Synthesis, dihydrofolate reductase inhibition, antitumor testing, and molecular modeling study of some new 4(3H)-quinazolinone analogs

Article abstract of DOI:10.1016/j.bmc.2006.08.030

In order to produce potent new leads for anticancer drugs, a new series of quinazoline analogs was designed to resemble methotrexate (MTX, 1) structure features and fitted with functional groups believed to enhance inhibition of mammalian DHFR activity. M

Full text of DOI:10.1016/j.bmc.2006.08.030

Bioorganic & Medicinal Chemistry 14 (2006) 8608–8621
Synthesis, dihydrofolate reductase inhibition, antitumor
testing, and molecular modeling study of some new
4(3H)-quinazolinone analogs
Sarah T. Al-Rashood,a Ihsan A. Aboldahab,a Mahmoud N. Nagi,b Laila A. Abouzeid,a
Alaa A. M. Abdel-Aziz,a Sami G. Abdel-hamide,a Khairia M. Youssef,a
Abdulrahman M. Al-Obaida and Hussein I. El-Subbagha,*
aDepartment of Pharmaceutical Chemistry, College of Pharmacy, King Saud University, PO Box 2457, Riyadh 11451, Saudi Arabia
bDepartment of Pharmacology, College of Pharmacy, King Saud University, PO Box 2457, Riyadh 11451, Saudi Arabia
Received 26 June 2006; revised 17 August 2006; accepted 21 August 2006
Available online 12 September 2006
Abstract—In order to produce potent new leads for anticancer drugs, a new series of quinazoline analogs was designed to resemble
methotrexate (MTX, 1) structure features and fitted with functional groups believed to enhance inhibition of mammalian DHFR
activity. Molecular modeling studies were used to assess the fit of these compounds within the active site of human DHFR. The
synthesized compounds were evaluated for their ability to inhibit mammalian DHFR in vitro and for their antitumor activity in
a standard in vitro tissue culture assay panel. Compounds 28, 30, and 31 were the most active DHFR inhibitors with IC50 values
of 0.5, 0.4, and 0.4 lM, respectively. The most active antitumor agents in this study were compounds 19, 31, 41, and 47 with median
growth inhibitory concentrations (GI50) of 20.1, 23.5, 26.7, and 9.1 lM, respectively. Of this series of compounds, only compound
31 combined antitumor potency with potent DHFR inhibition; the other active antitumor compounds (19, 41, and 47) all had
DHFR IC50 values above 15 lM, suggesting that they might exert their antitumor potency through some other mode of action.
Alternatively, the compounds could differ significantly in uptake or concentration within mammalian cells.
Ó 2006 Elsevier Ltd. All rights reserved.
1. Introduction
have shown antineoplastic11–13 and antifungal14 activi-
ties (Chart 1).
Dihydrofolate reductase (DHFR) is an enzyme of pivot-
al importance in biochemistry and medicinal chemistry.
DHFR catalyzes the reduction of folate or 7,8-dihydro-
folate to tetrahydrofolate and intimately couples with
thymidylate synthase (TS). Inhibition of DHFR or TS
A new series of quinazoline analogs is designed to
possess a sulfhydryl or thioether function at position
activity in the absence of salvage leads to ‘thymineless
1–3
O
COOH
death.’
Compounds that inhibit DHFR exhibit an
NH2
N
N
COOH
H
important role in clinical medicine as exemplified by
the use of methotrexate (MTX, 1) in neoplastic diseas-
es,4,5 inflammatory bowel diseases,6 and rheumatoid
N
N
N
N
CH3
H2N
arthritis,7 as well as in psoriasis,8,9 and in asthma.10
A
new generation of potent lipophilic DHFR inhibitors
such as trimetrexate (TMQ, 2) and piritrexim (PTX, 3)
MTX,1
OCH3
OCH3
NH2 CH3
N
NH2 CH3
OCH3
OCH3
N
N
H
OCH3
N
H2N
N
N
H2N
Keywords: 4(3H)-Quinazolinone; DHFR inhibitors; Antitumor activ-
ity; Molecular modeling studies.
PTX, 3
TMQ, 2
*
Corresponding author. Tel.: +966 1 467 7366; fax: +966 1 467
6383; e-mail: subbagh@yahoo.com
Chart 1.
0968-0896/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2006.08.030

S. T. Al-Rashood et al. / Bioorg. Med. Chem. 14 (2006) 8608–8621
8609
2-, an alkyl, aryl or arylalkyl group at position 3-, a ni-
tro or amino functional group at position 6-, (5–15). The
6-amino function was used to introduce Schiff’s base
(16–21), sulfonamide, amide or thioureido (34–57) func-
tions. Schiff’s bases were utilized to produce compounds
28–33, which carry N–CH3 side chain in resemblance to
MTX (1). Thioether,15 sulfonamide, Schiff’s base, and
amide and thioureido16 functional groups are known
to contribute to the enhancement of the antitumor activ-
ity. Combining the inherent DHFR inhibition activity of
the quinazolines16–24 and those functional groups in one
structure was expected to produce more active com-
pounds. Most of those functional groups are also known
to increase lipid solubility.25–28 The present study is a
continuation of our previous efforts29–33 aiming to create
novel synthetic lead compound(s) and its in vitro testing
for future development as DHFR inhibitor(s).
2. Chemistry
The synthetic strategy to synthesize the targets 5–57 is
depicted in Schemes 1 and 2. Nitroanthranilic acids
(4a,b) were allowed to react with ethyl, phenyl, and ben-
zyl isothiocyanates to produce 2-thioxo-3-substituted-6
or 7-nitro-3H-quinazolin-4-ones (5–8) adopting report-
ed procedure.32,33 The 2-thioxo function of 5–8 was then
methylated using methyl iodide to give the S-methyl thi-
oether derivatives 9–12. The 6- or 7-nitro function of 9–
12 was subjected to metal reduction using Fe/HCl to
O
O
COOH
R
S
R
S
N
CH3I
N
RNCS
O2N
O2N
O2N
CH3
CH3
CH3
NH2
N
H
N
4a, b
5-8
9-12
R = Et, Ph, Bn
Fe, HCl
CH3
N
O
(CH3O)n
R
S
N
O
CH3
N
R
N
28-33
R = Ph, Bn
n = 1,2,3
H2N
N
S
13-15
CHO
NaCNBH3
HCHO
(CH3O)n
H
N
O
N
O
N
(CH3O)n
(CH3O)n
R
S
N
R
NaBH4
N
N
CH3
S
16-21
22-27
R = Ph, Bn
n = 1,2,3
R = Ph, Bn
n = 1,2,3
Scheme 1.
R'
O
O
N
SO2Cl
H
N
H2N
R
N
R
S
N
SO2
R'
CH3
S
CH3
N
34- 39
13, 14
COCl
R`NCS
R'
O
N
R'
H
O
N
H
N
R' HN
N
R
S
R
N
N
CH3
S
CH3
O
S
R = Ph, Bn
46-57
40-45
Scheme 2.

8610
S. T. Al-Rashood et al. / Bioorg. Med. Chem. 14 (2006) 8608–8621
Table 1. Physicochemical properties and DHFR inhibition (IC50, lM) of the newly synthesized compounds 5–33
O
N
R
N
O
N
O
N
2
R
3
R
2
3
R
R
R
1
1
N
1
N
N
N
R
3
R
CH
CH
2
3
S
S
S
5-15
22-33
16-21
Compound R1 R2
R3
Yield (%) Mp (°C) Solvent
Moleculara formula m/z (%)
IC50 (lM)
5
6
Et
H
6-NO2
6-NO2
6-NO2
7-NO2
6-NO2
6-NO2
6-NO2
7-NO2
6-NH2
6-NH2
7-NH2
—
69
78
—
32
82
89
—
92
28
32
36
75
62
59
45
67
73
48
72
65
53
269–271
>300
—
EtOH/H2O
EtOH
—
C10H9N3O3S
C14H9N3O3S
Ref. 37
251 (100) 20.0
Ph
Bn
Bn
Et
H
299 (84)
—
30.0
20.0
30.0
25.0
7
H
8
H
CH3
220–222
130–132
200–202
—
EtOH/H2O
EtOH
MeOH
—
C15H11N3O3S
C11H11N3O3S
C15H11N3O3S
Ref. 37
313 (48)
265 (30)
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
Ph CH3
Bn CH3
Bn CH3
Ph CH3
Bn CH3
Bn CH3
Ph 4-OCH3
Bn 4-OCH3
313 (100) 25.0
—
327 (62)
25.0
30.0
139–141
175–176
210–212
160–162
220–222
109–110
118–120
133–135
113–115
188–190
120–122
131–133
188–190
174–175
138–140
152–154
185–187
172–174
202–204
123–125
104–106
72–75
EtOH/CHCl3 C16H13N3O3S
EtOH/CHCl3 C15H13N3OS
EtOH/CHCl3 C16H15N3OS
EtOH/CHCl3 C16H15N3OS
283 (100) 50.0
297 (73)
297 (60)
401 (3)
415 (10)
431 (6)
445 (1)
461 (1)
475 (4)
403 (9)
417 (7)
433 (11)
447 (14)
463 (10)
477 (3)
417 (6)
431 (1)
447 (8)
461 (15)
477 (9)
491 (4)
10.0
50.0
2.0
DMF
DMF
C23H19N3O2S
C24H21N3O2S
C24H21N3O3S
C25H23N3O3S
C25H23N3O4S
C26H25N3O4S
C23H21N3O2S
C24H23N3O2S
C24H23N3O3S
C25H25N3O3S
C25H25N3O4S
C26H27N3O4S
C24H23N3O2S
C25H25N3O2S
C25H25N3O3S
C26H27N3O3S
C26H27N3O4S
C27H29N3O4S
—
—
10.0
50.0
40.0
13.0
20.0
25.0
10.0
20.0
15.0
30.0
30.0
1.0
Ph 3,4-(OCH3)2
Bn 3,4-(OCH3)2
Ph 3,4,5-(OCH3)3
Bn 3,4,5-(OCH3)3
DMF/H2O
DMF/H2O
DMF
—
—
—
4-OCH3
4-OCH3
DMF
Ph
Bn
Ph
Bn
Ph
Bn
H
H
H
H
H
H
EtOH/Hex
EtOH/Hex
EtOH/Hex
EtOH/Hex
EtOH/Hex
EtOH/Hex
CHCl3/Hex
CHCl3/Hex
CHCl3/Hex
CHCl3/Hex
CHCl3/Hex
CHCl3/Hex
3,4-(OCH3)2
3,4-(OCH3)2
3,4,5-(OCH3)3 81
3,4,5-(OCH3)3 85
Ph CH3
Bn CH3
Ph CH3
Bn CH3
Ph CH3
Bn CH3
4-OCH3
4-OCH3
64
75
72
72
92
90
0.5
0.4
4-OCH3
4-OCH3
0.4
1.0
3,4-(OCH3)2
3,4-(OCH3)2
10.0
a Analyzed for C, H, N; results are within 0.4% of the theoretical values for the formulae given.
afford the corresponding 6- or 7-amino compounds 13–15
(Scheme 1 and Table 1). The amino function of com-
pounds 13 and 14 was then reacted with 4-methoxy-,
3,4-dimethoxy-, and 3,4,5-trimethoxybenzaldehyde in
boiling DMF to afford the benzylidineamino analogs
16–21. The formed azomethine function was then sub-
jected to NaBH4 reduction to produce the secondary
amines 22–27, which were then subjected to the reduc-
tive methylation process using formaldehyde and NaC-
NBH3 to produce the N-methylated tertiary amines,
6-[N-(substituted benzyl)-N-methyl-amino]-3H-quinaz-
olin-4-ones (28–33) (Scheme 1 and Table 1). The 2-meth-
ylthio-3-(phenyl or benzyl)-6-amino-3H-quinazolin-4-
ones (13 and 14) were reacted with a variety of substitut-
ed benzenesulfonyl chlorides producing the 6-sulfonami-
do analogs 34–39 (Scheme 2 and Table 2). The benzene
sulfonamide moiety is carrying 4-CH3 or 4-Br function
that represents electron-donating and electron-with-
drawing groups, respectively, in order to evaluate their
electronic effects on biological activity. The amino func-
tion of 13 and 14 was also reacted with ethyl, phenyl or
benzyl isothiocyanates to yield the 6-thioureido analogs
40–45 (Scheme 2 and Table 2). Compounds 13 and 14
were reacted with a variety of substituted benzoyl chlo-
rides to produce the 6-acylamino-derivatives 46–57,
adopting the same reaction conditions used to prepare
the 6-sulfonamido analogs 34–39 (Scheme 2 and
Table 2). Structure elucidation of the synthesized inter-
mediates and final products was attained by the aid of
elementary analyses (C, H and N), NMR, and mass
spectrometry.
3. Results and discussion
The synthesized compounds (5–57) were evaluated as
DHFR inhibitors following a reported procedure,16
using MTX (1) as a positive control. Results are report-
ed as IC50 values (Tables 1 and 2). Compounds 16, 28–
32, 34, and 36 proved to be the most active DHFR
inhibitors in this investigation with IC50 values of 2.0,
1.0, 0.5, 0.4, 0.4, 1.0, 2.0, and 2.0 lM, respectively. Com-
pounds 37, 39, 42, 45, 49, 51, and 55 showed moderate
inhibition activity with IC50 values of 5.0, 7.0, 5.0, 8.0,
8.0, 7.0 and 8.0 lM, respectively; while compounds 14,
17, 23, 33, 35, 38, 43, 44, 50, and 53 showed an equal
IC50 value of 10 lM.
The synthesized compounds were also subjected to the
NCI’s in vitro, one dose primary anticancer assay, using

S. T. Al-Rashood et al. / Bioorg. Med. Chem. 14 (2006) 8608–8621
8611
Table 2. Physicochemical properties and DHFR inhibition (IC50, lM) of the newly synthesized compounds 34–57
O
O
O
R2
R1
R1
SO2HN
R2
CH3
COHN
R1
R2HNCSHN
N
N
N
CH3
CH3
N
S
N
S
N
S
34-39
46-57
40-45
Compound
R1
R2
Yield (%)
Mp (°C)
Solvent
Moleculara formula
m/z (%)
IC50 (lM)
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
Ph
Bn
Ph
Bn
Ph
Bn
Ph
Bn
Ph
Bn
Ph
Bn
Ph
Bn
Ph
Bn
Ph
Bn
Ph
Bn
Ph
Bn
Ph
Bn
H
82
86
91
79
88
94
84
89
93
96
78
81
65
58
73
86
78
82
90
64
65
78
82
87
112–114
188–190
240–242
258–260
186–188
202–204
221–222
169–170
123–125
195–197
215–216
203–205
118–120
192–195
>300
MeOH
EtOH
C
C
21H17N3O3S2
22H19N3O3S2
423 (2)
437 (4)
502 (1)
516 (1)
437 (5)
451 (8)
370 (2)
375 (2)
418 (4)
432 (2)
432 (1)
446 (1)
387 (6)
401 (8)
466 (2)
480 (1)
401 (4)
415 (1)
417 (1)
431 (7)
447 (2)
461 (3)
477 (6)
491 (5)
2.0
10.0
2.0
H
Br
EtOH/H2O
EtOH/H2O
MeOH
C21H16BrN3O3S2
C22H18BrN3O3S2
C22H19N3O3S2
C23H21N3O3S2
C18H18N4OS2
C19H20N4OS2
C22H18N4OS2
C23H20N4OS2
C23H20N4OS2
C24H22N4OS2
C22H17N3O2S
Br
5.0
CH3
CH3
Et
10.0
7.0
MeOH/H2O
EtOAc
EtOAc
50.0
20.0
5.0
Et
Ph
CHCl3/EtOH
CHCl3/EtOH
CHCl3/EtOH
CHCl3/EtOH
EtOAc
Ph
Bn
10.0
10.0
8.0
Bn
H
20.0
15.0
50.0
8.0
H
EtOAc
C23H19N3O2S
4-Br
4-Br
4-CH3
4-CH3
4-OCH3
4-OCH3
CHCl3/Hex
CHCl3/Hex
DMF
C22H16BrN3O2S
C23H18BrN3O2S
C23H19N3O2S
C24H21N3O2S
C23H19N3O3S
C24H21N3O3S
C24H21N3O4S
C25H23N3O4S
C25H23N3O5S
C26H25N3O5S
>300
125–127
238–240
214–216
105–107
148–150
202–203
238–240
115–117
10.0
7.0
DMF
DMF
DMF
70.0
10.0
40.0
8.0
3,4-(OCH3)2
3,4-(OCH3)2
CHCl3/EtOH
CHCl3/EtOH
CHCl3/EtOH
CHCl3/EtOH
3,4,5-(OCH3)3
3,4,5-(OCH3)3
15.0
12.0
a Analyzed for C, H, N; results are within 0.4% of the theoretical values for the formulae given.
a 3-cell line panel consisting of MCF-7 (breast), NCI-
H460 (lung), and SF-268 (CNS) cancers. Compounds
which reduce the growth of any one of the cell lines to
32% or less are passed on for evaluation in the full panel
of 60 cell lines over a 5-log dose range.34–36 Three
response parameters, median growth inhibition (GI50),
total growth inhibition (TGI), and median lethal con-
centration (LC50), were calculated for each cell line,
using the known drug Melphalan as a positive control.
The NCI antitumor drug discovery screen has been de-
signed to distinguish between broad-spectrum antitumor
and tumor or subpanel-selective compounds.37 In the
present study, compounds 15, 19, 31, 41, and 47 passed
primary anticancer assay at an arbitrary concentration
of 100 lM. Consequently, those active compounds were
carried over and tested against a panel of 60 different
tumor cell lines. The tested quinazoline analogs showed
a distinctive potential pattern of selectivity as well as a
broad-spectrum antitumor activity. With regard to sen-
sitivity against individual cell lines, compound 19
showed GI50 effectiveness against leukemia CCRF-
CEM and SR; melanoma SK-MEL-5 cell lines at
concentrations of 1.0, 0.2, and 2.0 lM, respectively.
Compound 31 showed GI50 activity against leukemia
RPMI-8226, SR; melanoma SK-MEL-5 cell lines at con-
centrations of 2.9, 1.6, and 2.9 lM, respectively.
Compound 47 proved to be active at GI50 level against
colon HCT-116, ovarian OVCAR-3, and renal SN12C
cell lines at concentrations of 3.3, 4.3, and 2.4 lM,
respectively. With regard to broad-spectrum antitumor
activity, the tested compounds showed GI50, TGI, and
LC50 (MG-MID) values < 100 lM, against leukemia,
Table 3. Median growth inhibitory concentration (GI50, lM) of in vitro subpanel tumor cell lines
Compound
Subpanel tumor cell linesa
MG-MIDb
I
II
III
IV
V
VI
VII
VIII
IX
15
19
27.7
12.5
8.7
44.6
26.4
48.2
29.5
18.4
38.5
56.5
31.8
65.1
27.8
8.2
43.4
37.9
28.4
33.0
23.2
17.1
54.3
18.8
40.4
27.1
17.2
31.9
59.2
42.1
54.9
43.0
21.2
43.0
34.8
22.2
20.4
28.0
10.0
34.4
22.4
24.1
20.7
33.3
20.7
34.7
32.1
21.8
40.4
25.6
18.8
39.2
34.4
20.1
23.5
26.7
9.1
31
41
15.8
6.7
47
Melphalan
20.1
42.1
27.1
a 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.
b GI50 full panel mean-graph midpoint (lM).

8612
S. T. Al-Rashood et al. / Bioorg. Med. Chem. 14 (2006) 8608–8621
Table 4. Total growth inhibitory concentration (TGI, lM) of in vitro subpanel tumor cell lines
Compound
Subpanel tumor cell linesa
MG-MIDb
I
II
III
IV
V
VI
VII
VIII
95.2
IX
c
c
15
19
31
41
47
76.2
52.8
92.3
80.5
89.7
90.5
50.8
91.8
91.7
79.9
86.3
41.7
92.2
61.3
75.1
63.2
48.1
92.2
73.6
69.5
75.8
30.3
89.6
77.2
90.4 (c)d
90.7
c
82.2
93.8
84.8
40.6
87.1
c
68.0 (97.8)d
83.1 (c)d
c
c
c
59.2
c
71.4
64.5
69.3
45.7
74.7 (97.8)d
39.5 (78.1)d
87.2
a For subpanel tumor cell lines, see footnote a of Table 3.
b TGI full panel mean-graph midpoint (lM).
c Compound showed values >100 lM.
d Median lethal concentrations (LC50, lM) are shown in parentheses, Melphalan TGI and LC50 MG-MID values are 35.3 and 65.5 lM, respectively.
non-small cell lung, colon, CNS, melanoma, ovarian,
renal, prostate, and breast cancer subpanel cell lines.
Compounds 15 and 31 showed (MG-MID) values
<100 lM at only the GI50 and TGI levels, while com-
pounds 19, 41, and 47 showed potency at the three levels
of activity GI50, TGI, and LC50 (Tables 3 and 4). Com-
pound 47 proved to be the most active member in this
study with GI50, TGI and LC50 values of 9.1, 39.5,
and 78.1 lM, respectively.
in vacuo, bond dipole option for electrostatics, Polak–
Ribiere algorithm, and RMS gradient of 0.01 kcal/
A mol)38 as implemented in HyperChem 6.0.39 The most
˚
stable conformer was fully optimized with Amber force-
field40 followed by AM1 semi-empirical molecular orbi-
tal calculation.41 The global minimum was confirmed as
true minimum not saddle point by the absence of nega-
tive eigenvalue of the Hessian through frequency calcu-
lation. The calculations showed that, compounds 16,
28–32, 34, 36–37, and 42 exhibited strong structural sim-
ilarity to 1 as indicated by their molecular parameters
(Table 5). The lowest energy-minimized structures
exhibited a common arrangement of the aryl groups
(ring A and B, Figs. 1–3 and 5) around the quinazoline
core. Ring A was out of the plane of quinazoline ring
(torsional angles were 91° and 71° in case of the 3-phen-
yl and 3-benzyl moieties, respectively). The benzyl group
3.1. Molecular modeling study
As an attempt to gain a better insight into the molecular
structures of the active compounds 16, 28–32, 34, 36, 37,
and 42, and the inactive compounds 5–13 and 15, in
comparison to MXT (1), conformational analysis has
been formed by use of the MM+ force-field (calculations
Table 5. Molecular parameters of the active and the inactive compounds compared with MTX (1)
g
Compound
ClogPa
Rb
Pc
Vd
SAe
Df
4.8
IC50
5
6
1.8
3.2
3.0
3.0
2.1
3.6
3.3
3.3
3.3
3.0
0.5
5.2
6.0
5.8
5.8
5.5
5.4
4.7
5.6
5.4
4.6
64.1
24.2
30.2
32.0
32.0
26.1
32.0
33.9
33.9
32.0
33.4
46.1
45.1
47.4
49.2
49.8
51.7
52.3
42.4
45.0
46.8
47.6
658.2
410.9
471.7
495.1
491.3
443.2
507.1
521.9
523.9
486.7
509.1
704
20
30
20
20
25
25
25
30
50
50
79.1
83.9
767.6
814.0
815.8
4.4
4.8
4.8
5.1
5.8
5.0
5.2
5.2
5.8
8.8
5.4
5.3
3.4
4.8
3.1
5.2
9.9
10.6
8.6
2.7
7
8
83.9
68.9
9
716.2
830.7
10
11
12
13
15
MTX (1)
16
28
29
30
31
32
34
36
37
42
83.9
88.8
871.2
872.2
88.8
82.3
803.2
846.5
87.2
118.8
117.4
123.1
128.0
129.6
134.4
136.1
116.2
123.9
128.7
122.6
1200.8
1144.3
1199.4
1241.6
1275.4
1316.9
1353.9
1110.6
1175.5
1216.6
1141.7
0.001
2.0
1.0
0.5
0.4
0.4
1.0
2.0
2.0
5.0
5.0
681.2
700.7
715.27
738.8
759.9
783.7
658.9
690.3
710.7
668.2
a (ClogP) values were calculated by using the CS ChemDraw Ultra version 7.
b
3
Refractometry (A ).
˚
c
3
Polarizability (A ).
˚
d
e
3
Molecular volume (A ).
3
Surface area; gride (A ).
˚
˚
f Dipole moment (Debye).
g Data were taken from Tables 1 and 2.

S. T. Al-Rashood et al. / Bioorg. Med. Chem. 14 (2006) 8608–8621
8613
Compound 13
Compound 31
MTX (1)
Figure 3. RMS fit overlay of the active compound 31 (green) and
MTX, 1 (violet).
(ring B) was deviated from the plane of quinazoline ring
by ꢀ65° and arranged itself parallel with quinazoline
ring. Moreover, the freely rotated methoxy groups were
arranged in spatial manner approximately coplanar and
parallel with the aryl edges.42 Similarly, the Ring B of 1
arranged itself parallel with pteridine ring and deviated
from its plane by ꢀ67° (Figs. 3 and 5).
Further evidence of structural similarity and analogy of
binding of the most active compounds (28–32) and
MTX (1) was obtained from the QSAR-data using the
conformations depicted in Figure 1. Semi-empirical cal-
culations of the conformers demonstrated the closely
related charge distribution and electrostatic potentials
suggesting a similar interaction of those molecules with
the potential protein binding site. QSAR studies showed
an optimum hydrophobicity (ClogP) range of 5.2–6.0
Figure 1. Global minima of compounds 13, 31 and MTX (1) with their
dipole moment vector and electrostatic potential isosurface, negative
region colors red and positive region colors green.
Figure 2. Connolly molecular surface (upper panel) and solvent accessible surface (lower panel) of MTX, 1 (left) and the active compound 31 (right).

8614
S. T. Al-Rashood et al. / Bioorg. Med. Chem. 14 (2006) 8608–8621
for the active compounds, and ClogP range of 1.8–3.3
for the inactive compounds. Direct correlation could
be established between the ClogP and hDHFR inhibi-
tion of these series while it could not be correlated with
1 as its ClogP value is 0.5 (Table 5). It becomes appar-
ent that ClogP values are not the sole predicting factor
for biological activity in this study. In addition, despite
the variation of the molecular shapes of these ligands,
measurements of global molecular parameters such as
surface area, volume, and refractivity (Fig. 1) also reflect
their similarity to 1. As shown from Table 5, the biolog-
ical inefficacy of the inactive compounds 5–15 could be
attributed to the difficulty to cross the biological mem-
branes due to their physicochemical properties which
prevent their access to the putative cavity. Furthermore,
semi-empirical calculations of the conformers displayed
in Figure 2 described the Connolly molecular surface,
solvent accessible surface, and the electrostatic binding
characteristics of the surface of the molecules and dem-
onstrated their closely related molecular surface, charge
distribution, and electrostatic potential, suggesting a
similar interaction of those molecules and compound 1
with the potential hDHFR binding site. Moreover, the
energy-minimized conformer 31 and MTX (1), as a rep-
resentative example, were superimposed in order to re-
veal the similarities and the differences in their
structures (Fig. 3). The overlay fitting was carried out
with respect to quinazoline and pteridine rings. The re-
sults showed that deviation of the superimposed struc-
tures was observed in the region of the 6-amphiphilic
˚
moiety with rms value of 0.05 A. This is in addition to
what was displayed in Figure 1 which demonstrated
the closely related charge distribution and electrostatic
potential. Accordingly compounds 31 and 1 could adopt
conformations that have close distances and orienta-
tions of the assumed major hDHFR-binding groups
and showed sufficient agreement of the overall structure
electrostatic potential pattern.
The synthesized target compounds have been compara-
tively evaluated in terms of their binding mode to
human dihydrofolate reductase (hDHFR) pocket.
Molecular docking and molecular dynamic (MD) simu-
lations have been formed for the proposed compounds
(5–57), to evaluate their recognition profiles at the
hDHFR-binding pocket.43,44 Studying the pteridine ring
H-bonding interaction of 1 with the hDHFR active site
revealed that the nitrogen atoms at N1, N8 and the nitro-
gen atom of the 2-amino group contributed three stable
H-bonds with the key pocket residue Glu30. Besides, the
nitrogen atom of the 4-amino group of the pteridine ring
conferred trifurcated H-bonds with the ‘catalytic triad’
residues of hDHFR pocket Ile7, Tyr121, and Val115.
Furthermore, the nitrogen and oxygen of the glutamine
amide function formed two H-bonds with Asn64
(Fig. 4). Similarly, the binding mode of the investigated
quinazolines was formed. In general, the nitrogen atom
at N1and the 2-sulfur atom of the quinazoline analogs
Figure 4. Binding mode for MTX (1), and compounds 13 and 31 docked and minimized in the hDHFR binding pocket, showing residues involved in
its recognition. Solid lines indicate the H-bond formation.

S. T. Al-Rashood et al. / Bioorg. Med. Chem. 14 (2006) 8608–8621
8615
Hydrophobic Region
H-Bond Acceptor Region
O
13.5 Å
O
O
11.8-12.4 Å
B
A
N
−66ο
X
N
80ο(79ο)
71ο(91ο)
13.9 Å
N
S
Spacer Group as H-Bond Acceptor Region
H-Bond Acceptor Region
Figure 5. Deduced quinazoline pharmacophore requirement for DHFR binding activity.
formed two H-bonds with Glu30. The 6-amino function
of 13 was oriented in the cavity to form H-bond with
Asp21, while compound 31 showed three H-bonds with
Thr56, Trp57, and Phe58 via the methoxy groups locat-
ed at positions 3 and 4. Compounds possessing high
root mean square deviation (rmsd) values signify a neg-
ative correlation with the degree of binding affinity and
vice versa. The polar part of the active site appears to be
very rigid; the movements of the active site residues
upon binding to the tested compounds are limited. This
rigidity of the active site is evident from the low rmsd of
active site residues in the range of 0.58–1.12.
group of 1. The 2-amino function should be about
˚
13.7 A apart from the 3,4-dimethoxy groups, which is
similar to the non-bonded distance between the 2-amino
group of 1 and the carbonyl group of the amide function
(Fig. 5), (viii) the 3,4-dimethoxy function improves
binding with the active site residues, while the introduc-
tion of the third 5-methoxy moiety blunts this interac-
tion, and (ix) the 6-amino group interacts with the
surrounding amino acids better than the 7-amino
substituents.
3.2. Structure–activity correlation
The structure features essential for the DHFR binding
activity of this series (pharmacophore) as determined
by the experimental and molecular modeling data are
shown in Figure 5. The overall outcome of this molecu-
lar docking study revealed that: (i) the quinazoline ring
is a satisfactory backbone for inhibitors of mammalian
DHFR, establishing contact with the key amino acid
residues in the enzyme pocket, (ii) the presence of aryl
moiety, ring A and ring B, at the 3-position and 6-posi-
tion of quinazoline is necessary for the activity as hydro-
phobic and amphiphilic regions, respectively, (iii) the
two aryl groups at position 3- and 6- should be about
The obtained DHFR inhibition results revealed that, in
the 6-nitro series (5–12), the 2-SH or 2-SCH3 moieties
do not affect the magnitude of DHFR inhibition as evi-
denced by the IC50 values of the SH containing com-
pounds 5–8 and their SCH3 counterparts 9–12.
Reduction of the 6-nitro function into the 6-amino ana-
logs results in variable levels of inhibition. In the 3-phen-
yl series, reduction of the 6-nitro function decreased the
inhibition by 2-fold (10, 25.0 lM vs 13, 50.0 lM); while
in the 3-benzyl series, the inhibition increased by
2.5-fold (11, 25.0 lM vs 14, 10.0 lM), in accordance with
the molecular modeling predictions. In most cases, mov-
ing the 6-nitro group into position-7 did not markedly
affect the magnitude of activity. Meanwhile, the translo-
cation of the 6-amino moiety to position-7 reduced the
inhibition noticeably (14, 10.0 lM vs 15, 50.0 lM). In
the 6-benzylidinamino series (16–21), the combination
of phenyl group at position-3 and 4-methoxy-benzylidi-
namino at position-6 proved to remarkably enhance the
inhibition, as shown in compound 16 (IC50 value of
2.0 lM). Replacing the 3-phenyl group of 16 by 3-benzyl
moiety decreased the inhibition by 5-fold (17, 10.0 lM).
The introduction of 3,4-dimethoxy-phenyl replacing the
4-methoxy-phenyl moiety of 16 decreased the inhibition
activity by 25-fold (16, 2.0 lM vs 18, 50.0 lM), while the
3,4,5-trimethoxy-phenyl moiety restored some of the
DHFR inhibitory potency as shown in compounds 20
˚
11.8–12.4 A apart, (iv) The 3-benzyl group favors bind-
ing being accommodated within the enzyme binding
pocket without steric hindrance as compared to the
3-phenyl substituents, (v) introduction of two atoms spac-
er group, between quinazoline ring and ringB, which
contains small hydrophobic alkyl group, enhances the
lipophilic interaction with the hydrophobic pocket, such
as 6-[N-(substituted benzyl)-N-methyl-amino]-. Other
H-bond acceptor augments the binding interaction with-
in the enzyme pocket, such as 6-[N-arylsulfonylamino]-,
(vi) the torsional angle between the plane of the quinaz-
oline core and the spacer moiety is 79–80° while in case
of MTX (1) it is 77°, (vii) the presence of methylthio
group located on the 2-position of quinazoline is neces-
sary as H-bond acceptor region to mimic the 2-amino

8616
S. T. Al-Rashood et al. / Bioorg. Med. Chem. 14 (2006) 8608–8621
and 21 (IC50 13.0 and 20.0 lM, respectively). Reduction
of the 6-benzylidineamino function into the 6-benzyla-
mino analogs (22–27) showed a slight improvement in
DHFR inhibition activity. Upon reductive methylation,
using formaldehyde and NaCNBH3 to produce the
methotrexate (MTX, 1) like derivatives (28–33), the
activity increased dramatically producing the most ac-
tive members in this study (compounds 28–32), which
showed IC50 values of 1.0, 0.5, 0.4, 0.4, and 1.0 lM,
respectively, in accordance with the molecular modeling
predictions. In this particular group, the 3-phenyl substi-
tution at the quinazoline nucleus proved to favor the
activity, rather than the 3-benzyl, as shown in com-
pound 32 (IC50, 1.0 lM) and the 10-fold decrease in
potency in compound 33 (IC50, 10.0 lM). In the
4-substituted benzenesulfonamide series (34–39), unsub-
stitution or the presence of 4-bromo moiety favored the
activity producing compounds 34 and 36 with IC50 value
of 2.0 lM for both compounds. The introduction of
4-methyl group to 34 decreased the activity by 5-fold
as shown in compound 38 (IC50 value of 10.0 lM). In
the 6-thioureido series (40–45), aromatic substitution
on the thioureido moiety favored the activity rather than
the aliphatic substitution. The N-phenyl thioureido sub-
stitution proved to be 10-fold more active than the
N-ethyl analog, as seen in compounds 42 (5.0 lM) and
40 (50.0 lM). In the 6-benzamido series (46–57), the
3-benzyl substitution as in 49, 53, and 55 with IC50
values 8.0, 10.0, and 8.0 lM, respectively, favored the
activity rather than the 3-phenyl analogs (48, 52, and
54 with IC50 values of 50.0, 70.0, and 40.0 lM).
LC50 values of 26.7, 74.7, and 97.8 lM, respectively);
and 47 (GI50, TGI, and LC50 values of 9.1, 39.5, and
78.1 lM, respectively) are the most active antitumor
agents in this study, as compared with the known anti-
tumor drug ‘Melphalan’ (GI50, TGI, and LC50 values
of 27.1, 35.3, and 65.5 lM, respectively). It seemed that
compound 31 exerts its antitumor potency through
DHFR inhibition mode of action, while the other active
compounds, namely 19, 41, and 47, might exert their
antitumor potency through DHFR inhibition and/or
some other mechanism of action. Some of the differences
between DHFR inhibitory activity and whole cell anti-
tumor assays can be related to the ability of the com-
pounds to be taken up by cells or concentrated
therein. The comparison of the molecular parameters
of compound 31 and MTX (1) has led to a great deal
of model similarity for this new class of DHFRI. This
model suggests that key elements are in common
between 1 and the new quinazoline analogs which are
stabilized within the active site of the enzyme through
hydrophobic and H-bond interaction. Finally, EPS
(electrostatic potential isosurface) maps of both com-
pounds 1 and 31 were obtained. The two molecules
showed matching regions of high and low electrostatic
potential due to their structural resemblance. The results
of the computer modeling experiments indicate striking
similarities in the mode and strength of binding of the
two compounds due to their similar steric, volume,
and EPS maps. These studied quinazoline analogs could
be considered as useful templates for future develop-
ment and further derivatization or modification to
obtain more potent DHFR inhibitors.
The obtained antitumor testing results revealed that the
7-amino group of 15 (GI50, TGI—34.4, 90.4 lM, respec-
tively) favored the antitumor activity rather than the
6-amino analog 14. Comparing compound 19 (GI50, TGI,
LC50—20.0, 68.0, and 97.8 lM, respectively) with 3,4-
dimethoxyphenyl substitution to the inactive 4-methoxy
(17) and 3,4,5-trimethoxy (21) compounds showed that
the number and setting of the methoxy groups proved
crucial for antitumor activity. For DHFR inhibition,
aromatic substitution rather than aliphatic substitution
on the thioureido function was favored but the opposite
was true for anticancer activity: compound 41 (GI50,
TGI, and LC50—26.7 74.7, 97.8 lM, respectively) with
its 6-ethylthioureido function was active but compounds
43 and 45, with aromatic substitution were inactive. The
6-benzamido analog 47 (GI50, TGI, LC50—9.1, 39.5,
and 78.1 lM, respectively) represents another example
of the effect of substituent on activity in this study.
Introduction of the electron-withdrawing bromine atom
to 47 or electron-donating methyl function produced
compounds 49 and 51, respectively, with total loss of
activity.
5. Experimental
Melting points (°C) were determined on Mettler FP80
melting point 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 and 13C-
NMR spectra were recorded on
a Varian XL
500 MHz FT spectrometer, chemical shifts are expressed
in d parts per million with reference to TMS. Mass spec-
tral (MS) data were obtained on a Shimadzu GC/MS
QP 5000 apparatus. Thin-layer chromatography was
formed on precoated (0.25 mm) silica gel GF254 plates
(E. Merck, Germany), compounds were detected with
254 nm UV lamp. Silica gel (60–230 mesh) was em-
ployed for routine column chromatography separations.
DHFR inhibition activity experiments were formed at
Pharmacology Department, College of Pharmacy, King
Saud University. Bovine liver DHFR enzyme and meth-
otrexate (MIX, 3) were used in the assay (Sigma Chem-
ical Co., USA). Antitumor activity was formed at the
National Cancer Institute (NCI), Bethesda, Maryland,
USA. Compounds 7 and 11 were previously reported.33
4. Conclusion
Compounds 29, 30, and 31 are the most active members
of this study as DHFR inhibitors with IC50 values of
0.5, 0.4, and 0.4 lM, respectively. Compounds 19
(GI50, TGI, and LC50 values of 20.1, 68.0, 97.8 lM,
respectively): 31 (GI50, TGI, and LC50 values of 23.5,
83.1, <100.0 lM, respectively), 41 (GI50, TGI, and
All modeling studies were conducted with Hyperchem
6.03 package from Hypercube.39 Molecular dynamics
studies were formed using Amber force field40,41 within
Hyperchem 6.03. The docking of the candidates into

S. T. Al-Rashood et al. / Bioorg. Med. Chem. 14 (2006) 8608–8621
8617
hDHFR pocket was formed with docking protocol in
Hyperchem 6.03. Starting coordinate of hDHFR en-
zyme in tertiary complex with reduced-nicotinamide
adenine dinucleotide phosphate (NADPH) and MTX
(3), code ID 1DLS, was obtained from the Protein Data
Bank of Brookhaven National Laboratory.28
7.17 (m, 2H, ArH), 7.35–7.38 (m, 3H, ArH), 7.39–7.53
(m, 3H, ArH). Compound 14: d 2.52 (s, 3H, SCH3),
5.30 (s, 2H, CH2Ph), 5.61 (br s, 2H, NH2), 7.18–7.32
(m, 8H, ArH). Compound 15: d 2.52 (s, 3H, SCH3),
5.12 (s, 2H, CH2Ph), 6.17 (br s, 2H, NH2), 6.57 (s 1H,
ArH), 6.68 (d, 1H, J = 8.5 Hz, ArH), 7.21–7.34 (m,
5H, ArH), 7.77 (d, 1H, J = 8.5 Hz, ArH).
5.1. 2-Thioxo-3-substituted-6 or 7-nitro-3H-quinazolin-4-
ones (5, 6, and 8)
5.4. 2-Methylthio-3-(phenyl or benzyl)-6-[(substitut-
ed)benzylidineamino]-3H-quinazolin-4-ones (16–21)
A mixture of 4- or 5-nitroanthranilic acid (4a,b, 1.82 g,
0.01 mol), the appropriate isothiocyanate derivative
(0.012 mol), and TEA (2 mL) in ethanol (50 mL) was
heated under reflux for 4 h. The reaction mixture was
then cooled and the solvent was evaporated under vac-
uum. The obtained residue was washed with petroleum
ether 40:60, filtered, dried and recrystallized to give 5,
An equimolar amount (0.01 mol) of the 6-aminoquinaz-
oline analogs 13 or 14 and the appropriate methoxy-
benzaldehyde was dissolved in DMF (30 mL) and
heated under reflux for 4 h. Solvent was then removed
under reduced pressure. The obtained residue was fil-
tered, washed with water, dried, and recrystallized to af-
ford 16–21 (Table 1). 1H NMR (TFA), 16: d 2.51 (s, 3H,
SCH3), 3.86 (s, 3H, OCH3), 7.01 (d, 1H, J = 8.5 Hz,
ArH), 7.48–7.59 (m, 5H, ArH), 7.60–7.78 (dd, 4H,
J = 8.5, 2.5 Hz, ArH), 7.82 (s, 1H, ArH), 7.94 (d, 1H,
J = 8.5 Hz, ArH), 8.69 (s, 1H, CH@N). Compound 17:
d 2.53 (s, 3H, CH3), 3.85 (s, 3H, OCH3), 5.30 (s, 2H,
CH2Ph), 7.09–7.11 (m, 1H, ArH), 7.21–7.35 (m, 6H,
ArH), 7.62–7.95 (m, 5H, ArH), 8.71 (s, 1H, CH@N).
Compound 18: d 2.51 (s, 3H, SCH3), 3.86 (s, 6H,
OCH3), 7.11 (d, 1H, J = 8.5 Hz, ArH), 7.47–7.78 (m,
10H, ArH), 8.67 (s, 1H, CH@N). Compound 19: d
2.52 (s, 3H, SCH3), 3.87 (s, 6H, OCH3), 5.32 (s, 2H,
CH2Ph), 7.13 (d, 1H, J = 8.5 Hz, ArH), 7.45–7.81 (m,
9H, ArH), 8.71 (s, 1H, ArH), 8.69 (s, 1H, CH@N).
Compound 20: 2.51 (s, 3H, SCH3), 3.75 (s, 3H,
OCH3), 3.87 (s, 6H, OCH3), 7.35 (s, 2H, ArH), 7.43–
7.52 (m, 2H, ArH), 7.56–7.62 (m, 3H, ArH), 7.69 (d,
1H, J = 8.5 Hz, ArH), 7.78–7.82 (dd, 1H, J = 8.5,
2.5 Hz, ArH), 7.88 (d, 1H, J = 2.5 Hz, ArH), 8.71 (s,
1H, CH@N). Compound 21: 2.50 (s, 3H, SCH3), 3.76
(s, 3H, OCH3), 3.87 (s, 6H, OCH3), 5.30 (s, 2H, CH2Ph),
7.32 (s, 2H, ArH), 7.41–7.53 (m, 2H, ArH), 7.54–7.60
(m, 3H, ArH), 7.65–7.82 (m, 2H, ArH), 7.88 (s, 1H,
ArH), 8.70 (s, 1H, CH@N).
1
6 and 8 (Table 1). H NMR, 5 (CDCl3): d 1.24 (t, 3H,
J = 7.0 Hz, CH3CH2), 3.35 (br s, 1H, SH), 4.41–4.45
(q, 2H, J = 7.0 Hz, CH3CH2), 7.49 (d, 1H, J = 9.0 Hz,
ArH), 8.50–8.53 (dd, 1H, J = 9.0, 3.0 Hz, ArH), 8.60
(d, 1H, J = 3.0 Hz, ArH). Compound 6 (DMSO-d6): d
3.07 (br s, 1H, SH), 6.80 (d, 1H, J = 9.0 Hz, ArH),
7.30–7.58 (m, 5H, ArH), 7.98–7.99 (dd, 1H, J = 9.0,
3.0 Hz, ArH), 8.62 (d, 1H, J = 3.0 Hz, ArH). 8
(DMSO-d6): d 3.35 (s, 1H, SH), 5.34 (s, 2H, CH2Ph),
8.15–8.17 (dd, 1H, J = 9.0, 2.0 Hz, ArH), 8.24 (d, 1H,
J = 2.0 Hz, ArH), 8.31 (d, 1H, J = 9.0 Hz, ArH).
5.2. 2-Methylthio-3-substituted-6 or 7-nitro-3H-quinazo-
lin-4-ones (9, 10, and 12)
A mixture of the 2-thioxoquinazoline analogs 5, 6, or 8
(0.01 mol), methyl iodide (5 mL), and anhydrous potas-
sium carbonate (2 g) in acetone (50 mL) was heated
under reflux for 8 h. The separated solid was filtered
while hot, washed with acetone, the filtrate was then
evaporated, and the obtained residue was dried, and
1
recrystallized to give 9, 10, and 12 (Table 1). H NMR
(DMSO), 9: d 1.29 (t, 3H, J = 7.0 Hz, CH3CH2), 2.67
(s, 3H, SCH3), 4.08–4.12 (q, 2H, J = 7.0 Hz, CH3CH2),
7.64 (d, 1H, J = 9.0 Hz, ArH), 8.46 (d, 1H, J = 9.0 Hz,
ArH), 8.68 (s, 1H, ArH). Compound 10: d 2.51 (s, 3H,
SCH3), 7.49–7.60 (m, 5H, ArH), 7.61 (d, 1H,
J = 9.0 Hz, ArH), 8.52–8.55 (m, 1H, ArH), 8.71 (s,
1H, ArH). Compound 12: d 2.64 (s, 3H, SCH3), 5.36
(s, 2H, CH2Ph), 7.29–7.36 (m, 5H, ArH), 8.19 (d, 1H,
J = 9.0 Hz, ArH), 8.29 (s, 1H, ArH), 8.33 (d, 1H,
J = 9.0 Hz, ArH).
5.5. 2-Methylthio-3-(phenyl or benzyl)-6-(substituted
benzylamino)-3H-quinazolin-4-ones (22–27)
The 2-methylthio-3-(phenyl or benzyl)-6-(substituted
benzylidine)amino-3H-quinazolin-4-ones
(16–21,
0.01 mol) were dissolved in methanol (50 mL) and stir-
red. NaBH4 (1.0 g) was added portionwise over a period
of 0.5 h and stirred at room temperature for another 2 h.
Solvent was removed under reduced pressure. Water
(20 mL) was then added, and the undissolved solid
was filtered and recrystallized from the appropriate sol-
vent to afford 22–27 (Table 1). 1H NMR (CDCl3), 22: d
2.42 (s, 3H, SCH3), 3.72 (s, 3H, OCH3), 4.27 (d, 2H,
J = 5.2 Hz, CH2Ph), 6.71 (t, 1H, J = 5.2 Hz, NH), 6.89
(d, 1H, J = 8.0, Hz, ArH), 7.03 (s, 2H, ArH), 7.28–
7.53 (m, 9H, ArH), 23: d 2.50 (s, 3H, SCH3), 3.72 (s,
3H, OCH3), 4.26 (d, 2H, J = 5.8 Hz, CH2Ph), 5.27 (s,
2H, CH2Ph), 6.88 (t, 1H, J = 5.8 Hz, NH), 6.92 (d,
1H, J = 8.0 Hz, ArH), 7.20 (s, 1H, ArH), 7.28–7.31
(m, 10H, ArH). Compound 24: d 2.50 (s, 3H, SCH3),
3.72 (s, 6H, OCH3), 4.26 (s, 2H, CH2Ph), 6.90 (s, 1H,
5.3. 2-Methylthio-3-(phenyl or benzyl)-6 or 7-amino-3H-
quinazolin-4-ones (13–15)
A mixture of the 6 or 7-nitro-quinazoline derivatives 10,
11 or 12 (0.01 mol), Fe powder prewashed with dilute
HCl and water (0.5 g), concentrated HCl (10 mL), in
ethanol (50 mL), was heated under reflux for 2 h. Con-
centrated ammonia solution (5 mL) was added to pre-
cipitate Fe salts. The resulting mixture was filtered
while hot through Celite. Filtrate was concentrated to
give the crude products which were recrystallized to
1
yield compounds 13–15 (Table 1). H NMR (CDCl3),
13: d 2.50 (s, 3H, SCH3), 5.60 (br s, 2H, NH2), 7.1–

8618
S. T. Al-Rashood et al. / Bioorg. Med. Chem. 14 (2006) 8608–8621
NH), 6.88 (d, 1H, J = 8.0 Hz, ArH), 7.22–7.35 (m, 10H,
ArH). Compound 25: d 2.50 (s, 3H, SCH3), 3.71 (s, 6H,
OCH3), 4.25 (s, 2H, CH2Ph), 5.28 (m, 3H, CH2Ph and
NH), 6.89–7.38 (m, 11H, ArH). Compound 26: d 2.42
(s, 3H, SCH3), 3.63 (s, 3H, OCH3), 3.74 (s, 6H,
OCH3), 4.26 (d, 2H, J = 5.5 Hz, CH2Ph), 6.72–6.75
(m, 3H, NH and ArH), 7.08 (s, 1H, ArH), 7.21–7.23
(m, 1H, ArH), 7.37–7.43 (m, 3H, ArH), 7.52–7.53 (m,
3H, ArH). Compound 27: 2.41 (s, 3H, SCH3), 3.62 (s,
3H, OCH3), 3.75 (s, 6H, OCH3), 4.27 (d, 2H,
J = 5.6 Hz, CH2Ph), 5.29 (s, 2H, CH2Ph), 6.71–6.76
(m, 3H, NH and ArH), 7.08–7.23 (m, 2H, ArH), 7.38–
7.54 (m, 6H, ArH).
triturated with water and filtered. The obtained solid
was dried, and recrystallized to give 34–39 (Table 2).
1H NMR (DMSO-d6), 34: d 2.43 (s, 3H, SCH3), 7.28–
7.98 (m, 13H, ArH), 10.66 (s, 1H, NH). 35: d 2.54 (s,
3H, SCH3), 5.32 (s, 2H, CH2Ph), 7.19–7.98 (m, 13H,
ArH), 10.69 (br s, 1H, NH). Compound 36: d 2.43 (s,
3H, SCH3), 7.29–7.80 (m, 12H, ArH), 10.72 (br s, 1H,
NH). Compound 37: d 2.54 (s, 3H, SCH3), 5.31 (s,
2H, CH2Ph), 7.19–7.79 (m, 13H, ArH and NH). Com-
pound 38: d 2.33 (s, 3H, CH3Ar), 2.44 (s, 3H, SCH3),
7.28–7.86 (m, 12H, ArH), 10.54 (br s, 1H, NH). Com-
pound 39: d 2.32 (s, 3H, CH3Ar), 2.54 (s, 3H, SCH3),
5.28 (s, 2H, CH2Ph), 5.61 (br s, 1H, NH), 7.07–7.78
(m, 12H, ArH).
5.6. 2-Methylthio-3-(phenyl or benzyl)-6-[N-(substituted
benzyl)-N-methylamino]-3H-quinazolin-4-ones (28–33)
5.8. N-Substituted-N0-[2-methylthio-3-(phenyl or benzyl)-
4-oxo-3H-quinazolin-6-yl]-thioureas (40–45)
To a solution of 2-methylthio-3-(phenyl or benzyl)-6-
(substituted benzylamino)-3H-quinazoline-4-ones (22–
27, 0.05 mol) in 50 mL of acetonitrile, formaldehyde
(4 g, 0.11 mol) was added with constant stirring. To this
suspension, NaCNBH3 (5 g, 0.08 mol) was added, and
the pH of the mixture was adjusted to 2–3 using concen-
trated HCl. Five minutes later a bright yellow precipi-
tate was obtained. The acetonitrile was evaporated
under reduced pressure and the obtained residue was
suspended in 10 mL of water and neutralized using
NH4OH to afford 28–33 which were filtered, dried,
A solution of the 6-aminoquinazoline derivative 13 or 14
(0.002 mol), and the appropriate isothiocyanate
(0.0022 mol) in ethanol (10 mL) was heated under reflux
for 6 h. The separated solid was filtered, dried and
recrystallized to yield 40–45 (Table 2). 1H NMR
(DMSO-d6), 40: d 1.15 (t, 3H, CH2CH3), 2.43 (s, 3H,
SCH3), 3.45–3.51 (m, 2H, CH2CH3), 4.34 (br s, 1H,
NH), 5.58 (br s, 1H, NH), 7.18–8.17 (m, 8H, ArH).
Compound 41: d 1.19 (t, 3H, CH2CH3), 2.43 (s, 3H,
SCH3), 3.46–3.49 (m, 2H, CH2CH3), 5.37 (s, 2H,
CH2Ph), 6.31 (br s, 1H, NH), 7.28–8.03 (m, 8H, ArH),
8.18 (br s, 1H, NH). Compound 42: d 2.49 (s, 3H,
SCH3), 7.27–7.60 (m, 11H, ArH), 7.94 (s, 1H, ArH),
8.22 (s, 1H, ArH), 8.35 (br s, 1H, NH), 9.89 (br s, 1H,
NH). Compound 43: d 2.59 (s, 3H, SCH3), 5.34 (s,
2H, CH2Ph), 7.15 (t, 1H, ArH), 7.24–7.28 (m, 3H,
ArH), 7.33–7.37 (m, 4H, ArH), 7.49 (d, 2H,
J = 7.5 Hz, ArH), 7.57 (d, 1H, J = 9.0 Hz, ArH), 7.94–
7.96 (dd, 1H, J = 9.0, 2.5 Hz, ArH), 8.26 (d, 1H,
J = 2.5 Hz, ArH), 10.00 (br s, 1H, NH), 10.08 (br s,
1H, NH). Compound 44: d 2.49 (s, 3H, SCH3), 4.78
(s, 2H, CH2Ph), 7.27–7.60 (m, 11H, ArH), 7.94 (d, 1H,
J = 9.0 Hz, ArH), 8.22 (s, 1H, ArH), 8.35 (br s, 1H,
NH), 8.89 (br s, 1H, NH). Compound 45: d 2.53 (s,
3H, SCH3), 5.30 (s, 2H, CH2Ph), 5.61 (s, 2H, CH2Ph),
7.06–7.11 (dd, 2H, J = 9.0, 2.5 Hz, ArH), 7.19–7.37
(m, 13H, ArH and NH).
1
and recrystallized (Table 1). H NMR (DMSO-d6), 28:
d 2.43 (s, 3H, SCH3), 3.09 (s, 3H, NCH3), 3.72 (s, 3H,
OCH3), 4.61 (s, 2H, CH2Ph), 6.88 (d, 2H, J = 8.5 Hz,
ArH), 7.13–7.17 (m, 3H, ArH), 7.34–7.38 (m, 3H,
ArH), 7.48–7.54 (m, 4H, ArH). Compound 29: d 2.40
(s, 3H, SCH3), 2.98 (s, 3H, NCH3), 3.71 (s, 3H,
OCH3), 4.51 (s, 2H, CH2Ph), 5.08 (s, 2H, CH2Ph),
6.87 (d, 2H, J = 8.5 Hz, ArH), 7.07–7.13 (m, 4H,
ArH), 7.22–7.32 (m, 6H, ArH). Compound 30: d 2.43
(s, 3H, SCH3), 3.11 (s, 3H, NCH3), 3.71 (s, 6H,
OCH3), 4.60 (s, 2H, CH2Ph), 6.68 (d, 1H, J = 8.0 Hz,
ArH), 6.85–6.88 (m, 2H, ArH), 7.16 (d, 1H,
J = 3.0 Hz, ArH), 7.36–7.40 (m, 3H, ArH), 7.48–7.56
(m, 4H, ArH). Compound 31: 2.55 (s, 3H, SCH3), 3.11
(s, 3H, NCH3), 3.71 (s, 6H, OCH3), 4.62 (s, 2H, CH2Ph),
5.33 (s, 2H, CH2Ph), 6.71 (s, 1H, ArH), 6.89–6.91 (m,
2H, ArH), 7.19 (s, 1H, ArH), 7.41–7.45 (m, 3H, ArH),
7.51–7.57 (m, 4H, ArH). Compound 32: d 2.43 (s, 3H,
SCH3), 3.13 (s, 3H, NCH3), 3.62 (s, 3H, OCH3), 3.70
(s, 6H, OCH3), 4.60 (s, 2H, CH2Ph), 6.52 (s, 2H,
ArH), 7.18 (d, 1H, J = 2.5 Hz, ArH), 7.37–7.40 (s, 3H,
ArH), 7.50–7.55 (s, 4H, ArH). Compound 33: d 2.56
(s, 3H, SCH3), 3.13 (s, 3H, NCH3), 3.63 (s, 3H,
OCH3), 3.70 (s, 6H, OCH3), 4.60 (s, 2H, CH2Ph), 5.31
(s, 2H, CH2Ph), 6.54 (s, 2H, ArH), 7.22–7.46 (m, 8H,
ArH).
5.9. 2-Methylthio-3-(phenyl or benzyl)-6-[(substituted
phenyl)carbonylamino]-3H-quinazolin-4-ones (46–57)
A solution of the 6-aminoquinazoline derivative 13 or 14
(0.022 mol) and the appropriate benzoyl chloride analog
(0.003 mol) in pyridine (10 mL) was heated under reflux
for 1 h and continued as mentioned under 34–39 to pro-
1
duce the N-acylated compounds 46–57 (Table 2). H
NMR (DMSO-d6), 46: d 2.50 (s, 3H, SCH3), 7.31–7.72
(m, 10H, ArH), 8.01 (d, 1H, J = 7.0 Hz, ArH), 8.28 (d,
1H, J = 7.5 Hz, ArH), 8.58 (s, 1H, ArH), 10.59 (br s,
1H, NH). Compound 47: d 2.50 (s, 3H, SCH3), 5.36
(s, 2H, CH2Ph), 7.22–7.36 (m, 5H, ArH), 7.51–7.65
(m, 5H, ArH), 7.98–8.01 (m, 1H, ArH), 8.20 (dd, 1H,
J = 9.0, 2.5 Hz, ArH), 8.64 (d, 1H, J = 2.5 Hz, ArH),
10.59 (s, 1H, NH). Compound 48: d 2.52 (s, 3H,
SCH3), 5.31 (br s, 1H, NH), 7.05–7.36 (m, 12H, ArH).
5.7. 2-Methylthio-3-(phenyl or benzyl)-6-[(substituted
phenyl)sulfonyl]amino-3H-quinazolin-4-ones (34–39)
A solution of the 6-aminoquinazoline analog 13 or 14
(0.002 mol) and the appropriate phenylsulfonyl chloride
derivative (0.003 mol) in pyridine (10 mL) was heated
under reflux for 1 h. Solvent was evaporated under re-
duced pressure and the remaining residue was then

S. T. Al-Rashood et al. / Bioorg. Med. Chem. 14 (2006) 8608–8621
8619
Compound 49: d 2.51 (s, 3H, SCH3), 5.54 (s, 2H,
CH2Ph), 7.00–7.51 (m, 12H, ArH), 12.82 (br s, 1H,
NH). Compound 50: d 2.40 (s, 3H, CH3Ph), 2.51 (s,
3H, SCH3), 7.32–7.69 (m, 9H, ArH), 7.88–7.97 (m,
1H, ArH), 8.23–8.32 (m, 1H, ArH), 8.56 (s, 1H, ArH),
10.46 (s, 1H, NH). Compound 51: d 2.40 (s, 3H,
CH3Ph), 2.51 (s, 3H, SCH3), 5.36 (s, 2H, CH2Ph),
7.22–7.31 (m, 3H, ArH), 7.33–7.39 (m, 4H, ArH), 7.62
(d, 1H, J = 8.5 Hz, ArH), 7.92 (d, 2H, J = 8.5 Hz,
ArH), 8.19–8.25 (dd, 1H, J = 8.5, 2.5 Hz, ArH), 8.65
(d, 1H, J = 2.5 Hz, ArH), 10.50 (br s, 1H, NH). Com-
pound 52: d 2.49 (s, 3H, SCH3), 3.89 (s, 3H, OCH3),
6.82–7.18 (m, 2H, ArH), 7.40–7.68 (m, 6H, ArH),
7.86–8.12 (m, 3H, ArH), 8.27 (d, 1H, J = 8.5 Hz,
ArH), 8.56 (s, 1H, ArH), 10.40 (br s, 1H, ArH). Com-
pound 53: d 2.50 (s, 3H, SCH3), 3.86 (s, 3H, OCH3),
5.36 (s, 2H, CH2Ph), 7.09 (d, 2H, J = 9.0 Hz, ArH),
7.24–7.37 (m, 5H, ArH), 7.61 (d, 1H, J = 9.0 Hz,
ArH), 8.01 (d, 2H, J = 8.5 Hz, ArH), 8.21–8.23 (dd,
1H, J = 8.5, 2.5 Hz, ArH), 8.62 (d, 1H, J = 2.5 Hz,
ArH), 10.43 (s, 1H, NH). Compound 54: d 2.49 (s,
3H, SCH3), 3.86 (s, 6H, OCH3), 7.10–7.12 (m, 2H,
ArH), 7.45–7.69 (m, 7H, ArH), 8.27–8.29 (m, 1H,
ArH), 8.53–8.54 (m, 1H, ArH), 10.37 (br s, 1H, NH).
Compound 55: d 2.50 (s, 3H, SCH3), 3.86 (s, 6H,
OCH3), 5.36 (s, 2H, CH2Ph), 7.11 (d, 1H, J = 8.5 Hz,
ArH), 7.25–7.34 (m, 3H, ArH), 7.58–7.68 (m, 5H,
ArH), 8.23 (d, 1H, J = 8.5 Hz, ArH), 8.60 (s, 1H,
ArH), 10.40 (s, 1H, NH). Compound 56: d 2.49 (s,
3H, SCH3), 3.75 (s, 3H, OCH3), 3.89 (s, 6H, OCH3),
7.35 (s, 2H, ArH), 7.46–7.49 (m, 2H, ArH), 7.56–7.59
(m, 3H, ArH),7.67 (d, 1H, J = 9.0 Hz, ArH), 8.23–8.31
(m, 1H, ArH), 8.52 (d, 1H, J = 2.5 Hz, ArH), 10.42 (s,
1H, NH). Compound 57: d 2.50 (s, 3H, SCH3), 3.75
(s, 3H, OCH3), 3.89 (s, 6H, OCH3), 5.35 (s, 2H, CH2Ph),
7.22–7.31 (m, 2H, ArH), 7.33–7.41 (m, 5H, ArH), 7.51
(d, 1H, J = 9.0 Hz, ArH), 8.17–8.25 (dd, 1H, J = 9.0,
2.5 Hz, ArH), 8.53 (d, 1H, J = 2.5 Hz, ArH), 10.45 (s,
1H, NH).
of MTX (1) was also formed showing IC50 value of
0.006 lM.
5.11. Molecular dynamics calculations
5.11.1. Enzyme structure. Starting coordinate of
hDHFR enzyme in tertiary complex with reduced-nico-
tinamide adenine dinucleotide phosphate (NADPH)
and MTX (1), code ID 1DLS, was obtained from the
Protein Data Bank of Brookhaven National Laborato-
ry. The structure of human DHFR was the L22Y
mutant enzyme in complex with NADPH and metho-
trexate (PDB entry 1DLS) from which the wild type en-
zyme was modeled prior to simulation.28 All the
hydrogens were added and enzyme structure was sub-
jected to a refinement protocol in which the constraints
on the enzyme were gradually removed and 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 molecular dynamics studies.
5.11.2. Molecular structure of the synthesized quinazo-
lines. The quinazoline analogs 5–57 were constructed
from fragment libraries in the Hyperchem program.39
The partial atomic charges for each analog were as-
signed with the semiempirical mechanical calculation
method ‘AM1’40,41 implemented in Hyperchem 6.03.
Conformational search was formed around all the rotat-
able bonds with an increment of 10° using conforma-
tional search module as implemented in HyperChem
6.03. All the conformers were minimized until the rms
˚
deviation was 0.01 kcal/mol A.
5.11.3. Docking and molecular dynamics simulations. The
docking and MD simulations were carried out on the
hDHFR enzyme for each compound. The active site
˚
of the enzyme was defined using a radius of 12.0 A
around MTX (1).MD simulations were carried out for
each ligand, with the rest of the enzyme kept fixed.
Non-standard charges of NADPH as a part of the en-
zyme structure were not fixed during the simulations.
The lowest energy conformer ‘global-minima’ was pre-
positioned using the crystal structure ligand ‘MTX’ as
a template for the initial placement at the enzyme-bind-
ing domain using a flexible fitting of the selected atoms
within the MTX ligand. Distance between the hydrogen
bond donor and acceptor atoms within the pocket sur-
rounding the ligand and the ligand itself was defined
and restrained. Once prepositioned, MTX was deleted
and each inhibitor was optimally docked in the
enzyme-binding pocket. For each of the quinazoline
analogs, energy minimizations (EM) were formed using
1000 steps of steepest descent, followed by conjugate
gradient minimization to a RMS energy gradient of
5.10. Dihydrofolate reductase (DHFR) inhibition assay
The assay mixture contained 50 mM Tris–HCl buffer
(pH 7.4), 50 lM NADPH, 20 lL DMSO or the same
volume of DMSO solution containing the test com-
pounds to a final concentration of 10ꢀ11–10ꢀ5 M, and
0.02 U of bovine liver DHFR, in a final volume of
2.0 mL. After addition of the enzyme, the mixture was
incubated at room temperature for 2.0 min, and the
reaction was initiated by adding 25 lM dihydrofolic
acid, the change in absorbance (DA/min) was measured
at 340 nm. The activity under these conditions was line-
ar for 10 min.16 Results are reported as % inhibition of
enzymatic activity calculated using the following
formula:
˚
0.01 kcal/mol A. MD simulation was formed using time
steps of 0.001 pico-second (ps), a distance dependent
dielectric of 4.00, and a non-bonded cut-off distance of
ꢀ
ꢁ
DA=mintest
DA=minDMSO
% inhibition ¼ 1 ꢀ
ꢁ 100
˚
8.0 A at 300 K. Complexes were first equilibrated for
10 ps and then simulated for 90 ps. Trajectory frames
were collected every 200 steps for detailed analysis on
the basis of potential energy, hydrogen bond interac-
tions, and rms deviation of the active site residues were
The % inhibition values were plotted versus drug con-
centration (log scale). The 50% inhibitory concentration
(IC50) of each compound was obtained. Inhibition plot

8620
S. T. Al-Rashood et al. / Bioorg. Med. Chem. 14 (2006) 8608–8621
16. Pignatello, R.; Spampinato, G.; Sorrenti, V.; Vicari, L.;
Giacomo, C. Di.; Vanella, A.; Puglisi, G. Pharm. Phar-
macol. Commun. 1999, 5, 299–304.
17. Wyss, P. C.; Gerber, P.; Hartman, P. G.; Hubschwerlen,
C.; Locher, H.; Marty, H.; Stahl, M. J. Med. Chem. 2003,
46, 2304–2312.
determined. Energy of binding was calculated as the dif-
ference between the energy of the complex and individ-
ual energies of the enzyme and ligand:
Ebinding ¼ EComplex ꢀ ðELigand þ EenzymeÞ:
where EComplex is the energy of the ligand-enzyme com-
plex, ELigand is the energy of the ligand corresponding to
the binding conformation, and Eenzyme is the energy of
the enzyme.40 The conformational energy difference
(<E) was calculated as the difference between the energy
of the conformation within the active site and the energy
of the global minimum conformation for each com-
pound. Root-mean-square deviation (rmsd) values of
active-site residues were determined as a validation for
the nature and flexibility of the active site.
18. Rastelli, G.; Pacchioni, S.; Sirawaraporn, W.; Sirawarap-
orn, R.; Parenti, M. D.; Ferrari, A. M. J. Med. Chem.
2003, 46, 2834–2845.
19. Gebauer, M. G.; McKinaly, C.; Gready, J. E. Eur. J. Med.
Chem. 2003, 38, 719–728.
20. Donkor, I. O.; Li, H.; Queener, S. F. Eur. J. Med. Chem.
2003, 38, 605–611.
21. Rosowsky, A.; Chen, H.; Fu, H.; Queener, S. F. Bioorg.
Med. Chem. 2003, 11, 59–67.
22. Zink, M.; Lanig, H.; Troschutz, R. Eur. J. Med. Chem.
2004, 39, 1079–1088.
23. Pignatello, R.; Guccione, S.; Forte, S.; Giacomo,
C.; Sorrenti, V.; Vicari, L.; Barretta, G. U.;
Balzano, F.; Puglisi, G. Bioorg. Med. Chem.
2004, 12, 2951–2964.
Acknowledgments
The financial support of King Abdulaziz City for
Science and Technology, Grant AT-12-18, is acknowl-
edged. Thanks are due to the NCI, Bethesda, MD, for
performing the antitumor testing of the synthesized
compounds. The technical assistance of Mr. Tanvir A.
Butt is greatly appreciated.
24. Raffa, D.; Edler, M. C.; Daidone, G.; Maggio,
B.; Merickech, M.; Plescia, S.; Schillaci, D.; Bai,
R.; Hamel, E. Eur. J. Med. Chem. 2004, 39,
299–304.
25. Foye, W. O.; Lemke, T. L.; Williams, D. A. Principles of
Medicinal Chemistry, 2nd ed.; Williams and Wilkins:
Media, PA, 2002.
26. Masur, H.; Polis, M. A.; Tuazon, C. V.; Ogota, A. D.;
Kovacs, J.; Katz, D.; Hilt, D.; Simmons, T.; Feuerstein, I.;
Lundgren, B.; Lane, H. C.; Chabner, B. A.; Allegra, J. C.
J. Infect. Dis. 1993, 167, 1422–1426.
References and notes
1. Masur, H. J. Infect. Dis. 1990, 161, 858–864.
2. Berman, E. M.; Werbel, L. M. J. Med. Chem. 1991, 34,
479–485.
27. Nordberg, M. G.; Kolmodin, K.; Aqvist, J.; Queener,
S. F.; Hallberg, A. J. Med. Chem. 2001, 44, 2391–
2402.
3. Roth, B. Fed. Proc. 1986, 45, 2665–2772.
4. Jolivet, J.; Cowan, K. H.; Curt, G. A.; Clendninn, N. J.;
Chabner, B. A. N. Eng. J. Med. 1983, 309, 1094–1104.
5. Borst, P.; Quellette, M. Annu. Rev. Microbiol. 1995, 49,
427–460.
6. Kozarek, R. A.; Patterson, D. J.; Gelfand, M. D.;
Botoman, V. A.; Ball, T. J.; Wilske, K. R. Ann. Intern.
Med. 1989, 110, 353–356.
7. Weinblatt, M. E.; Coblyn, J. S.; Fox, D. A.; Fraser, P. A.;
Holdsworth, D. E.; Glass, D. N.; Trentham, D. E. N. Eng.
J. Med. 1985, 312, 818–822.
8. Weinstein, G. D.; Frost, P. Arch. Dermatol. 1971, 103, 33–
38.
28. Nordberg, M. G.; Kolmodin, K.; Aqvist, J.; Queener,
S. F.; Hallberg, P. Eur. J. Pharm. Sci. 2004, 22, 43–
53.
29. Abdel Hamid, S. G.; El-Obaid, H. A.; Al-Majed, A. A.;
El-Kashef, H. A.; El-Subbagh, H. I. Med. Chem. Res.
2001, 10, 378–389.
30. El-Subbagh, H. I.; Abu-Zaid, S. M.; Mahran, M. A.;
Badria, F. A.; Al-Obaid, A. M. J. Med. Chem. 2000, 43,
2915–2921.
31. Abdel Hamid, S. G.; El-Obaid, H. A.; Al-Rashood, K. A.;
Khalil, A. A.; El-Subbagh, H. I. Sci. Pharm. 2001, 69,
351–366.
32. Khalil, A. A.; Abdel Hamide, S. G.; Al-Obaid, A. M.; El-
Subbagh, H. I. Arch. Pharm. Pharm. Med. Chem. 2003,
336, 95–103.
33. Al-Omar, M. A.; Abdel Hamide, S. G.; Al-Khamees,
H. A.; El-Subbagh, H. I. Saudi Pharm. J. 2004, 12, 63–
71.
9. Abu-Shakra, M.; Gladman, D. D.; Thorne, J. C.; Long, J.;
Gough, J.; Faresell, V. T. J. Rheumatol. 1995, 22, 241–245.
10. Mullarkey, M. F.; Blumenstein, B. A.; Andrade, W. P.;
Bailey, G. A.; Olason, I.; Wetzel, C. E. N. Eng. J. Med.
1988, 318, 603–607.
11. Maroun, J. Semin. Oncol. 1988, 15, 17–21.
12. Lin, J. T.; Cashmore, A. R.; Baker, M.; Dreyer, R. N.;
Ernstoff, M.; Marsh, J. C.; Bertino, J. R.; Whitfield, L. R.;
Delap, R.; Griloo-Lopez, A. Cancer Res. 1987, 47, 609–616.
13. Kovacs, J. A.; Allegra, C. A.; Swan, J. C.; Drake, J. C.;
Parillo, J. E.; Chabner, B. A.; Masur, H. Antimicrob.
Agents Chemother. 1988, 43, 430–433.
14. Allegra, C. J.; Chanber, B. A.; Tuazon, C. U.; Ogata
Arakaki, D.; Baird, B.; Darke, J. C.; Simmons, J. T.;
Lack, E. E.; Shelhamer, J. H.; Balis, F.; Walker, R.;
Kovacs, J. A.; Cliffordlane, H.; Masur, H. N. Eng. J. Med.
1987, 317, 978–985.
34. Grever, M. R.; Schepartz, S. A.; Chabner, B. A. Semin.
Oncol. 1992, 19, 622–638.
35. Monks, A.; Scudiero, D.; Skehan, P. J. Natl. Cancer. Inst.
1991, 83, 757–766.
36. Boyd, M. R.; Paull, K. D. Drug Rev. Res. 1995, 34,
91–109.
37. Skehan, P.; Storeng, R.; Scudiero, D.; Monks, A.;
McMahon, J.; Vistica, D.; Warren, J. R.; Bokesch, H.;
Kenney, S.; Boyd, M. R. J. Natl. Cancer Inst. 1990, 82,
1107–1112.
38. (a) Profeta, S.; Allinger, N. L. J. Am. Chem. Soc. 1985,
107, 1907–1918; (b) Dosen-Micovic, L.; Jeremic, D.;
Allinger, N. L. J. Am. Chem. Soc. 1983, 105, 1716–1722;
(c) Allinger, N. L. J. Am. Chem. Soc. 1977, 99, 8127–
8134.
15. El-Subbagh, H. I.; El-Sherbeny, M. A.; Nasr, M. N.;
Goda, F. E.; Badria, F. A. Boll. Chim. Farm. 1995, 134,
80–84.

S. T. Al-Rashood et al. / Bioorg. Med. Chem. 14 (2006) 8608–8621
8621
39. Hyperchem: Molecular Modeling System, Hypercube,
Inc., Release 6, Florida, USA, 1999.
40. Cornell, W. D.; Cieplak, P.; Bayly, C. I.; Gould, I. R.;
Merz, K. M., Jr.; Ferguson, D. M.; Spellmeyer, D. C.;
Fox, T.; Caldwell, J. W.; Kollman, P. A. J. Am. Chem.
Soc. 1995, 117, 5179–5197.
42. Abdel-Aziz, A. A.-M.; El-Subbagh, H. I.; Kunieda, T.
Bioorg. Med. Chem. 2005, 13, 4929–4935, and the refer-
ences cited therein.
43. Klon, A. E.; Heroux, A.; Ross, L. J.; Pathak, V.; Johnson,
C. A.; Piper, J. R.; Borhanti, D. w. J. Mol. Biol. 2002, 320,
677–682.
41. Dewar, M. J. S.; Zoebisch, E. G.; Healy, E. F.; Stewart, J.
J. P. J. Am. Chem. Soc. 1985, 107, 3902–3909.
44. Gokhale, V. M.; Kulkarni, V. M. J. Comput. Aided Mol.
Des. 2000, 14, 495–506.