mixture of 1H-thieno[3,2-d]pyrimidine-2,4-dione(105 g,
0.63 mol), phosphorous oxychloride (1000 mL) and catalytic
amount DMF (2 mL) was heated at reflux for 8 h and the reaction
was monitored by TLC. The reaction mixture was concentrated
under reduced pressure and the residue was poured onto ice/water
with vigorous stirring yielding a precipitate. The mixture was then
filtered to yield 2,4-dichloro-thieno[3,2-d]pyrimidine as a wþhite
solid (93 g, 72%). m.p. 142e144 ꢀC. ESI-MS m/z: 205.3 [M þ H] .
Fig. 6. Calculated pIC50 versus experimental pIC50 values for the 31 molecules ob-
tained by PLS analysis using CoMFA and CoMSIA models.
5.4. Preparation of 2-chloro-4-morpholine-4-yl-thieno[3,2-d]
pyrimidine (12)
To the mixture of 2,4-dichloro-thieno[3,2-d]pyrimidine (93 g,
0.45 mol) and MeOH (1500 mL), morpholine (81 mL, 2.1 eq) was
Fig. 7. CoMFA contour maps based on compounds 15f (shown in capped sticks) and 17d (shown in ball and stick). (A) CoMFA steric contours: green contours indicate regions where
steric interaction is favored. Yellow contours are areas where the steric interaction is disfavored. (B) CoMFA electrostatic contours: the blue region represents the area where an
electropositive group is favorable and the red region refers to the area where an electronegative group is favorable. (For interpretation of the references to color in this figure legend,
the reader is referred to the web version of this article.)

170
W. Zhu et al. / European Journal of Medicinal Chemistry 57 (2012) 162e175
Fig. 8. CoMSIA contour maps based on compounds 15f (shown in capped sticks) and 17d (shown in ball & stick). (A) CoMSIA steric contour maps: green contour favours steric or
bulky group and yellow contour denotes disfavored region. (B) CoMSIA electrostatic contour maps: blue contour indicates electropositive charge and red contour electronegative
charge. (C) CoMSIA hydrophobic contour maps: yellow contours represent regions where hydrophobic groups increase activity, while white contours highlight regions that would
favour hydrophilic groups. (D) CoMSIA hydrogen bond donor contours: cyan and purple contours represent favorable and unfavorable hydrogen bond donor regions, respectively.
(E) CoMSIA hydrogen bond acceptor contours: magenta and red contours represent favorable and unfavorable hydrogen bond acceptor regions, respectively. (For interpretation of
the references to color in this figure legend, the reader is referred to the web version of this article.)
added drop-wise at 0 ꢀC. The reaction mixture then was stirred at
room temperature for 1.5 h. After completion of reaction as
indicated by TLC, the mixture was then filtered, washed with
water and MeOH, to yield the title compound as a white solid
(114 g, 98%). m.p. 204e205 ꢀC. ESI-MS m/z: 255.9 [M þ H]þ,
5.6. General procedure for the preparation of compounds 14ae14p
1H-Indole-3-carbaldehyde 21 (14.5 g, 0.1 mol), together with
the respective substituted benzyl chloride (0.11 mol) and anhy-
drous K2CO3 (27.6 g, 0.2 mol) were dissolved in N,N-dime-
thylformamide (30 mL) and heated at 80 ꢀC for 5e8 h. Upon cooling
to room temperature, the mixture was poured into ice-water and
then the precipitated was collected by filtration and dried. The
crude product was purified by recrystallized from ethanol to give
the solids 14ae14p.
277.6 [M þ Na]þ. 1H NMR (300 MHz, DMSO-d6)
d: 3.76 (s, 4H),
3.90 (d, J ¼ 4.2 Hz, 4H), 7.41 (d, J ¼ 5.4 Hz, 1H), 8.31 (d, J ¼ 5.4 Hz,
1H).
5.5. Preparation of 4-(2-hydrazinylthieno[3,2-d]pyrimidin-4-yl)
morpholine (13)
A solution of 80% hydrazine hydrate (NH2NH2$H2O, 170 mL,
4.9 mol) and 12 (20.0 g, 0.078 mol) was stirred at 80 ꢀC for 5 h.
The precipitate was collected by filtration and washed with
water, and dried to give 13 as a light yellow solid (13.4 g, 68%).
5.7. General procedure for the preparation of compounds 16ae16f
5.7.1. General procedure for the preparation of compounds 23ae23f
1H-Benzo[d]imidazole 22 (11.8 g, 0.1 mol), together with the
respective substituted benzyl chloride (0.11 mol) and anhydrous
K2CO3 (27.6 g, 0.2 mol) were dissolved in DMF (30 mL) and heated
at 50 ꢀC for 5e6 h. Upon cooling to room temperature, the mixture
was poured into ice-water and then the precipitated was collected
m.p. 160e161 ꢀC. ESI-MS m/z: 251.8 [M
(600 MHz, DMSO-d6)
þ
H]þ. 1H NMR
: 3.74e3.70 (m, 4H), 3.86-3.80 (m, 4H),
d
4.27 (s, 2H), 7.16 (d, J ¼ 5.5 Hz, 1H), 7.54 (s, 1H), 8.00 (d, J ¼ 5.5 Hz,
1H).

W. Zhu et al. / European Journal of Medicinal Chemistry 57 (2012) 162e175
171
by filtration and dried. The crude product was purified by recrys-
tallized from ethanol to give the solids 23ae23f.
159.00, 158.76, 141.15, 137.43, 136.99, 133.96, 133.86, 131.39, 131.28,
128.21, 127.65, 126.53, 125.96, 124.47, 123.46, 123.35, 121.09, 113.56,
111.06, 106.27, 66.77 (2C), 49.25, 46.58 (2C). Anal. Calcd for
5.7.2. General procedure for the preparation of compounds 16ae16f
To a stirred, cooled (ꢁ78 ꢀC) solution of substituted 1-benzyl-
1H-benzo[d]imidazole 23ae23f (0.03 mol) in 60 mL THF under
nitrogen atmosphere was added dropwise n-butyllithium solution
(0.04 mol, 2.5 M in hexanes) for half an hour. When addition was
complete, stirring at ꢁ75 ꢀC was continued for 1 h. The reaction
mixture was then treated with 0.1 mol dried DMF within 10 min
and stirring at ꢁ75 ꢀC for 1 h. The mixture was then allowed to
warm to ambient temperature and stirred for a future 2 h. The
reaction was quenched with saturated NH4Cl (50 mL) and extracted
with ether (100 mL ꢃ 2). The organic phase was washed with brine,
dried (MgSO4), filtered and concentrated. The crude product was
purified by chromatography on silica gel using petroleum/EtOAc to
give substituted 1-benzyl-1H-benzo[d]imidazole-2-carbaldehyde
16ae16f.
C26H23ClN6OS (%): C, 62.08; H, 4.61; N, 16.71. Found (%): C, 62.02; H,
4.63; N, 16.78.
5.8.4. (E)-4-(2-(2-((1-(4-Chlorobenzyl)-1H-indol-3-yl)methylene)
hydrazinyl)thieno[3,2-d]pyrimidin-4-yl)morpholine (15d)
This compound was obtained as light yellow solid in 87% yield.
m.p. 221e223 ꢀC. ESI-MS m/z: 503.3 (Cl ¼ 35), 505.3 (Cl ¼ 37)
[M þ H]þ. IR (KBr) cmꢁ1: 3436.7, 2845.7, 1574.0, 1535.5, 1374.3,
1164.9, 1120.0, 1013.8, 745.5. 1H NMR (300 MHz, DMSO-d6)
d: 3.80
(s, 4H), 3.97 (s, 4H), 5.44 (s, 2H), 7.19 (m, 5H), 7.38 (d, J ¼ 8.2 Hz, 2H),
7.47 (d, J ¼ 7.7 Hz, 1H), 7.81 (s, 1H), 8.06 (d, J ¼ 5.5 Hz, 1H), 8.27 (s,
1H), 8.48 (d, J ¼ 7.5 Hz, 1H), 10.51 (s, 1H). 13C NMR (DMSO-d6)
d:
163.84, 158.90, 158.48, 137.47, 137.34, 133.94, 132.87, 131.41,
129.71(2C), 129.30 (2C), 128.90, 125.93, 124.31, 123.49, 123.19,
121.13, 113.28, 111.05, 106.43, 66.73 (2C), 49.25, 46.55 (2C). Anal.
Calcd for C26H23ClN6OS (%): C, 62.08; H, 4.61; N, 16.71. Found (%): C,
62.02; H, 4.58; N, 16.79.
5.8. General procedure for preparation of compounds 15ae15o and
17ae17f
5.8.5. (E)-4-(2-(2-((1-(2-Fluorobenzyl)-1H-indol-3-yl)methylene)
hydrazinyl)thieno[3,2-d]pyrimidin-4-yl)morpholine (15e)
This compound was obtained as light yellowþsolid in 80% yield.
A mixture of 4-(2-hydrazinylthieno[3,2-d]pyrimidin-4-yl)mor-
pholine 13 (0.12 g, 0.48 mmol), substitued benzyl-1H-indole-3-
carbaldehyde 14ae14o or 1-benzyl-1H-benzo[d]imidazole-2-
carbaldehyde 16ae16f (0.48 mmol) and a drop of glacial acetic
acid in absolute ethanol (8 mL) was refluxed for 4e8 h. The mixture
was cooled, separated by filtration and washed with EtOH to give
off a white to light yellow solid. The crude product was recrystal-
lized from ethanol or purified by chromatography on silica gel using
MeOH/CH2Cl2 to afford the solids 15ae15o and 17ae17f.
m.p. 134e135 ꢀC. ESI-MS m/z: 487.1 [M þ H] . IR (KBr) cmꢁ1
:
3428.8, 3051.1, 2856.3, 1539.6, 1380.4, 1168.4, 1115.0, 1016.1, 888.0,
753.6. 1H NMR (300 MHz, DMSO-d6)
d
: 3.81 (d, J ¼ 4.7 Hz, 4H), 3.94
(d, J ¼ 4.5 Hz, 4H), 5.50 (s, 2H), 7.32e6.87 (m, 6H), 7.41e7.27 (m,1H),
7.50 (d, J ¼ 8.0 Hz, 1H), 7.75 (s, 1H), 8.05 (d, J ¼ 5.5 Hz, 1H), 8.27 (s,
1H), 8.49 (d, J ¼ 7.4 Hz, 1H), 10.51 (s, 1H). Anal. Calcd for
C26H23FN6OS (%): C, 64.18; H, 4.76; N, 17.27. Found (%): C, 64.14; H,
4.74; N, 17.29.
5.8.1. (E)-4-(2-(2-((1-benzyl-1H-indol-3-yl)methylene)hydrazinyl)
thieno[3,2-d]pyrimidin-4-yl)morpholine (15a)
5.8.6. (E)-4-(2-(2-((1-(3-Fluorobenzyl)-1H-indol-3-yl)methylene)
hydrazinyl)thieno[3,2-d]pyrimidin-4-yl)morpholine (15f)
This compound was obtained as off-white sþolid in 85% yield.
m.p. 256e258 ꢀC. ESI-MS m/z: 469.1 [M þ H] . IR (KBr) cmꢁ1
:
This compound was obtained as off-white sþolid in 82% yield.
3437.2, 3171.5, 2964.6, 2854.3, 1537.0, 1364.4, 1343.8, 1008.9, 786.6,
m.p. 248e250 ꢀC. ESI-MS m/z: 486.6 [M þ H] . IR (KBr) cmꢁ1
:
744.5, 727.9. 1H NMR (300 MHz, DMSO-d6)
d
: 3.80 (d, J ¼ 4.7 Hz,
3436.4, 3174.1, 2965.1, 2853.7, 1537.0, 1489.5, 1365.8, 1342.3, 1248.9,
4H), 3.97 (d, J ¼ 4.5 Hz, 4H), 5.43 (s, 2H), 7.37e7.07 (m, 8H), 7.48 (d,
J ¼ 7.8 Hz, 1H), 7.81 (s, 1H), 8.04 (t, J ¼ 6.7 Hz, 1H), 8.28 (s, 1H), 8.49
(d, J ¼ 7.6 Hz, 1H), 10.51 (s, 1H). Anal. Calcd for C26H24N6OS (%): C,
66.64; H, 5.16; N, 17.94. Found (%): C, 66.56; H, 5.18; N, 17.98.
1166.5, 1119.8, 785.0, 744.1. 1H NMR (300 MHz, DMSO-d6)
d: 3.80 (s,
4H), 3.97 (s, 4H), 5.45 (s, 2H), 7.30-6.91 (m, 6H), 7.35 (d, J ¼ 7.2 Hz,
1H), 7.49 (d, J ¼ 8.0 Hz, 1H), 7.84 (s, 1H), 8.06 (d, J ¼ 5.4 Hz, 1H), 8.28
(s, 1H), 8.49 (d, J ¼ 7.3 Hz, 1H), 10.55 (s, 1H). Anal. Calcd for
C26H23FN6OS (%): C, 64.18; H, 4.76; N, 17.27. Found (%): C, 64.11; H,
5.8.2. (E)-4-(2-(2-((1-(2-Chlorobenzyl)-1H-indol-3-yl)methylene)
hydrazinyl)thieno[3,2-d]pyrimidin-4-yl)morpholine (15b)
4.77; N, 17.32.
This compound was obtained as light yellow powder in 87%
yield. m.p. 247e249 ꢀC. ESI-MS m/z: 503.1 (Cl ¼ 35), 505.1 (Cl ¼ 37)
[M þ H]þ. IR (KBr) cmꢁ1: 3435.6, 3253.6, 1539.2, 1372.8, 1274.2,
1172.7, 1120.2, 1013.7, 885.7, 789.7, 734.4. 1H NMR (300 MHz,
5.8.7. (E)-4-(2-(2-((1-(4-Fluorobenzyl)-1H-indol-3-yl)methylene)
hydrazinyl)thieno[3,2-d]pyrimidin-4-yl)morpholine (15g)
This compound was obtained as light yellowþsolid in 79% yield.
m.p. 134e136 ꢀC. ESI-MS m/z: 487.1 [M þ H] . IR (KBr) cmꢁ1
:
DMSO-d6)
d
: 3.81 (d, J ¼ 4.3 Hz, 4H), 3.98 (s, 4H), 5.53 (s, 2H), 6.78
3433.5, 3070.9, 2925.6, 2857.1, 1616.1, 1559.9, 1381.8, 1282.1, 1252.6,
(d, J ¼ 7.5 Hz, 1H), 7.38-7.12 (m, 5H), 7.41 (d, J ¼ 8.1 Hz, 1H), 7.52 (d,
J ¼ 7.8 Hz, 1H), 7.74 (s, 1H), 8.06 (d, J ¼ 5.5 Hz, 1H), 8.33 (s, 1H), 8.53
(d, J ¼ 7.1 Hz, 1H), 10.58 (s, 1H). Anal. Calcd for C26H23ClN6OS (%): C,
62.08; H, 4.61; N, 16.71. Found (%): C, 62.04; H, 4.63; N, 16.65.
1177.0, 1115.9, 1016.8, 888.3, 780.4, 743.6. 1H NMR (300 MHz,
DMSO-d6)
d
: 3.80(d, J ¼ 4.6 Hz, 4H), 3.96 (d, J ¼ 4.7 Hz, 4H), 5.42 (s,
2H), 7.17 (t, J1 ¼ 15.1 Hz, J2 ¼ 7.1 Hz, 5H), 7.31 (dd, J1 ¼ 8.5 Hz,
J2 ¼ 5.6 Hz, 2H), 7.49 (d, J ¼ 7.9 Hz, 1H), 7.81 (s, 1H), 8.05 (d,
J ¼ 5.5 Hz, 1H), 8.27 (s, 1H), 8.48 (d, J ¼ 7.5 Hz, 1H), 10.49 (s, 1H).
Anal. Calcd for C26H23FN6OS (%): C, 64.18; H, 4.76; N, 17.27. Found
(%): C, 64.13; H, 4.74; N, 17.32.
5.8.3. (E)-4-(2-(2-((1-(3-Chlorobenzyl)-1H-indol-3-yl)methylene)
hydrazinyl)thieno[3,2-d]pyrimidin-4-yl)morpholine (15c)
This compound was obtained as off-white solid in 86% yield.
m.p. 238e240 ꢀC. ESI-MS m/z: 503.1 (Cl ¼ 35), 505.1 (Cl ¼ 37)
[M þ H]þ. IR (KBr) cmꢁ1: 3442.3, 3173.5, 2852.7, 1536.9, 1432.7,
1366.3, 1339.7, 1166.0, 1120.3, 785.1, 746.9. 1H NMR (300 MHz,
5.8.8. (E)-3-((3-((2-(4-Morpholinothieno[3,2-d]pyrimidin-2-yl)
hydrazono)methyl)-1H-indol-1-yl)methyl)benzonitrile (15h)
This compound was obtained as pale yellowþpowder in 8ꢁ31%
:
DMSO-d6)
d
: 3.81 (d, J ¼ 4.4 Hz, 4H), 3.98 (s, 4H), 5.45 (s, 2H), 7.27e
yield. m.p. 227e229 ꢀC. ESI-MS m/z: 494.1 [M þ H] . IR (KBr) cm
7.06 (m, 4H), 7.34 (dd, J1 ¼ 12.2 Hz, J2 ¼ 6.1 Hz, 3H), 7.49 (d,
J ¼ 7.9 Hz, 1H), 7.84 (s, 1H), 8.05 (d, J ¼ 5.5 Hz, 1H), 8.28 (s, 1H), 8.50
3443.6, 2854.5, 1539.3, 1372.5, 1172.4, 1118.3, 1016.2, 886.1, 789.5,
737.3. 1H NMR (300 MHz, DMSO-d6)
(s, 2H), 7.31-7.01 (m, 3H), 7.62e7.42 (m, 3H), 7.82e7.65 (m, 2H), 7.86
d: 3.81 (s, 4H), 3.98 (s, 4H), 5.50
(d, J ¼ 7.3 Hz, 1H), 10.53 (s, 1H). 13C NMR (DMSO-d6)
d: 164.31,

172
W. Zhu et al. / European Journal of Medicinal Chemistry 57 (2012) 162e175
(s, 1H), 8.05 (d, J ¼ 5.5 Hz, 1H), 8.28 (s, 1H), 8.50 (d, J ¼ 7.5 Hz, 1H),
10.53 (s, 1H). Anal. Calcd for C27H23N7OS (%): C, 65.70; H, 4.70; N,
19.86. Found (%): C, 65.66; H, 4.69; N, 19.92.
21.39. Anal. Calcd for C27H26N6OS (%): C, 67.20; H, 5.43; N, 17.41.
Found (%): C, 67.05; H, 5.42; N, 17.49.
5.8.14. (E)-4-(2-(2-((1-(4-tert-butylbenzyl)-1H-indol-3-yl)
5.8.9. (E)-4-((3-((2-(4-Morpholinothieno[3,2-d]pyrimidin-2-yl)
hydrazono)methyl)-1H-indol-1-yl)methyl)benzonitrile (15i)
This compound was obtained as light yellow þsolid in 80% yield.
methylene)hydrazinyl)thieno[3,2-d]pyrimidin-4-yl)morpholine (15n)
This compound was obtained as light yellowþsolid in 71% yield.
m.p. 143e145 ꢀC. ESI-MS m/z: 525.2 [M þ H] . IR (KBr) cmꢁ1
:
m.p. 248e250 ꢀC. ESI-MS m/z: 494.1 [M þ H] . IR (KBr) cmꢁ1
:
3428.9, 2960.6, 1652.9, 1540.7, 1378.3, 1248.9, 1170.9, 1014.9, 744.2.
3436.3, 3242.3, 2951.7, 1614.0, 1538.3, 1510.6, 1352.8, 1170.6, 1014.0,
1H NMR (300 MHz, DMSO-d6)
d: 1.22 (s, 9H), 3.81 (s, 4H), 4.00 (s,
883.8, 789.6, 735.1. 1H NMR (300 MHz, DMSO-d6)
d
: 3.81 (s, 4H),
4H), 5.41 (s, 2H), 7.28-7.03 (m, 5H), 7.32 (d, J ¼ 8.1 Hz, 2H), 7.50 (d,
3.98 (s, 4H), 5.72 (s, 2H), 6.96 (d, J ¼ 7.7 Hz, 1H), 7.35-7.09 (m, 3H),
7.47 (dd, J1 ¼12.2 Hz, J2 ¼ 7.5 Hz, 2H), 7.61 (t, J ¼ 7.4 Hz, 1H), 7.77 (s,
1H), 7.91 (d, J ¼ 7.3 Hz, 1H), 8.06 (d, J ¼ 5.5 Hz, 1H), 8.28 (s, 1H), 8.53
(d, J ¼ 7.1 Hz, 1H), 10.54 (s, 1H). Anal. Calcd for C27H23N7OS (%): C,
65.70; H, 4.70; N, 19.86. Found (%): C, 65.62; H, 4.68; N, 19.93.
J ¼ 8.0 Hz, 1H), 7.80 (s, 1H), 8.09 (d, J ¼ 5.4 Hz, 1H), 8.31 (s, 1H), 8.48
(d, J ¼ 7.8 Hz, 1H), 10.50 (s, 1H). 13C NMR (DMSO-d6)
d: 164.46,
164.28, 158.98, 158.74, 150.62, 137.49, 137.17, 135.58, 133.84, 131.40,
127.65 (2C), 126.07 (2C), 125.93, 124.46, 123.27, 120.90, 113.25,
111.14, 106.29, 66.76 (2C), 49.60, 46.57 (2C), 34.93, 31.80 (3C). Anal.
Calcd for C30H32N6OS (%): C, 68.67; H, 6.15; N, 16.02. Found (%): C,
68.59; H, 6.12; N, 16.09.
5.8.10. (E)-4-(2-(2-((1-(2,4-Dichlorobenzyl)-1H-indol-3-yl)
methylene)hydrazinyl)thieno[3,2-d]pyrimidin-4-yl)morpholine (15j)
This compound was obtained as light yellow solid in 88% yield.
m.p. 138e140 ꢀC. ESI-MS m/z: 537.1 (Cl ¼ 35), 538.9 (Cl ¼ 37)
[M þ H]þ. IR (KBr) cmꢁ1: 3436.6, 3069.8, 2856.8, 1559.7, 1382.7,
1252.6, 1177.8, 1115.3, 1016.5, 782.1, 742.8. 1H NMR (300 MHz,
5.8.15. (E)-4-(2-(2-((1-(3-(Trifluoromethyl)benzyl)-1H-indol-3-yl)
methylene)hydrazinyl)thieno[3,2-d]pyrimidin-4-yl)morpholine (15o)
This compound was obtained as light yellowþsolid in 78% yield.
m.p. 136e137 ꢀC. ESI-MS m/z: 537.2 [M þ H] . IR (KBr) cmꢁ1
:
DMSO-d6)
d
: 3.81 (d, J ¼ 4.6 Hz, 4H), 3.97 (d, J ¼ 4.4 Hz, 4H), 5.55 (s,
3435.8, 3071.1, 1619.1, 1560.1, 1384.0, 1334.3, 1110.9, 1075.4, 888.8,
2H), 6.75 (d, J ¼ 8.4 Hz, 1H), 7.27e7.09 (m, 3H), 7.34 (dd, J1 ¼ 8.4 Hz,
J2 ¼ 2.1 Hz, 1H), 7.41 (d, J ¼ 7.3 Hz, 1H), 7.81e7.63 (m, 2H), 8.06 (d,
J ¼ 5.5 Hz, 1H), 8.27 (s, 1H), 8.53 (d, J ¼ 7.1 Hz, 1H), 10.54 (s, 1H).
Anal. Calcd for C26H22Cl2N6OS (%): C, 58.10; H, 4.13; N, 15.64. Found
(%): C, 58.04; H, 4.12; N, 15.70.
781.0, 742.6, 700.8. 1H NMR (300 MHz, DMSO-d6)
d: 3.80 (s, 4H),
3.98 (s, 4H), 5.56 (s, 2H), 7.35e7.02 (m, 3H), 7.79e7.38 (m, 5H),
7.92 (s, 1H), 8.07 (d, J ¼ 5.4 Hz, 1H), 8.29 (s, 1H), 8.50 (d,
J ¼ 7.3 Hz, 1H), 10.59 (s, 1H). 13C NMR (DMSO-d6)
d: 172.79,
158.94, 158.48, 140.10, 137.45, 137.28, 134.00, 131.90, 131.50, 130.51,
125.95, 125.05, 125.0, 124.42, 124.29, 123.52, 123.38, 121.15, 113.58,
111.02, 106.36, 66.76 (2C), 49.30, 46.60 (2C), 21.80. Anal. Calcd for
5.8.11. (E)-4-(2-(2-((1-(2,6-Dichlorobenzyl)-1H-indol-3-yl)
methylene)hydrazinyl)thieno[3,2-d]pyrimidin-4-yl)morpholine(15k)
This compound was obtained as pale yellow solid in 89% yield.
m.p. 239e241 ꢀC. ESI-MS m/z: 537.4 (Cl ¼ 35), 539.3 (Cl ¼ 37)
[M þ H]þ. IR (KBr) cmꢁ1: 3428.4, 2854.2, 1539.7, 1371.6, 1171.8,
C27H23F3N6OS (%): C, 60.21; H, 4.68; N, 15.66. Found (%): C, 60.07;
H, 4.66; N, 15.79.
5.8.16. (E)-4-(2-(2-((1-benzyl-1H-benzo[d]imidazol-2-yl)
methylene)hydrazinyl)thieno[3,2-d]pyrimidin-4-yl)
morpholine(17a)
787.8, 741.7. 1H NMR (300 MHz, DMSO-d6)
d
: 3.80 (d, J ¼ 4.4 Hz, 4H),
3.96 (d, J ¼ 4.2 Hz, 4H), 5.57 (s, 2H), 7.17 (dd, J1 ¼11.7 Hz, J2 ¼ 6.3 Hz,
2H), 7.25 (d, J ¼ 11.2 Hz, 2H), 7.54e7.41 (m, 1H), 7.60 (d, J ¼ 7.7 Hz,
3H), 8.04 (d, J ¼ 5.5 Hz,1H), 8.18 (s,1H), 8.49 (d, J ¼ 7.7 Hz,1H),10.48
(s, 1H). Anal. Calcd for C26H22Cl2N6OS (%): C, 58.10; H, 4.13; N, 15.64.
Found (%): C, 58.05; H, 4.16; N, 15.67.
This compound was obtained as pale yellow þsolid in 68% yield.
m.p. 268e270 ꢀC. ESI-MS m/z: 470.3 [M þ H] . IR (KBr) cmꢁ1
:
3432.6, 3245.7, 1572.9, 1539.4, 1353.4, 1276.8, 1177.5, 888.8, 740.3.
1H NMR (300 MHz, DMSO-d6)
2H), 7.25 (dt, J1 ¼ 12.0 Hz, J2 ¼ 6.6 Hz, 6H), 7.38 (d, J ¼ 7.0 Hz,
2H), 7.48 (d, 5.7 Hz, 1H), 7.79e7.58 (m, 1H), 8.12 (d,
J ¼ 5.5 Hz, 1H), 8.30 (s, 1H), 11.43 (s, 1H). Anal. Calcd for
25H23N7OS (%): C, 63.95; H, 4.94; N, 20.88. Found (%): C,
d: 3.56 (s, 4H), 3.73 (s, 4H), 6.20 (s,
5.8.12. (E)-4-(2-(2-((1-(3,4-Dichlorobenzyl)-1H-indol-3-yl)
J
¼
methylene)hydrazinyl)thieno[3,2-d]pyrimidin-4-yl)morpholine (15l)
This compound was obtained as light yellow solid in 84% yield.
m.p. 208e210 ꢀC. ESI-MS m/z: 537.0 (Cl ¼ 35), 539.1 (Cl ¼ 37)
[M þ H]þ. IR (KBr) cmꢁ1: 3436.4, 2951.2, 1534.2, 1374.4, 1166.4,
C
63.86; H, 4.92; N, 20.94.
1118.1, 1014.0, 747.3. 1H NMR (300 MHz, DMSO-d6)
d: 3.81 (s, 4H),
5.8.17. (E)-4-(2-(2-((1-(2-Chlorobenzyl)-1H-benzo[d]imidazol-2-yl)
methylene)hydrazinyl)thieno[3,2-d]pyrimidin-4-yl)morpholine (17b)
This compound was obtained as pale yellþow solid in 71% yþield.
m.p. 281e283 ꢀC. ESI-MS m/z: 504.2 [M þ H] , 526.2 [M þ Na] . IR
(KBr) cmꢁ1: 3425.4, 2963.6, 2854.4, 1571.3, 1538.0, 1443.0, 1361.6,
3.98 (s, 4H), 5.46 (s, 2H), 7.34e7.10 (m, 4H), 7.50 (d, J ¼ 7.9 Hz, 1H),
7.66e7.54 (m, 2H), 7.85 (s, 1H), 8.06 (d, J ¼ 5.5 Hz, 1H), 8.28 (s, 1H),
8.49 (d,
J
¼
7.4 Hz, 1H), 10.54 (s, 1H). Anal. Calcd for
C
26H22Cl2N6OS (%): C, 58.10; H, 4.13; N,15.64. Found (%): C, 58.02; H,
4.11; N, 15.73.
1177.7, 788.0, 750.3. 1H NMR (300 MHz, DMSO-d6)
d: 3.61 (d,
J ¼ 4.4 Hz, 4H), 3.71 (d, J ¼ 4.3 Hz, 4H), 6.29 (s, 2H), 6.62 (d,
J ¼ 7.3 Hz, 1H), 7.34e7.08 (m, 6H), 7.55 (d, J ¼ 7.9 Hz, 1H), 7.73e7.64
(m, 1H), 8.11 (d, J ¼ 5.5 Hz, 1H), 8.31 (s, 1H), 11.38 (s, 1H). Anal. Calcd
for C25H22ClN7OS (%): C, 59.58; H, 4.40; N, 19.45. Found (%): C,
59.51; H, 4.39; N, 19.52.
5.8.13. (E)-4-(2-(2-((1-(4-Methylbenzyl)-1H-indol-3-yl)methylene)
hydrazinyl)thieno[3,2-d]pyrimidin-4-yl)morpholine (15m)
This compound was obtained as light yellow þsolid in 82% yield.
m.p. 224e226 ꢀC. ESI-MS m/z: 483.1 [M þ H] . IR (KBr) cmꢁ1
:
3436.5, 3048.7, 2922.5, 2857.9, 1561.3, 1382.4, 1252.2, 1176.5, 1115.1,
1017.2, 888.5, 780.5, 737.4. 1H NMR (300 MHz, DMSO-d6)
d: 2.24 (s,
5.8.18. (E)-4-(2-(2-((1-(3-Chlorobenzyl)-1H-benzo[d]imidazol-2-yl)
methylene)hydrazinyl)thieno[3,2-d]pyrimidin-4-yl)morpholine (17c)
This compound was obtained as white solid in 70% yield. m.p.
266e268 ꢀC. ESI-MS m/z: 504.0 [M þ H]þ, 525.8 [M þ Na]þ. IR (KBr)
cmꢁ1: 3438.0, 3246.1, 1573.2, 1539.9, 1486.8, 1442.0, 1353.7, 888.7,
3H), 3.80 (d, J ¼ 4.6 Hz, 4H), 3.96 (d, J ¼ 4.4 Hz, 4H), 5.37 (s, 2H), 5.76
(s,1H), 7.22-6.97 (m, 6H), 7.46 (d, J ¼ 7.9 Hz,1H), 7.78 (s,1H), 8.05 (d,
J ¼ 5.5 Hz, 1H), 8.27 (s, 1H), 8.47 (d, J ¼ 7.5 Hz, 1H), 10.50 (s, 1H). 13
C
NMR (DMSO-d6) d: 164.30, 158.99, 158.75, 137.45, 137.17, 135.48,
133.83, 131.40, 129.86 (2C), 127.93 (2C), 125.97, 124.46, 123.24,
120.90, 113.21, 111.16, 106.25, 66.77 (2C), 55.66, 49.77, 46.57 (2C),
789.4, 740.1. 1H NMR (300 MHz, DMSO-d6)
d: 3.56 (s, 4H), 3.73 (d,
J ¼ 4.0 Hz, 4H), 6.20 (s, 2H), 7.23 (dd, J1 ¼ 11.3 Hz, J2 ¼ 5.8 Hz, 4H),

W. Zhu et al. / European Journal of Medicinal Chemistry 57 (2012) 162e175
173
7.29 (d, J ¼ 6.8 Hz,1H), 7.38 (d, J ¼ 7.1 Hz, 2H), 7.47 (d, J ¼ 4.9 Hz,1H),
5.11. General procedure for preparation of compounds 20ce20k
and 20p
7.75e7.57 (m, 1H), 8.11 (d, J ¼ 5.5 Hz, 1H), 8.31 (s, 1H), 11.42 (s, 1H).
Anal. Calcd for C25H22ClN7OS (%): C, 59.58; H, 4.40; N, 19.45. Found
(%): C, 59.51; H, 4.41; N, 19.51.
They were prepared in a similar procedure as described for 15ae
15o and 17ae17f. According to the procedure used to prepare 15ae
15o and 17ae17f, starting from 14ce14k or 14p, compounds 20ce
20k and 20p were obtained.
5.8.19. (E)-4-(2-(2-((1-(4-Chlorobenzyl)-1H-benzo[d]imidazol-2-yl)
methylene)hydrazinyl)thieno[3,2-d]pyrimidin-4-yl)morpholine (17d)
This compound was obtained as pale yellow þsolid in 69% yield.
m.p. 265e267 ꢀC. ESI-MS m/z: 504.2 [M þ H] . IR (KBr) cmꢁ1
:
5.11.1. (E)-2-(2-(2-((1-(3-Chlorobenzyl)-1H-indol-3-yl)methylene)
hydrazinyl)-4-morpholinothieno[3,2-d]pyrimidin-6-yl)propan-2-ol (20c)
This compound was obtained as light yellowþsolid in 74% yield.
:
3429.2, 3211.7, 2966.1, 2852.4, 1578.5, 1540.0, 1443.5, 1367.7, 1177.3,
1014.2, 898.1, 788.7, 740.3. 1H NMR (300 MHz, DMSO-d6)
d: 3.61 (d,
J ¼ 4.3 Hz, 4H), 3.74 (d, J ¼ 4.4 Hz, 4H), 6.17 (s, 2H), 7.28e7.20 (m,
3H), 7.36 (d, J ¼ 8.5 Hz, 2H), 7.42 (d, J ¼ 8.5 Hz, 2H), 7.54e7.47 (m,
1H), 7.66 (dd, J1 ¼ 5.9 Hz, J2 ¼ 3.0 Hz, 1H), 8.12 (d, J ¼ 5.5 Hz, 1H),
8.30 (s, 1H), 11.43 (s, 1H). Anal. Calcd for C25H22ClN7OS (%): C,
59.58; H, 4.40; N, 19.45. Found (%): C, 59.54; H, 4.37; N, 19.49.
m.p. 233e234 ꢀC. ESI-MS m/z: 561.1 [M þ H] . IR (KBr) cmꢁ1
3406.3, 2973.8, 2855.0, 1578.2, 1516.7, 1382.2, 1275.1, 1169.6, 888.2,
745.7. 1H NMR (300 MHz, DMSO-d6)
d: 1.56 (s, 6H), 3.80 (d,
J ¼ 4.3 Hz, 4H), 3.96 (s, 4H), 5.45 (s, 2H), 5.75 (s, 1H), 7.00 (s, 1H),
7.27-7.09 (m, 3H), 7.34 (dd, J1 ¼ 13.2 Hz, J2 ¼ 7.1 Hz, 3H), 7.49 (d,
J ¼ 8.0 Hz, 1H), 7.82 (s, 1H), 8.26 (s, 1H), 8.50 (d, J ¼ 7.6 Hz, 1H), 10.47
(s,1H). Anal. Calcd for C29H29ClN6O2S (%): C, 62.08; H, 5.21; N,14.98.
Found (%): C, 62.04; H, 5.19; N, 15.02.
5.8.20. (E)-4-(2-(2-((1-(4-tert-butylbenzyl)-1H-benzo[d]imidazol-
2-yl)methylene)hydrazinyl)thieno[3,2-d]pyrimidin-4-yl)morpholine
(17e)
This compound was obtained as light yellow solid in 61% yield.
m.p. 253e255 ꢀC. ESI-MS m/z: 526.5 [M þ H]þ, 548.5 [M þ Na]þ,
565.3 [M þ K]þ. IR (KBr) cmꢁ1: 3432.7, 2963.1, 1576.5, 1544.9,
5.11.2. (E)-2-(2-(2-((1-(4-Chlorobenzyl)-1H-indol-3-yl)methylene)
hydrazinyl)-4-morpholinothieno[3,2-d]pyrimidin-6-yl)propan-2-ol (20d)
This compound was obtained as off-white solid in 75% yield.
1361.0, 1184.6, 744.9. 1H NMR (300 MHz, DMSO-d6)
d: 1.19 (s, 9H),
m.p. 251e252 ꢀC. ESI-MS m/z: 561.1 [M þ H]þ. IR (KBr) cmꢁ1
:
3.53 (s, 4H), 3.71 (d, J ¼ 4.2 Hz, 4H), 6.15 (s, 2H), 7.22 (dd, J1 ¼7.9 Hz,
J2 ¼ 4.5 Hz, 3H), 7.31 (s, 4H), 7.50 (d, J ¼ 8.1 Hz, 1H), 7.64 (dd,
J1 ¼ 6.1 Hz, J2 ¼ 2.7 Hz, 1H), 8.12 (d, J ¼ 5.5 Hz, 1H), 8.30 (s, 1H), 11.44
(s, 1H). Anal. Calcd for C29H31N7OS (%): C, 66.26; H, 5.94; N, 18.65.
Found (%): C, 66.19; H, 5.91; N, 18.67.
3418.2, 3224.3, 1554.8, 1511.0, 1381.4, 1169.6, 885.6, 743.1. 1H NMR
(300 MHz, DMSO-d6) d: 1.56 (s, 6H), 3.79 (s, 4H), 3.95 (s, 4H), 5.43 (s,
2H), 5.75 (s, 1H), 7.00 (s, 1H), 7.22-7.07 (m, 2H), 7.26 (d, J ¼ 8.2 Hz,
2H), 7.38 (d, J ¼ 8.3 Hz, 2H), 7.46 (d, J ¼ 7.7 Hz, 1H), 7.80 (s, 1H), 8.26
(s, 1H), 8.49 (d, J ¼ 7.5 Hz, 1H), 10.45 (s, 1H). 13C NMR (DMSO-d6)
d:
164.41, 163.91, 158.78, 158.58, 137.61, 137.42, 136.81, 132.82, 131.30,
129.73(2C), 129.32(2C), 125.98, 123.39, 121.01, 118.73, 113.56, 111.05,
105.36, 100.99, 70.99, 66.82(2C), 49.20, 46.64(2C), 32.65(2C). Anal.
Calcd for C29H29ClN6O2S (%): C, 62.08; H, 5.21; N, 14.98. Found (%):
C, 62.05; H, 5.19; N, 15.01.
5.8.21. (E)-4-(2-(2-((1-(2,3-Dichlorobenzyl)-1H-benzo[d]imidazol-
2-yl)methylene)hydrazinyl)thieno[3,2-d]pyrimidin-4-yl)morpholine
(17f)
This compound was obtained as pale yellow þsolid in 71% yield.
m.p. 296e298 ꢀC. ESI-MS m/z: 538.0 [M þ H] . IR (KBr) cmꢁ1
:
3425.9, 2964.8, 2854.8, 1569.7, 1537.8, 1363.2, 1179.9, 788.0, 743.9.
5.11.3. (E)-2-(2-(2-((1-(2-Fluorobenzyl)-1H-indol-3-yl)methylene)
hydrazinyl)-4-morpholinothieno[3,2-d]pyrimidin-6-yl)propan-2-ol (20e)
This compound was obtained as white solid in 77% yield. m.p.
218e220 ꢀC. ESI-MS m/z: 568.4 [M þ Na]þ. IR (KBr) cmꢁ1: 3442.4,
1632.5, 1543.4, 1511.4, 1382.5, 1171.8, 1116.7, 745.9. 1H NMR
1H NMR (300 MHz, DMSO-d6)
d
: 3.64 (s, 4H), 3.74 (d, J ¼ 3.8 Hz, 4H),
6.27 (s, 2H), 7.34e7.17 (m, 6H), 7.54 (d, J ¼ 8.1 Hz, 1H), 7.70 (d,
J ¼ 7.9 Hz, 1H), 8.11 (d, J ¼ 5.4 Hz, 1H), 8.30 (d, J ¼ 5.1 Hz, 1H),
11.34 (s, 1H). Anal. Calcd for C25H21Cl2N7OS (%): C, 55.76; H, 3.93;
N, 18.21. Found (%): C, 55.71; H, 3.92; N, 18.26.
(300 MHz, DMSO-d6) d: 1.57 (s, 6H), 3.80 (s, 4H), 3.96 (s, 4H), 5.46 (s,
2H), 5.79 (s, 1H), 7.03 (s, 1H), 7.20 (d, J ¼ 8.4 Hz, 4H), 7.60e7.49 (m,
3H), 7.86 (s, 1H), 8.28 (s, 1H), 8.51 (d, J ¼ 7.4 Hz, 1H), 10.51 (s, 1H).
Anal. Calcd for C29H29FN6O2S (%):C, 63.95; H, 5.37; N, 15.43. Found
(%): C, 63.89; H, 5.39; N, 15.48.
5.9. Preparation of 2-(2-chloro-4-morpholinothieno[3,2-d]
pyrimidin-6-yl)propan-2-ol (18)
To a suspension of 2-chloro-4-morpholine-4-yl-thieno[3,2-d]
pyrimidine (20 g, 0.078 mol) in dry THF (500 mL) at ꢁ78 ꢀC was
added a 2.5 M solution of n-BuLi in hexane (56 mL, 0.14 mol). After
stirring for 1 h, dry acetone (8.5 mL, 0.12 mol) was added. The
reaction mixture was stirred for 2 h at ꢁ78 ꢀC and then warmed
slowly to room temperature. After a further 2 h at room tempera-
ture the mixture poured onto ice/water yielding a yellow precipi-
tate. This was collected by filtration and dried to yield the title
compound (20 g, 83%). m.p.174e176 ꢀC. ESI-MS m/z: 313.2 (Cl ¼ 35),
5.11.4. (E)-2-(2-(2-((1-(3-Fluorobenzyl)-1H-indol-3-yl)methylene)
hydrazinyl)-4-morpholinothieno[3,2-d]pyrimidin-6-yl)propan-2-ol
(20f)
This compound was obtained as whitþe solid in 79% yield. m.p.
243e244 ꢀC. ESI-MS m/z: 545.2 [M þ H] . IR (KBr) cmꢁ1: 3428.8,
3223.5, 1556.3, 1510.9, 1379.5, 1174.1, 745.5. 1H NMR (300 MHz,
DMSO-d6) d: 1.57 (s, 6H), 3.80 (s, 4H), 3.96 (s, 4H), 5.46 (s, 2H), 5.75
(s, 1H), 7.01 (s, 1H), 7.26e7.03 (m, 5H), 7.36 (dd, J1 ¼ 14.0 Hz,
J2 ¼ 8.1 Hz, 1H), 7.49 (d, J ¼ 8.0 Hz, 1H), 7.82 (s, 1H), 8.27 (s, 1H), 8.50
(d, J ¼ 7.4 Hz, 1H), 10.46 (s, 1H). Anal. Calcd for C29H29FN6O2S (%):C,
63.95; H, 5.37; N, 15.43. Found (%): C, 63.90; H, 5.35; N, 15.36.
315.1 (Cl ¼ 37) [M þ H]þ. 1H NMR (300 MHz, DMSO-d6)
d: 1.57 (s,
6H), 3.75 (s, 4H), 3.90 (s, 4H), 5.95 (s, 1H), 7.22 (s, 1H).
5.10. Preparation of 2-(2-hydrazinyl-4-morpholinothieno[3,2-d]
pyrimidin-6-yl)propan-2-ol (19)
5.11.5. (E)-2-(2-(2-((1-(4-Fluorobenzyl)-1H-indol-3-yl)methylene)
hydrazinyl)-4-morpholinothieno[3,2-d]pyrimidin-6-yl)propan-2-ol (20g)
This compound was obtained as off-white sþolid in 73% yield.
According to the procedure used to prepare 13, starting from 18,
key intermediate 19 was obtained as an off-white solid. Yield: 79%.
m.p. 92e93 ꢀC. ESI-MS m/z: 340.1 [M þ H]þ. 1H NMR (300 MHz,
m.p. 225e227 ꢀC. ESI-MS m/z: 545.2 [M þ H] . IR (KBr) cmꢁ1
:
3425.7, 3223.7, 1555.6, 1509.8, 1379.7, 1266.9, 1170.7, 742.6. 1H NMR
DMSO-d6)
d
: 1.53 (s, 6H), 3.71 (s, 4H), 3.80 (s, 4H), 4.05 (s, 2H), 5.71
(300 MHz, DMSO-d6)
d: 1.56 (s, 6H), 3.80 (s, 4H), 3.94 (s, 4H), 5.49 (s,
(s, 1H), 6.94 (s, 1H), 7.42 (s, 1H).
2H), 5.74 (s, 1H), 7.00 (s, 1H), 7.40e7.06 (m, 6H), 7.50 (d, J ¼ 8.1 Hz,

174
W. Zhu et al. / European Journal of Medicinal Chemistry 57 (2012) 162e175
1H), 7.73 (s,1H), 8.25 (s,1H), 8.49 (d, J ¼ 7.6 Hz,1H),10.45 (s,1H). 13
C
[M þ H]þ. IR (KBr) cmꢁ1: 3225.7, 2972.1, 1578.7, 1553.3, 1508.1,
NMR (DMSO-d6)
d
: 164.44, 163.89, 158.75, 158.60, 137.46, 136.78,
1377.6, 1174.6, 771.3, 744.8. 1H NMR (300 MHz, DMSO-d6)
d: 1.57 (s,
131.24, 130.66, 130.58, 130.36, 130.31, 125.45, 123.43, 121.04, 118.80,
116.42,116.11,113.58,110.83,105.34,100.99, 70.99, 66.81(2C), 48.85,
46.64(2C), 32.63(2C). Anal. Calcd for C29H29FN6O2S (%): C, 63.95; H,
5.37; N, 15.43. Found (%): C, 63.89; H, 5.34; N, 15.49.
6H), 3.81 (d, J ¼ 4.3 Hz, 4H), 3.96 (d, J ¼ 4.2 Hz, 4H), 5.58 (s, 2H), 5.75
(s,1H), 6.60 (d, J ¼ 7.6 Hz,1H), 7.01 (s,1H), 7.28e7.15 (m, 3H), 7.41 (d,
J ¼ 7.4 Hz, 1H), 7.58 (d, J ¼ 7.8 Hz, 1H), 7.73 (s, 1H), 8.26 (s, 1H), 8.54
(d, J ¼ 7.0 Hz, 1H), 10.48 (s, 1H). 13C NMR (DMSO-d6)
d: 164.41,
163.92, 158.79, 158.58, 138.62, 137.65, 136.65, 132.85, 131.37, 130.44,
130.33, 129.20, 127.39, 125.87, 123.50, 121.22, 118.73, 113.93, 110.93,
105.40, 101.00, 70.99, 66.82(2C), 48.37, 46.64(2C), 32.65(2C). Anal.
Calcd for C29H28Cl2N6O2S (%): C, 58.49; H, 4.74; N, 14.11. Found (%):
C, 58.42; H, 4.72; N, 14.17.
5.11.6. (E)-3-((3-((2-(6-(2-Hydroxypropan-2-yl)-4-morpholinothieno
[3,2-d]pyrimidin-2-yl)hydrazono)methyl)-1H-indol-1-yl)methyl)
benzonitrile (20h)
This compound was obtained as light yellow þsolid in 73% yield.
m.p. 232e234 ꢀC. ESI-MS m/z: 552.1 [M þ H] . IR (KBr) cmꢁ1
:
3427.9, 3231.2, 1556.3, 1508.9, 1380.4, 1171.5, 746.3. 1H NMR
5.12. Cytotoxicity assay in vitro
(300 MHz, DMSO-d6) d: 1.57 (s, 6H), 3.80 (s, 4H), 3.96 (s, 4H), 5.50 (s,
2H), 5.75 (s, 1H), 7.01 (s, 1H), 7.28e7.08 (m, 2H), 7.63e7.40 (m, 3H),
7.75 (d, J ¼ 7.8 Hz, 2H), 7.84 (s, 1H), 8.28 (s, 1H), 8.51 (d, J ¼ 7.5 Hz,
1H), 10.47 (s, 1H). Anal. Calcd for C30H29N7O2S (%): C, 65.32; H, 5.30;
N, 17.77. Found (%): C, 65.19; H, 5.28; N, 17.84.
The cytotoxic activities of compounds (15ae15o, 17ae17f, 20ce
20k, 20p) were evaluated with H460, HT-29, MDA-MB-231, U87MG
and H1975 cell lines by the standard MTT assay in vitro, with
compound 5, PAC-1 and GDC-0941 as the positive controls. The
most promising compound 15f was further evaluated against four
cancer cell lines A549, H226, HepG2 and SGC-7901. The cancer cell
lines were cultured in minimum essential medium (MEM)
supplement with 10% fetal bovine serum (FBS).
5.11.7. (E)-4-((3-((2-(6-(2-Hydroxypropan-2-yl)-4-morpholinothieno
[3,2-d]pyrimidin-2-yl)hydrazono)methyl)-1H-indol-1-yl)methyl)
benzonitrile (20i)
This compound was obtained as light yellþow solid in 75% yþield.
m.p. 156e158 ꢀC. ESI-MS m/z: 552.2 [M þ H] , 574.2 [M þ Na] . IR
(KBr) cmꢁ1: 3420.7, 2973,4, 2855.5, 1667.7, 1578.8, 1517.4, 1384.5,
Approximately 4 ꢃ 103 cells, suspended in MEM medium, were
plated onto each well of a 96-well plate and incubated in 5% CO2 at
37 ꢀC for 24 h. The test compounds at indicated final concentrations
were added to the culture medium and the cell cultures were
continued for 72 h. Fresh MTT was added to each well at a terminal
1275.9, 1168.9, 1115.1, 765.9, 746.2. 1H NMR (300 MHz, DMSO-d6)
d:
1.57 (s, 6H), 3.81 (s, 4H), 3.96 (s 4H), 5.68 (s, 2H), 5.75 (s, 1H), 7.07e
6.88 (m, 2H), 7.31-7.10 (m, 2H), 7.47 (dd, J1 ¼ 13.8 Hz, J2 ¼ 7.3 Hz,
2H), 7.61 (t, J ¼ 7.6 Hz, 1H), 7.75 (s, 1H), 7.91 (d, J ¼ 7.5 Hz, 1H), 8.27
(s, 1H), 8.54 (d, J ¼ 7.2 Hz, 1H), 10.48 (s, 1H). Anal. Calcd for
concentration of 5
The formazan crystals were dissolved in 100
m
g/mL and incubated with cells at 37 ꢀC for 4 h.
L DMSO each well,
m
and the absorbency at 492 nm (for absorbance of MTT formazan)
and 630 nm (for the reference wavelength) was measured with the
ELISA reader. All of the compounds were tested three times in each
of the cell lines. The results expressed as IC50 (inhibitory concen-
tration 50%) were the averages of three determinations and
calculated by using the Bacus Laboratories Incorporated Slide
Scanner (Bliss) software.
C
30H29N7O2S (%): C, 65.32; H, 5.30; N, 17.77. Found (%): C, 65.21; H,
5.28; N, 17.80.
5.11.8. (E)-2-(2-(2-((1-(2,4-Dichlorobenzyl)-1H-indol-3-yl)
methylene)hydrazinyl)-4-morpholinothieno[3,2-d]pyrimidin-6-yl)
propan-2-ol (20j)
This compound was obtained as off-white solid in 76% yield.
m.p. 231e233 ꢀC. ESI-MS m/z: 595.1 (Cl ¼ 35), 597.1 (Cl ¼ 37)
[M þ H]þ. IR (KBr) cmꢁ1: 3293.5, 3196.2, 2924.1, 2927.9, 2853.4,
1545.7, 1513.6, 1375.2, 1173.4, 1123.6, 741.2. 1H NMR (300 MHz,
5.13. 3D-QSAR on H460 cell line
5.13.1. Data set
DMSO-d6) d: 1.57 (s, 6H), 3.80 (s, 4H), 3.96 (s, 4H), 5.51 (s, 2H), 5.75
In this study, in vitro cytotoxic data against H460 cells were
selected to construct the 3D-QSAR models, Table 1 list their
structures and biological data. For CoMFA and CoMSIA analysis, 31
compounds with determinate IC50 values of H460 cell line were
employed and their IC50 values were converted to pIC50 values
according to the following formula:
(s, 1H), 6.75 (d, J ¼ 8.4 Hz, 1H), 7.01 (s, 1H), 7.27e7.11 (m, 2H), 7.34
(d, J ¼ 8.3 Hz, 1H), 7.41 (d, J ¼ 7.6 Hz, 1H), 7.70 (d, J ¼ 5.5 Hz, 2H),
8.26 (s, 1H), 8.53 (d, J ¼ 6.7 Hz, 1H), 10.47 (s, 1H), Anal. Calcd for
C
29H28Cl2N6O2S (%): C, 58.49; H, 4.74; N, 14.11. Found (%): C, 58.42;
H, 4.71; N, 13.97.
5.11.9. (E)-2-(2-(2-((1-(2,6-Dichlorobenzyl)-1H-indol-3-yl)
methylene)hydrazinyl)-4-morpholinothieno[3,2-d]pyrimidin-6-yl)
propan-2-ol (20k)
1
pIC50 ¼ lg
IC50
This compound was obtained as white solid in 73% yield. m.p.
241e243 ꢀC. ESI-MS m/z: 595.4 (Cl ¼ 35), 597.3 (Cl ¼ 37) [M þ H]þ.
IR (KBr) cmꢁ1: 3422.1, 2970.1, 1550.5, 1510.5, 1377.8, 1172.3, 744.6.
5.13.2. Structure optimization and alignment
Construction and optimization of the compounds, CoMFA and
CoMSIA modeling were all performed by the molecular modeling
software SYBYL 6.91 (TRIPOS Associates Inc.) on Silicon Graphics
workstation. The alignment of the molecules was derived by using
SKETCH in SYBYL. Energy minimization were performed using
Tripos force field and conjugate gradient method with a conver-
1H NMR (300 MHz, DMSO-d6)
d: 1.57 (s, 6H), 3.80 (s, 4H), 3.96 (s,
4H), 5.58 (s, 2H), 5.75 (s, 1H), 6.60 (d, J ¼ 7.7 Hz, 1H), 7.01 (s, 1H),
7.33e7.08 (m, 3H), 7.41 (d, J ¼ 7.7 Hz, 1H), 7.58 (d, J ¼ 8.0 Hz, 1H),
7.73 (s, 1H), 8.26 (s, 1H), 8.54 (d, J ¼ 7.1 Hz, 1H), 10.48 (s, 1H).
Anal. Calcd for C29H28Cl2N6O2S (%): C, 58.49; H, 4.74; N, 14.11.
Found (%): C, 58.39; H, 4.71; N, 14.18.
ꢀ
gence criterion of 0.05 kcal/(mol$A). Charges were calculated by the
Gasteiger-hücker method in the software. Subsequently, all
compounds were aligned using a 2-hydrazinyl-4-morpholino-
thieno[3,2-d]pyrimidine nucleus as a template, which is a common
substructure in all molecules and the minimum scaffold required
for active molecules, and compound 15f with the highest pIC50
value was chosen as the template molecule [21e22].
5.11.10. (E)-2-(2-(2-((1-(2,3-Dichlorobenzyl)-1H-indol-3-yl)
methylene)hydrazinyl)-4-morpholinothieno[3,2-d]pyrimidin-6-yl)
propan-2-ol (20p)
This compound was obtained as off-white solid in 75% yield.
m.p. 243e245 ꢀC. ESI-MS m/z: 595.0 (Cl ¼ 35), 597.1 (Cl ¼ 37)

W. Zhu et al. / European Journal of Medicinal Chemistry 57 (2012) 162e175
175
5.13.3. CoMFA and CoMSIA studies
Appendix A. Supplementary material
CoMFA calculation was carried out using default values. The
steric (Lennard-Jones) and electrostatic (Coulomb) fields were
calculated at each intersection on the regularly spaced grid. 3D
Supplementary data related to this article can be found at http://
dx.doi.org/10.1016/j.ejmech.2012.09.002.
ꢀ
ꢀ
cubic lattice with grid spacing of 2 A and extending to 4 A beyond
the aligned molecules in all directions was used. Steric and elec-
trostatic fields energies were calculated using a sp3 carbon as the
steric probe atom and þ1 charge (default probe atom in SYSYL) for
the electrostatic probe with the cutoff energy of 30 kcal/mol.
CoMSIA similarity index descriptors were derived using the
same lattice box as that used in CoMFA calculations. The standard
settings were used in CoMSIA to calculate five different fields viz.
steric, electrostatic, hydrophobic, acceptor and donor. The default
value of 0.3 was used as the attenuation factor (R). The statistical
evaluation for the CoMSIA analyses were performed in the same
way as described for CoMFA.
References
[1] http://www.who.int/mediacentre/factsheets/fs297/en/ (WHO Media Centre
Cancer Feb 8, 2012, last accessed: 26.03.12).
[2] M. Lindvall, C. McBride, M. McKenna, T.G. Gesner, A. Yabannavar, K. Wong,
S. Lin, A. Walter, C.M. Shafer, ACS Med. Chem. Lett. 2 (2011) 720e723.
[3] Y. Ni, A. Gopalsamy, D. Cole, Y. Hu, R. Denny, M. Lpek, J. Liu, J. Lee, J.P. Hall,
M. Luong, J.B. Telliez, L.L. Lin, Bioorg. Med. Chem. Lett. 21 (2011) 5952e5956.
[4] T.P. Heffron, M. Berry, G. Castanedo, C. Chang, I. Chuckowree, J. Dotson, A. Folkes,
J. Gunzner, J.D. Lesnick, C. Lewis, S. Mathieu, J. Nonomiya, A. Olivero, J. Pang,
D. Peterson, L. Salphati, D. Sampath, S. Sideris, D.P. Sutherlin, V. Tsui, N.C. Wan,
S. Wang, S. Wong, B. Zhu, Bioorg. Med. Chem. Lett. 20 (2010) 2408e2411.
[5] D.P. Sutherlin, D. Sampath, M. Berry, G. Castanedo, Z. Chang, I. Chuckowree,
J. Dotson, A. Folkes, L. Friedman, R. Goldsmith, T. Heffron, L. Lee, J. Lesnick,
C. Lewis, S. Mathieu, J. Nonomiya, A. Olivero, J. Pang, W.W. Prior, L. Salphati,
S. Sideris, Q. Tian, V. Tsui, N.C. Wan, S. Wang, C. Wiesmann, S. Wong, B. Zhu,
J. Med. Chem. 53 (2010) 1086e1097.
[6] N.S. Shetty, R.S. Lamani, I.A.M. Khazi, J. Chem. Sci. 121 (2009) 301e307.
[7] A.B.A. El-gazzar, H.A.R. Hussein, H.N. Hafez, Acta Pharm. 57 (2007) 395e411.
[8] A.J. Folkes, K. Ahmadi, W.K. Alderton, S. Alix, S.J. Baker, G. Box,
I.S. Chuckowree, P.A. Clarke, P. Depledge, S.A. Eccles, L.S. Friedman, A. Hayes,
T.C. Hancox, A. Kugendradas, L. Lensun, P. Moore, A.G. Olivero, J. Pang, S. Patel,
G.H. Pergl-Wilson, F.I. Raynaud, A. Robson, N. Saghir, L. Salphati, S. Sohal,
M.H. Ultsch, M. Valenti, H.J. Wallweber, N.C. Wan, C. Wiesmann, P. Workman,
A. Zhyvoloup, M.J. Zvelebil, S.J. Shuttleworth, J. Med. Chem. 51 (2008) 5522e
5532.
[9] T.T. Junttila, R.W. Akita, K. Parsons, C. Fields, G.D. Lewis Phillips, L.S. Friedman,
D. Sampath, M.X. Sliwkowski, Cancer Cell 15 (2009) 429e440.
[10] D. Sarker, R. Kristeleit, K.E. Mazina, J.A. Ware, Y. Yan, M. Dresser, M.K. Derynck,
J. De-Bono, J. Clin. Oncol. 27 (15S) (2009) 3538 (Meeting Abstracts).
[11] P.J. Rohan, P. Davis, C.A. Moskaluk, M. Kearns, H. Krutzsch, U. Siebenlist,
K. Kelly, Science 259 (1993) 1763e1766.
[12] K.A. Menear, S. Gomez, K. Malagu, Bioorg. Med. Chem. 19 (2009) 5898e5901.
[13] G. Zhang, Y. Liu, S. Wang, C. Zhou, Q. Huang, P. Gong, Arch. Pharm. Chem. Life
Sci. 345 (2012) 49e56.
5.13.4. Partial least squares (PLS) analysis
The CoMFA or CoMSIA descriptors were used as independent
variables and pIC50 values as dependant variables in partial least
square regression analysis. SAMPLS leave-one-out (LOO) Cross-
validated PLS was performed to determine the optimal number of
components to be used in the subsequent final analysis. Column
filtering value was set 2.0 kcal/mol by omitting those lattice points
those energy variation is below this threshold to improve the
signal-to-noise ratio. Finally, no-cross-validated analysis was per-
formed and employed to analyze the result of the CoMFA and
CoMSIA models [22e23]. The predictive ability of the models is
expressed by conventional correlation coefficient r2 value, which is
analogous to cross-validated r2 (q2). The cross-validated coefficient
q
2 (or r2) was evaluated as:
ꢀ
ꢁ
2
2
P
Ypred ꢁ Yexp
q2 ¼ 1 ꢁ
[14] G.G. Zhang, Y.J. Liu, X.G. Ma, H. Dong, J. Li, P. Gong, Chin. Chem. Lett. 22 (2011)
1223e1225.
[15] W. Guo, S.H. Wu, L. Wang, R. Wang, X. Wei, J. Liu, B. Fang, Mol. Cancer Ther. 8
(2009) 441e448.
[16] D. Kumar, N.M. Kumar, K.H. Chang, R. Gupta, K. Shah, Bioorg. Med. Chem. Lett.
21 (2011) 5897e5900.
[17] D. Kumar, N.M. Kumar, S. Ghosh, K. Shah, Bioorg. Med. Chem. Lett. 22 (2012)
212e215.
ꢂ
ꢃ
P
Yexp ꢁ Ymean
where Ypred, Yexp, Ymean are the predicted, experimental, and mean
values of the pIC50, respectively.
The standard error of estimate SE and the Fisher test F value
were also derived to evaluate the predictive quality of the models.
[18] K.K. Roy, S. Singh, S.K. Sharma, R. Srivastava, V. Chaturvedi, A.K. Saxena,
Bioorg. Med. Chem. Lett. 21 (2011) 5589e5593.
[19] C. Garino, S. Ghiani, R. Gobetto, C. Nervi, L. Salassa, G. Croce, M. Milanesio,
E. Rosenberg, J.B. Alexander Ross, Eur. J. Inorg. Chem. 2006 (2006) 2885e2893.
[20] T.H. Quang, T.T. Ha, C.V. Minh, P.V. Kiem, H.T. Huong, N.T.T. Ngan, N.X. Nhiem,
N.H. Tung, B.H. Tai, D.T.T. Thuy, S.B. Song, H. Kang, Y.H. Kim, Bioorg. Med.
Chem. 19 (2011) 2625e2632.
[21] R. Cao, X. Guan, B. Shi, Z. Chen, Z. Ren, W. Peng, H. Song, Eur. J. Med. Chem. 45
(2010) 2503e2515.
[22] A. Politi, S. Durdagi, P. Moutevelis-Minakakis, G. Kokotos, M.G. Papadopoulos,
T. Mavromoustakos, Eur. J. Med. Chem. 44 (2009) 3703e3711.
[23] A.Y. Shaw, C.Y. Chang, M.Y. Hsu, P.J. Lu, C.N. Yang, H.L. Chen, C.W. Lo,
C.W. Shiau, M.K. Chern, Eur. J. Med. Chem. 45 (2010) 2860e2867.
Acknowledgments
We thank professor Maosheng Cheng and Dr Jian Wang for
offering help in 3D-QSAR analysis. This work was supported by
a grant from National S & T Major Project (No. 2009ZX09301-012)
and National Natural Science Foundation of China (NSFC No.
21002065).