Novel and versatile methodology for synthesis of cyclic imides and evaluation of their cytotoxic, DNA binding, apoptotic inducing activities and molecular modeling study

Article abstract of DOI:10.1016/j.ejmech.2006.12.003

Versatile method has been developed for synthesis of N-substituted imides. Thus, acid anhydrides, imides and dicarboxylic acids were successfully subjected to dehydrative cyclization with substituted amines using DPPOx and Et₃N to afford N-substituted imides under mild conditions. The DNA binding and apoptosis induction were investigated with regard to their potential utility as cytotoxic agents. Molecular modeling methods are used to study the cytotoxic activity of the active compounds by means of molecular and quantum mechanics.

Full text of DOI:10.1016/j.ejmech.2006.12.003

European Journal of Medicinal Chemistry 42 (2007) 614e626
http://www.elsevier.com/locate/ejmech
Original article
Novel and versatile methodology for synthesis of cyclic imides and
evaluation of their cytotoxic, DNA binding, apoptotic inducing
activities and molecular modeling study
*
Alaa A.-M. Abdel-Aziz
Department of Pharmaceutical Chemistry, College of Pharmacy, King Saud University, P.O. Box 2457, Riyadh 11451, Saudi Arabia
Received 31 May 2006; received in revised form 25 November 2006; accepted 1 December 2006
Available online 10 December 2006
Abstract
Versatile method has been developed for synthesis of N-substituted imides. Thus, acid anhydrides, imides and dicarboxylic acids were
successfully subjected to dehydrative cyclization with substituted amines using DPPOx and Et3N to afford N-substituted imides under mild
conditions. The DNA binding and apoptosis induction were investigated with regard to their potential utility as cytotoxic agents. Molecular
modeling methods are used to study the cytotoxic activity of the active compounds by means of molecular and quantum mechanics.
Ó 2006 Elsevier Masson SAS. All rights reserved.
Keywords: N-Substituted imides; Novel synthesis; Apoptosis induction; DNA-binding; Molecular modeling
1. Introduction
reported [6]. However, Westaway and Gedye [7], did not see
any differences when the reaction was carried out by micro-
wave or conventional heating in DMF. Therefore, synthesis
of functionalized imides is still a challenging endeavor.
We describe herein an efficient and a mild approach for the
synthesis of N-substituted imides from the parent imides, anhy-
drides and dicarboxylic acids under mild conditions. This ap-
proach requires only a few minutes of reaction time, in contrast
to conventional method that requires a long reaction time and ex-
pensiveLewisacid,whichisdifficulttohandle.Weenvisagedthat
the reaction of an anhydride with an appropriately substituted
amine and subsequent in situ cyclization of the resulting amic
acid in the presence of DPPOx [8] and Et3N would give the cor-
responding imide derivatives which were investigated for their
cytotoxic activity which may lead to tumor cell apoptosis.
Imide derivatives are a valuable group of bioactive com-
pounds showing androgen receptor antagonists, anti-inflam-
matory, anxiolytic, antiviral, antibacterial, and antitumor
properties [1]. In spite of their wide applicability, available
procedures for their synthesis are limited [2]. Among them,
the dehydrative condensation of an anhydride and an amine
at high temperature [3] and the cyclization of the amic acid
in the presence of acidic reagents are the typical methods of
choice [4]. The direct N-alkylation of maleimide with alcohols
under Mitsunobu reaction conditions is an alternative method
for the synthesis of imide [2a]. Many catalysts including Lewis
acids and hexamethyldisilazane [5] have been proposed for the
synthesis of N-alkyl and N-arylimide derivatives. However,
each of these routes has its own synthetic problems when
applied to a range of derivatives, such as, low yields, numer-
ous by-product formations and only a narrow range of imide
derivatives can be synthesized. Moreover, microwave assisted
addition of amines to phthalic anhydride has been also
2. Results and discussion
2.1. Chemistry
Activation of the carboxyl groups under mild conditions is
apparently of great value as a fundamental process in a wide
* Tel.: þ966 562947305; fax: þ966 14676220.
E-mail address: alaa_moenes@yahoo.com
0223-5234/$ - see front matter Ó 2006 Elsevier Masson SAS. All rights reserved.
doi:10.1016/j.ejmech.2006.12.003

A.A.-M. Abdel-Aziz / European Journal of Medicinal Chemistry 42 (2007) 614e626
615
scope of chemical conversions including amide, imide and es-
ter formations [8a]. The 5- and 6-membered heterocycles such
as imidazoles, triazoles, 2-thiazolidinethiones and 2-pyridine-
thiol have been successfully used in the acylation and conden-
sation reactions as the bifunctional leaving moieties [9].
In order to substantiate the concept, N-benzylphthalimide
generated from phthalimide and benzylamine in a MeCN solu-
tion was heated at 40 ꢀC with an equimolar amounts of ben-
zylamine, Et3N and diphenyl 2-oxo-3-oxazolinylphosphonate
[DPPOx] for 60 min to afford the corresponding N-substituted
imide derivatives (Table 1). This favorable result encouraged
us to further improve the efficiency of the reaction to get
high yields of imide derivatives in a short reaction time. In
order to optimize the reaction conditions, we briefly investi-
gated the stoichiometric ratio of the reagents and reactants.
The imides’ formation was observed almost quantitatively in
0.25e1 h, stirring in MeCN with equimolar amounts of an an-
hydride or imide and an amine, whereas the molar ratios of
Et3N and DPPOx showed a significant influence on the
reaction time and the yield of imides. Although, theoretically,
1 mol equiv of Et3N seems to be sufficient to complete the
reaction, practically the isolated yield of imides was higher
using 1.5 mol equiv of Et3N and 1.5 equiv of DPPOx. When
the reaction was carried out with an excess amount of Et3N
(>1.5 equiv) and/ or DPPOx (>1.5 equiv), the yield of the
imides was reduced drastically and formation of undesired
products was observed upon prolonged heating. In the presence
of less than an equimolar quantity of Et3N or DPPOx, the
reaction was not completed even after long reaction time.
The imides’ formation was not observed even in a trace amount
in the absence of either DPPOx or Et3N.
Such a cyclizing condensation of phthalimide and benzyl-
amine promoted by DPPOxeEt3N was also observed for the
reaction of phthalic anhydride or phthalic acid. The phthalic
anhydride was smoothly converted to the corresponding N-ben-
zylphthalimide with nearly quantitative yield when the reaction
proceeded at 40 ꢀC for 15e30 min (Table 1, entries 9 and 10).
On the other hand, the reaction of phthalic acid with benzyl-
amine gave good yield after increasing the reaction temperature
to 60 ꢀC for 45 min using 2.0 mol equiv of both Et3N and
DPPOx (Table 1, entries 11e13). Moreover, other bases such
as diisopropylethylamine, DABCO, HMDS and dimethylamio-
pyridine were ineffective for accelerating the reaction to get
high yields (Table 1, footnote). There seems to be little relation-
ship between the accelerating effect and basic properties of
working base. Furthermore, the nature of the solvent also
affected the reaction yield. However, the reactions proceeded in
the presence of solvents, such as THF and DMF afforded mod-
erate yields (Table 1, entries 1e4) while the result of the reaction
in toluene and benzene (Table 1, entries 7 and 8) was completely
ineffective to accelerate the reaction yield, in contrast to the
reaction proceeded in MeCN which gave N-benzylphthalimide
exclusively (Table 1, entries 5 and 6).
Table 1
DPPOx promoted the synthesis of N-benzylphthalimide by varying solvents,
bases and reaction time
COOH
COOH
O
BnNH2, Et3N (1.5eq)a
O
N
DPPOx (1.5 eq.), 40°C
Solvent, Time
X
O
To generalize the scope and limitations of DPPOx and Et3N
promoted imides synthesis, the reaction was examined with
various structurally diverse amines and anhydrides or imides.
These results are summarized in Table 2. In most cases, the
amic acid ester cyclization (Fig. 1) proceeded smoothly with
1.5 mol equiv of DPPOx and 1.5 mol equiv of Et3N in hot
MeCN, giving good to excellent yields of the desired bicy-
clo[2.2.2]-oct-5-ene-2,3-dicarboximides 1, 2,3-pyrazinedicar-
boximides 2, 2,3-pyridinedicarboximides 3, or phthalimides
4 (Table 2). Yields of imides derived from an anhydride or im-
ide and an alkylamine or arylalkylamine, such as ethylamine,
tert-butylamine, phenethylamine and benzylamine were found
to be better than an aromatic amine such as aniline which may
be attributed to the lower nucleophilicity of aromatic amines
(Table 2, entries 1e10 vs 11e13). Generally, high yield was
observed when less hindered primary amines were used which
decreased when sterically congested amines were investigated
(Table 2, entries 1e11 vs 12 and 13). Apparently, structural
variation in the arylamine in o-position with steric bulky group
affecting the yields of the product, such as aniline and p-pen-
tylaniline (Table 2, entries 11 and 14) gave higher yield than
o,o-dimethylaniline (Table 2, entry 12). However, increasing
the bulkiness of o-position from 2,6-dimethyl to 2,6-diisopro-
pylaniline may induce a tremendous effect on the reaction
O
X= O, NH
Entry
Acid
Time (min)
Solvent
Yield (%)b
1
2
3
4
5
6
7
8
60
120
30
THF
THF
40
53
61
67
94
95
33
27
O
DMF
DMF
60
NH
20
30
MeCN
MeCN
Toluene
Benzene
O
120
180
9
15
MeCN
97
O
O
10
11
O
O
30
30
MeCN
MeCN
98
50
OH
OH
12c
13c
30
45
MeCN
MeCN
71
75
O
aLower yields have been observed using bases other than Et3N, such as, diiso-
propylethylamine (60%), HMDS (44%), Dimethylaminopyridine (52%),
DABCO (33%).
b
Isolated yields.
The reaction temperature was 60 ꢀC.
c

616
A.A.-M. Abdel-Aziz / European Journal of Medicinal Chemistry 42 (2007) 614e626
Table 2
DPPOx and Et3N promoted synthesis of N-substituted imide derivatives from anhydrides or acid imides using various amines
O
O
O
O
O
N
N
RNH2, Et3N (1.5eq),
N R
N R
N R
N
X
R
N
DPPOx (1.5 eq.), 40°C,
MeCN, Time
O
O
O
O
O
1
2
3
4
X= O, NH
R= Et (a), t-But (b), 1-adamantyl (c), Phenethyl (d), Benzyl (e), 4-methoxyBenzyl (f), 3,4-dimethoxybenzyl (g),
4-chloroBenzyl (h), (R)-α-Methylbenzyl (i), (S)-α-Methylbenzyl (j), Phenyl (k), 2,6-Dimethylphenyl (l),
2,6-Diisopropylphenyl (m), 4-pentylphenyl (n), 4-chlorophenyl (o), 4-trifluoromethylphenyl (p)
Entry
Anhydride
Time (min)a
Imide
Yields (%)b
Entry
Anhydride
Time (min)a
Imide
Yields (%)b
1
2
3
4
5
15
20
30
15
15
1a
1b
1c
1d
1e
98
94
93
99
96
28
29
30
31
32
45 (45)
45 (45)
15 (30)
20 (30)
30 (45)
2l
74 (75)
65 (62)
93 (89)
86 (82)
83 (84)
2m
2n
2o
2p
6
20
15
30
30
30
30
45
45
20
30
30
1f
97
95
93
91
90
89
76
67
95
87
85
33
34
35
36
37
38
39
40
41
42
43
15 (30)
15 (30)
20 (30)
15 (30)
15 (30)
20 (45)
20 (45)
30 (45)
30 (30)
30 (30)
30 (45)
3a
3b
3c
3d
3e
3f
99 (96)
93 (89)
89 (85)
96 (94)
95 (95)
92 (93)
90 (88)
91 (86)
89 (88)
91 (92)
87 (85)
7
1g
1h
1i
O
8
9
O
10
11
12
13
14
15
16
1j
O
1k
1l
O
3g
3h
3i
1m
1n
1o
1p
O
N
3j
3k
O
17
18
19
20
21
20 (30)
30 (45)
30 (45)
15 (30)
15 (30)
2a
2b
2c
2d
2e
96 (93)
91 (90)
87 (83)
97 (95)
96 (97)
44
45
46
47
48
45 (45)
45 (45)
15 (30)
30 (30)
30 (30)
3l
70 (71)
61 (60)
91 (87)
86 (85)
84 (82)
3m
3n
3o
3p
O
N
O
N
22
23
20 (45)
20 (45)
2f
2g
94 (91)
95 (94)
49
50
20 (30)
15 (30)
4c
4d
92 (90)
98 (99)
O
O
24
25
26
27
a
30 (45)
30 (45)
30 (45)
30 (45)
2h
2i
92 (89)
93 (89)
94 (93)
90 (88)
51
52
53
54
15 (20)
15 (30)
30 (45)
30 (45)
4e
4i
97 (94)
93 (92)
91 (91)
74 (69)
O
2j
2k
4k
4m
O
Time in parentheses was the reaction time using acid imides.
Yield in parentheses was the isolated yield using acid imides.
b
products (Table 2, entry 13). It is clear that the o-substituents
could significantly affect the amic acid ester formation which
might play a crucial role in condensation step. Moreover, the
presence of an electron-withdrawing group in an arylamine,
such as a chloro or trifluoromethyl group leads to a decrease
in the yield, compared to an arylamine containing an electron-
donating group, which may be attributable to its poor nucleophilic
character (Table 2, entries 14e16). When chiral amines such as
(þ)-(R)-a-methylbenzylamine or (ꢁ)-(S)-a-methylbenzylamine
were used, the corresponding imide derivatives were isolated
O
Et3NH
O
O
P
X
O
OPh
OPh
OPh
O
X
O
O
X
P
RNH2
-H2X
-Et3N
OPh
Ox
Acid
+
HN
O
O
NR
Ox
NH
R
N
Et
O
P
O
O
O
Et
Et
Ion pair
PhO
PhO
N
O
Amic acid ester
1- 4
Ox
O
DPPOx
Fig. 1. Plausible mechanism of DPPOxeEt3N promoted N-substituted imides formation.

A.A.-M. Abdel-Aziz / European Journal of Medicinal Chemistry 42 (2007) 614e626
617
without racemization under the described reaction conditions
(Table 2, entries 9, 10, 25, 26, 41 and 42). It is clear from
Table 2, the nature of starting acid anhydride or imide did
not affect the reaction yields such as the nonplanar bicy-
clo[2.2.2]-oct-5-ene-2,3-dicarboxylic acid anhydride (Table 2,
entries 1e16) gave nearly similar results with other planar acid
anhydride including the 2,3-pyrazinedicarboxylic anhydride
(Table 2, entries 17e32), 2,3-pyridinedicarboxylic anhydride
(Table 2, entries 33e48) and finally phthalic anhydride (Table 2,
entries 49e54).
The plausible mechanism of the direct formation of imides
may involve the coupling in a concerted manner through pen-
tavalent phosphorus ion pair and amic acid ester formation as
depicted in Fig. 1. We believe that the DPPOxeEt3N pro-
moted formation of N-substituted imide could function by
the reaction with acid anhydride or imide to generate a penta-
valent phosphorus ion pair [8]. The ion pair is attacked by
a substituted amine to afford the corresponding amic acid ester
via a nucleophilic substitution. This amic acid ester can un-
dergo intramolecular nucleophilic substitution to afford the
corresponding N-substituted imide, oxazolone and H2O or
NH3 (in case of acid imide). Moreover, the direct formation
of an imide from phthalic acid may involve the activation
through phthalic anhydride intermediate followed by amic
acid ester formation as described above [8].
compounds in this investigation with IC50 values in the range
of 23e10 mM. Compounds 1b, 1he1j and 2b showed moderate
inhibition activity with IC50 values of 27, 51, 43, 49 and 32 mM,
respectively; while other compounds showed weak activities.
2.2.2. DNA as an affinity probe for evaluation
of biologically active compounds
DNA is the pharmacologic target of many drugs currently in
clinical use or in advanced clinical trials. Small molecules that
bind genomic DNA have proven to be effective anticancer, anti-
biotic, and antiviral therapeutic agents. It is known that small
molecules can bind to DNA through various modes of interac-
tion. They are (i) minor groove binding, (ii) major groove bind-
ing, (iii) intercalation (between two base pairs) and (iv) surface
binding [11]. Avariety of methods have been utilized for study-
ing the interaction of small molecular weight compounds with
DNA, such as equilibrium dialysis [12].
Briefly [13], a fixed amount of ligand is spotted on the RP-
18 TLC plates followed by the addition of known amount of
DNA on the same spot. The plate was then developed and
the position of the DNA was determined by spraying the plates
with anisaldehyde reagent. It is important to establish if the re-
sponse of the test system is dependent on the dose of the test
substance. In the presence of increasing quantities of DNA
intercalators, a greater portion of DNA is bound to form a com-
plex, and consequently, the free DNA was detected as a blue
spot (Rf; MeOHeH2O, 8:2) on RP-18 TLC after spraying
with anisaldehyde reagent. On the other hand, compounds
with high binding affinity for DNA retained on the base line.
However, when the DNA was mixed with compounds with
which it is known to interact (ethidium bromide), the complex
was retained at the origin when MeOHeH2O (8:2) was used
for elution. Inactive compounds did not cause the DNA to
be retained at the origin.
2.2. Biological assays
2.2.1. Cytotoxic activity
Cultured mammalian cells have been utilized recently to
determine the response of the tumor cells isolated from cancer
patients to various chemotherapeutic [10] agents with very
high sensitivity (concentration as low as 0.25e1.7 mM). Mam-
malian cell culture systems have the benefit to be faster, less
expensive and more sensitive than using the intact animals.
Cytotoxicity assay of the prepared compounds expressed in
IC50 values (Table 3) revealed that compounds 1c, 1le1n, 2c,
2le2n, 2c, 3le3n, 4c and 4m proved to be the most active
Moreover, methyl green reversibly binds polymerised DNA
forming a stable complex at neutral pH. The maximum absorp-
tion for the DNA/methyl green complex is 642e645 nm. Incu-
bation for 24 h, in the buffer used for displacement reactions in
Table 3
Cytotoxic activities of the N-substituted cyclic imides (mM) against Vero African green monkey kidney cells
Compound
IC50
Compound
IC50
Compound
IC50
Compound
IC50
20
1a
1b
1c
1d
1e
1f
>100
27
14
87
74
67
63
51
43
49
67
13
16
10
77
75
2a
2b
2c
2d
2e
2f
>100
32
18
3a
3b
3c
3d
3e
3f
>100
33
18
4c
4d
100
100
>100
82
4e
>100
>100
>100
>100
88
>100
>100
>100
>100
95
4j
4k
4m
5-Fluorouracil
23
0.2
1g
1h
1i
2g
2h
2i
3g
3h
3i
79
85
81
92
1j
2j
3j
1k
1l
2k
2l
71
18
3k
3l
72
20
1m
1n
1o
1p
2m
2n
2o
2p
20
16
3m
3n
3o
3p
21
18
86
80
92
87

618
A.A.-M. Abdel-Aziz / European Journal of Medicinal Chemistry 42 (2007) 614e626
this study, resulted in virtually a complete loss of methyl green
absorbance. This colorimetric assay [14] was used to measure
the displacement of methyl green from DNA by compounds
having ability to bind with DNA. The degree of displacement
was determined spectrophotometrically by measuring the
change in the initial absorbance of the DNA/methyl green so-
lution in the presence of reference compound.
Results from DNA binding assay revealed that compounds
1n, 2n, 3n, 1l, 2l, 3l and 1c showed the highest affinity for
DNA, which was demonstrated by retaining the complex at
the origin or by migrating for a very short distances, and by mea-
suring C50 (concentration required for 50% decrease in the ini-
tial absorbance of the DNA/methyl green solution). Compounds
1d, 1e, and 2d showed moderate activity while compounds 1a,
2a, 3a, 1b, 2b, 3b and 1g showed weak activity.
As mentioned above, development of drugs that can
effectively trigger apoptosis in cancer cells has been receiving
considerable attention [15h]. It is a well-known fact that the
regulating mechanisms of apoptosis are extremely complex,
but it is also known that the way in which the apoptotic pro-
cesses function differs, depending on whether the cells
involved are tumoral or healthy. These differences are due to
the fact that the tumoral lines are more sensitive when these
mechanisms are activated, while the healthy cells possess
the ability to make repairs, which often counteracts this apo-
ptotic process, maintaining survival of the cell [18].
However, all active compounds, subjected to cytotoxic
assay, have been subjected to apoptosis methodology for
studying the mechanism of cytotoxic activity using blood
neutrophils and compared with positive control quercetin
(Table 4). The data revealed that the human neutrophils
derived from a peripheral blood of normal subjects in culture
undergo morphological and chromatin fragmentation changes
of programmed cell death. Moreover, the cytotoxic activities
were correlated to apoptotic mechanism as indicated by the
results obtained in Table 4.
2.2.3. Apoptosis assay
Progressive elucidation of the molecular mechanisms in-
volved in cancer has opened up a new horizon for the develop-
ment of new antitumor compounds. Apoptosis or programmed
cell death is the prevalent mechanism complementary to
proliferation that is critical for the normal development and
function of multicellular organisms. Rapid proliferation needs
to be balanced by apoptosis to maintain a constant cell num-
ber. Retarded cell death contributes to a wide variety of human
cancers [15]. Apoptotic pathways might be significantly
altered in cancer cells with respect to untransformed cells,
and these differences might present a therapeutic window
that can be exploited for the development of cancer drugs
[16]. In contrast to necrosis, this tightly regulated and complex
process exhibits some typical morphological changes, such as
chromatin condensation, membrane blebbing, formation of
apoptotic bodies, and in most cases, DNA fragmentation [17].
2.3. Quantitative structureeactivity relationship
The obtained cytotoxic results revealed that (Table 3), in
the nonplanar (T-shaped imides) series (1ae1p), the N-ethyl
1a, N-phenethyl 1d and N-benzyl 1e moieties were the least
active as evidenced by their IC50 values of >100, 87 and
74 mM, respectively. Introduction of one or two methoxy
groups and even a chloro moiety in the benzyl group resulted
in a slight increase in the cytotoxic activity. Moreover, the re-
placement of one hydrogen atom from the a-carbon of benzyl
group with a methyl moiety lead to a sharp increase of activity
Table 4
The effects of the active compounds on the apoptosis of peripheral blood neutrophils
Apoptotic neutrophilsb (%)
a
Compound
IC50
0 h
24 h
48 h
72 h
1b
1c
27
14
13
16
10
32
18
18
20
16
33
18
20
21
18
20
23
ndc
ndc
0.31 ꢂ 0.12
0.89 ꢂ 0.23
0.92 ꢂ 0.20
0.87 ꢂ 0.10
0.95 ꢂ 0.04
0.32 ꢂ 0.13
0.72 ꢂ 0.10
0.65 ꢂ 0.23
0.44 ꢂ 0.10
0.77 ꢂ 0.30
0.36 ꢂ 0.02
0.80 ꢂ 0.10
0.39 ꢂ 0.01
0.36 ꢂ 0.21
0.69 ꢂ 0.03
0.37 ꢂ 0.20
0.34 ꢂ 0.23
0.37 ꢂ 0.01
0.88 ꢂ 0.03
12.82 ꢂ 0.51
35.5 ꢂ 0.42
38.40 ꢂ 0.20
32.40 ꢂ 0.50
39.11 ꢂ 0.01
12.00 ꢂ 0.2
30.01 ꢂ 0.13
28.00 ꢂ 0.34
24.07 ꢂ 0.09
31.10 ꢂ 0.10
11.00 ꢂ 0.30
31.10 ꢂ 0.23
22.21 ꢂ 0.31
19.10 ꢂ 0.33
32.11 ꢂ 0.01
21.11 ꢂ 0.23
15.56 ꢂ 0.04
0.39 ꢂ 0.01
39.90 ꢂ 0.02
29.76 ꢂ 0.04
5321 ꢂ 0.28
5322 ꢂ 0.2
35.64 ꢂ 0.06
68.11 ꢂ 2.29
67.98 ꢂ 1.20
59.20 ꢂ 1.40
70.01 ꢂ 1.30
33.91 ꢂ 0.60
54.05 ꢂ 0.02
50.10 ꢂ 0.10
46.11 ꢂ 0.03
60.00 ꢂ 1.10
30.23 ꢂ 0.03
56.10 ꢂ 0.70
44.16 ꢂ 0.30
41.62 ꢂ 0.50
57.87 ꢂ 1.10
42.01 ꢂ 0.21
39.10 ꢂ 0.03
0.36 ꢂ 0.01
1l
1m
1n
49.11 ꢂ 0.80
58.06 ꢂ 0.02
27.60 ꢂ 0.05
45.10 ꢂ 0.11
42.03 ꢂ 0.10
38.10 ꢂ 0.20
4920 ꢂ 0.03
25.07 ꢂ 0.21
46.00 ꢂ 0.04
35.00 ꢂ 0.24
32.77 ꢂ 0.24
47.05 ꢂ 0.11
34.66 ꢂ 0.31
3021 ꢂ 0.04
0.39 ꢂ 0.01
49.90 ꢂ 0.02
2b
2c
2l
2m
2n
3b
3c
31
3m
3n
4c
4m
DMSO
Quercetin
a
71.30 ꢂ 0.07
Data was taken from Table 3.
b
c
Values represent the percentage of apoptotic neutrophils (mean ꢂ SD, n ¼ 3e5 separate determinations).
No data available.

A.A.-M. Abdel-Aziz / European Journal of Medicinal Chemistry 42 (2007) 614e626
619
as indicated by their IC50 (1i; 43 mM, 1j; 49 mM vs 1e;
2.4. Molecular modeling study
74 mM). Replacement of N-ethyl moiety with the correspond-
ing N-tert-butyl 1b or N-(1-adamantyl) 1c analogue resulted in
controversial values of inhibition with IC50 values of 27 mM
and 14 mM, respectively. In the N-phenyl series, introduction
of an electron-donating functional group resulted in a sharp
increase in activity by 6-fold (1l; 13 mM, 1m; 16 mM and
1n; 10 mM vs 1k; 67 mM); while introduction of an electron-
withdrawing group lead to a decrease in activity as indicated
by their IC50 (1o; 77 mM and 1p; 75 mM vs 1k; 67 mM). Re-
placing the T-shaped imides (structure 1) with the planar
form such as series 2, 3 or 4 decreased the activity such as
IC50 values of 2c, 3c, and 4c were 18 mM, 18 mM and
20 mM, respectively, while IC50 value of compound 1c was
14 mM. It is obvious from the structureeactivity profile of
N-substituted imides that substituents with steric bulky 1-ada-
mantyl moiety or aryl groups greatly influence the cytotoxic
activity (Table 3). Apparently, a small structural variation in
the N-substituted imides may induce a tremendous effect on
cytotoxic activity. It is noteworthy that the observed cytotoxic
activity was highly dependent on the bulkiness of the N-aryl
substituents on the imide core, in which ortho-disubstituents
played an important role in achieving an excellent level of cy-
totoxic activity indicating that the steric bulk of the substitu-
ents played a critical role in the cytotoxic enhancing
activity. The 1-adamantyl, 2,6-dimethylphenyl and 4-n-pentyl-
phenyl groups represent the N-substituents of choice for ex-
pressing significant activity, such as compounds 1c, 1l, 1n
and their analogues in series 2 and 3 which were five times
more active than the phenyl derivatives 1k, 2k and 3k and
10-fold more active than the less bulky ethyl and arylalkyl de-
rivatives 1a, 2a and 3a.
The results of the DNA binding assay revealed that com-
pounds 1n, 2n, 3n, 1l, 2l, 3l and 1c exhibited marked affinity
for DNA as indicated by their concentration required for 50%
decrease in the initial absorbance of the DNA/methyl green
solution (31, 42, 49, 53, 59, 64, 69 mg/ml, respectively). More-
over, low levels of DNA binding affinity were achieved for 1d
(75 mg/ml) and 1e (77 mg/ml), while compound 1a showed
weak activity (90 mg/ml). The obtained results might suggest
that DNA binding is partially correlated with the cytotoxic
activity of target compounds (Table 3). It would appear that
affinity for DNA is not the sole determinant of cytotoxic
potency in this series of compounds. The existence of a dual
mechanism is possible whereby DNA binding contributes as
one mechanism only.
An attempt to gain a better insight on the molecular struc-
tures of the active compounds 1c, 1l, 1n, 2c, 2l and 2n and
nonactive compounds 1d and 1e, conformational analysis of
the target compounds has been performed by use of the
MMþ [19] force-field (calculations in vacuo, bond dipole
option for electrostatics, PolakeRibiere algorithm, and RMS
˚
gradient of 0.01 kcal/A mol) as implemented in HyperChem
5.1 [20]. As clear from the calculation, most of the target
compounds exhibit structural similarity as indicated by
their molecular parameters (Table 5). The results show that
compounds 1d and 1e exhibited the same arrangement of
the aryl groups around the imide core showing two main con-
formers syn and anti in which the aryl moiety was deviated
from the planar imide core by ꢁ88 to ꢁ94ꢀ in case of syn-con-
former and by 89e102ꢀ in case of anti-conformer (Fig. 2). The
lower activity of such compounds may be attributed to the
occurrence of these two conformers caused by low barrier of
Table 5
Molecular parameters of the active compounds compared with nonactive one
O
O
N
N
O
O
T-shaped cyclic imide
planar cyclic imide
f
Compound
Log Pa
MVb
MSAc
Rd
Plane
anglee (tꢀ)
IC50
1b
1c
1.3
2.2
3.1
4.6
4.2
0.9
1.8
2.7
4.2
3.8
1.4
2.3
3.2
4.6
4.3
2.7
5.1
0.8
704
870
806
971
1006
592
754
707
891
892
601
765
717
901
902
774
912
628
417
497
469
537
599
370
449
428
518
549
373
453
434
522
551
456
526
387
65.8
87.1
59.9
59.9
27
14
1l
82.6
101.1
95.1
52.4
73.7
69.2
87.9
81.7
54.8
76.1
71.7
90.4
84.2
78.6
92.9
56.5
91.4
81.6
13
16
1m
1n
2b
2c
90.0
57.9
10
32
60.4
94.9
18
18
2l
2m
2n
3b
3c
88.6
87.1
20
16
57.1
60.3
33
18
3l
95.1
88.7
20
21
3m
3n
4c
87.2
60.2
18
20
4m
1a
88.7
89.3
23
>100
The obtained apoptotic testing results (Table 4), for the
most active cytotoxic compounds along with value obtained
for quercetin, the reference substance, showed that compounds
1c, 1le1n, 2c, 2l, 2n, 3c and 3n achieved significant apoptosis
levels as indicated by their percentages for inducing apoptosis
in blood neutrophils after 72 h (68.1, 67.9, 59.2, 70.0, 54.0,
50.1, 60.0, 56.1, 57.8, respectively). Furthermore, good levels
of apoptosis induction were achieved for 2m (46.1), 3l (44.1),
3m (41.6) and 4c (42.0) in this cell line after 72 h. With regard
to the obtained results in Table 4, there is a quite correlation
between the cytotoxic activity and apoptotic inducing ability.
(ꢁ87.5)g
1d
1e
a
2.5
2.3
845
786
508
471
80.8
76.2
89.3
87
74
(ꢁ88.3)g
102.9
(ꢁ94)g
Partition coefficient.
Molecular volume.
b
c
d
e
f
Molecular surface area.
Refractometry.
Torsional angle between the plane of the imide ring and N-substituent group.
Data was taken from Table 3.
g
Torsional angle of the syn-conformer.

620
A.A.-M. Abdel-Aziz / European Journal of Medicinal Chemistry 42 (2007) 614e626
1d (Anti-conformer)
1d (Syn-conformer)
1e (Anti-conformer)
1e (Syn-conformer)
Fig. 2. Lowest energy conformers of the selected nonactive compounds 1d and 1e with balls and cylinders rendering (anti and syn conformers).
rotations, which were lower than 1 kcal/mol. Both conformers
of compound 1d or 1e can exist. On the contrary, the more
active compounds showed only one conformer in which the
N-substituent group was deviated from the imide core by
57e60ꢀ in case of aliphatic substituent group such as tert-butyl
or 1-adamantyl (Fig. 3, Table 5; 1b, 1c, 2b, 2c, 3b, 3c and 4c)
and 81e95ꢀ in case of N-aryl derivatives such as compounds
1l, 1m, 1n, 2l, 2m, 2n, 3l, 3m and 3n (Table 5). As evident
from the experimental data, the structural features (pharmaco-
phore) essential for the cytotoxic activity of this series are as
follows; (i) nonplanar imide core is more favorable for the
best activity, (ii) the presence of a hydrophobic aryl moiety
directly attached to imide backbone enhances the activity,
(iii) this aryl group adapts only one conformer which may
be responsible for the activity, in contrary to benzyl moiety,
(iv) the presence of steric bulky o,o-dialkyl or p-pentyl groups
in the aromatic group or its steric equivalent 1-adamantyl is
necessary for the activity, (v) the steric and hydrophobic pa-
rameters are essential factors affecting the activity as indicated
by the experimental and theoretical data (Fig. 3, Table 5). The
stable conformers resulting from computational chemistry
analysis as a representative example were superimposed in
order to reveal the similarities and differences in structure
(Fig. 4). The strategy of overlay fit was to match imide rings
and to examine any spatial differences between the atoms of
the N-substituent groups. The results show that atoms of the
aryl groups occupy slightly different spatial position relative
to the plane of imide core. Further evidence for identical bio-
logical activity [21] comes from QSAR data that were derived
from the QSAR module of HyperChem using the conforma-
tions depicted in Figs. 2 and 3. As a result, the QSAR displays
calculated data for ClogP (the hydrophobic parameters),
refractivity (steric and polarizability parameters), molecular
volume (steric and polarizability parameters), grid surface
area (Table 5) [22]. Despite a variation of the molecular shape
of the active compounds, measurements of global molecular
parameters (surface area, volume, and refractivity) reflect their
similarity. As can be shown, ClogP values of the most active
compounds are well in the range of 2.7e4.6 while for com-
pounds 1c and 2c, ClogP values are somewhat lower i.e. 2.2
and 1.8, respectively, so the optimum hydrophobicity of the
active compounds is close to 2.0e5.1. However, direct corre-
lation could be established between the ClogP and cytotoxic
activity of the series as indicated by their IC50 values (Table
5), such as the ClogP values of 1a (0.8), 1b (1.3), 1c (2.2),
1l (3.1) and 1n (4.2) are well correlated with their cytotoxicity
(>100, 27, 14, 13 and 10 mM, respectively). It becomes appar-
ent that the criteria relating to favorable ClogP value range
maybe the sole predicting factor for cytotoxic activity. More
interestingly, the biological inefficacy of the inactive com-
pounds 1a, 1d and 1e (Fig. 2) could be attributed to the diffi-
culty to cross the biological membranes due to their
physicochemical properties which prevent their access to the
putative binding cavity [21]. It is noteworthy to say that the
structural similarity among these series was responsible for
similar biological activity as evident from the experimental
data (Table 5) as for example the molecular volume of the
3
most active compounds was higher than 800 A while that
˚
3
of the inactive species was lower than 600 A .
˚
3. Experimental
Melting points were recorded on a FishereJohns apparatus
(ꢀC, uncorrected). IR spectra (KBr) were recorded on a JASCO
1
IR Report-100 spectrophotometer (n in cmꢁ1) and H NMR
spectra on a JEOL ALPHA 500 MHz and Varian XL-200 MHz

A.A.-M. Abdel-Aziz / European Journal of Medicinal Chemistry 42 (2007) 614e626
621
Fig. 3. Lowest energy conformers of the most active compounds 1c, 1l, 1n, 2c, 2l and 2n with balls and cylinders rendering.
spectrometers using TMS as internal standard (chemical shift,
d ppm). Microanalytical data (C, H, N) agreed with the pro-
posed structures within ꢂ0.4% of the theoretical values. All
solvents were distilled before use. All reactions were run un-
der nitrogen atmosphere. The following materials and instru-
ments were used in biological evaluation; hemocytometer
(Bright line, AO instrument Co., Buffalo, NY, USA), inverted
microscope (Nikon Inc, Instrument division, Garden City, NY,
USA), Dulbecco’s modified Eagles medium (Gibco, Grand Is-
land, NY, USA), supplement with 10% calf serum (Gibco),
Giemsa stain (Fisher Scientific Co., Fair Lawn, NY, USA),
TLC plates (RP-18F254; 0.25 mm, Merck), DNA (calf thymus
type I, Sigma, 100 mg/ml), DNA/methyl green (Sigma, St.
Louis, MO, USA).
amine in MeCN in addition to Et3N (1.5 equiv) and the mix-
ture was stirred at 40 ꢀC for 15e60 min. After removal of
the solvent, the residue was taken up in organic solvent such
as EtOAc, and washed successively with HCl aq and NaHCO3
aq. Evaporation of the dried organic solvent gave the imide
which was further purified by chromatography on silica gel
or by recrystallization (Tables 1 and 2). The reported com-
pounds, including series 1 (1a, 1e, 1k and 1o), 2 (2a, 2b,
2d, 2e, 2k, 2l, 2o and 1p), 3 (3a, 3b, 3d, 3e, 3f, 3h, 3i, 3j,
3k, 3l, 3o and 3p) and 4 (4c, 4d, 4e, 4i, 4k and 4m), were iden-
tical with authentic samples by melting points, TLC and NMR
determinations [23].
3.1.1. Compound 1b
Mp 121e122 ꢀC (EtOH). 1H NMR (CDCl3, 500 MHz):
d 6.08e6.04 (dd, 2H, J ¼ 3.2, 1.3 Hz), 3.14e3.13 (d, 2H,
J ¼ 1.4 Hz), 2.83 (s, 2H), 1.61e1.56 (t, 2H, J ¼ 1.4 Hz),
1.40e1.38 (t, 2H, J ¼ 1.6 Hz), 1.42 (s, 9H). Anal. Calcd
for C14H19NO2: 72.07, 8.21, 6.00. Found: 71.88, 8.37,
5.94.
3.1. General and typical procedure for DPPOx-promoted
synthesis of N-substituted imides
The DPPOx (1.5 equiv) was added to the equimolar
solution of a carboxylic acid anhydride or imide and a primary

622
A.A.-M. Abdel-Aziz / European Journal of Medicinal Chemistry 42 (2007) 614e626
Fig. 4. Overlay of the selected active compounds and nonactive one showing the importance of the T-shaped structure for biological activity; 1c colors white, 1d
colors violet, 1e colors yellow, 1l colors red, 1n colors cyan, 2c colors green, 2l colors blue, T-shaped imides 1 was out of the plane of planar imides 2. (For
interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
3.1.2. Compound 1c
J ¼ 8.6 Hz), 6.12e6.10 (d, 2H, J ¼ 3.3 Hz), 4.80 (s, 2H),
3.20e3.18 (d, 2H, J ¼ 1.3 Hz), 2.79 (s, 2H), 1.60e1.56 (t,
2H, J ¼ 1.4 Hz), 1.39e1.36 (t, 2H, J ¼ 1.3 Hz). Anal. Calcd
for C17H16ClNO2: 67.66, 5.34, 4.64. Found: 67.42, 5.30, 4.61.
Mp 199e200 ꢀC (EtOH). 1H NMR (CDCl3, 500 MHz):
d 6.21e6.18 (dd, 2H, J ¼ 3.4, 1.6 Hz), 2.72e2.70 (d, 2H,
J ¼ 1.4 Hz), 2.87 (s, 2H), 2.04 (s, 4H), 1.78 (s, 4H), 1.66 (s,
4H), 1.63e1.58 (t, 2H, J ¼ 1.3 Hz), 1.55e1.53 (d, 3H,
J ¼ 12 Hz), 1.43e1.39 (t, 2H, J ¼ 1.6 Hz). Anal. Calcd for
C20H25NO2: 77.14, 8.09, 4.50. Found: 77.00, 8.15, 4.40.
3.1.7. Compound 1i
Mp 173e174 ꢀC (EtOH). 1H NMR (CDCl3, 500 MHz):
d 7.30e7.21 (m, 5H), 6.16e6.14 (d, 2H, J ¼ 3.6 Hz), 5.43e
5.41 (q, 1H, J ¼ 7.6 Hz), 3.25e3.21 (d, 2H, J ¼ 3.2 Hz),
2.87 (s, 2H), 1.81e1.79 (d, 3H, J ¼ 7.6 Hz), 1.64e1.60 (t,
2H, J ¼ 1.6 Hz), 1.40e1.37 (t, 2H, J ¼ 1.6 Hz). Anal. Calcd
for C18H19NO2: 76.84, 6.81, 4.98. Found: 76.99, 7.01, 5.11.
3.1.3. Compound 1d
Mp 81e82 ꢀC (EtOH). 1H NMR (CDCl3, 200 MHz):
d 7.45e7.39 (m, 5H), 6.11e6.07 (dd, 2H, J ¼ 3.4, 1.2 Hz),
4.85 (s, 2H), 3.46 (s, 2H), 3.17e3.15 (d, 2H, J ¼ 1.3 Hz),
2.85 (s, 2H), 1.63e1.59 (t, 2H, J ¼ 1.3 Hz), 1.42e1.39 (t,
2H, J ¼ 1.4 Hz). Anal. Calcd for C18H19NO2: 76.84, 6.81,
4.98. Found: 76.69, 6.84, 4.90.
3.1.8. Compound 1j
Mp 166e167 ꢀC (EtOH). 1H NMR (CDCl3, 500 MHz):
d 7.32e7.25 (m, 5H), 6.18e6.16 (d, 2H, J ¼ 3.4 Hz), 5.45e
5.43 (q, 1H, J ¼ 7.3 Hz), 3.27e3.24 (d, 2H, J ¼ 3.1 Hz),
2.86 (s, 2H), 1.80e1.78 (d, 3H, J ¼ 7.7 Hz), 1.62e1.59 (t,
2H, J ¼ 1.6 Hz), 1.41e1.39 (t, 2H, J ¼ 1.4 Hz). Anal. Calcd
for C18H19NO2: 76.84, 6.81, 4.98. Found: 76.71, 6.69, 4.86.
3.1.4. Compound 1f
Mp 165e166 ꢀC (EtOH). 1H NMR (CDCl3, 200 MHz):
d
6.98e6.96 (d, 2H, J ¼ 8.8 Hz), 6.72e6.70 (d, 2H,
J ¼ 8.8 Hz), 6.00e5.96 (dd, 2H, J ¼ 3.4, 1.4 Hz), 4.76 (s,
2H), 3.81 (s, 3H), 3.19e3.17 (d, 2H, J ¼ 1.6 Hz), 2.81 (s,
2H), 1.61e1.57 (t, 2H, J ¼ 1.4 Hz), 1.40e1.37 (t, 2H,
J ¼ 1.6 Hz). Anal. Calcd for C18H19NO3: 72.71, 6.44, 4.71.
Found: 72.96, 6.21, 4.57.
3.1.9. Compound 1l
1
Mp 216e217 ꢀC (AcOH). H NMR (CDCl3, 500 MHz):
d 6.98e6.97 (d, 2H, J ¼ 7.9 Hz), 6.87 (s, 1H), 6.09e6.05 (d,
2H, J ¼ 3.7 Hz), 3.13e3.10 (d, 2H, J ¼ 1.7 Hz), 2.81 (s,
2H), 2.21 (s, 6H), 1.63e1.61 (t, 2H, J ¼ 1.6 Hz), 1.41e1.38
(t, 2H, J ¼ 1.7 Hz). Anal. Calcd for C18H19NO2: 76.84, 6.81,
4.98. Found: 76.70, 6.99, 4.82.
3.1.5. Compound 1g
Mp 170e171 ꢀC (EtOH). 1H NMR (CDCl3, 500 MHz):
d 7.22 (s, 1H), 7.14e7.12 (d, 1H, J ¼ 8.0 Hz), 7.01e6.99 (d,
1H, J ¼ 8.0 Hz), 6.13e6.11 (d, 2H, J ¼ 3.6 Hz), 4.84 (s,
2H), 3.81 (s, 3H), 3.79 (s, 3H), 3.21e3.19 (d, 2H,
J ¼ 1.4 Hz), 2.85 (s, 2H), 1.64e1.60 (t, 2H, J ¼ 1.3 Hz),
1.43e1.38 (t, 2H, J ¼ 1.6 Hz). Anal. Calcd for C19H21NO4:
69.71, 6.47, 4.28. Found: 69.69, 6.51, 4.32.
3.1.10. Compound 1m
1
Mp 243e244 ꢀC (AcOH). H NMR (CDCl3, 500 MHz):
d 7.15e7.12 (d, 2H, J ¼ 6.8 Hz), 7.01 (s, 1H), 6.24e6.22 (d,
2H, J ¼ 3.1 Hz), 3.23e3.21 (d, 2H, J ¼ 1.5 Hz), 3.11e3.03
(m, 2H), 2.90 (s, 2H), 1.66e1.64 (t, 2H, J ¼ 1.7 Hz), 1.45e
1.32 (t, 2H, J ¼ 1.7 Hz), 1.23e1.20 (d, 12H, J ¼ 7.2 Hz).
Anal. Calcd for C22H27NO2: 78.30, 8.06, 4.15. Found:
78.35, 7.97, 4.23.
3.1.6. Compound 1h
Mp 120e121 ꢀC (EtOH). 1H NMR (CDCl3, 500 MHz):
d
7.46e7.43 (d, 2H, J ¼ 8.7 Hz), 7.29e7.27 (d, 2H,

A.A.-M. Abdel-Aziz / European Journal of Medicinal Chemistry 42 (2007) 614e626
623
3.1.11. Compound 1n
Mp 128e129 ꢀC (EtOH). 1H NMR (CDCl3, 500 MHz):
3.1.19. Compound 2m
Mp 262e263 ꢀC (AcOH). 1H NMR (DMSO-d6, 500 MHz):
d 8.91 (s, 2H), 7.01e6.99 (d, 2H, J ¼ 7.7 Hz), 6.90 (s, 1H),
3.09e32.94 (m, 2H), 1.17e1.15 (d, 12H, J ¼ 6.9 Hz). Anal.
Calcd for C18H19N3O2: 69.88, 6.19, 13.58. Found: 69.76,
6.06, 13.47.
d
7.61e7.59 (d, 2H, J ¼ 7.7 Hz), 7.43e7.40 (d, 2H,
J ¼ 7.6 Hz), 6.13e6.11 (d, 2H, J ¼ 3.4 Hz), 3.19e3.16 (d,
2H, J ¼ 3.2 Hz), 3.11e3.03 (m, 2H), 2.79 (s, 2H), 2.31e
2.29 (t, 2H, J ¼ 4.4 Hz), 1.62e1.59 (t, 2H, J ¼ 1.4 Hz),
1.56e1.51 (m, 2H), 1.39e1.37 (t, 2H, J ¼ 1.6 Hz), 1.23e
1.18 (m, 2H), 1.03e1.00 (t, 3H, J ¼ 4.8 Hz). Anal. Calcd for
C21H25NO2: 77.98, 7.79, 4.33. Found: 78.04, 8.00, 4.41.
3.1.20. Compound 2n
1
Mp 197e198 ꢀC (MeCN). H NMR (CDCl3, 500 MHz):
d 8.76 (s, 2H), 7.36e7.34 (d, 2H, J ¼ 7.1 Hz), 7.28e7.26 (d,
2H, J ¼ 7.7 Hz), 2.46e2.43 (t, 2H, J ¼ 4.6 Hz), 1.49e1.43
(m, 2H), 1.19e1.14 (m, 4H), 1.00e0.97 (t, 3H, J ¼ 4.6 Hz).
Anal. Calcd for C17H17N3O2: 69.14, 5.80, 14.23. Found:
70.01, 5.97, 14.38.
3.1.12. Compound 1p
Mp 116e117 ꢀC (EtOH). 1H NMR (CDCl3, 500 MHz):
d
7.63e7.61 (d, 2H, J ¼ 8.0 Hz), 7.42e7.40 (d, 2H,
J ¼ 8.0 Hz), 6.31e6.29 (dd, 2H, J ¼ 3.2, 1.6 Hz), 3.27 (s,
2H), 3.04e3.03 (d, 2H J ¼ 1.2 Hz), 1.69e1.67 (d, 2H,
J ¼ 1.2 Hz), 1.47e1.44 (d, 2H, J ¼ 1.6 Hz). Anal. Calcd for
C17H14F3NO2: 63.55, 4.39, 4.36. Found: 63.69, 4.52, 4.43.
3.1.21. Compound 3c
Mp 151e152 ꢀC (MeOH). 1H NMR (DMSO-d6,
200 MHz): d 8.91e8.87 (m, 3H), 1.97 (s, 4H), 1.69 (s, 4H),
1.60 (s, 4H), 1.56e1.53 (s, 3H). Anal. Calcd for
C17H18N2O2: 72.32, 6.43, 9.92. Found: 72.11, 6.36, 10.09.
3.1.13. Compound 2c
Mp 241e242 ꢀC (AcOH). 1H NMR (DMSO-d6, 500 MHz):
d 8.50 (s, 1H), 8.44 (s, 1H), 2.02 (s, 4H), 1.76 (s, 4H), 1.64 (s,
4H), 1.59e1.56 (d, 3H, J ¼ 12.0 Hz). Anal. Calcd for
C16H17N3O2: 67.83, 6.05, 14.83. Found: 68.00, 6.15, 14.70.
3.1.22. Compound 3g
Mp 140e141 ꢀC (EtOH). 1H NMR (CDCl3, 500 MHz):
d 9.02e8.97 (m, 3H), 7.35 (s, 1H), 6.89e6.87 (d, 1H,
J ¼ 7.2 Hz), 6.79e6.77 (d, 1H, J ¼ 7.1 Hz), 4.51 (s, 2H),
3.67 (s, 3H), 3.37 (s, 3H). Anal. Calcd for C16H14N2O4:
64.42, 4.73, 9.39. Found: 64.40, 4.57, 9.26.
3.1.14. Compound 2f
Mp 149e150 ꢀC (EtOH). 1H NMR (CDCl3, 500 MHz):
d 8.42 (s, 2H), 7.43e7.41 (d, 2H, J ¼ 8.0 Hz), 6.98e6.96 (d,
2H, J ¼ 8.0 Hz), 4.81 (s, 2H), 3.74 (s, 3H). Anal. Calcd for
C14H11N3O3: 62.45, 4.12, 15.61. Found: 62.51, 4.10, 15.73.
3.1.23. Compound 3m
Mp 212e213 ꢀC (MeCN). 1H NMR (DMSO-d6, 500 MHz):
d 8.84e8.80 (m, 3H), 6.97e6.901 (m, 3H), 3.12e3.07 (m,
2H), 1.11e1.08 (d, 12H, J ¼ 7.9 Hz). Anal. Calcd for
C19H20N2O2: 74.00, 6.54, 9.08. Found: 73.89, 6.46, 9.00.
3.1.15. Compound 2g
Mp 183e184 ꢀC (EtOH). 1H NMR (CDCl3, 500 MHz):
d 8.91 (s, 2H), 7.26 (s, 1H), 7.08e6.06 (d, 1H, J ¼ 8.0 Hz),
6.82e6.80 (d, 1H, J ¼ 8.1 Hz), 4.91 (s, 2H), 3.88 (s, 3H),
3.85 (s, 3H). Anal. Calcd for C15H13N3O4: 60.20, 4.38,
14.04. Found: 60.16, 4.42, 14.21.
3.1.24. Compound 3n
1
Mp 179e180 ꢀC (MeCN). H NMR (CDCl3, 500 MHz):
d 8.93e8.87 (m, 3H), 7.43e7.41 (d, 2H, J ¼ 7.0 Hz), 7.35e
7.32 (d, 2H, J ¼ 7.2 Hz), 2.40e2.37 (t, 2H, J ¼ 3.9 Hz),
1.42e1.36 (m, 2H), 1.20e1.16 (m, 4H), 0.98e0.96 (t, 3H,
J ¼ 4.0 Hz). Anal. Calcd for C18H18N2O2: 73.45, 6.16, 9.52.
Found: 73.29, 6.04, 9.49.
3.1.16. Compound 2h
Mp 251e252 ꢀC (AcOH). 1H NMR (DMSO-d6, 500 MHz):
d 8.98 (s, 2H), 7.58e7.56 (d, 2H, J ¼ 8.2 Hz), 7.23e7.21 (d,
2H, J ¼ 8.4 Hz), 5.01 (s, 2H). Anal. Calcd for C13H8ClN3O2:
57.05, 2.95, 15.35. Found: 57.10, 3.06, 15.46.
3.2. Biological assay
3.2.1. Cytotoxicity assay
3.1.17. Compound 2i
3.2.1.1. Sample and culture preparation. Samples were pre-
pared for assay by dissolving in 50 ml of DMSO and diluting
aliquots into sterile culture medium at 0.4 mg/ml. These solu-
tions were subdiluted to 0.02 mg/ml in sterile medium and the
two solutions were used as stocks to test samples at 500, 300,
200, 100, 50, 20, 10, 5, 2, and 1 mM in triplicate in the wells of
microtiter plates. Vero African green monkey kidney cells
were purchased from Viromed Laboratories, Minnetonka,
MN, and grown in Dulbecco’s modified Eagle’s medium sup-
plemented with 10% (v/v) calf serum (HyClone Laboratories,
Mp 173e174 ꢀC (EtOH). 1H NMR (CDCl3, 500 MHz):
d 8.74 (s, 2H), 7.10e6.86 (m, 5H), 5.13e5.09 (q, 1H,
J ¼ 7.1 Hz), 1.69e1.67 (d, 3H, J ¼ 7.3 Hz). Anal. Calcd for
C14H11N3O2: 66.40, 4.38, 16.59. Found: 66.35, 4.50, 16.39.
3.1.18. Compound 2j
1
Mp 180e181 ꢀC (AcOH). H NMR (CDCl3, 500 MHz):
d 8.78 (s, 2H), 7.09e6.87 (m, 5H), 5.12e5.09 (q, 1H,
J ¼ 7.2 Hz), 1.70e1.68 (d, 3H, J ¼ 7.1 Hz). Anal. Calcd for
C14H11N3O2: 66.40, 4.38, 16.59. Found: 66.29, 4.36, 16.43.
Ogden, UT), 60 mg/ml penicillin
G
and 100 mg/ml

624
A.A.-M. Abdel-Aziz / European Journal of Medicinal Chemistry 42 (2007) 614e626
streptomycin sulfate maintained at 37 ꢀC in a humidified at-
mosphere containing about 15% (v/v) CO2 in air. All compo-
nents in the medium were obtained from Sigma Chemical Co.,
St. Louis, MO, unless otherwise indicated. Vero stocks were
maintained at 34 ꢀC in culture flasks filled with medium sup-
plemented with 1% (v/v) calf serum. Subcultures for cytotox-
icity screening were grown in the wells of microtiter trays
(Falcon Microtest III 96-wells trays, Becton Dickinson Lab-
ware, Lincolin Park, NJ) by suspending Vero cells in medium
following trypsineEDTA treatment, counting the suspension
with a hemocytometer, diluting in medium containing 10%
calf serum to 2 ꢃ 104 cells per 200 ml culture, aliquoting
into each well of a tray and culturing until confluent.
of the untreated DNA/methyl green absorbance value and C50
were determined for each compound. Daunomycin was used
as a positive control.
3.2.3. Apoptosis technique
3.2.3.1. Preparation of blood neutrophils. Neutrophils (>98%
pure on May-Giemsa stain) were isolated from peripheral blood
of normal healthy volunteer donors and from oesteoarthritic
and rheumatoid arthritic patients by a combination of dextran
sedimentation and centrifugation through discontinuous plasma
percoll gradients as described by Haslett [24]. In brief, neutro-
phils were prepared as follows: freshly drawn venous blood was
citrated (1.1 ml of 3.8% sodium citrate to 10 ml blood), centri-
fuged at 3000g for 20 min at 20 ꢀC and the platelet rich plasma
aspirated and centrifuged at 2500g for 10 min [for production of
platelet poor plasma (PPP)] or recalcified by adding 20 mM final
concentration of calcium chloride to prepare platelet rich
plasma derived serum (PRPDS). To red and white cells remain-
ing in each tube 5 ml of 6% dextran (500,000 mol wt) in 0.9%
saline mixed gently and then allowed to stand for erythrocyte
sedimentation for 30 min. The leukocyte-rich plasma was aspi-
rated, centrifuged at 275g for 6 min, suspended in 2 ml. The
leukocytes were then mixed with 2 ml of 42% percol (9:1 v/v
percole0.9% saline) in PPP followed by adding 2 ml of 51%
percol in PPP. The gradients were centrifuged at 275g for
10 min and neutrophils were then aspirated from the interface
of the 51% and 42% percol.
3.2.1.2. Procedure. Microtiter trays with confluent monolayer
cultures of Vero cells were inverted, the medium shaken out,
and replaced with serial dilutions of sterile extracts in tripli-
cate in 100 ml medium followed by titered virus in 100 ml me-
dium containing 10% (v/v) calf serum in each well. In each
tray, the last row of wells was reserved for controls that
were not treated with compounds. The trays were cultured
for 48 h. The trays were inverted onto a pad of paper towels,
the remaining cells rinsed carefully with medium, and fixed
with 3.7% (v/v) formaldehyde in saline for at least 20 min.
The fixed cells were rinsed with water, and examined visually.
Cytotoxic activity is identified as confluent, relatively unal-
tered monolayers of stained Vero cells treated with com-
pounds. Cytotoxicity was estimated as the concentration that
caused approximately 50% loss of the monolayer.
3.2.3.2. Culture of neutrophils. Neutrophils were prepared as
described above, then resuspended in an appropriate volume
of RPMI 1640 medium with 10% autologous PRPDS,
100 mg/ml of penicillin and streptomycin and then divided
into five equal volumes each put in culture tube. Cells were in-
cubated (37 ꢀC in a 5% carbon dioxide) as follows: (1) Only
cells; (2) Cells and DMSO at 0.01% v/v; (3) Cells and each
compound in DMSO at dose of 50 mM culture. The age of
neutrophils in culture was calculated at the start of culture at
time zero (or base line), 24 h, 48 h and 72 h.
3.2.2. DNA binding affinity
3.2.2.1. DNA binding assay. Analysis of the DNA binding af-
finity of the tested compounds was performed using RP-TLC
plates (RP-18 F254; 0.25 mm; Merck). TLC plates were pre-de-
veloped with MeOHeH2O (8:2). Test compounds were then ap-
plied (5 mg/ml in MeOH) at the origin, followed by the addition
of DNA (1 mg/ml in H2O and MeOH mixture) at the same posi-
tions at the origin. The plates were then developed with the same
solvent system and the position of the DNA was determined by
spraying with anisaldehyde reagent. The reagent yields a blue
color spot with DNA, and the intensity of the color was propor-
tional to the quantity of DNA added to the plate. Ethidium bro-
mide was used as positive control.
3.2.3.3. Assessment of cell viability. At time zero and at sub-
sequent times, cells were removed from culture and counted
on a hemocytometer. Cell viability was determined by trypan
blue dye exclusion test; one volume of trypan blue (0.4%
GiBCo) was added to five volumes of cells at room tempera-
ture for 5 min.
3.2.2.2. Methyl green/DNA displacement assay. DNA/methyl
green (20 mg, Sigma, St. Louis, MO, USA) was suspended
in 100 ml of 0.05 M Tris/HCl buffer, pH 7.5, containing
7.5 mM MgSO4 and stirred at 37 ꢀC with a magnetic stirrer
for 24 h. Unless otherwise indicated, samples to be tested
were dissolved in EtOH in Eppendorff tubes. Solvent was re-
moved under vacuum and 200 ml of the DNA/methyl green so-
lution was added to each tube. The absorption maxima for the
DNA/methyl green complex is at 642e645 nm. Samples were
incubated in the dark at ambient temperature and after 24 h the
final absorbance of samples was determined. Readings were
corrected for initial absorbance and normalized as a percentage
3.2.3.4. Measurements of apoptosis
3.2.3.4.1. Morphological assessment of apoptosis (Giemsa
and Acridine orange stains). At time zero and at subsequent
times, cells were removed from each culture, fixed in methanol,
harvested on slides and slides were stained with May Grunwald
Giemsa and examined by oil immersion light microscope. For
assessment of the percentage of cells showing morphology of
apoptosis 500 cells/slide were examined for each case at differ-
ent times (0, 24, 48, and 72 h) in the presence or absence of the
drugs used. Neutrophils were considered apoptotic if they

A.A.-M. Abdel-Aziz / European Journal of Medicinal Chemistry 42 (2007) 614e626
625
exhibited the highly characteristic morphological features of
chromatin aggregation, nuclear pyknosis and cytoplasmic va-
culation [17]. The apoptotic neutrophils’ percentage at differ-
ent times was calculated for normal, oesteoarthritic and
rheumatoid arthritis in the presence or after addition of natural
products and the results were then compared statistically using
F-test and Students t- test. One drop from cell suspension was
added to one drop of AO solution (10 mg/ml in PBS), mixed
gently on a slide, and immediately examined with an Olympus
HB-2 microscope with fluorescence attachment. Green fluores-
cence was detected between 500 and 525 nm. Cells exhibiting
bright green fluorescent condensed nuclei (intact or frag-
mented) were interpreted as apoptotic cells.
torsional space was performed using the multiconformer
method [26]. Energy minima for compounds 1c, 1d, 1e, 1l,
1n, 2c, 2l and 2n were determined by a semi-empirical method
AM1 [27] (as implemented in HyperChem 5.1). The confor-
mations thus obtained were confirmed as minima by vibra-
tional analysis. Atom-centred charges for each molecule
were computed from the AM1 wave functions (HyperChem
5.1) by the procedure of Orozco and Luque [28], which pro-
vides derived charges that closely resemble those obtainable
from ab initio 6-31G* calculations.
4. Conclusion
3.2.3.4.2. DNA fragmentation assay. Assessment of chro-
matin fragmentation in neutrophils was done by modification
of methods previously used for thymocytes [25]. Cells
(2.5X10.7) were washed three times and resuspended in
0.15 M of NaCl solution. The cells were chilled at 4 ꢀC and
lyzed by adding 4.5 ml of 10 mM of Tris/HCl buffer (pH 8.0)
containing 100 mM of EDTA and 0.2% v/v Triton X-100 (lysis
buffer). After 4 h, the lysate was centrifuged at 3500g at 4 ꢀC
for 20 min. The supernatants were collected into tubes and pre-
cipitated with 0.1 volume of 5 M of NaCl and two volumes of
absolute ethanol. The DNA was precipitated for 24 h at 4 ꢀC.
The precipitate was centrifuged at 12,500g at 6 ꢀC for
15 min. The pellet was resuspended in 1 ml of 10 mM of
Tris/HCl buffer (pH 8.0) containing 100 mM of EDTA and
0.1 mM of sodium dodecyl sulfate. Proteinase K was added
to a final concentration of 20 mg/ml, and the sample was incu-
bated for a further 24 h at 37 ꢀC. The DNA was extracted with
phenol and chloroform and reprecipitated with absolute etha-
nol. The pellet was redissolved in 20 ml of lyses buffer and
10 ml of RNase. Each sample of the purified DNA (20 ml)
was subjected to electrophoresis in 1% agarose gel containing
200 mg/ml ethidium bromide and were visualized under ultravi-
olet light. The size of the fragments was confirmed by reference
to a 1-Kb DNA ladder (Gibco/BRL). Aging neutrophils exhibit
morphological features of apoptosis as nuclear pyknosis, chro-
matin condensation and cytoplasmic vaculation by light micro-
scope. A bright green-fluorescent condensed nuclei (intact or
fragmented), reduction of cell volume, were interpreted as ap-
optotic cells and expressed as a percentage of the total cell num-
ber by fluorescent microscopy. Viable cells were interpreted as
cells, which exhibited a green, diffusely stained intact nucleus.
3.2.3.4.3. Chromatin fragmentation. Classical 180e200
base pair integer oligonucleosome ‘‘ladder’’ was observed
by using gel electrophoresis of the extracted DNA as a typical
degeneration pattern of apoptosis. This confirms the previ-
ously obtained results from morphological assessment.
We successfully developed an efficient, economical and
practical method for the synthesis of a wide range of imide de-
rivatives by using inexpensive and readily available reagents
under mild conditions. In light of its operational simplicity, sim-
ple purification procedure, and high efficiency, the procedure is
expected to have broad utility, especially in the synthesis of
functionalized imide derivatives, for future biological and
chemical applications. Based on a general pattern inspired by
the reference literature, we have completed biological evalua-
tion of the target compounds as potential cytotoxic, DNA
binder and apoptosis inducers. These compounds were per-
formed in a cytotoxicity in vitro assay against Vero African
green monkey kidney cells. Some compounds showed good cy-
totoxic activity against the cell line tested, with IC50 values in
the low medium micro-molar range. Compounds which showed
cytotoxicity were evaluated in a DNA binding and apoptosis as-
say. Some compounds exhibited high DNA affinity and ability
in inducing apoptosis in the cell line tested. The best DNA af-
finity and apoptosis levels were achieved for compounds 1n,
2n, 3n, 1l, 2l, 3l, 1c, 2c and 3c. Although a precise struc-
tureeactivity relationship can not be defined, it is possible to
point out some general tendencies observed for the active com-
pounds, the greatest DNA binding and apoptotic activity is
found in compounds with a nonplanar imide nucleus. With re-
gard to the N-functional group of connection, the best results
were obtained for the 4-pentylphenyl, 1-adamantyl, and 2,6-di-
methylphenyl groups. In general, a direct relationship has been
found between the apoptotic activity and the cytotoxicity, while
the affinity for DNA was partially correlated with cytotoxic ac-
tivity, thereby suggesting the existence of two different mecha-
nisms of action controlling cytotoxic activity.
Acknowledgement
The author would like to thank Prof. Farid Badria (Depart-
ment of Pharmacognosy, Faculty of Pharmacy, Mansoura
University, Egypt) for providing the biological data.
3.3. Molecular modeling
References
Initial structures for the molecules 1c, 1d, 1e, 1l, 1n, 2c, 2l
and 2n were constructed using the HyperChem program ver-
sion 5.1. The MMþ [19] (calculations in vacuo, bond dipole
option for electrostatics, PolakeRibiere algorithm, and RMS
[1] (a) M.E. Salvati, A. Balog, W. Shan, D.D. Wei, D. Pickering, R.M. Attar,
J. Geng, C.A. Rizzo, M.M. Gottardis, R. Weinmann, S.R. Krystek,
J. Sack, Y. An, K. Kish, Bioorg. Med. Chem. Lett. 15 (2005) 271;
(b) J. Kossakowski, M. Jarocka, Il Farmaco 56 (2001) 785;
˚
gradient of 0.01 kcal/A mol) conformational searching in

626
A.A.-M. Abdel-Aziz / European Journal of Medicinal Chemistry 42 (2007) 614e626
(c) K. Ishizumi, A. Kojma, F. Antoku, Chem. Pharm. Bull. 39 (1991) 2288; [12] J.M. Pezzuto, S.K. Antosiak, W.M. Messmer, M.B. Slaytor, G.R. Honig,
(d) Y. Sabata, M. Shichita, K. Sasaki, Y. Hashimoto, S. Iwasaki, Chem.
Pharm. Bull. 43 (1995) 177;
Chem. Biol. Interact. 43 (1983) 323.
[13] J.M. Pezzuto, C.T. Che, D.D. McPherson, J.-P. Zhu, G. Topcu,
C.A.J. Erdelmeier, G.A. Cordell, J. Nat. Prod. 54 (1991) 1522.
[14] N.S. Burres, A. Frigo, R.R. Rasmussen, J.B. McAlpine, J. Nat. Prod. 55
(1992) 1582.
(e) D.P. Jindal, V. Bedi, B. Jit, N. Karkra, S. Guleria, R. Bansal,
A. Palusczak, R.W. Hartmann, Il Farmaco 60 (2005) 283;
(f) J.J. Wang, S.S. Wang, C.F. Lee, M.A. Chung, Y.T. Chern, Chemotherapy
43 (1997) 182;
[15] (a) Z. Huang, Chem. Biol. 9 (2002) 1059;
´
(g)A.L. Machado, L.M. Lima, J.X. AraujoJr., C.A.M. Fraga, V.L.G. Koatz,
(b) R.A. Schlegel, P. Williamson, Cell Death Differ. 8 (2001) 551;
(c) S.J. Martin, C.P. Reutelingsperger, A.J. McGahon, J.A. Rader,
R.C. Van Scie, D.M. LaFace, D.R. Green, J. Exp. Med. 182 (1995)
1545;
E.J. Barreiro, Bioorg. Med. Chem. Lett. 15 (2005) 1169;
(h) S. Kenji, N. Hideko, U. Yoshihiro, S. Yoshikazu, N. Kazuharu,
W. Motoji, W. Konstanty, T. Tadafumi, A. Tetsuji, Y. Yuji, K. Kenji,
H. Hitoshi, Bioorg. Med. Chem. 13 (2005) 4014;
(d) F.R. Kerr, J. Pathol. 105 (1971) 13;
(i) A. Mayer, S. Neuenhofer, Angew. Chem. Int. Ed. Engl. 33 (1994) 1044;
(j) F.B. Miguel, D. Gema, S. Beatriz, R. Cynthia, R. Simmon, B. Teresa,
Eur. J. Med. Chem. 37 (2002) 541;
(e) F.E. Sarrol, I.D. Bowen, Cell Tissue Res. 21 (1988) 45;
(f) D.A. Carson, J.M. Ribeiro, Lancet 341 (1993) 1251;
(g) H.-Z. Zhang, J. Drewe, B. Tseng, S. Kasibhatla, S.X. Cai, Bioorg.
Med. Chem. 12 (2004) 3649;
(k) H. Miyachi, A. Azuma, A. Ogasawara, E. Uchimura, N. Watanabe,
Y. Kobayashi, F. Kato, M. Kato, H. Hashimoto, J. Med. Chem. 40 (1997)
2858.
(h) H. Corwin, J. Ali, B.M. Suresh, G. Rajni, B. Benjamin, Bioorg.
Med. Chem. 11 (2003) 3015.
[2] (a) M.A. Walker, J. Org. Chem. 60 (1995) 5352; Tetrahedron Lett. 35
(1994) 665;
[16] (a) K. Debatin, Toxicol. Lett. 112 (2000) 41;
(b) D.W. Nicholson, Nature 407 (2000) 810;
´
(b) R.L. Dorta, C.G. Francisco, E. Suarez, Tetrahedron Lett. 35 (1994)
1083.
(c) P. Huang, A. Oliff, Trends Cell Biol. 11 (2001) 343.
[17] M.K. Pec, A. Aguirre, K. Moser-Thier, J.J. Fernandez, M.L. Souto,
J. Dorta, F. Diaz-Gonzales, J. Villar, Biochem. Pharmacol. 65 (2003)
1451.
[3] A. Da Settimo, G. Primofiore, F. Da Settimo, F. Simorini, C. La Motta,
A. Martinelli, E. Boldrini, Eur. J. Med. Chem. 31 (1996) 49.
[4] N.B. Mehta, A.P. Phillips, F.F. Lui, R.E. Brooks, J. Org. Chem. 25 (1960)
1012.
[18] G. Evan, K.H. Vousden, Nature 411 (2001) 342.
[19] (a) S. Profeta, N.L. Allinger, J. Am. Chem. Soc. 107 (1985) 1907;
(b) N.L. Allinger, J. Am. Chem. Soc. 99 (1977) 8127.
[20] HyperChem version 5.1, Hypercube Inc.
[5] P.Y. Reddy, S. Kondo, T. Toru, Y. Ueno, J. Org. Chem. 62 (1997) 2652.
[6] A.K. Bose, M.S. Manhas, M. Ghush, V.S. Raju, K. Tabei, Z. Urbanczyk-
Lykewaska, Heterocycles 30 (1990) 741.
[21] S.T. Al-Rashood, I.A. Aboldahab, M.N. Nagi, L.A. Abouzeid,
A.A.M. Abdel-Aziz, S.G. Abdul-hamide, K.M. Youssef, A.M. Al-Obaid,
H.I. El-Subbagh, Bioorg. Med. Chem. 14 (2006) 8608.
[22] R.P. Verma, A. Kurup, C. Hansch, Bioorg. Med. Chem. 13 (2005) 237.
[23] (a) M. Ishikawa, M. Fujimoto, M. Sakai, A. Matsumoto, Chem. Pharm.
Bull. 16 (1968) 622;
[7] K.C. Westaway, R.N. Gedye, J. Microw. Power Electromagn. Energy
(1995) 219.
[8] For the synthesis and application of DPPOx, see: (a) T. Kunieda,
T. Higuchi, Y. Abe, M. Hirobe, Tetrahedron 39 (1983) 3253;
(b) T. Kunieda, T. Nagamatsu, T. Higuchi, M. Hirobe, Tetrahedron Lett.
29 (1988) 2203.
(b) A. Meijer, S. Otto, J.B.F.N. Engberts, J. Org. Chem. 63 (1998) 8989;
(c) M.R. Baar, K. Wustholz, J. Chem. Educ. 82 (2005) 1393;
(d) V.V. Kovalishin, I.V. Tetko, A.I. Luik, J.R. Chretien, D.J. Livingstone,
Russ. J. Bioorganic Chem. 27 (2001) 267;
[9] (a) H.A. Staab, Angew. Chem. Int. Ed. Engl. 1 (1962) 351;
(b) H.C. Beyerman, W.M. Van denBrink, Rec. Trav. Chim. 80 (1961)1372;
(c) T. Izawa, T. Mukaiyama, Chem. Lett. (1977) 1443;
(d) T. Mukaiyama, Angew. Chem. Int. Ed. Engl. 15 (1976) 94.
[10] (a) I. Antonini, P. Polucci, A. Magnano, D. Cacciamani, M.T. Konieczny,
J. Paradziej-qukowicz, S. Martelli, Bioorg. Med. Chem. 11 (2003) 399;
(b) A. Kamal, O. Srinivas, P. Ramulu, G. Ramesh, P. Praveen Kumar,
M. Shiva Kumar, Bioorg. Med. Chem. 12 (2004) 4337;
(c) B.K. Banik, F.F. Becker, I. Banik, Bioorg. Med. Chem. 12 (2004) 2523;
(d) E.R. El-Bendary, F.A. Badria, Arch. Pharm. Med. Chem. 333(2000)99;
(e) M.A. El-Sherbeny, F.A. Badria, S.A. Kheira, Med. Chem. Res. 6
(1996) 28.
(e) T. Goto, M. Konno, M. Saito, R. Sato, Bull. Chem. Soc. Jpn. 62
(1989) 1205;
(f) M.M. Blanco, G.J. Levin, C.B. Schapira, I.A. Perillo, Heterocycles 57
(2002) 1881;
(g) B. Gutkowska, Z. Kabzinska, J. Wasiak, Acta Pol. Pharm. 44 (1987)
242;
(h) P.M. Harrington, Heterocycles 35 (1993) 683;
(i) T. Noguchi, H. Fujimoto, H. Sano, A. Miyajima, H. Miyachi,
Y. Hashimoto, Bioorg. Med. Chem. Lett. 15 (2005) 5509.
[24] (a) J.S. Savill, A.H. Wyllie, J.E. Henson, C. Haslett, J. Clin. Invest. 83
(1989) 865;
[11] (a) A. Amutha, V. Subramanian, B. Unni Nair, Chem. Phys. Lett. 344
(2001) 40;
(b) J.P. Dheyongera, W.J. Geldenhuys, T.G. Dekker, C.J. Van der Schyf,
Bioorg. Med. Chem. 13 (2005) 689;
(b) C. Haslett, M. Guthrie, M. Henson, Am. J. Pathol. 199 (1985) 101.
[25] (a) H. Wyla, Nature 284 (1980) 555;
(c) W. Zhang, Y. Dai, U. Schmitz, T.W. Bruice, FEBS Lett. 509 (2001) 85;
(d) D.F.S. Kehrer, O. Soepenberg, W. Loos, J. Verweij, A. Sparreboom,
Anticancer Drugs 12 (2001) 89;
(b) G. Maino, G. Jotris, Am. J. Pathol. 146 (1995) 3.
[26] M. Lipton, W.C. Still, J. Comput. Chem. 9 (1988) 345.
[27] M.J.S. Dewar, E.G. Zoebisch, E.F. Healy, J.J.P. Stewart, J. Am. Chem.
Soc. 107 (1985) 3902.
(e) K. Chen, S.J. Adelstein, A.I. Kassis, J. Mol. Struct. (Theochem) 711
(2004) 49.
[28] M. Orozco, F.J. Luque, J. Comput. Chem. 11 (1990) 909.