Nonclassical antifolates, part 5. Benzodiazepine analogs as a new class of DHFR inhibitors: Synthesis, antitumor testing and molecular modeling study

Article abstract of DOI:10.1016/j.ejmech.2014.01.004

A new series of tetrahydro-quinazoline and tetrahydro-1H-dibenzo[b,e][1,4] diazepine analogs were synthesized and tested for their DHFR inhibition and in vitro antitumor activity. Compound 35 showed a remarkable DHFR inhibitory potency (IC₅₀, 0.004 μM) which is twenty fold more active than methotrexate (MTX). Compounds 17 and 23 proved to be fifteen fold more active than the known antitumor 5-FU, with MG-MID GI₅₀, TGI, and LC ₅₀ values of 1.5, 46.8, 93.3 and 1.4, 17.4, 93.3 μM, respectively. Computer modeling studies allowed the identification that methoxy and methyl substituents, the π-system of the chalcone core, the nitrogen atoms, on the dibenzodiazepine ring as pharmacophoric features essential for activity. These mark points could be used as template model for further future optimization.

Full text of DOI:10.1016/j.ejmech.2014.01.004

European Journal of Medicinal Chemistry 74 (2014) 234e245
Contents lists available at ScienceDirect
European Journal of Medicinal Chemistry
journal homepage: http://www.elsevier.com/locate/ejmech
Original article
Nonclassical antifolates, part 5. Benzodiazepine analogs as a new class
of DHFR inhibitors: Synthesis, antitumor testing and molecular
modeling studyq
,
b
c
Hussein I. El-Subbagh a , Ghada S. Hassan , Shahenda M. El-Messery ,
*
Sarah T. Al-Rashood d, Fatmah A.M. Al-Omary d, Yasmin S. Abulfadl e, Marwa I. Shabayek e
a Department of Pharmaceutical Chemistry, Faculty of Pharmaceutical Sciences & Pharmaceutical Industries, Future University, 12311 Cairo, Egypt
b Department of Medicinal Chemistry, Faculty of Pharmacy, Mansoura University, P.O. Box 35516, Mansoura, Egypt
c Department of Pharmaceutical Organic Chemistry, Faculty of Pharmacy, Mansoura University, P.O. Box 35516, Mansoura, Egypt
d Department of Pharmaceutical Chemistry, College of Pharmacy, King Saud University, P.O. Box 2457, Riyadh 11451, Saudi Arabia
e Department of Pharmacology (Biochemistry Section), Faculty of Pharmaceutical Sciences & Pharmaceutical Industries, Future University, 12311 Cairo,
Egypt
a r t i c l e i n f o
a b s t r a c t
Article history:
A new series of tetrahydro-quinazoline and tetrahydro-1H-dibenzo[b,e][1,4]diazepine analogs were
synthesized and tested for their DHFR inhibition and in vitro antitumor activity. Compound 35 showed a
Received 15 November 2013
Received in revised form
3 January 2014
Accepted 6 January 2014
Available online 13 January 2014
remarkable DHFR inhibitory potency (IC50, 0.004
trexate (MTX). Compounds 17 and 23 proved to be fifteen fold more active than the known antitumor 5-
FU, with MG-MID GI50, TGI, and LC50 values of 1.5, 46.8, 93.3 and 1.4, 17.4, 93.3 M, respectively. Com-
puter modeling studies allowed the identification that methoxy and methyl substituents, the -system of
mM) which is twenty fold more active than metho-
m
p
the chalcone core, the nitrogen atoms, on the dibenzodiazepine ring as pharmacophoric features
essential for activity. These mark points could be used as template model for further future optimization.
Ó 2014 Elsevier Masson SAS. All rights reserved.
Keywords:
Synthesis
Tetrahydro-quinazolines
Dibenzo[b,e][1,4]diazepines
DHFR inhibition
Molecular modeling study
1. Introduction
through DHFR inhibition as non-classical antifolates such as tri-
metrexate (TMQ) and piritrexim (PTX), Chart 1 [5e11]. On the other
Dihydrofolate reductase (DHFR) is the key enzyme in folate
metabolism; it plays a major role in the biosynthesis of nucleic
acids through the catalysis of the NADPH reduction of 7,8-
dihydrofolate to 5,6,7,8-tetrahydrofolate and intimately couples
with thymidylate synthase for purine and pyrimidine production
[1,2]. As a result, DHFR becomes an important target for anticancer
and antimicrobial agents [3]. Pre-clinical and clinical studies
identified a plethora of mechanisms of drug resistance that are the
primary hindrance for the allocation of curative cancer chemo-
therapy. Drug discovery of novel antifolates with improved prop-
erties and superior activities remains an attractive strategy both in
academic research and in the pharmaceutical industry [3,4]. Many
quinazoline analogs are known to possess anticancer activity
hand, benzodiazepine derivatives have been reported to possess
antiviral, antimicrobial, and antitumor activities [12]. Meanwhile,
the
a,b-unsaturated ketone function condensed with variety of
cyclizing agents such as hydrazine hydrate, thiourea, malononitrile,
ethylcyanoacetate, and 2-aminothiazoles produced compounds
possessing broad spectrum antineoplastic activity against variety of
tumor cell lines [13e18].
Recently, a new series of substituted-quinazolin-4-ones was
designed, synthesized, and evaluated for their in vitro DHFR inhi-
bition in our laboratories. This study allowed the allocation of active
DHFR inhibitors with IC50 values around 0.4
mM, bearing aromatic
p-systems. Molecular modeling study of this class of compounds
revealed the importance of the main pharmacophoric groups (the
4-carbonyl fragment, the basic nitrogen atom at N-1, and the hy-
drophobic
p-system regions) as well as of their relative spatial
distances in regard to the quinazoline nucleus. The substitution
q
For parts 1e4 see Refs. [19e22].
pattern and spatial considerations of the
p-systems proved to be
* Corresponding author. Tel./fax: þ20 2 26186111.
critical for DHFR inhibition [19e21].
E-mail address: subbagh@yahoo.com (H.I. El-Subbagh).
0223-5234/$ e see front matter Ó 2014 Elsevier Masson SAS. All rights reserved.
http://dx.doi.org/10.1016/j.ejmech.2014.01.004

H.I. El-Subbagh et al. / European Journal of Medicinal Chemistry 74 (2014) 234e245
235
and Mass spectrometry) data and microanalyses. Starting with the
1H NMR spectra of 2-methylpyrimdines derivatives 13e17, proton
singlets of CH3 group were observed at around 1.21e1.25 ppm. The
only olefinic proton, CH]C appeared as a singlet in the range of
d 7.57e7.60 ppm. The alicyclic and aromatic protons were observed
at their usual chemical shifts. On the other hand, the 1H NMR
spectra of dibenzodiazepine analogs 27e31 showed the appearance
of broad peak for NH group observed at 6.65e6.76 ppm. This peak is
characteristic to the other analogs 32e41, in addition to the
appearance of singlet that refers to the 2 CH3 group of 32e36 at
1.68e1.86 ppm. Elemental microanalysis, in conjunction with MS
confirmed and further differentiated the incorporation of either
halogens such as 27 or methoxy groups as 34, along with the major
protonated molecular ion peak at m/z 520 (9.4 Mþ) and m/z 450
(11.2 Mþ) respectively. 13C NMR spectra were also in accordance to
the depicted structures.
OCH3
OCH3
NH2 CH3
OCH3
OCH3
NH2 CH3
N
N
N
H
OCH3
N
H2N
H2N
N
N
PTX
TMQ
Chart 1. Structures of literature antifolate lead compounds.
In view of these facts, and in continuation to our previous efforts
[19e28], the present study reports an efficient and reproducible
synthesis of new series of tetrahydroquinazoline and tetrahydro-
1H-dibenzo[b,e][1,4]diazepine analogs accommodating
urated imine function (the isosteric analog of the -unsaturated
ketone) in their structures. Those moieties are known to contribute
to the anticancer activity [13e18]. Structure modifications of the
quinazoline nucleus through its partial saturation into tetrahy-
droquinazoline or ring expansion into benzodiazepine have been
performed to define the structure requirements and features that
enhance selectivity and specificity for the tight binding to DHFR
active site. The aim of this study is to identify novel synthetic lead
compound(s) tackling the folate pathway. Compounds possessing
DHFR inhibition activity might be suitable candidates for treating
cancer. The new synthesized derivatives were tested for their
in vitro DHFR inhibition and in vitro antitumor activity using the
NCI’s disease-oriented human cell lines assay [29e32].
a,b-unsat-
a,b
2.2. Dihydrofolate reductase (DHFR) inhibition
The synthesized compounds 7e41 were evaluated as inhibitors
of bovine liver DHFR using reported procedures [34e36]. Results
were reported as IC50 values (Table 1). Compound 35 showed a
remarkable DHFR inhibitory potency (IC50, 0.004
mM) which is
twenty fold more active than methotrexate (MTX, IC50, 0.08
mM)
which used as a positive control. Compounds 30 and 31 proved to
be almost equipotent to MTX activity with IC50 values of 0.06 and
0.09
showed weak to moderate DHFR inhibition with IC50 values range
of 0.1e21.0 M.
mM, respectively (Fig. 1, Table 1). The rest of test compounds
m
2. Results and discussion
2.3. In vitro antitumor screening
2.1. Chemistry
The synthesized compounds 7e41 were subjected to the Na-
tional Cancer Institute (NCI) in vitro disease-oriented human cells
screening panel assay for in vitro antitumor activity. A single dose
The synthetic strategy to prepare the target compounds is
depicted in Scheme 1. The reported intermediate compounds
(2E,6E)-2,6-dibenzylidenecyclohexanone derivatives 7e11 [18]
were prepared by the reaction of cyclohexanone (1) with variety
of substituted benzaldehydes such as 4-bromo- (2), 4-chloro- (3),
4-methoxy (4), 3,4-dimethoxy- (5), 3,4,5-trimethoxy-benzalde-
hyde (6) in ethanolic solution of sodium hydroxide. The target
compounds 2-methyl-tetrahydroquinazolines (13e17) were ob-
(10 mM) of the test compounds were used in the full NCI 60 cell lines
panel assay which includes nine tumor subpanels namely; Leuke-
mia, Non-small cell lung, Colon, CNS, Melanoma, Ovarian, Renal,
Prostate, and Breast cancer cells [29e32]. The data reported as
mean-graph of the percent growth of the treated cells, and pre-
sented as percentage growth inhibition (GI %) caused by the test
compounds (Table 2). Among the tested compounds, only com-
pounds 13e17, 19e23, 28, 29, and 31 exhibited antitumor potency
of various magnitudes.
Compounds 16, 17, 20, 22, 23, 28, and 31 showed remarkable
broad spectrum antitumor potency and considered to be the most
active members in this study. Meanwhile, compounds 13, 14 and 15
displayed modest potency against the tested tumor cell lines and
considered to be the least effective members. Concerning tumor
cell line panel selectivity, compound 13 proved selective towards
Leukemia, Colon, and Breast cancer panels while compound 15
exhibited selectivity towards Leukemia and Melanoma cancer
panels. Regarding potency against individual cell lines, compound
14 proved lethal to Leukemia RPMI-8226 and Colon HCT-116 cancer
cell lines; 19 proved lethal to Leukemia RPMI-8226, Colon HCT-116
and Melanoma LOX IMVI cancer cell lines; while 21 proved lethal to
breast MDA-MB-468. Compound 19 showed 93.4 GI% towards Non-
small cell lung cancer NCI-H522 cell line (Table 2).
tained in considerable yields by reacting the a,b-unsaturated ke-
tones 7e11 with acetamidine HCl (12). Similarly, the reported 2-
amino-tetrahydroquinazolines 19e23 were synthesized by react-
ing 7e11 with guanidine HCl (18). Unlike earlier reports for the
synthesis of 2-aminopyrimidines from chalcones and guanidine
hydrochloride where an oxidizing agent is required to convert
dihydropyrimidines to pyrimidines, in the present study the
aromatization took place spontaneously with simple aerial oxida-
tion [33]. The reaction mechanism suggested to get the target
compounds involves Michael addition of either acetamindine (12),
guanidine (18), or phenylenediamines (24e26) on one of the two
olefinic bonds of enone moiety (7e11) to give an intermediate
followed by reaction between the amine group and the keto group
of enone to give the cyclic products, upon dehydration the dihy-
dropyrimidine skeleton is formed. The latter on tautomerization
aromatizes itself by the aerial oxidation to give finally the target
compounds. On the other hand, the target compounds dibenzo[b,e]
[1,4]diazepines (27e31), 7,8-dimethyl-dibenzo[b,e][1,4]diazepines
(32e36), and 7,8-dichloro-dibenzo[b,e][1,4]diazepines (37e41)
were synthesized by reacting 7e11 with the appropriate
substituted phenylenediamines 24e26 in glacial acetic acid
(Scheme 1, Table 1).
The most active members of this study, compounds 16, 17, 20,
22, 23, 28 and 31passed the primary anticancer assay at an arbitrary
concentration of 10
mM. Consequently, those active compounds
were carried over and tested against a panel of 60 different tumor
cell lines at a 5-log dose range [29e32]. Three response parameters,
GI50, TGI, and LC50 were monitored for each cell line, using the
known drug 5-Fluorouracil (5-FU) as a positive control. Compounds
All compounds were purified by recrystallization and were
characterized on the basis of their spectroscopic (1H NMR, 13C NMR

236
H.I. El-Subbagh et al. / European Journal of Medicinal Chemistry 74 (2014) 234e245
CH3
N
NH2
N
N
N
R1
R2
R1
R2
R1
R2
R1
R2
R3
R3
R3
R3
19-23
13-17
H2N
H2N
H3C
NH
NH
12
AcOH
AcOH
O
H2N
18
O
CHO
R1
R2
R1
R2
NaOH, EtOH
+
R1
R3
R2
R3
R3
2-6
1
7-11
R4
R4
NH2
AcOH
NH2
24-26
R1 = H, OCH3
R2 = Br, Cl, OCH3,
R3 = H, OCH3
R4 = H, CH3, Cl
R4
R4
N
NH
R3
R1
R2
R1
R2
R3
27-41
Scheme 1. Synthesis of the target compounds 13e17, 19e23 and 27e41.
17 and 23 proved to be fifteen fold more active than 5-FU, with MG-
MID GI50, TGI, and LC50 values of 1.5, 46.8, 93.3 and 1.4, 17.4,
group produced compounds 19e23 with diminished DHFR inhibi-
tion activity with IC50 range of 14e21 M. Ring expansion of 16 and
17 produced the dibenzo[b,e][1,4]diazepine derivatives 30 and 31
m
93.3
more active than 5-FU, with MG-MID GI50, TGI, and LC50 values of
2.0, 12.6, 61.7 M, respectively (Fig. 1, Table 3).
mM, respectively; while compound 31 proved to be eleven fold
with remarkable DHFR inhibition potency of 0.06 and 0.09
respectively.
mM,
m
In general, the electron withdrawing halogen atom did not
contribute to the DHFR inhibition activity as evidenced by com-
3. Structureeactivity correlation
pounds 27 (IC50 65.0 mM) and 33 (IC50 70.0 mM). Meanwhile, the
Two series of compounds were synthesized where structure
modifications of the quinazoline nucleus was attained through its
partial saturation into tetrahydroquinazoline (13e17, 19e23) and
ring expansion into dibenzo[b,e][1,4]diazepine (27e31, 32e36, 37e
41) nuclei. Those alterations were performed to define the structure
requirements and features that enhance selectivity and specificity
for the tight binding to DHFR active site. It seems that the type of
substituent attached to those two ring systems manipulate the
biological activity. The 2-methyl-tetrahydroquinazoline analogs 16
and 17 with 8-(dimethoxy- or trimethoxy-benzylidene)- and 4-
(dimethoxy- or trimethoxy-phenyl)-substituents showed DHFR
introduction of electron donating 4-methoxyphenyl function
increased the potency as exemplified by compounds 29 (IC50
0.6 mM) and 34 (IC50 0.1 mM). It is proven that the introduction of di-
methoxy function produced 3,4-dimethoxyphenyl and 3,4-
dimethoxy-benzylidene analogs with a dramatic increase in the
potency as shown in 30 (IC50 0.06
Compound 35 is twenty fold more active than the DHFR inhibitor
methotrexate (MTX, IC50, 0.08 M). The introduction of tri-methoxy
mM) and 35 (IC50 0.004 mM).
m
group gave 3,4,5-trimethoxyphenyl and 3,4,5-trimethoxy-benzyli-
dene derivatives with decreased activity as noticed in 31 and 36.
Concerning the antitumor activity, electron donating methoxy
function proved to contribute to the magnitude of potency rather
than the electron withdrawing halogen atoms. Compounds 20 and
inhibition potency with IC50 values of 0.1 and 0.5
mM, respectively.
Replacement of the 2-methyl function of 16 and 17 by 2-amino

H.I. El-Subbagh et al. / European Journal of Medicinal Chemistry 74 (2014) 234e245
237
Table 1
Physicochemical properties and DHFR inhibition activity results (IC50
,
m
M) of the synthesized compounds 13e41.
R4
R4
CH3
NH2
N
N
N
N
N
NH
R1
R2
R1
R2
R1
R2
R1
R2
R1
R2
R1
R2
R3
R3
R3
R3
R3
R3
13-17
19-23
27-41
Compound
R1
R2
R3
R4
Yield, %
Mp, ꢀ
C
Molecular formulaea
DHFR inhibition
13
14
15
16
17
19
20
21
22
23
27
28
29
30
31
32
33
34
35
36
37
38
39
40
H
H
H
OCH3
OCH3
H
H
H
OCH3
OCH3
H
H
H
OCH3
OCH3
H
H
H
OCH3
OCH3
H
Br
Cl
OCH3
OCH3
OCH3
Br
H
H
H
H
OCH3
H
H
H
H
OCH3
H
H
H
H
OCH3
H
H
H
H
e
e
e
e
e
e
e
e
e
e
H
H
H
H
H
CH3
CH3
CH3
CH3
CH3
Cl
Cl
Cl
Cl
Cl
55
54
68
64
59
48
40
32
52
49
81
72
43
51
59
62
75
66
44
57
50
75
70
66
61
e
191e193
132e135
211e214
176e178
144e146
207e209
187e188
227e229
204e206
220e223
112e115
137e139
188e190
158e160
215e217
133e135
168e170
184e186
140e142
110e113
177e179
154e156
124e126
152e155
162e164
e
C22H18Br2N2
nt
nt
nt
0.1
C22H18Cl2N2
C24H24N2O2
C26H28N2O4
C28H32N2O6
0.5
C
C
21H17Br2N3
21H17Cl2N3
16.0
17.0
18.0
14.0
21.0
65.0
0.3
Cl
OCH3
OCH3
OCH3
Br
C23H23N3O2
C25H27N3O4
C27H31N3O6
C
C
26H20Br2N2
26H20Cl2N2
Cl
OCH3
OCH3
OCH3
Br
C28H26N2O2
C30H30N2O4
C32H34N2O6
C28H24Br2N2
C28H24Cl2N2
C30H30N2O2
C32H34N2O4
C34H38N2O6
0.6
0.06
0.09
21.0
70.0
0.1
0.004
1.0
0.1
12.0
nt
nt
nt
0.08
Cl
OCH3
OCH3
OCH3
Br
OCH3
H
H
H
H
C
C
26H18Br2Cl2N2
26H18Cl4N2
H
H
OCH3
OCH3
e
Cl
OCH3
OCH3
OCH3
e
C28H24Cl2N2O2
C30H28Cl2N2O4
C32H32Cl2N2O6
e
41
OCH3
e
Methotrexate
e
a
Compounds analyzed for C,H,N; results were within ꢁ0.4% of the theoretical values for the formulae given. nt, not tested.
28 with 4-chlorophenyl substitution showed anticancer activity
with MG-MID GI50 values of 9.1 and 4.0 M, respectively. The
introduction of tri-methoxyphenyl group produced compounds 17,
23 and 31 with MG-MID GI50 values of 1.5, 1.4 and, 2.0 M,
respectively; which proved to be the most active antitumor
members of this study.
comparative modeling study of the most active DHFR inhibitor 35
and the least active compound 27 against MTX. The tertiary com-
plex of human dihydrofolate reductase (hDHFR) crystal structure
(pdb ID: 1DLS obtained from the protein data bank), NADPH and
MTX were used as references for modeling and docking [37].
The binding of MTX to hDHFR is considered to be a complex
interaction where hDHFR undergo isomerization and conforma-
tional changes resulting in tight binding to Glu30 [38e42]. The
most active compound 35 and the inactive compound 27 were
docked into the hDHFR binding site as a trial to explain their
different destabilizing activity against DHFR assembly. The amino
acid Lys46 is not one of the key residues involved in the recognition
of the parent ligand MTX but it plays a critical role in the binding of
the most active compound 35. The 2D binding mode of 35 (IC50
m
m
It is worth mentioning that the active anticancer compounds 16,
17, 28 and 31 (MG-MID GI50 14.5,1.5, 4.0, 2.0
to exert their antitumor activity through DHFR inhibition (IC50 0.1,
0.5, 0.3, 0.09 M, respectively). The other active anticancer com-
pounds 20, 22, and 23 (MG-MID GI50 9.1, 5.2, 1.4 M, respectively)
proved to be inactive DHFR inhibitors (IC50 17.0, 14.0, 21.0 M,
mM, respectively) seem
m
m
m
respectively) which suggest that they might exert their antitumor
activity through some other mechanism(s).
0.004 mM) docked and minimized in the hDHFR binding pocket was
4. Molecular modeling study
shown in Fig. 2a, b. Lys46 residue is linked to one of the methoxy
groups of the side chain of 35 which suggest a similar interaction
with the other active compounds 30, 31 in addition to a hydrogen
bonding network, this binding explains the diminished activity of
27 where the methoxy group is replaced by a bromine atom. This
finding showed the necessity of such methoxy group for binding to
The DHFR inhibitory activity of the new synthesized compounds
7e41 was experimentally determined. Compounds 30, 31, and 35
proved to be the most active members in the present study (IC50
0.06, 0.09 and 0.004 M, respectively) compared to the used pos-
itive control MTX (IC50, 0.08 M). Compound 35 in particular,
,
m
m
hDHFR binding pocket. Compound 35 (IC50 0.004
mM) and 27 (IC50
possessed extraordinary potency being twenty fold more active
than MTX. Molecular modeling study was essentially needed to
understand and interpret the unusual DHFR inhibitory pattern of
this new class of compounds. It was interesting to start a
65.0 M) are similar in structure but significantly different in their
m
hDHFR inhibitory activity, each compound has its own unique
feature of binding profile. The most active 35 is located in and
seems to be deeply buried inside the pocket site so it utilizes the

238
H.I. El-Subbagh et al. / European Journal of Medicinal Chemistry 74 (2014) 234e245
NH2
N
CH3
N
N
N
OCH3
OCH3
OCH3
OCH3
CH3O
CH3O
CH3O
CH3O
OCH3
OCH3
CH3O
CH3O
µ
17: (a) IC50 0.5 M; (b) MG-MID GI50, TGI, and
µ
23: (a) IC50 21.0 M; (b) MG-MID GI50, TGI, and
µ
LC50 1.5, 46.8, 93.3
M
µ
LC50 1.4, 17.4, 93.3 M
N
N
NH
NH
CH3O
CH3O
CH3O
OCH3
OCH3
OCH3
OCH3
CH3O
OCH3
CH3O
31
µ
: (a) IC50 0.09 M; (b) MG-MID GI50, TGI, and
30
µ
M
: (a) IC50 0.06
µ
M
LC50 2.0,12.6, 61.7
H3C
N
CH3
NH
CH3O
CH3O
OCH3
OCH3
35
µ
M
: (a) IC50 0.004
Fig. 1. Structures of (a) the active DHFR inhibitors 30, 31 and 35 and (b) the active antitumor agents 17, 23 and 31.
pocket space efficiently. In the contrary, compound 27 seems to
bind in a flipped fashion; one of the two 4-(4-bromo-benzylidene)
moieties is placed into the pocket and the rest of the molecule is
directed outwards from the active site (Fig. 3). Compounds 35 and
36 have a very similar structure and quite different DHFR inhibition
activity. This could be explained through docking experiment
which revealed that 36 and in contrary to 35, could not bind to the
active site of 1DLS through Lys46 leading to the loss of its activity.
Furthermore, 36 binds in a flipped fashion directed outwards from
the active site, similar to the binding mode of the least active
compound 27.
A comparative modeling study of both 35 and 27 against MTX
was performed. Ligand-based active site alignment is a widely
adopted technique for structure analysis of proteineligand com-
plexes. An approved alignment is acceptable when the molecules
have similar shapes; aromatic moieties overlap; and low strain
energy. The objective of computing a multiple ligand alignment for
a set of input ligands is to maximize the similarity among ligands,
while keeping their sound conformations. To probe similarity be-
tween the 3D structures of 35 and MTX, flexible alignment was
performed. MOE/MMFF94 was employed to automatically generate
superposition of the compounds under investigation with minimal
user bias. 200 conformers of each compound were generated and
minimized with a distance-dependant dielectric model. A low en-
ergy set of 100 was selected for further analysis [43,44]. Interest-
ingly, compound 35 adopted nearly the same binding mode in a
similar fashion as MTX (Fig. 4a), while 27 with its (4-bromophenyl)-
deduce specific structural requirements for biological activity
[45,46]. Structure superposition of the active compounds 30, 31 and
35 with their mark points such as methoxy substituents, the
p-
system of the chalcone core, N and NH atoms, methyl groups and
the dibenzodiazepine ring are shown in Fig. 4c. The three com-
pounds aligned fairly well, with almost complete superposition of
ꢀ
the benzodiazepine rings (0.4 A). This superposition suggests the
possible existence of a region on the receptor suited for specific
recognition of the aforementioned selected features. This hypoth-
esis is consistent with the experimental data which strongly sug-
gests that these mark points could be used as template model for
further optimization.
Further investigation was conducted to explore the reasons
behind the diminished DHFR inhibition activity of compound 27, in
comparison to 35. Hydrophobic surface mapping study revealed
that compounds bearing methoxy groups showed more lipophilic
character than their bromo counterparts and hence were able to
achieve more contact with the lipophobic pocket of the enzyme. In
addition, compound 35, pointed out more lipophilic regions
attributed to the presence of two methyl groups (Fig. 5a). Moreover,
35 possess a more flat structure which allow the compound to
adopt conformation that utilizes more hydrophobic space inside
the pocket. On the other hand, the hydrophobic distributions of the
least active compound 27, showed less lipophilic moieties (no
methoxy or methyl groups) and hence the required lipophilicity for
effective binding to DHFR is diminished (Fig. 5b). Meanwhile,
electrostatic potential isosurface maps of both compounds 35 and
27 were obtained. The two molecules showed different regions of
high and low electrostatic potentials although of their structure
resemblance. The electrostatic potential can be mapped onto the
electron density surface by using color to represent the value of the
potential. Colors toward green or blue represent positive potential
tetrahydro-1H-dibenzodiazepine moiety seemed to adopt
a
different angle than that shown for MTX and could not reach as far
into the enzyme pocket (Fig. 4b). Atomic-level details of the
structure of pharmaceutically relevant receptors are not available;
in such cases, 3D superposition of putative ligands can be used to

H.I. El-Subbagh et al. / European Journal of Medicinal Chemistry 74 (2014) 234e245
239
Table 2
Percentage growth inhibition (GI%) of in vitro subpanel tumor cell lines at 10
m
M concentration of compounds 13e17, 19e23, 28, 29, 31.
Tumor cell lines
% Growth inhibition (GI %)a
13
14
15
16
17
19
20
21
22
23
28
29
31
Leukemia
CCRF-CEM
HL-60(TB)
K-562
MOLT-4
RPMI-8226
SR
e
e
e
e
e
L
e
L
L
94.1
L
L
L
L
88.9
L
L
e
e
e
e
84.7
57.7
44.0
46.9
84.8
71.9
97.1
L
93.6
L
L
99.0
e
e
95.7
98.5
94.7
99.4
L
e
e
e
35.2
56.5
35.1
L
32.2
e
e
e
e
e
e
52.8
34.2
L
49.1
e
e
e
e
e
73.7
e
50.1
e
L
54.1
42.3
e
74.3
46.1
e
e
79.2
78.5
97.3
Non-small cell lung cancer
A549/ATCC
EKVX
HOP-62
HOP-92
NCI-H226
NCI-H23
NCI-322M
NCIeH460
NCIeH522
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
82.1
e
83.8
e
e
33.3
e
e
41.2
e
L
e
e
94.1
40.2
L
e
e
44.0
82.2
L
L
88.4
e
e
e
67.9
69.9
L
70.9
52.6
99.4
L
78.0
e
e
e
e
43.0
e
e
e
e
e
31.4
e
39.4
e
e
e
62.4
66.1
58.9
L
e
e
65.9
50.7
e
e
L
91.5
e
e
38.9
e
e
e
e
e
e
e
e
e
e
e
35.4
83.3
L
L
e
e
L
L
48.2
51.6
81.0
e
80.1
93.4
Colon cancer
COLO 205
HCC-2998
HCT-116
HCT-15
HT29
KM12
SW-620
e
e
e
e
e
e
e
e
e
L
L
L
L
L
93.8
L
80.9
88.1
L
95.8
92.1
97.1
L
e
e
e
e
e
e
e
e
e
54.0
e
L
L
L
L
L
L
L
31.9
e
e
99.8
L
L
93.9
L
L
e
58.5
L
51.5
L
L
L
41.7
92.3
97.7
81.1
e
55.8
e
92.1
64.1
75.8
67.9
59.3
L
61.9
e
e
e
36.4
84.1
82.7
58.6
e
45.6
62.0
e
67.8
53.0
e
e
e
72.5
e
e
94.3
CNS cancer
SF-268
SF-295
SF-539
SNB-19
SNB-75
U251
e
e
e
e
e
e
e
e
e
e
e
e
e
80.0
97.0
82.9
39.6
64.8
81.8
78.8
L
86.5
38.8
78.1
90.2
e
e
e
e
e
e
e
e
36.2
e
84.9
L
74.1
38.5
54.3
L
e
e
e
e
e
e
e
74.0
L
64.9
35.6
45.5
L
e
e
32.5
e
e
e
e
e
e
e
e
e
e
e
e
e
e
36.1
62.2
e
44.0
65.2
86.1
79.2
Melanoma
LOX IMVI
MALME-3M
M14
MDA-MB-435
SK-MEL-2
SK-MEL-28
SK-MEL-5
UACC-257
UACC-62
e
e
e
e
e
e
e
e
e
63.2
e
e
L
L
L
98.5
94.1
77.0
93.8
79.8
89.1
L
69.8
L
94.8
73.4
68.1
80.9
56.2
91.0
L
e
L
e
e
87.1
47.5
42.7
69.2
43.2
34.7
60.3
53.8
47.2
L
L
L
L
L
75.1
L
L
L
L
e
42.6
e
L
L
L
96.8
82.1
70.6
L
83.8
L
e
e
34.7
77.7
e
e
38.1
63.2
e
89.0
L
e
70.9
88.2
e
e
40.5
e
e
86.2
e
e
e
e
e
e
e
e
e
e
e
e
e
e
43.8
e
e
e
e
e
e
e
e
e
e
e
e
e
32.1
e
32.0
Ovarian cancer
IGORV1
OVCAR-3
OVCAR-4
OVCAR-5
OVCAR-8
NCI/ADR-RES
SK-OV-3
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
75.6
97.1
79.1
95.3
96.9
88.8
93.0
69.3
72.0
57.8
89.2
94.1
85.6
56.1
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
42.9
89.2
46.7
e
L
L
e
e
e
e
e
e
e
e
L
L
60.9
e
33.0
e
73.1
93.2
L
96.1
92.8
66.5
83.5
81.4
94.1
54.6
e
e
58.8
e
89.4
74.9
55.9
34.3
e
e
e
Renal cancer
786-0
A498
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
90.8
94.1
90.2
94.9
L
94.1
87.8
65.7
82.0
83.7
87.5
88.6
77.4
89.2
55.5
48.9
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
L
99.5
L
L
L
e
e
e
e
e
e
e
e
e
98.2
L
L
93.5
L
L
45.3
e
e
32.8
e
ACHN
72.7
44.2
79.1
42.6
e
CAKI-1
RXF 393
SN12C
TK-10
e
e
65.5
e
34.7
e
L
e
e
e
e
UO-31
32.2
37.1
65.7
e
94.4
Prostate cancer
PC-3
DU-145
e
e
e
e
e
e
68.0
92.7
57.4
95.1
e
e
e
e
e
e
38.5
e
74.1
L
e
e
e
e
61.8
95.1
Breast cancer
MCF7
MDA-MB-231/ATCC
HS 578T
BT-549
T-47D
56.7
e
68.7
e
e
e
e
e
e
e
73.7
92.9
41.0
L
88.8
L
77.0
86.9
51.4
97.1
e
61.4
e
66.9
e
e
e
e
e
e
L
72.4
e
57.0
71.1
63.0
L
72.5
L
64.2
e
44.0
e
63.1
75.9
46.8
L
76.2
L
e
e
e
e
e
e
e
e
34.0
e
e
e
63.9
e
e
e
e
e
e
e
e
MDA-MB-468
e
e
L
e
37.8
L
40.0
56.4
a
GI < 30%; nt, not tested; L, compound proved lethal to the cancer cell line.

240
H.I. El-Subbagh et al. / European Journal of Medicinal Chemistry 74 (2014) 234e245
Table 3
Compounds 16, 17, 20, 22, 23, 28, 31 median growth inhibitory (GI50, mM), total growth inhibitory (TGI, mM) and median lethal (LC50, mM) concentrations of in vitro subpanel
tumor cell lines.
Compound
Activity
I
II
III
IV
V
VI
VII
VIII
IX
MG-MIDa
14.5
16
GI50
TGI
LC50
GI50
TGI
LC50
GI50
TGI
LC50
GI50
TGI
LC50
GI50
TGI
LC50
GI50
TGI
LC50
GI50
TGI
LC50
GI50
TGI
6.8
37.8
28.9
34.2
26.9
25.5
23.7
98.7
96.9
2.2
26.9
15.7
80.6
99.2
2.6
b
b
b
b
92.7
90.2
88.9
87.1
b
b
b
b
b
b
b
b
17
1.1
2.7
0.9
34.5
1.4
1.9
1.8
1.5
b
b
b
b
62.5
67.3
76.2
4.0
12.5
51.3
17.4
75.8
89.8
1.1
2.6
68.3
2.4
6.4
38.8
1.6
3.1
42.0
80.8
46.8
93.3
9.1
33.1
72.5
5.2
61.7
95.5
1.4
17.4
93.3
4.0
18.6
56.2
2.0
12.6
61.7
b
b
b
b
b
b
b
b
20
7.8
15.2
55.7
88.6
22.1
71.1
94.3
1.7
13.7
53.1
91.9
26.7
15.4
36.5
80.0
5.5
73.4
95.6
1.3
16.0
71.9
94.1
14.7
40.6
78.4
7.8
66.7
93.6
1.8
12.3
51.2
95.0
23.6
70.9
92.2
3.9
61.1
b
22
2.1
19.6
5.1
b
b
65.1
85.7
87.3
b
b
b
b
b
23
1.3
3.3
1.9
2.7
2.0
b
b
36.5
42.5
20.0
82.4
20.8
69.2
b
b
b
b
b
b
b
b
28
4.1
9.0
4.9
9.5
4.3
3.3
5.1
4.6
49.4
35.2
81.3
2.8
52.4
84.3
6.8
24.3
64.2
1.7
5.6
71.4
43.5
93.8
2.6
29.1
64.0
2.3
20.0
68.4
61.9
58.2
88.1
2.9
48.2
89.4
b
b
31
1.4
2.3
5.0
28.0
5.2
36.4
94.3
54.2
53.4
b
b
b
b
73.6
b
5-FU
15.1
8.4
72.1
70.6
61.4
45.6
22.7
76.4
22.6
b
b
b
b
b
b
b
b
b
b
b
b
b
b
b
b
b
b
b
b
LC50
I, leukemia; II, non-small cell lung cancer; III, colon cancer; IV, CNS cancer; V, melanoma; VI, ovarian cancer; VII, renal cancer; VIII, prostate cancer; IX, breast cancer.
a
Full panel mean-graph midpoint (
Compounds showed values >100
m
m
M).
M.
b
values, while colors toward red indicate less positive value of the
electrostatic potential (Fig. 5c, d). It is pretty easy to distinguish that
35, have more positive total charge density due to the presence of
two methoxy groups, on the contrary to 27 which showed less
positive total charge due to the absence of both methoxy and
methyl groups (Fig. 5e, f), lacking of such hydrophobic moieties
leading to diminished lipophilicity. The obtained charge distribu-
tion, electrostatic, hydrophobic mapping and conformations
emphasize a distinct different interaction between the active and
the inactive molecules with the potential protein-binding site and
confirm the hypothesis that as the hydrophobicity of the molecule
increases, the interaction to the DHFR binding site increases. The
results of the computer modeling studies indicate striking differ-
ences in the mode of binding, hydrophobic, and electrostatic po-
tential isosurface maps between the most active DHFR inhibitor 35
versus the least active 27, inspite of their structure similarity which
in turn explain their different DHFR inhibition pattern.
5. Conclusion
Compounds 7e41 represent
droquinazoline and tetrahydro-1H-dibenzo[b,e][1,4]diazepine an-
alogs bearing -unsaturated imine function. Compound 35
showed a remarkable DHFR inhibitory potency (IC50, 0.004 M)
a
new series of tetrahy-
a,b
m
which is twenty fold more active than the DHFR inhibitor metho-
trexate (MTX). Compounds 17 and 23 proved to be fifteen fold more
active than 5-FU, with MG-MID GI50, TGI, and LC50 values of 1.5,
46.8, 93.3 and 1.4, 17.4, 93.3 mM, respectively; while compound 31
Fig. 3. (a) The aligned conformations of the most active compound 35 (IC50 0.004
occupying the DHFR binding pocket (orange) and (b) the most inactive compound 27
(IC50 65.0 M) exposing out of the pocket (cyan). (For interpretation of the references
to color in this figure legend, the reader is referred to the web version of this article.)
mM);
Fig. 2. (a) 2D binding mode and residues involved in the recognition for the most
active compound 35 (IC50 0.004
mM); (b) best docking pose for 35 docked and mini-
m
mized in the hDHFR binding pocket.

H.I. El-Subbagh et al. / European Journal of Medicinal Chemistry 74 (2014) 234e245
241
Fig. 4. (a) Flexible alignment of the most active compound 35 (orange) and MTX (gray); (b) the least active compound 27 (cyan), and MTX (gray) and (c) Superposition of the most
active compounds 30 (green), 31 (violet) and 35 (orange). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
Fig. 5. Surface map for (a) the most active compound 35 (orange); (b) the least active compound 27 (cyan). Pink: hydrogen bond, blue: mild polar, green: hydrophobic region.
Electrostatic map of (c) the lowest energy conformers for 35; (d) the lowest energy conformers for 27. Red: hydrophilic region, green: hydrophobic region. Total density charge (e)
for the most active compound 35; (f) the least active compound 27. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this
article.)

242
H.I. El-Subbagh et al. / European Journal of Medicinal Chemistry 74 (2014) 234e245
proved to be eleven fold more active than 5-FU, with MG-MID GI50
,
130.7,131.0,131.3,131.5,132.2,132.3,133.7,134.5,134.6,134.9,135.3,
TGI, and LC50 values of 2.0, 12.6, 61.7 M, respectively. Structure
m
136.9, 138.4, 139.2, 188.8, 191.7.MS m/z (%): 470 (14.2, Mþ). 14: 1H
activity correlation revealed that ring expansion of the tetrahydro-
quinazolines 16 and 17 produced the dibenzo[b,e][1,4]diazepine
derivatives 30 and 31 with remarkable increase in the DHFR inhi-
NMR (DMSO-d6) d 1.25 (s, 3H, CH3) 1.74 (m, 2H, cyclohexaneeH),
2.88 (m, 4H, cyclohexaneeH), 7.52 (d, 4H, J ¼ 8.5, AreH), 7.57 (s, 1H,
CH]C), 7.59 (d, 4H, J ¼ 8.5, AreH), 13C NMR
d 22.2, 23.7, 27.7, 28.0,
34.2, 127.8, 128.6, 131.1, 132.0, 132.1, 132.3, 133.4, 133.5, 133.7, 134.1,
bition potency at concentrations of 0.06 and 0.09
mM, respectively.
Substitution of the said nuclei with electron donating methoxy
function proved to contribute to the magnitude of potency rather
than the electron withdrawing halogen atoms. Computer modeling
134.5, 134.9, 136.8, 138.4, 139.2, 188.7, 191.6. MS m/z (%): 380 (20.2,
Mþ). 15: 1H NMR (DMSO-d6)
d 1.24 (s, 3H, CH3), 1.73 (m, 2H, cyclo-
hexaneeH), 2.89 (m, 4H, cyclohexaneeH), 3.82 (s, 6H, OCH3), 7.02
(d, 4H, J ¼ 8.5, AreH), 7.53 (d, 4H, J ¼ 9, AreH), 7.59 (s, 1H, CH]C).
studies allowed the identification of methoxy substituents, the
p-
system of the chalcone core, N atoms, methyl groups and the
dibenzodiazepine ring as pharmacophoric features essential for
activity. These mark points could be used as template model for
further future optimization.
13C NMR
d 22.5, 24.0, 27.9, 28.2, 34.7, 55.2, 55.4, 113.2, 114.1, 127.9,
128.3, 130.4, 131.5, 132.2, 133.8, 134.2, 134.8, 135.8, 136.2, 136.5,
138.0, 159.7, 188.6, 191.9. MS m/z (%): 372 (12.4, Mþ). 16: 1H NMR
(DMSO-d6)
(m, 4H, cyclohexaneeH), 3.81 (s, 12H, OCH3), 7.05 (d, 4H, J ¼ 8.5,
AreH), 7.15 (s, 2H, AreH), 7.60 (s, 1H, CH]C). 13C NMR
22.4, 24.0,
d 1.21 (s, 3H, CH3), 1.74 (m, 2H, cyclohexaneeH), 2.92
6. Experimental part
d
27.8, 28.2, 34.9, 55.4, 55.5, 56.1, 56.3, 110.9, 111.5, 113.5, 114.1, 123.7,
128.3, 134.3, 135.6, 136.3, 136.8, 147.7, 148.5, 149.1, 149.4, 149.5,
188.5, 191.8. MS m/z (%): 432 (5.8, Mþ). 17: 1H NMR (DMSO-d6)
Melting points (ꢀC) were determined on Mettler FP80 melting
point apparatus and are uncorrected. Microanalyses were per-
formed on a PerkineElmer 240 elemental analyzer at the Central
Research Laboratory, College of Pharmacy, King Saud University. All
of the new compounds were analyzed for C, H and N and agreed
with the proposed structures within ꢁ0.4% of the theoretical
values. 1H, 13C NMR spectra were recorded on a Bruker 500 MHz FT
spectrometer (the Central Research Laboratory, College of Phar-
d
1.25 (s, 3H, CH3), 1.75 (m, 2H, cyclohexaneeH), 2.96 (m, 4H,
cyclohexaneeH), 3.83 (s, 18H, OCH3), 6.87 (s, 2H, AreH), 7.60 (s, 1H,
CH]C). MS m/z (%): 492 (11.9, Mþ).
6.1.2. (4E,4aZ,11Z)-4-(Substituted benzylidene)-11-(substituted
phenyl)-2,3,4,10-tetrahydro-1H-dibenzo[b,e][1,4]diazepines (27e31)
A solution of 1,2-diaminobenzene (24, 1.08 g, 0.01 mol), the
macy, King Saud University); chemical shifts are expressed in d ppm
with reference to TMS. Mass spectral (MS) data were obtained on a
Perkin Elmer, Clarus 600 GC/MS and Joel JMS-AX 500 mass spec-
trometers. Thin layer chromatography was performed on precoated
(0.25 mm) silica gel GF254 plates (E. Merck, Germany), compounds
were detected with 254 nm UV lamp. Silica gel (60e230 mesh) was
employed for routine column chromatography separations. All the
fine chemicals and reagents used were purchased from Aldrich
Chemicals Co, USA. Compounds 7e11, 19e23 were previously re-
ported [18,33]. DHFR inhibition activity experiments were per-
formed at Pharmacology Department, Faculty of Pharmacy; Future
University in Egypt. Bovine liver DHFR enzyme, methotrexate
(MTX) was used in the assay (Sigma Chemical Co, USA). Elisa reader
SN208125 Bio Tek MQX200, Software program GEN5 wave length
340 nm was used to measure the changes in absorbance. In vitro
antitumor testing was conducted at the NCI’s disease-oriented
human cell lines assay facility, Bethesda, MD, USA. Concerning
the molecular modeling study, all experiments were conducted
with MOE software except electrostatic potential and total density
maps were done using Hyperchem 8.05 package from Hypercube
running on a PC computer. Enzyme structure, starting coordinate of
hDHFR enzyme in tertiary complex with reduced-nicotinamide
adenine dinucleotide phosphate (NADPH) and MTX, code ID
1DLS, was obtained from the Protein Data Bank of Brookhaven
National Laboratory [37].
appropriate a,b-unsaturated ketone (7e11, 0.01 mol) in glacial
acetic acid (20 ml) was heated under reflux for 20 h, and continued
as mentioned under compounds 13e17. 27: 1H NMR (DMSO-d6)
d
1.73 (m, 2H, cyclohexaneeH), 2.87 (m, 4H, cyclohexaneeH), 6.72
(brs,1H, NH), 7.39 (d, 2H, J ¼ 8, AreH), 7.49e7.51 (m, 6H, AreH), 7.58
(s, 1H, CH]C), 7.65 (d, 4H, J ¼ 8, AreH). 13C NMR
d 2.5, 22.2, 23.7,
27.7, 28.0, 34.2, 62.9, 122.2, 130.7, 130.9, 131.1, 131.3, 131.5, 132.3,
132.8,133.7,134.5,134.6,134.9,136.9,137.1,138.2,138.5,139.0,188.8,
191.4. MS m/z (%): 520 (9.4, Mþ). 28: 1H NMR (DMSO-d6)
d 1.73 (m,
2H, cyclohexaneeH), 2.88 (m, 4H, cyclohexaneeH), 6.76 (brs, 1H,
NH), 7.36 (d, 2H, J ¼ 8, AreH), 7.51e7.53 (m, 6H, AreH), 7.57 (s, 1H,
CH]C), 7.60 (d, 4H, J ¼ 8, AreH). MS m/z (%): 430 (12.7, Mþ). 29: 1H
NMR (DMSO-d6)
d 1.73 (m, 2H, cyclohexaneeH), 2.89 (m, 4H,
cyclohexaneeH),3.80 (s, 6H, OCH3), 6.76 (brs, 1H, NH), 6.88 (d, 2H,
J ¼ 8, AreH), 7.03(d, 4H, J ¼ 8.5, AreH), 7.54e7.52 (m, 6H, AreH),
7.57 (s, 1H, CH]C). 13C NMR
d 22.5, 23.9, 27.9, 28.2, 34.7, 38.9, 55.1,
55.3, 113.2, 114.1, 127.9, 128.0, 128.6, 129.5, 129.8, 130.4, 131.5, 132.0,
132.2, 133.1, 134.2, 134.8, 135.4, 136.2, 137.5, 159.7, 188.6, 191.2. MS
m/z (%): 422 (3.0, Mþ). 30: 1H NMR (DMSO-d6)
d 1.75 (m, 2H,
cyclohexaneeH), 2.92 (m, 4H, cyclohexaneeH),3.78 (s, 12 H, OCH3),
6.65 (brs, 1H, NH), 6.92 (d, 2H, J ¼ 8, AreH), 7.04 (d, 2H, J ¼ 7.5, Are
H), 7.11e7.19 (m, 4H, AreH), 7.40 (s, 2H, AreH), 7.60 (s, 1H, CH]C).
13C NMR
d 22.5, 24.0, 27.9, 28.2, 34.9, 38.8, 55.4, 55.5, 61.0, 62.9,
110.9, 111.5, 113.4, 114.0, 123.6, 123.8, 128.2, 128.4, 134.3, 135.1, 135.7,
135.9, 136.3, 137.4, 147.7, 148.4, 149.1, 149.5, 188.6, 191.5. MS m/z (%):
6.1. Chemistry
482 (18.5, Mþ). 31: 1H NMR (DMSO-d6)
d 1.86 (m, 2H, cyclohexanee
H), 2.95 (m, 4H, cyclohexaneeH), 3.81 (s, 18 H, OCH3), 6.68 (brs, 1H,
NH), 6.80 (s, 2H, AreH), 6.87 (s, 2H, AreH), 6.97 (s, 2H, AreH), 7.36
(s, 2H, AreH), 7.60 (s, 1H, CH]C). MS m/z (%): 542 (30.4, Mþ).
6.1.1. (E)-8-Substituted benzylidene-2-methyl-4-substituted
phenyl-5,6,7,8-tetrahydroquinazolines (13e17)
A solution of acetamidine HCl (12, 0.94 g, 0.01 mol), the
appropriate diarylidene compounds (7e11, 0.01 mol) in glacial
acetic acid (20 ml) was heated under reflux for 20 h. Solvent was
evaporated under vacuum; the obtained residue was dissolved in
chloroform, washed with water, and the organic layer was sepa-
rated, dried and evaporated. The obtained solid was recrystallized
from ethanol to yield compounds 13e17. 13: 1H NMR (DMSO-d6)
6.1.3. (4E,4aZ,11Z)-7,8-Dimethyl-4-(substituted benzylidene)-11-
(substituted phenyl)-2,3,4,10-tetrahydro-1H-dibenzo[b,e][1,4]
diazepines (32e36)
A solution of 1,2-diamino-4,5-dimethylbenzene (25, 1.36 g,
0.01 mol), the appropriate a,b-unsaturated ketone (7e11, 0.01 mol)
in glacial acetic acid (20 ml) was heated under reflux for 20 h, and
d
1.22 (s, 3H, CH3), 1.73 (m, 2H, cyclohexaneeH), 2.87 (m, 4H,
continued as mentioned under compounds 13e17. 32: 1H NMR
cyclohexaneeH), 7.50 (d, 4H, J ¼ 8, AreH), 7.58 (s, 1H, CH]C), 7.65
(DMSO-d6)
d 1.69 (s, 6H, CH3), 1.71 (m, 2H, cyclohexaneeH), 2.87
(d, 4H, J ¼ 8.5, AreH), 13C NMR
d 22.2, 23.7, 27.7, 34.2, 121.0, 122.2,
(m, 4H, cyclohexaneeH), 6.75 (brs, 1H, NH), 7.32 (s, 1H, AreH), 7.41

H.I. El-Subbagh et al. / European Journal of Medicinal Chemistry 74 (2014) 234e245
243
(s, 1H, AreH), 7.49 (d, 4H, J ¼ 8, AreH), 7.57 (s, 1H, CH]C), 7.65 (d,
4H, J ¼ 8.5, AreH). MS m/z (%): 548 (10.2, Mþ). 33: 1H NMR (DMSO-
aliquots), NADPH (50 mg in 10 ml of buffer (A) in 0.4 ml aliquots),
and DHFR (2.1 U in 10 ml of buffer (B) in 0.5 ml aliquots) stored
ml FH2 reaction solutions, 0.25 ml aliquot of FH2 stock
solution in 8.0 ml of buffer B, yielding a final working concentration
of 0.104 g/l, was added to each well of the 96-well flat-bottom
d6)
d
1.68 (s, 6H, CH3), 1.75 (m, 2H, cyclohexaneeH), 2.87 (m, 4H,
at ꢂ70 ꢀC. 130
cyclohexaneeH), 6.74 (brs, 1H, NH), 7.38 (s, 1H, AreH), 7.50 (s, 1H,
AreH), 7.53 (d, 4H, J ¼ 8.5, AreH), 7.57 (s, 1H, CH]C), 7.58 (d, 4H,
J ¼ 8.5, AreH). 13C NMR
d
22.2, 23.7, 27.7, 28.0, 34.2, 38.8, 62.9,127.8,
microplate reader. 20 ml MTX calibrators or unknown samples
128.5,128.9,130.5,131.0,131.9,132.0,132.1,132.3,133.5,133.6,134.1,
and blank of different concentrations 10ꢂ11 to 10ꢂ5 (adjusted by
DMSO) were added into duplicate wells. The microplate was
134.5,134.8,134.9,136.8,138.1,138.4,139.2,188.8,191.5. MS m/z (%):
458 (5.2, Mþ). 34: 1H NMR (DMSO-d6)
d
1.68 (s, 6H, CH3), 1.76 (m,
shaken in a shaking incubator for 1 min, after which 50 ml NADPH/
2H, cyclohexaneeH), 2.89 (m, 4H, cyclohexaneeH), 3.82 (s, 6H,
OCH3), 6.65 (brs,1H, NH), 6.82 (s,1H, AreH), 7.02 (d, 4H, J ¼ 8.5, Are
H), 7.39 (s, 1H, AreH), 7.52 (d, 4H, J ¼ 8.5, AreH), 7.60 (s, 1H, CH]C).
DHFR reaction solution, 0.4-ml aliquot of NADPH stock solution and
0.5 ml aliquot of DHFR stock solution in 6.0 ml of buffer B, yielding a
final working concentration of 0.29 g NADPH/l and 15u DHFR/l was
added to each well, and the microplate was again shaken in a
shaking incubator for 1 min. The absorbance of each well was read
in the microplate reader at room temperature at wavelengths of
340 nm [34e36], using the kinetic mode with a reading interval of
1 min for duration of 10 min. The changes in absorbance were
downloaded directly into an ELISA reader SN208125 Bio Tek
MQX200 computer and analyzed with software Gen5 wave length
340 nm. Results are reported as % inhibition of enzymatic activity
calculated using the following formula:
MS m/z (%): 450 (11.2, Mþ). 35: 1H NMR (DMSO-d6)
d 1.70 (s, 6H,
CH3), 1.76 (m, 2H, cyclohexaneeH), 2.92 (m, 4H, cyclohexaneeH),
3.82 (s, 12 H, OCH3), 6.65 (brs, 1H, NH), 6.91 (s, 1H, AreH), 7.04 (d,
2H, J ¼ 8.5, AreH), 7.39 (m, 4H, AreH), 7.40 (s,1H, AreH), 7.60 (s,1H,
CH]C). 13C NMR
d 22.5, 24.0, 27.9, 28.2, 34.9, 38.8, 55.4, 55.5, 59.1,
61.2, 110.9, 111.5, 113.4, 114.0, 120.1, 122.9, 123.7, 123.8, 128.1, 128.4,
134.3,135.1, 135.7, 135.9, 136.3,136.9,147.7, 148.4,149.1,149.5, 188.6,
191.9. MS m/z (%): 510 (19.4, Mþ). 36: 1H NMR (DMSO-d6)
d 1.86 (s,
6H, CH3), 2.72 (m, 2H, cyclohexaneeH), 2.95 (m, 4H, cyclohexanee
H), 3.83 (s, 18 H, OCH3), 6.69 (brs, 1H, NH), 6.81 (s, 2H, AreH), 6.87
(s, 2H, AreH), 6.97 (s, 2H, AreH), 7.60 (s, 1H, CH]C). MS m/z (%):
570 (24.1, Mþ).
ꢀ
ꢁ
D
A=mintest
% Inhibition
¼
1 ꢂ
ꢃ 100
A/min) between 0 and
D
A=minDMSO
The linear decrease of absorbance (
D
6.1.4. (4E,4aZ,11Z)-7,8-Dichloro-4-(substituted benzylidene)-11-
(substituted phenyl)-2,3,4,10-tetrahydro-1H-dibenzo[b,e][1,4]
diazepines (37e41)
A solution of 1,2-diamino-4,5-dichlorobenzene (26, 1.75 g,
0.01 mol), the appropriate a,b-unsaturated ketone (7e11, 0.01 mol)
10 min for each MTX or sample was calculated and presented by %
of inhibition then plotted against their concentrations (log scale) to
obtain a calibration curve [36]. The 50% inhibitory concentration
(IC50) of each compound was obtained.
in glacial acetic acid (20 ml) was heated under reflux for 20 h, and
continued as mentioned under compounds 13e17. 37: 1H NMR
6.3. Antitumor screening
(DMSO-d6)
d 1.73 (m, 2H, cyclohexaneeH), 2.87 (m, 4H, cyclo-
Under a sterile condition, cell lines were grown in RPMI 1640
media (Gibco, NY, USA) supplemented with 10% fetal bovine serum
(Biocell, CA, USA), 5 ꢃ 105 cell/ml was used to test the growth in-
hibition activity of the synthesized compounds. The concentrations
hexaneeH), 6.74 (brs,1H, NH), 7.32 (s,1H, AreH), 7.39 (s,1H, AreH),
7.49 (d, 4H, J ¼ 8, AreH), 7.57 (s, 1H, CH]C), 7.65 (d, 4H, J ¼ 8.5, Are
H). 13C NMR
d 22.2, 23.7, 27.7, 28.0, 34.1, 38.6, 62.9, 122.2, 123.4,
123.9, 124.0, 130.7, 131.3, 131.5, 132.3, 133.7, 134.5, 134.9, 135.2,
of the compounds ranging from 0.01 to 100 mM were prepared in
136.5, 136.9, 138.4, 138.8, 139.2, 188.8, 191.4. MS m/z(%): 588 (6.3,
Mþ). 38: 1H NMR (DMSO-d6)
d 1.73 (m, 2H, cyclohexaneeH), 2.88
phosphate buffer saline. Each compound was initially solubilized in
dimethyl sulfoxide (DMSO), however, each final dilution contained
less than 1% DMSO. Solutions of different concentrations (0.2 ml)
were pipetted into separate well of a microtiter tray in duplicate.
Cell culture (1.8 ml) containing a cell population of 6 ꢃ 104 cells/ml
was pipetted into each well. Controls, containing only phosphate
buffer saline and DMSO at identical dilutions, were also prepared in
the same manner. These cultures were incubated in a humidified
incubator at 37 ꢀC. The incubator was supplied with 5% CO2 at-
mosphere. After 48 h, cells in each well were diluted 10 times with
saline and counted by using a coulter counter. The counts were
corrected for the dilution [29e32].
(m, 4H, cyclohexaneeH), 6.74 (brs, 1H, NH), 7.37 (s, 1H, AreH), 7.47
(s, 1H, AreH), 7.49 (d, 4H, J ¼ 8, AreH), 7.53 (s, 1H, CH]C), 7.59 (d,
4H, J ¼ 8, AreH). 13C NMR
d 22.1, 23.7, 27.7, 28.0, 34.2, 38.4, 61.0,
123.0, 124.8, 127.7, 127.9, 128.6, 129.5, 131.0, 131.9, 132.1, 133.4, 134.1,
134.4, 134.9, 135.7, 136.8, 137.5, 139.2, 188.7, 191.5. MS m/z (%): 500
(13.0, Mþ). 39: 1H NMR (DMSO-d6)
d 1.73 (m, 2H, cyclohexaneeH),
2.88 (m, 4H, cyclohexaneeH), 3.80 (s, 6H, OCH3), 6.74 (brs, 1H, NH),
6.89 (s,1H, AreH), 7.03 (d, 4H, J ¼ 8.5, AreH), 7.41 (s,1H, AreH), 7.53
(d, 4H, J ¼ 8.5, AreH), 7.60 (s, 1H, CH]C), MS m/z (%): 490 (2.2, Mþ).
40: 1H NMR (DMSO-d6)
d 1.75 (m, 2H, cyclohexaneeH), 2.91 (m, 4H,
cyclohexaneeH), 3.80 (s, 12 H, OCH3), 6.74 (brs, 1H, NH), 6.93 (s, 1H,
AreH), 7.04 (d, 4H, J ¼ 8.5, AreH), 7.15 (s, 1H, AreH), 7.42 (s, 2H, Are
H), 7.60 (s, 1H, CH]C). 13C NMR
d
22.5, 24.1, 27.9, 28.2, 34.9, 38.9,
6.4. Docking and molecular modeling study
55.4, 55.6, 59.2, 60.4, 110.9, 111.5, 113.4, 114.1, 122.2, 123.7, 124.1,
125.8, 128.1, 130.4, 134.3, 135.1, 135.6, 135.9, 136.3, 136.8, 148.4,
149.5, 188.5, 191.5. MS m/z (%): 550 (14.6, Mþ). 41: 1H NMR (DMSO-
The three-dimensional structures of the substituted dibenzo-
diazepine chalcone derivatives, which presented best and worst
biological profiles, in their neutral forms were constructed using
the MOE of Chemical Computing Group Inc software. Lowest en-
ergy conformer of each new analog ‘global-minima’ was docked
into the hDHFR enzyme-binding domain. Enzyme structure. Start-
ing coordinate of hDHFR enzyme in tertiary complex with reduced-
nicotinamide adenine dinucleotide phosphate (NADPH) and MTX,
code ID 1DLS, was obtained from the Protein Data Bank of Broo-
khaven National Laboratory [37]. All of the hydrogens were added
and enzyme structure was subjected to refinement protocol in
which the constraints on the enzyme were gradually removed and
d6)
d 1.72 (m, 2H, cyclohexaneeH), 2.95 (m, 4H, cyclohexaneeH),
3.83 (s, 18 H, OCH3), 6.63 (brs, 1H, NH), 6.75 (s, 1H, AreH), 6.85
(s, 1H, AreH), 6.97 (s, 2H, AreH), 7.52 (s, 2H, AreH), 7.67
(s, 1H, CH]C). MS m/z (%): 610 (10.3, Mþ).
6.2. Dihydrofolate reductase (DHFR) inhibition assay
The assay mixture contained 0.5 mol/l Tris buffer (A), pH 7.5 and
0.05 mol/l Tris buffer (B), pH 7.5. Stock solutions of FH2 (25 mg in
1.5 ml of 2-mercaptoethanol and 6.0 ml of buffer A in 0.25 ml

244
H.I. El-Subbagh et al. / European Journal of Medicinal Chemistry 74 (2014) 234e245
[5] M.F. Mullarkey, B.A. Blumenstein, W.P. Andrade, G.A. Bailey, I. Olason,
C.E. Weizel, Methotrexate in the treatment of corticosteroid-dependent
asthma. A double-blind crossover study, N. Engl. J. Med. 318 (1988) 603e607.
[6] E.F. Elslager, J.L. Johnson, L.M. Werbel, Folate antagonists 20. Synthesis, anti-
tumor, and antimalarial properties of trimetrexate and related 6-[(phenyl-
amino)methyl]2,4-quinazolin-diamines, J. Med. Chem. 26 (1983) 1753e1760.
[7] E.M. Grivsky, S. Lee, C.W. Sigel, D.S. Duch, C.A. Nichol, Synthesis and antitumor
activity of 2,4-diamino-6(2,5-dimethyloxybenzyl)-5-methyl[2,3-d] pyrimi-
dine, J. Med. Chem. 23 (1980) 327e329.
[8] V. Bavetsias, A.L. Jackman, J.H. Marriott, R. Kimbell, W. Gibson, F.T. Boyle,
G.M. Bisset, Folate-based inhibitors of thymidylate synthase, J. Med. Chem. 40
(1997) 1495e1510.
[9] V. Bavetsias, J.H. Marriott, C. Melin, R. Kimbell, Z.S. Matusiak, F.T. Boyle,
A.L. Jackman, Design and synthesis of cyclopenta[g]quinazoline-based anti-
folate as inhibitors of thymidylate synthase and potential antitumor agents,
J. Med. Chem. 43 (2000) 1910e1926.
[10] C. Sheng-Li, F. Yu-Ping, J. Yu-Yang, L. Shi-Ying, D. Guo-Yu, L. Run-tao, Syn-
thesis and in vitro antitumor activity of 4(3H)-quinazolione derivatives with
dithiocarbamate side chains, Bioorg. Med. Chem. Lett. 15 (2005) 1915e1917.
[11] P.C. Wyss, P. Gerber, P.G. Hartman, C. Hubschwerlen, H. Locher, H. Marty,
M. Stahl, Novel dihydrofolate reductase inhibitors. Structure-based versus
diversity-based library design and high-throughput synthesis and screening,
J. Med. Chem. 46 (2003) 2304e2312.
[12] K.M. El-Shaieb, A.A. Hassan, A.S. Abdel-Aal, Synthesis of dibenzo[b,e][1,4]
diazepine derivatives, J. Chem. Res. 35 (2011) 592e594.
[13] J.R. Dimmock, P. Kumar, A.J. Nazarali, N.L. Motaganahalli, T.P. Kowalchuk,
M.A. Beazely, J.W. Quail, E.O. Oloo, T.M. Allen, J. Szydlowski, E. Declerq,
J. Balzarini, Cytotoxic 2,6-bis-(arylidene)-cyclohexanones and related com-
pounds, Eur. J. Med. Chem. 35 (2000) 967e977.
minimized until the rms gradient was 0.01 kcal/mol A. The energy
minimization was carried out using the molecular mechanics force
field ‘AMBER.’ The energy-minimized structure was used for mo-
lecular modeling studies keeping all the heavy atoms fixed until an
ꢀ
RMSD gradient of 0.05 kcal/mol A was reached. Ligand structures
were built with MOE and minimized using the MMFF94x force field
ꢀ
until an RMSD gradient of 0.05 kcal/mol A was reached. For each of
chalcone analog, energy minimizations (EM) were performed using
1000 steps of steepest descent, followed by conjugate gradient
ꢀ
minimization to an RMS energy gradient of 0.01 kcal/mol A. The
ꢀ
active site of the enzyme was defined using a radius of 10.0 A
around MTX. Energy of binding was calculated as the difference
between the energy of the complex and individual energies of the
enzyme and ligand [43,44,47,48]. The docking was performed using
the Alpha Triangle placement method and the London dG scoring
method. 300 results for each ligand were generated, discarding the
ꢀ
results with an RMSD value >3 A. The best scored result of the
remaining conformations for each ligand was further analyzed. The
protein/ligand complexes were minimized using the MMFF94x
force field, until an RMSD gradient 0.1 kcal mol/Aꢀ was reached.
6.5. Flexible alignment and superposition
[14] J.R. Dimmock, M.P. Podmanilayam, G.A. Zello, K.H. Nienaber, T.M. Allen,
C.L. Santos, E. Declerq, J. Balzarini, E.K. Manavathu, J.P. Stables, Cytotoxic an-
alogues of 2,6-bis(arylidene) cyclohexanones, Eur. J. Med. Chem. 38 (2003)
169e177.
The investigated compounds were subjected to flexible align-
ment and superposition experiments using ‘Molecular Operating
Environment’ software (MOE of Chemical Computing Group Inc., on
a Core 2 duo 2.3 GHz workstation). The molecules were built using
the Builder module of MOE. Their geometry was optimized by using
the MMFF94 force field followed by a flexible alignment using
systematic conformational search. Lowest energy aligned confor-
mation(s) were identified [43,44,47,48].
[15] H.I. El-Subbagh, S.M. Abu-Zaid, M.A. Mahran, F.A. Badria, A.M. Al-Obaid,
Synthesis and biological evaluation of certain
a,b-unsaturated ketones and
their corresponding fused pyridines as antiviral and cytotoxic agents, J. Med.
Chem. 43 (2000) 2915e2921.
[16] B.K. Adams, E.M. Ferstl, M.C. Davis, M. Herold, S. Kurtkaya, R.F. Camalier,
M.G. Hollingshead, G. Kaur, E.A. Sauville, F.R. Rickles, J.P. Synder, D.C. Liotta,
M. Shoji, Synthesis and biological evaluation of novel curcumin analogs as anti-
cancer and antiangiogenesis agents, Bioorg. Med. Chem. 12 (2004) 3871e3883.
[17] S.A.F. Rostom, G.S. Hassan, H.I. Subbagh, Synthesis and biological evaluation of
some polymethoxylated fused pyridine ring systems as antitumor agents,
Arch. Pharm. Chem. Life Sci. 342 (2009) 584e590.
6.6. Electrostatic potential isosurface maps
[18] F.A. Al-Omary, G.S. Hassan, S.M. El-Messery, H.I. El-Subbagh, Substituted thi-
azoles V. Synthesis and antitumor activity of novel thiazolo[2,3-b]quinazoline
and pyrido[4,3-d]thiazolo[3,2-a]pyrimidine analogues, Eur. J. Med. Chem. 47
(2012) 65e72.
[19] S.T. Al-Rashood, I.A. Aboldahab, L.A. Abouzeid, A.A.-M. Abdel-Aziz, M.N. Nagi,
S.G. Abdul-hamide, K.M. Youssef, A.M. Al-Obaid, H.I. El-Subbagh, Synthesis,
dihydrofolate reductase inhibition, and molecular modeling study of
some new 4(3H)-quinazolinone analogues, Bioorg. Med. Chem. 14 (2006)
8608e8621.
[20] F.A.M. Al-Omary, L.A. Abou-zeid, M.N. Nagi, E.E. Habib, A.A.-M. Abdel-Aziz,
A.S. El-Azab, S.G. Abdel-Hamide, M.A. Al-Omar, A.M. Al-Obaid, H.I. El-Subbagh,
Non-classical antifolates. Part 2: synthesis, biological evaluation, and molec-
ular modeling study of some new 2,6-substituted-quinazolin-4-ones, Bioorg.
Med. Chem. 18 (2010) 2849e2863.
Molecular structure of the selected synthesized chalcones was
constructed from fragment libraries in the Hyperchem program
[49]. The partial atomic charges for each analog were assigned with
the semiempirical mechanical calculation method ‘AM1’ [50,51]
implemented in Hyperchem 8.05. Conformational search was
formed around all the rotatable bonds with an increment of 10ꢀ
using conformational search module as implemented in Hyper-
Chem 6.03. All the conformers were minimized until the rms de-
ꢀ
viation was 0.01 kcal/mol A.
[21] F.A.M. Al-Omary, G.S. Hassan, S.M. El-Messery, M.N. Nagi, E.E. Habib, H.I. El-
Subbagh, Nonclassical antifolates, part 3: synthesis, biological evaluation and
molecular modeling study of some new 2-heteroarylthio-quinazolin-4-ones,
Eur. J. Med. Chem. 63 (2013) 33e45.
[22] G.S. Hassan, S.M. El-Messery, F.A.M. Al-Omary, S.T. Al-Rashood, M.I. Shabayek,
Y.S. Abulfadl, E.E. Habib, S.M. El-Hallouty, W. Fayad, K.M. Mohamed, B.S. El-
Menshawi, H.I. El-Subbagh, Nonclassical antifolates, part 4. 5-(2-aminothiazol-
4-yl)-4-phenyl-4H-1,2,4-triazole-3-thiols as a new class of DHFR inhibitors:
synthesis, biological evaluation and molecular modeling study, Eur. J. Med.
Chem. 66 (2013) 135e145.
[23] H.I. El-Subbagh, W.A. El-Naggar, F.A. Badria, Synthesis and biological testing of
2,4-disubstituted thiazole derivatives as potential antitumor antibiotics, Med.
Chem. Res. 3 (1994) 503e516.
[24] H.I. El-Subbagh, A.M. Al-Obaid, 2,4-Disubstituted thiazoles, II. A novel class of
antitumor agents, synthesis and biological evaluation, Eur. J. Med. Chem. 31
(1996) 1017e1021.
[25] H.I. El-Subbagh, A.H. Abadi, J. Lehmann, 2,4-Disustituted thiazoles, III. Syn-
thesis and antitumor activity of ethyl 2-substituted-aminothiazole-4-
carboxylate analogs, Arch. Pharm. Pharm. Med. Chem. 332 (1999) 137e142.
[26] H.I. El-Subbagh, I.E. Al-Khawad, E.R. El-Bendary, A.M. Al-Obaid, Substituted
thiazoles IV. Synthesis and antitumor activity of new substituted imidazo[2,1-
b]thiazole analogs, Saudi Pharm. J. 9 (2001) 14e20.
[27] S.M. El-Messery, G.S. Hassan, F.A. Al-Omary, H.I. El-Subbagh, Substituted thi-
azoles VI. Synthesis and antitumor activity of new 2-acetamido- and 2 or 3-
propanamido-thiazole analogs, Eur. J. Med. Chem. 54 (2012) 615e625.
[28] G.S. Hassan, S.M. El-Messery, F.A. Al-Omary, H.I. El-Subbagh, Substituted
thiazoles VII. Synthesis and antitumor activity of certain 2-(substituted
Acknowledgments
The authors extend their appreciation to the Deanship of Sci-
entific Research at King Saud University for funding this work
through the research grant number NFG2-08-33.
Appendix A. Supplementary material
Supplementary data related to this article can be found online at
http://dx.doi.org/10.1016/j.ejmech.2014.01.004.
References
[1] H. Masur, Problems in the management of opportunistic infections in patients
infected with human immunodeficiency virus, J. Infect. Dis. 161 (1990) 858e
864.
[2] E.M. Berman, L.M. Werbel, The renewed potential for folate antagonists in
contemporary cancer chemotherapy, J. Med. Chem. 34 (1991) 479e485.
[3] N. Gonen, Y.G. Assaraf, Antifolates in cancer therapy: structure, activity and
mechanisms of drug resistance, Drug Resist. Updates 15 (2012) 183e210.
[4] P. Borst, M. Quellette, New mechanism of drug resistance in parasitic pro-
tozoa, Annu. Rev. Microbiol. 49 (1995) 427e460.

H.I. El-Subbagh et al. / European Journal of Medicinal Chemistry 74 (2014) 234e245
245
amino)-4-phenyl-1,3-thiazole analogs, Bioorg. Med. Chem. Lett. 22 (2012)
6318e6323.
[39] R.L. Blakley, L. Cocco, Role of isomerization of initial complexes in the binding
of inhibitors to dihydrofolate reductase, Biochemistry 24 (1985) 4772e4777.
[40] J.F. Davies, T.J. Delcamp, N.J. Prendergast, V.A. Ashford, J.H. Freisheim, J. Kraut,
Crystal structures of recombinant human dihydrofolate reductase complexed
with folate and 5-deazafolate, Biochemistry 29 (1990) 9467e9479.
[41] C. Oefner, A. D’Arcy, F.K. Winkle, Crystal structure of human dihydrofolate
reductase complexed with folate, Eur. J. Biochem. 174 (1988) 377e385.
[42] J. Stockman, N.R. Nirmala, G. Wagner, T.J. Delcamp, M.T. DeYarman,
J.H. Freisheim, Methotrexate binds in a non-productive orientation to human
dihydrofolate reductase in solution based on NMR spectroscopy, FEBS Lett.
283 (1991) 267e269.
[43] P. Labute, C. Williams, M. Feher, E. Sourial, J.M. Schmidt, Flexible alignment of
small molecules, J. Med. Chem. 44 (2001) 1483e1490.
[44] S. Kearsley, G.M. Smith, An alternative method for the alignment of molecular
structures: maximizing electrostatic and steric overlap, Tetrahedron Comput.
Methodol. 3 (1990) 615e633.
[29] M.R. Grever, S.A. Schepartz, B.A. Chabner, The National Cancer Institute cancer
drug discovery and development program, Semin. Oncol. 19 (1992) 622e638.
[30] A. Monks, D. Scudiero, P. Skehan, Feasibility of a high flux anticancer drug
screen utilizing a derive panel of human tumor cell lines in culture, J. Natl.
Cancer Inst. 83 (1991) 757e766.
[31] M.R. Boyd, K.D. Paull, Some practical considerations and applications of the
National Cancer Institute in vitro anticancer drug discovery screen, Drug. Rev.
Res. 34 (1995) 91e109.
[32] P. Skehan, R. Storeng, D. Scudiero, A. Monks, J. McMahon, D. Vistica,
J.R. Warren, H. Bokesch, S. Kenney, M.R. Boyd, New colorimetric cytotoxicity
assay for anticancer-drug screening, J. Natl. Cancer Inst. 82 (1990) 1107e1112.
[33] N. Singh, S.K. Pandey, N. Anand, R. Dwivedi, S. Singh, S.K. Sinha, V. Chaturvedi,
N. Jaiswal, A.K. Srivastava, P. Shah, M.I. Siddiqui, R.P. Tripathi, Synthesis,
molecular modeling and bio-evaluation of cycloalkyl fused 2-
aminopyrimidines as antitubercular and antidiabetic agents, Bioorg. Med.
Chem. Lett. 21 (2011) 4404e4408.
[45] C.-M. Chu, S. Gao, M.N.V. Sastry, C.-F. Yao, Iodine-catalyzed Michael addition
of mercaptans to
a,b-unsaturated ketones under solvent-free conditions,
[34] C. Adamson, M. Balis, L. McCully, S. Godwin, G. Poplack, Methotrexate phar-
macokinetics following administration of recombinant carboxypeptidase-G2
in Rhesus monkeys, J. Clin. Oncol. 10 (1992) 1359e1364.
[35] C. Falk, R. Clark, M. Kalman, Enzymatic assay for methotrexate in serum and
cerebrospinal fluid, Clin. Chem. 22 (1976) 785e788.
Tetrahedron Lett. 46 (2005) 4971e4974.
[46] V. Cody, C.H. Schwalbe, Structural characteristics of antifolate dihydrofolate
reductase enzyme interactions, Crystallogr. Rev. 12 (2006) 301e333.
[47] S. Profeta, N.L. Allinger, Molecular mechanics calculations on aliphatic amines,
J. Am. Chem. Soc. 107 (1985) 1907e1918.
[36] R. Pignatello, G. Sapmpinato, V. Sorrenti, L. Vicari, C. Di-Giacomo, A. Vanella,
[48] N.L. Allinger, Conformational analysis. 130. MM2. A hydrocarbon force field
utilizing V1 and V2 torsional terms, J. Am. Chem. Soc. 99 (1977) 8127e8134.
[49] Hyperchem: Molecular Modeling System, Hypercube, Inc., Florida, USA, 1999.
Release 6.
G. Puglisi, Aliphatic a,g-bis(amides) of methotrexate. Influence of chain length
on in vitro activity against sensitive and resistant tumour cells, Pharm.
Pharmacol. Commun. 5 (1999) 299e305.
[37] M.G. Nordberg, K. Kolmodin, J. Aqvist, S.F. Queener, P. Hallberg, Design, syn-
thesis, and computational affinity prediction of ester soft drugs as inhibitors of
dihydrofolate reductase from Pneumocystis carinii, Eur. J. Pharm. Sci. 22 (2004)
43e53.
[50] W.D. Cornell, P. Cieplak, C.I. Bayly, I.R. Gould, K.M. Merz Jr., D.M. Ferguson,
D.C. Spellmeyer, T. Fox, J.W. Caldwell, P.A. Kollman, A second generation forth
field for the simulation of proteins and nucleic acids, J. Am. Chem. Soc. 117
(1995) 5179e5197.
[38] J.R. Appleman, W.A. Beard, T.J. Delcamp, N.J. Prendergast, J.H. Freisheim,
R.L. Blakley, Kinetics of the formation and isomerization of methotrexate
complexes of recombinant human dihydrofolate reductase, J. Biol. Chem. 263
(1988) 10304e10313.
[51] M.J.S. Dewar, E.G. Zoebisch, E.F. Healy, J.J.P. Stewart, AM1: a new general
purpose quantum mechanical molecular model, J. Am. Chem. Soc. 107 (1985)
3902e3909.