Optimization of (phenylmethylidene)-hydantoins as prostate cancer migration inhibitors: SAR-directed design, synthesis, and pharmacophore modeling

Article abstract of DOI:10.1002/cbdv.201000248

Prostate cancer is one of the most common cancer forms among males of Western countries. Natural products proved to be an unparalleled source of molecular diversity. The 4-(hydroxyphenylmethylidene)hydantoin (PMH; 1), (5Z)-5-(4-hydroxybenzylidene)imidazolidine-2,4-dione, was isolated from the Red Sea sponge Hemimycale arabica, and recently showed junctional complexes stabilization, anti-invasive, and antimetastatic activities in vitro and in vivo. The related synthetic analogue, (5Z)-5-[4-(ethylsulfanyl)benzylidene] imidazolidine-2,4-dione (2), showed several-fold-improved in vivo antimetastatic properties against the highly invasive prostate cancer. To further optimize the activity of PMHs, various ligand-based strategies were used including the extension of the structure, structural simplification, linker extension, and computer-assisted CoMFA (Comparative Molecular Field Analysis) results. These strategies yielded thirty 2nd-generation PMHs, designed based on the 1st-generation PMHs, such as 1 and 2. Wound-healing assay was selected to evaluate the in vitro anti-migratory potential of these new PMHs against the PC-3 cell line. Several active PMHs, including 10, 13, 24, 29, with nearly twelvefold enhancement of activity vs. 2, were identified. Active compounds were then used to build a pharmacophore model using the SYBYL's DIStance COmparison technique (DISCOtech). Active PMHs were also screened for fragment-based drug likeness using the OSIRIS program, and an overall drug score was also calculated. Interestingly, the overall drug scores of 24 and 29 along with their anti-migratory activity were significantly greater than those of 1 and 2. In conclusion, PMHs can be the appropriate scaffolds for the urgently needed drug candidates for the control of androgen independent prostate cancer. Copyright

Full text of DOI:10.1002/cbdv.201000248

1470
CHEMISTRY & BIODIVERSITY – Vol. 8 (2011)
Optimization of (Phenylmethylidene)-hydantoins as Prostate Cancer
Migration Inhibitors: SAR-Directed Design, Synthesis, and Pharmacophore
Modeling
by Mudit Mudit1) and Khalid A. El Sayed*
Department of Basic Pharmaceutical Sciences, College of Pharmacy, University of Louisiana at Monroe,
1800 Bienville Drive, Monroe, Louisiana 71201, USA
(phone: þ13183421725; fax: þ13183421737; e-mail: elsayed@ulm.edu)
Prostate cancer is one of the most common cancer forms among males of Western countries. Natural
products proved to be an unparalleled source of molecular diversity. The 4-(hydroxyphenylmethylide-
ne)hydantoin (PMH; 1), (5Z)-5-(4-hydroxybenzylidene)imidazolidine-2,4-dione, was isolated from the
Red Sea sponge Hemimycale arabica, and recently showed junctional complexes stabilization, anti-
invasive, and antimetastatic activities in vitro and in vivo. The related synthetic analogue, (5Z)-5-[4-
(ethylsulfanyl)benzylidene]imidazolidine-2,4-dione (2), showed several-fold-improved in vivo antime-
tastatic properties against the highly invasive prostate cancer. To further optimize the activity of PMHs,
various ligand-based strategies were used including the extension of the structure, structural
simplification, linker extension, and computer-assisted CoMFA (Comparative Molecular Field Analysis)
results. These strategies yielded thirty 2nd-generation PMHs, designed based on the 1st-generation
PMHs, such as 1 and 2. Wound-healing assay was selected to evaluate the in vitro anti-migratory potential
of these new PMHs against the PC-3 cell line. Several active PMHs, including 10, 13, 24, 29, with nearly
twelvefold enhancement of activity vs. 2, were identified. Active compounds were then used to build a
pharmacophore model using the SYBYLꢀs DIStance COmparison technique (DISCOtech). Active
PMHs were also screened for fragment-based drug likeness using the OSIRIS program, and an overall
drug score was also calculated. Interestingly, the overall drug scores of 24 and 29 along with their anti-
migratory activity were significantly greater than those of 1 and 2. In conclusion, PMHs can be the
appropriate scaffolds for the urgently needed drug candidates for the control of androgen independent
prostate cancer.
1. Introduction. – Prostate cancer is the second leading cause of cancer-related
death in men in the United States [1]. Surgery and radiation therapy constitute the
currently available treatment options for localized and early-stage prostate cancer [2].
However, in patients with advanced metastatic disease, relapse of the condition is very
common [2]. Treatment with androgen ablation therapy can cause the disease to
become androgen-independent in nature, highly metastatic, chemoresistant, and life-
threatening [3–5]. The poor clinical outcome with chemotherapeutic agents such as
doxorubicin, mitoxantrone, and docetaxel in patients with hormone refractory prostate
cancer urges the critical need for the discovery and development of alternative effective
therapeutic agents.
1
)
Current Address: Department of Pharmaceutical Sciences, DꢀYouville College, School of Pharmacy,
320 Porter Avenue, Buffalo, New York 14201, USA.
ꢁ 2011 Verlag Helvetica Chimica Acta AG, Zꢂrich

CHEMISTRY & BIODIVERSITY – Vol. 8 (2011)
1471
(Phenylmethylidene)-hydantoins (PMHs) are marine-derived compounds report-
ed earlier from the Red Sea sponge Hemimycale arabica and showed good in vitro and
in vivo activities against the highly metastatic androgen-independent human prostate
cancer cells [6][7]. Several synthetic analogues related to the natural PMH 1 (Table 1)
Table 1. Synthesized (Phenylmethylidene)hydantoin Analogues
Compound
Type
Substituents
Reference
R1
R2
R3
R4
1
2
3
4
5
6
7
8
I
I
I
I
I
I
I
I
I
I
I
H
H
H
H
H
H
H
H
H
Cl
H
H
H
H
H
H
H
H
H
OH
EtS
HOꢀ(CH2)2O
Me2N(CH2)3O
tBuO
H
H
H
H
H
H
H
H
H
H
H
[6]
[6]
this work
this work
this work
this work
this work
this work
this work
this work
this work
this work
HCꢁC
Me(CH2)6O
Me(CH2)15O
1H-Tetrazol-5-yl
NH2
9
H
H
NO2
10
11
12
BuO
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
II
II
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
MeO
H
H
H
H
H
H
H
H
H
H
OH
MeO
MeO
MeO
MeO
MeO
Et2N
OH
MeSO2
HOOC
MeCONH
Me2N
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
this work
this work
[18]
[19]
a
)
[20][21]
b
H2C¼CHCH2O
)
c
NC(CH2)2N(Me)
)
(HOꢀ(CH2)2)2N
[22][23]
[24]
Bu2N
d
Me(CH2)5O
(2,5-Dioxoimidazolidin-4-ylidene)methyl
)
)
e
OH
OH
BuO
H
OH
OH
[25]
[20]
f
)
MeO
MeO
Cl
[26]
[27][28]
[29]
I

1472
CHEMISTRY & BIODIVERSITY – Vol. 8 (2011)
Table 1 (cont.)
Compound
Type
Substituents
R1 R2
Reference
R3
R4
g
31
32
)
[30]
a) CAS registry No. 709023-48-3. b) CAS registry No. 432014-76-1. c) CAS registry No. 883101-39-1.
d) CAS registry No. 894264-48-3. e) CAS registry No. 709620-57-5. f) CAS registry No. 462626-00-2.
g) CAS registry No. 709644-05-3.
were also reported to show significant activities [6]. The natural PMH 1 and the
synthetic (5Z)-5-[4-(ethylsulfanyl)benzylidene]imidazolidine-2,4-dione (2) showed
potent anti-invasive and anti-migratory properties against PC-3M prostate cancer cells
in various pharmacological assays including the spheroid disaggregation assay [6][7]. A
50 mm dose of PMH 1 significantly increased transepithelial resistance (TER) of diluent
and CT- (calcitonin)-treated PC-3M cell cultures, suggesting that 1 promotes tight
junctions (TJs) formation and completely reverses the action of CT on TER [6][7].
PMH 1 decreased baseline paracellular permeability and also abolished CT-induced
increase in paracellular permeability of polarized PC-3M cell layers [6][7]. This effect
further supported the fact that 1 promotes TJs formation. PMHs 1 and 2 showed
prominent anti-invasive and anti-metastatic activities in an orthotopic xenograft model
of PC-3M cells in nude mice [7]. They also decreased orthotopic tumor growth and
inhibited the formation of tumor micrometastases in distant organs [7]. Further, PMH
2 also showed potent activity in LPB (long probasin promoter)-Tag (large T-antigen)
transgenic mice model [7]. It remarkably reduced the growth of primary tumors and
their metastasis in reproductive organs, decreased morbidity, and increased mice
survival average. Thus, PMHs 1 and 2 were considered as novel leads appropriate for
future optimizations as treatment for metastatic prostate cancers. The present study
builds on the previous studies and targets further optimization of PMHs as a lead anti-
migratory class to improve their potency level and to establish the structureꢀactivity
relationship by using diverse strategies.
2. Results and Discussion. – 2.1. Design of New PMHs. Previous CoMFA
(Comparative Molecular Field Analysis) results generated by using a data set of 35
PMHs were used in part to design the 2nd-generation PMHs [6]. The results of the
CoMFA are summarized as follows: 1) Areas of high steric bulk tolerance (80%
contribution) in p-position of the benzylidene group in PMH 2 were observed, and,
therefore, the activity can be significantly enhanced by bulky groups in this position. 2)

CHEMISTRY & BIODIVERSITY – Vol. 8 (2011)
1473
Bulky groups are sterically unfavorable in the o-positions of the benzylidene group;
therefore, it should not possess any bulky groups. 3) Electronegative (high electron
density) groups in the m- and p-positions of the benzylidene group may show better
activity (presence of substituted O, N, or S groups might increase the activity). 4) Low-
electron-density groups in o-position can improve the activity.
Several other strategies have also been used to design additional PMHs, for
example extension of the structure, which involves the addition of extra-functional
groups to the primary scaffold to probe for unused binding interactions with the target
[8]. This can better target hydrophobic regions in the binding pocket of the receptor.
Additional H-bonding interactions or ionic interactions can also be achieved by the
substitution with appropriate functional groups [8–12]. The linking group or an
extender sometimes play a significant activity role; especially if the terminal segments
are known to improve the activity, extension of the linker might improve the
interactions of these segments with their respective binding regions [8][13]. In
homologation technique, chain length can be used for optimization. Extending the
length of a saturated carbon side chain to five to nine atoms can sometimes increase the
pharmacological activity, while further lengthening reduces the activity [14–16].
Pharmacophore modeling is a useful tool to find new pharmacophore moieties, when a
structureꢀactivity relationship (SAR) of a known compound is available. Pharmaco-
phoric group determination can also be used in database searching to identify other
novel classes sharing the common pharmacophoric features [17]. Pharmacophoric
feature identification was illustrated successfully for hydroxamate inhibitors of the
matrix metalloproteinases (MMPs) [17].
This study describes the use of ligand-based drug optimization techniques including
CoMFA, dual substitutions at m- and p-positions, extension of the structure, linker
extension, and homologation techniques. Pharmacophore modeling was then used for
mapping essential pharmacophoric features needed for PMHsꢀ activity and to validate
the CoMFA results in terms of electrostatic requirements for the activity. Drug likeness
was also evaluated for the active PMHs to assess their future potential as drug
candidates.
2.2. Chemistry. Twelve new and eighteen known (3–14 and 15–32, resp.) PMHs
were synthesized using the feasible regioselective and cost-effective base-catalyzed
condensation reaction of hydantoin with various substituted aldehydes in the presence
of 2-aminoethanol (Table 1, Scheme) [6].
Scheme. General Synthesis of PMHs 1–32 [6]
For example, the HR-EI-MS data of 3 showed a [MꢀH]þ ion peak at m/z 247.0720,
suggesting the molecular formula C12H12N2O4. The identity of 3 as (5Z)-5-[4-(2-
hydroxyethoxy)benzylidene]imidazolidine-2,4-dione was established via analysis of its

1474
CHEMISTRY & BIODIVERSITY – Vol. 8 (2011)
13
1H- and C-NMR data. The singlet of the olefinic HꢀC(6)2) (d(H) 6.39) showed
HMBCs with the signals of aromatic C(8) and C(12) (d(C) 131.7), and with the one of
C(4) (d(C) 166.2). Furthermore, the signal of HꢀC(6) showed HMBCs with the signals
of quaternary C(5) and C(7) (d(C) 125.9 and 126.6, resp.). In the COSY spectrum, the
aromatic H-atom doublet at d(H) 6.96 (HꢀC(9) and HꢀC(11)) correlated with the
signal of HꢀC(8) and HꢀC(12) at d(H) 7.58. The latter H-atoms also showed HMBCs
with the signal of quaternary O-bearing C(10) (d(C) 159.5) and C(7). The triplet
CH2(1’) (d(H) 4.02) showed a HMBC with C(10) and COSY coupling with the
neighboring triplet CH2(2’) (d(H) 3.72).
The HR-EI-MS data of 4 showed a [MꢀH]þ ion peak at m/z 288.1353, which
1
suggested a molecular formula of C15H19N3O3. The H- and 13C-NMR data of 4
confirmed the identity of the 3-(dimethylamino)propoxy moiety at C(10)2).
The HR-EI-MS data of 5 and 6 showed [MꢀH]þ ion peaks at m/z 259.1084 and
211.0514, suggesting the molecular formulae C14H16N2O3 and C12H8N2O2, respectively.
The 1H- and 13C-NMR data of 5 and 6 further confirmed their identity to be (5Z)-5-(4-
tert-butoxybenzylidene)imidazolidine-2,4-dione and (5Z)-5-(4-ethynylbenzylidene)-
imidazolidine-2,4-dione, respectively.
The HR-EI-MS data of 7 and 8 showed [MꢀH]þ ion peaks at m/z 301.1550 and
427.2965, consistent with the molecular formulae C17H22N2O3 and C26H40N2O3,
respectively. Their 1H- and 13C-NMR data confirmed their identity as (5Z)-5-[4-
(heptyloxy)benzylidene]imidazolidine-2,4-dione and (5Z)-5-[4-(hexadecyloxy)benzyl-
idene]imidazolidine-2,4-dione, respectively.
1
The H- and 13C-NMR data of 9–11, along with their HR-EI-MS, suggested the
proposed structures.
Analysis of various spectral data indicated that 12 is (5Z)-5-(7-hydroxy-3,7-
dimethyloctylidene)imidazolidine-2,4-dione. The olefinic HꢀC(6)2) (d(H) 5.50)
showed HMBCs with the amide CO of the hydantoin ring (d(C) 164.9) and the CH
group C(8) (d(C) 33.2).
A
COSY experiment connected the segment
HꢀC(6)ꢀCH2(11)/Me(13). The doublet of Me(13) (d(H) 0.86) showed HMBCs with
C(7) (d(C) 33.8), C(9) (d(C) 37.3), and with C(8). Similarly, the Me singlets Me(14)
and Me(15) showed HMBCs with C(12) (d(C) 69.3) and C(11) (d(C) 44.4), confirming
the structure of the side chain.
1
The HR-MS, H- and 13C-NMR data identified 13 as (5Z)-5-{(2E)-3-[4-(diethyl-
amino)phenyl]prop-2-en-1-ylidene}imidazolidine-2,4-dione. The doublet for the ole-
finic HꢀC(6)2) appeared at d(H) 6.22, and showed HMBCs with the signal for a CO
group (d(C) 165.2) and the one for the olefinic CH group C(8) (d(C) 138.6). A COSY
experiment connected the segment HꢀC(6)ꢀHꢀC(8). The signal for HꢀC(10) and
HꢀC(14) (d(H) 7.31) showed HMBCs with those of C(8) and N-bearing quaternary
C(12) (d(C) 148.3). Similarly, the doublet for HꢀC(11) and HꢀC(13) (d(H) 6.66)
showed a HMBC with the signal for quaternary aromatic C(9) (d(C) 124.1). The triplet
for Me(2’) (d(H) 1.10) showed COSY coupling with the quartet for CH2(1’) (d(H)
3.37). This quartet also showed a HMBC with C(12). Compound 14 showed similar
NMR data to that of 13 with different R2 and R3 substitution patterns.
2
)
Arbitrary numbering; see Table 1. For systematic names, see Exper. Part.

CHEMISTRY & BIODIVERSITY – Vol. 8 (2011)
1475
2.3. Biological Evaluation and SAR Studies. During the process of metastasis, the
primary tumor tends to spread from its primary site to its preferable secondary tissues
or organs [31][32]. In this multistep process, cell migration comprises a major
contributing factor. Any failure in this multistep process, including the cell migration,
could block the entire metastatic process. Wound-healing assay is a classical and
commonly used method for studying cell migration [33]. A scratched cell monolayer
heals the wound in a characteristic manner; therefore, this assay is widely used to study
cell migration rates, cell polarization, and matrix remodeling studies [34–36]. The
wound-healing assay can be used for the validation of small molecule leads that affect
cell migration and for detailed cell biological studies [37–40]. This assay proved useful
as a proxy for metastasis, angiogenesis, and other pathophysiological and physiological
processes. Therefore, the wound-healing assay is used in this study to identify and rank
the anti-migratory potential of synthesized PMHs [41–46].
In this study, 30 PMHs were designed and synthesized to identify an optimized lead
for the control of androgen independent prostate cancer. The design of these
compounds was based in part on the earlier CoMFA results [6]. Different tactics were
utilized to modify PMHs and optimize their anti-migratory activity. The anti-migratory
activities of PMHs 3–32 using the wound-healing assay are shown in Table 2. These
compounds did not show any cytotoxicity at the tested doses, as evidenced by no
significant affect on cell viability in MTT assay. The IC50 values were calculated for
those compounds more active than PMH 2 (Table 2). Figs. 1 and 2 show the
doseꢀresponse effects for the PMHs more active than 2.
Type I compounds contain electronegative groups in the m- and p-positions. In
order to expand the series, sterically demanding substitutents were introduced in these
positions. Additionally, compounds with substitutents both in m- and p-position were
prepared, 11, 25–27, 29, 30, as well as a compound substituted in both m-positions, 28.
Fig. 1. Anti-migratory activity of 10, 13, 20, 24, 28, and 29 on highly metastatic PC-3 prostate
adenocarcinoma cells in the wound-healing assay. Active compounds are shown with respect to PMH 2.
PMH 2 is taken as the positive control, C (þ). Vehicle control is represented by C (ꢀ). Column height
indicates the mean value for three replicates in each treatment group.

1476
CHEMISTRY & BIODIVERSITY – Vol. 8 (2011)
Table 2. Anti-migratory Activity of PMHs 1–32 against PC-3 Cells using Wound-Healing Assay
Compound
Anti-migratory activity
% Migration (IC50 [mm]a)
Activity rankb)
1
2
3
4
5
6
7
8
9
74.0 (139.2)
60.8 (51.4)
95.9
94.3
79.0
100.0
97.2
99.4
89.5
33.2 (9.5)
68.7
100.1
15.9 (4.2)
65.8
95.1
76.7
87.1
69.2
94.0
47.8 (50.1)
76.8
100.3
96.6
28.4 (5.7)
72.7
65.0
99.7
54.5 (50.1)
43.6 (21.6)
67.2
–
–
1
1
1
1
1
1
1
3
2
1
3
2
1
1
1
2
1
3
1
1
1
3
2
2
1
3
3
2
1
2
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
99.3
73.2
a) IC50 values were calculated for compounds with activity level better than 2. b) Anti-migratory activity
ranking: rank-1: Activity ꢂcompared to the activity of 1. rank-2: Activity >activity of 1 and <than the
activity of 2. rank-3: Activity >than the activity of 2.
The active functional groups reported earlier were used. Among these, 11, 25, 26, and
30 were more active than PMH 1, while 28 and 29 were even more active than the most
active PMH 2 earlier reported. In 7, 8, and 23, C(10) was substituted with a long
saturated carbon side chain attached to an O-atom, but this homologation did not
improve the activity. These results were consistent with earlier results that showed that
increasing the number of saturated C-atoms attached to an O-atom even to two or four
C-atoms can reduce the activity [6]. Similarly, the saturated carbon side chain attached
to N-atoms was modified to one- and four-C-atom length in the case of 18 and 22,
respectively. In an earlier report, enhanced activity of PMHs with two C-atoms in the

CHEMISTRY & BIODIVERSITY – Vol. 8 (2011)
1477
Fig. 2. Wound-healing assay using PC-3 cells after 24 h of incubation with various doses of 24 (upper row)
and 29 (lower row) versus vehicle control (VC). X¼100.
saturated side chain attached to an N-atom at C(10) was described [6]. The lack of
activity of 22 suggests the complete loss of activity with a side chain with four C-atoms.
Type I PMHs retained the PMH moiety. An overall extension of the structure was
attempted with substituents to possibly increase the extent of binding interactions with
the binding regions at the target receptor pocket.
The necessity of a C(10) free OH group for the activity was investigated.
Interestingly, a significant increase in activity was observed for 10, where the OH group
was replaced by an NH2 group at this position, and an additional Cl substituted at C(8),
which counteracted the difference of lipophilicity between the OH and NH2 groups
(Table 2). In an earlier report [6], a PMH with a ClꢀC(8) group had a similar activity as
PMH 1 with a HOꢀC(10) group. The ClꢀC(8) moiety was presumed to sufficiently
increase the lipophilicity, without significantly compromising the steric effects at this
position due to its smaller atomic radius. Activity improvement of 10 can also be
explained by the presence of an H-bond acceptor (HBA) N-atom in p-position (C(10),
as also evidenced by pharmacophore modeling for activity increment). It can also be
interpreted that H-bond donor (HBD) groups are not crucial at this position, as
concluded by the increased activity of 10 versus 1 (the NH2 group has lower H-bond
donating ability than the OH group). The same conclusion can further be supported by
the improved activities of compounds with EtS, Me2N, and Et2N groups, which have no
HBD groups at the p-position (C(10)). This was also consistent with the pharmaco-
phore modeling results, which suggested the necessity of the presence of HBA
pharmacophore at C(10) for the activity.
In 31 and 32, the connection of the two rings contains a chain of two and three C-
atoms, respectively, with a Me substituent. While 31 was completely inactive, 32 was
comparable to 1 in activity. Therefore, type II molecules 13 and 14 were synthesized
with a three-C-atom linker in conjunction with earlier active groups in the m- and p-
positions [6]. Compound 13 showed the most promising anti-migratory activity, which
highlights the potential of 5-[3-phenylprop-2-en-1-ylidene]imidazolidine-2,4-diones as

1478
CHEMISTRY & BIODIVERSITY – Vol. 8 (2011)
a novel anti-migratory class. The justification of activity enhancement of 13 might be
attributed to the presence of the three-C-atom linker in addition to the terminal active
Et2N group.
In 12, an aliphatic side chain replaced the Ph ring, while the OH group was
maintained in the terminal part. The complete loss of anti-migratory activity of 12 in
highly metastatic PC-3 prostate cells further suggest that the Ph moiety plays a crucial
role for this activity.
2.4. Pharmacophore Modeling. DIStance COmparison technique (DISCOtech)
based on 3D pharmacophore mapping methodology was used to build the pharma-
cophore for the five active analogues 10, 20, 24, 28, and 29 [47][48]. DISCOtechTM is a
well-established module to design pharmacophoric maps and is based on the
assumption that a given set of active molecules, which are related by their
pharmacological activities, may possess consensus features in structural distance space
[49][50]. The identification of common pharmacophoric model features in the
molecule can be accomplished by using SYBYLꢀs DISCOtechTM. The pharmacophoric
elements taken to construct this model include H-bond donor (HBD) atoms, H-bond
acceptor (HBA) atoms, and hydrophobic centers. DISCOtechTM can be efficiently used
with as low as three to five active compounds to generate a pharmacophore model [49–
52].
In the present study, 100 conformers were generated for each compound with the
individual pair-wise tolerance value ranging from 0.5 to 2.0. A stochastic search method
was used to create these conformers. Ten DISCOtech pharmacophoric runs (Table 3)
were performed by varying the tolerance and range of required features. The pair-wise
tolerance value was optimized to accommodate the range of pharmacophoric features
with the highest score. Highest scores for top 5 pharmacophoric models in any
individual DISCOtech run are listed in Table 3. The final DISCOtech pharmacophore
model with the highest score, reflecting a maximum structural overlap, and a maximum
number of pharmacophoric features was proposed (Fig. 3). This model has three HBA
ligands (AL-1 to AL-3), two HBD ligands (DL-1 and DL-2), and two hydrophobic
centers (HY-1 and HY-2). It is worth noting that pharmacophore modeling of PMHs
was reported earlier for the search for glycogen synthase kinase-3b inhibitors [53]. In
Table 3. Number of Pharmacophoric Features and Top 5 High Score Ranges for Each of the DISCO
Pharmacophoric Run
Run Name
Feature
Range
Tolerance
Models
Total
Features
Top 5 Score
Range
Features in Top
Scored Model
DISC-1 (a)
DISC-1 (b)
DISC-1 (c)
DISC-1 (d)
DISC-1 (e)
DISC-2 (a)
DISC-2 (b)
DISC-2 (c)
DISC-2 (d)
DISC-2 (e)
3–7
3–7
3–7
3–7
0.5
1.0
1.5
1.85
2.0
0.5
1.0
1.5
1.85
2.0
50
100
52
57
153
49
49
49
52
57
5
5
6
7
7
6
6
6
7
7
2.2874–2.2877
2.7901–2.7902
2.8406–3.1067
3.1137–4.1223
3.6326–4.1324
4.3496–4.3497
4.3496–4.3497
3.3498–3.3499
4.0532–4.5373
4.1238–4.5095
2DL, 2AL, 1 HY
2DL, 1AL, 2 HY
2DL, 2AL, 2 HY
2DL, 2AL, 3 HY
2DL, 2AL, 3 HY
2DL, 2AL, 1 HY
2DL, 2AL, 1 HY
2DL, 2AL, 1 HY
2DL, 3AL, 2 HY
2DL, 3AL, 2 HY
3–7
7–15
7–15
7–15
7–15
7–15

CHEMISTRY & BIODIVERSITY – Vol. 8 (2011)
1479
the present pharmacophore modeling study, the tolerance parameters were set more
relaxed (1.0–1.85) to accommodate maximum relevant pharmacophoric features, since
the exact molecular target is not known. Distance relationships between various
pharmacophoric features are denoted in Fig. 4.
Fig. 3. Pharmacophoric features generated by using overlaid active PMHs 10, 20, 24, 28, and 29. Color
codes: blue, hydrophobic center; green, HBA site; pink, HBD site.
Fig. 4. Pharmacophoric features of active PMHs and their distance relationship generated by DISCO-
techTM module. AL-1/AL-2/AL-3, H-bond acceptor ligands; DL-1/DL-2, H-bond donor ligands; HY-1/
HY-2, hydrophobic centers.

1480
CHEMISTRY & BIODIVERSITY – Vol. 8 (2011)
2.5. In silico ADMET Screening. Drug-likeness of a compound can be assessed by
many approaches based on topological descriptors, fingerprints of molecular drug
likeness (MDL) structure keys, or other common properties such as clogP and
molecular weights [54]. Approaches for early ADMETox profiling, and new under-
standings with respect to molecular pharmacodynamics, have been rapidly evolving,
providing guidance and enabling early conclusions that were not available several years
ago, when several clinical trials failed due to poor ADMETox characters of drug
candidates. In the present study, the OSIRIS program was used to explicit these
properties. OSIRIS involves the database of traded drugs and supposedly non-drug-like
Fluka compounds to assess the occurrence frequency of each fragment in the individual
structure. Fragment-based drug-likeness of the four most active PMHs were evaluated
and compared with the reported anti-metastatic PMHs 1 and 2 (Fig. 5). Interestingly,
the potential drug-likeness value of 29 was significantly higher than PMHs 1 and 2,
while 13 showed a negative value with possible induction of toxicity. Furthermore, 24
also showed significant drug-likeness.
Fig. 5. Drug-likeness of active PMHs 10, 13, 24, and 29 compared to PMHs 1 and 2
In addition, the overall drug score was evaluated, which is a combined outcome of
drug-likeness, molecular weight, logS, clogP, and toxicity risks, encompassing muta-
genicity, tumorigenicity, irritating, and reproductive effects [55]. Drug score is
important in predicting a compoundꢀs potential to meet the criteriaꢀs of a possible
drug candidate [55]. Fig. 6 shows comparable drug scores for the PMHs 24 and 29,
which were significantly better than the reported PMHs 1 and 2. Interestingly, 24 was
never correlated with any biological activity, while a patent was awarded on anti-
allergic and anti-inflammatory activities of related structures including 29, which
revealed that it was devoid of both activities [27][28]. The structural features of PMHs
24 and 29 are commonly present in the traded drugs, and therefore they might qualify as
potential drug candidates for the control and prevention of metastatic androgen
independent prostate cancer.

CHEMISTRY & BIODIVERSITY – Vol. 8 (2011)
1481
Fig. 6. Overall drug score values of active PMHs 10, 13, 24, and 29 compared to those of PMHs 1 and 2
3. Summary of SAR and Conclusions. – The SAR studies suggest the following: 1.
R3 Substitution is critical for the anti-migratory activity based on the activity of 28
versus 29. A HBA atom is required at this position as evidenced by the pharmacophore
modeling. 2. If O is the HBA atom, the R3 position can not hold alkoxy substituents
with more than one C-atom, while one or two C-atoms attached to an N- or S-
containing substituent seem reasonable for the activity. 3. An electronegative R3 group
contributes more by behaving as a HBA moiety as opposed to HBD for the activity. 4.
Combined electronegative groups at R2, R3, and R4 positions seem to have an additive
effect for the activity as seen in 28 and 29. 5. A linker containing a three-C-atoms chain,
with (E)-oriented C¼C bond geometry, between the hydantoin and the Ph ring proved
superior vs. a linker containing only one or two C-atoms, as suggested by the active
phenylallylidene hydantoin 13. 6. The presence of the Ph ring is crucial for the activity
as evidenced by the lack of activity of 12.
PMHs proved effective in inhibiting cell migration, and therefore we can suggest
that they interfere with metastasis. Hence, PMHs could be valuable leads not only to
limit the spread of existing tumors, but also can control new tumor formation and
micrometastases. Therefore, PMHs can be used for chemoprevention in addition to
possible concomitant use with chemotherapeutic agents for the control of metastatic
prostate cancer.
This study was supported in part by the Louisiana Biomedical Research Network.
Experimental Part
General Experimental Procedures. TLC Analyses: precoated silica gel 60 F254 500 mm TLC plates
(EMD Chemicals), using MeOH/CHCl3 1:9 as a developing eluent. Column chromatography (CC):
silica gel 60 (SiO2; particle size 63–200 mm; Natland International). Optical rotation: Rudolph Research
Analytical Autopol III polarimeter. IR Spectra: Varian 800 FT-IR spectrophotometer. 1H- and 13C-NMR
spectra: in (D6)DMSO; with TMS as an internal standard; JEOL Eclipse-400 NMR spectrometer

1482
CHEMISTRY & BIODIVERSITY – Vol. 8 (2011)
operating at 400 MHz for 1H and 100 MHz for 13C. HR-EI-MS: conducted at Louisiana State University
on a 6200-TOF LCMS (Agilent) instrument equipped with a multimode source (mixed source that can
ionize the samples alternatively by ESI and APCI).
Preparation of PMHs. A two-neck round bottom flask was used to dissolve hydantoin (1.0 g) in 10 ml
H2O by heating at 708 in an oil bath with continuous stirring [6]. A sat. soln. of NaHCO3 was used to
maintain the pH at 7.0 [6]. 2-Aminoethanol (0.9 ml) was then added and the temp. was raised to 908 [6].
An equimolar quantity of substituted benzaldehyde dissolved in 10 ml of EtOH was then added dropwise
[6]. The mixture was kept under reflux for 10 h. The reaction was monitored by TLC every hour and the
endpoint of the reaction was visualized by the formation of a precipitate. The mixture was then cooled to
48, and the precipitate was filtered, washed with EtOH/H2O 1:5, and then recrystallized from EtOH.
Yields of the product ranged from 60–90%, based on the nature of individual aldehyde.
(5Z)-5-[4-(2-Hydroxyethoxy)benzylidene]imidazolidine-2,4-dione (3). White amorphous solid. IR
(CHCl3): 3274, 1734, 1717, 1653. 1H-NMR: 7.58 (d, J¼8.4, HꢀC(8), HꢀC(12)2)); 6.96 (d, J¼8.8, HꢀC(9),
HꢀC(11)); 6.39 (s, HꢀC(6)); 4.02 (t, J¼4.8, CH2(1’)); 3.72 (t, J¼4.8, CH2(2’)). 13C-NMR: 166.2 (s, C(4));
159.5 (s, C(10)); 156.2 (s, C(2)); 131.7 (d, C(8), C(12)); 126.6 (s, C(7)); 125.9 (s, C(5)); 115.4 (d, C(9),
C(11)); 109.3 (d, C(6)); 70.2 (t, C(1’)); 60.1 (t, C(2’)). HR-EI-MS: 247.0720 ([MꢀH]þ , C12H11N2O4þ ; calc.
247.0719).
(5Z)-5-{4-[3-(Dimethylamino)propoxy]benzylidene}imidazolidine-2,4-dione (4). White amorphous
solid. IR (CHCl3): 3159, 1716, 1670, 1654. 1H-NMR: 7.57 (d, J¼8.8); 6.94 (d, J¼8.8); 6.38 (s); 4.03 (t, J¼
6.6); 2.34 (t, J¼7.0); 2.14 (s); 1.84 (tt, J¼6.6). 13C-NMR: 166.2 (s); 159.4 (s); 156.3 (s); 131.7 (d); 126.6
(s); 125.9 (s); 115.3 (d); 109.2 (d); 66.5 (t); 56.2 (t); 45.8 (q); 27.4 (t). HR-EI-MS: 288.1353 ([MꢀH]þ
,
C15H18N3Oþ3 ; calc. 288.1348).
(5Z)-5-(4-tert-Butoxybenzylidene)imidazolidine-2,4-dione (5). Yellow amorphous solid. IR
(CHCl3): 3209, 1748, 1725. 1H-NMR: 7.55 (d, J¼8.8); 6.98 (d, J¼8.4); 6.40 (s); 1.32 (s). 13C-NMR:
166.2 (s); 156.2 (s); 156.1 (s); 131.0 (d); 128.2 (s); 127.4 (s); 124.0 (d); 108.9 (d); 79.2 (s); 29.1 (q). HR-EI-
MS: 259.1084 ([MꢀH]þ , C14H15N2O3þ ; calc. 259.1083).
(5Z)-5-(4-Ethynylbenzylidene)imidazolidine-2,4-dione (6). Yellow amorphous solid. IR (CHCl3):
1734, 1717, 1653. 1H-NMR: 7.63 (d, J¼8.4); 7.50 (d, J¼8.4); 6.41 (s); 4.32 (s). 13C-NMR: 165.8 (s); 156.1
(s); 134.1 (s); 132.5 (d); 130.0 (d); 129.1 (s); 121.8 (s); 107.7 (d); 83.9 (s); 82.8 (d). HR-EI-MS: 211.0514
([MꢀH]þ , C12H7N2Oþ2 ; calc. 211.0508).
(5Z)-5-[4-(Heptyloxy)benzylidene]imidazolidine-2,4-dione (7). White amorphous solid. IR
1
(CHCl3): 3429, 2928, 2856, 1747, 1734. H-NMR: 7.57 (d, J¼8.8); 6.94 (d, J¼8.8); 6.37 (s); 3.99 (t, J¼
6.2); 1.71 (tt, J¼6.6); 1.40 (tt, J¼6.2); 1.27–1.29 (m); 0.87 (t, J¼6.6). 13C-NMR: 166.4 (s); 159.4 (s); 157.4
(s); 131.6 (d); 125.9 (s); 124.2 (s); 115.3 (d); 109.1 (d); 68.1 (t); 31.8 (t); 29.2 (t); 29.0 (t); 26.0 (t); 22.6 (t);
14.5 (q). HR-EI-MS: 301.1550 ([MꢀH]þ , C17H21N2O3þ ; calc. 301.1552).
(5Z)-5-[4-(Hexadecyloxy)benzylidene]imidazolidine-2,4-dione (8). White amorphous solid. IR
1
(CHCl3): 2926, 2852, 1718, 1661. H-NMR: 7.34 (d, J¼8.8); 6.91 (d, J¼8.4); 6.60 (s); 3.96 (t, J¼6.6);
1.76 (tt, J¼7.0); 1.42 (tt, J¼7.0); 1.20–1.24 (m); 0.84 (t, J¼6.6). 13C-NMR: 163.5 (s); 160.0 (s); 155.9 (s);
130.7 (d); 125.3 (s); 115.2 (d); 112.2 (d); 68.3 (t); 32.0 (t); 29.8 (t); 29.7 (t); 29.6 (t); 29.4 (t); 29.2 (t); 26.1
(t); 22.8 (t); 14.2 (q). HR-EI-MS: 427.2965 ([MꢀH]þ ; C26H39N2O3þ ; calc. 427.2961).
(5Z)-5-[4-(1H-Tetrazol-5-yl)benzylidene]imidazolidine-2,4-dione (9). Yellow amorphous solid. IR
(CHCl3): 1716, 1699, 1653. 1H-NMR: 8.00 (d, J¼8.0); 7.65 (d, J¼8.4); 6.43 (s). 13C-NMR: 166.2 (s); 160.7
(s); 156.3 (s); 132.8 (s); 132.1 (s); 130.2 (d); 128.0 (s); 126.6 (d); 109.1 (d). HR-EI-MS: 255.0634 ([Mꢀ
H]þ ; C11H7N6Oþ2 ; calc. 255.0630).
(5Z)-5-(4-Amino-2-chlorobenzylidene)imidazolidine-2,4-dione (10). Yellow amorphous solid. IR
1
(CHCl3): 3420, 3338, 1733, 1717, 1654. H-NMR: 7.47 (d, J¼8.8); 6.68 (d, J¼2.6); 6.55 (s); 6.53 (s, J¼
2.2). 13C-NMR: 166.1 (s); 156.1 (s); 151.2 (s); 135.2 (s); 131.2 (d); 126.0 (s); 117.6 (s); 114.1 (d); 113.4 (d);
105.5 (d). HR-EI-MS: 236.0227 ([MꢀH]þ ; C10H7ClN3O2þ ; calc. 236.0227).
(5Z)-5-(4-Butoxy-3-nitrobenzylidene)imidazolidine-2,4-dione (11). Yellow amorphous solid. IR
(CHCl3): 1750, 1716, 1654. 1H-NMR: 8.08 (d, J¼2.2); 7.82 (dd, J¼2.2, 8.8); 7.40 (d, J¼8.8); 6.44 (s); 4.19
(t, J¼6.2); 1.70 (tt, J¼6.6); 1.42 (qt, J¼7.3); 0.92 (t, J¼7.3 ) . 13C-NMR: 165.8 (s); 156.2 (s); 151.1 (s);
140.5 (s); 135.5 (d); 128.6 (s); 126.0 (s); 125.3 (d); 115.7 (d); 106.5 (d); 69.6 (t); 31.0 (t); 19.1 (t); 14.1 (q).
HR-EI-MS: 304.0939 ([MꢀH]þ ; C14H14N3O5þ ; calc. 304.0933).

CHEMISTRY & BIODIVERSITY – Vol. 8 (2011)
1483
(5Z)-5-(7-Hydroxy-3,7-dimethyloctylidene)imidazolidine-2,4-dione (12). White amorphous solid.
IR (CHCl3): 3365, 2963, 2938, 1748, 1718, 1684. 1H-NMR: 5.50 (dd, J¼8.4, 8.1); 2.12–2.14 (m); 2.01–2.05
(m); 1.56–1.58 (m); 1.27–1.29 (m); 1.22–1.24 (m); 1.05 (s); 0.86 (d, J¼6.6). 13C-NMR: 164.9 (s); 155.5
(s); 131.7 (s); 111.1 (d); 69.3 (s); 44.4 (t); 37.3 (t); 33.8 (t); 33.2 (d); 29.8 (q); 21.9 (t); 19.9 (q). HR-EI-MS:
253.1556 ([MꢀH]þ ; C13H21N2Oþ3 ; calc. 253.1552).
(5Z)-5-{(2E)-3-[4-(Diethylamino)phenyl]prop-2-en-1-ylidene}imidazolidine-2,4-dione (13). Red
amorphous solid. IR (CHCl3): 1771, 1734, 1589. 1H-NMR: 7.31 (d, J¼8.4); 6.97 (dd, J¼12.1, 14.6);
6.79 (d, J¼15.4); 6.66 (d, J¼8.4); 6.22 (d, J¼11.7); 3.37 (q, J¼7.0); 1.10 (t, J¼7.0 ) . 13C-NMR: 165.2 (s);
154.9 (s); 148.3 (s); 138.6 (d); 129.0 (d); 127.5 (s); 124.1 (s); 117.5 (d); 111.9 (d); 111.4 (d); 40.5 (t); 13.1
(q). HR-EI-MS: 284.1404 ([MꢀH]þ ; C16H18N3O2þ ; calc. 284.1399).
(5Z)-5-[(2E)-3-(4-Hydroxy-3-methoxyphenyl)prop-2-en-1-ylidene]imidazolidine-2,4-dione (14).
Orange amorphous solid. IR (CHCl3): 3446, 3153, 1716, 1699, 1653. 1H-NMR: 7.12 (d, J¼1.8); 7.07
(dd, J¼12.1, 15.4); 6.91 (dd, J¼1.5, 8.2); 6.82 (d, J¼15.4); 6.77 (d, J¼8.1); 6.23 (d, J¼11.7); 3.90 (s).
13C-NMR: 165.3 (s); 155.1 (s); 148.5 (s); 148.2 (s); 137.9 (d); 128.9 (s, C(5,9)); 121.6 (d); 120.0 (d); 116.2
(d); 110.3 (d); 110.0 (d); 56.0 (q). HR-EI-MS: 259.0725 ([MꢀH]þ ; C13H11N2O4þ ; calc. 259.0719).
DISCOtech Pharmacophore Generation. Pharmacophore modeling studies were performed using
the SYBYL 8.1 suit of programs (Tripos Discovery Informatics, St. Louis, Missouri) installed on a DELL
desktop workstation equipped with a dual 2.0 GHz Intelꢃ Xeonꢃ processor running the Red Hat
Enterprise Linux (version 5) operating system. The chemical structures of hydantoin derivatives were
sketched in SYBYL 8.1 and assigned Gasteiger partial charges, and energy minimized using Tripos Force
Field to a final root-mean-square gradient of 0.01 kcal/mol to obtain the local minimum. Diverse
conformations of each compound were generated using stochastic conformational analysis tool in
SYBYL. The DISCOtech module on SYBYL 8.1 was used to build pharmacophore models based on
common chemical features. DISCOtech-generated initial pharmacophore features were edited to
remove the receptor interaction sites like H-bond donor and acceptor sites present in the receptor.
Overall alignment score and the associated pharmacophore features were analyzed for the selection of
appropriate pharmacophore model.
DISCOtech Pharmacophore Model Scores. The score of individual pharmacophore model was
calculated according to the following equation:
P
ðkꢀ0:5
10
ijÞ
X
dij ꢀ d0
H
M
alloverlaps
Score ¼
ð1Þ
allfeat k=ðk ꢀ 1Þ
where: H, number of molecules that match the model; M, number of targets (molecules in the input set);
k, number of features; dij, interface distance; d0, 2 A0; all feat, all features; all overlaps, all overlapping
features.
All the features were calculated in the model, features that are closer to each other than 2 A0 will
attract a penalty, while features that are further apart than 2 A0 will attract a positive score term. Penalty
and bonus were involved in the scoring formula, which constitutes an important aspect in the scoring
function. Penalty occurs for any mismatched molecules and for features closer than 2 A0. Bonus occurs
for features further than 2 A0 and for features further than tolerance. However, there was a lack of bonus
(not a penalty) for features closer than tolerance.
Drug-Likeness Score Predictions. The compoundsꢀ overall potential to qualify for a drug can be
determined by the combination of factors such as drug-likeness, cLogP, logS, molecular weight, and
toxicity risks [54][55]. This value was calculated by multiplying individual properties using the following
equation:
ꢀ
ꢁ
1
2
1
2
ds ¼ p
þ
si ꢃ p ꢃ ti
ð2Þ
where ds is the drug score, si are the contributions calculated directly from cLogP, logS, molecular weight,
and drug-likeness (pi) via Eqn. 3 describing a spline curve. Parameters a and b are (1, ꢀ5), (1, 5), (0.012,
ꢀ6) and (1, 0) for cLogP, logS, molecular weight, and drug-likeness, resp. [54][55]. ti are the

1484
CHEMISTRY & BIODIVERSITY – Vol. 8 (2011)
contributions taken from four toxicity risk types. The ti values of 1.0, 0.8, and 0.6 stand for no risk, medium
risk, and high risk, resp.
1
s ¼
ð3Þ
1 þ eapþb
Wound-Healing Assay. Prostate adenocarcinoma (human) cell line (PC-3) were maintained in
RPMI-1640 supplemented with 2 mmol/l glutamine, 10% fetal bovine serum (FBS), 100 mg/ml penicillin
G, and 100 mg/ml streptomycin [34–36]. Briefly, the cells were allowed to grow to 100% confluency in 24-
well plates. Once a monolayer developed, the same area of each well was then displaced by scratching a
line through the layer and a wound made with a sterile 100 ml pipette tip. The detached/floating cells were
then washed two times with PBS and finally with serum-free medium (SFM). The cells were then re-fed
with fresh medium RPMI-1640 containing 0.5% FBS and were treated with or without each of the tested
compounds for a period of 24 h. The cells were then fixed, stained with Diff-QuickTM (Dade Behring,
Newark, DE) and measurements of the wound area were photographed digitally with an inverted
microscope [34–40]. The total number of cells migrated was calculated for each condition. All
experiments were conducted independently in triplicate.
REFERENCES
[1] A. Jemal, R. Siegel, E. Ward, Y. Hao, J. Xu, M. J. Thun, CA Cancer J. Clin. 2009, 59, 225.
[2] American Cancer Society, http://www.cancer.org, accessed on March 9, 2010.
[3] J. Jongsma, M. H. Oomen, M. A. Noordzij, W. M. Van Weerden, G. J. M. Martens, T. H.
van der Kwast, F. H. Schrçder, G. J. van Steenbrugge, Prostate 2002, 50, 203.
[4] Y. Furuya, K. Akakura, S. Akimoto, H. Inomiya, H. Ito, Int. J. Urol. 1999, 6, 240.
[5] N. Segawa, I. Mori, H. Utsunomiya, M. Nakamura, Y. Nakamura, L. Shan, K. Kakudo, Y. Katsuoka,
Pathol. Int. 2001, 51, 452.
[6] M. Mudit, M. Khanfar, A. Muralidharan, S. Thomas, G. V. Shah, R. W. M. van Soest, K. A.
El Sayed, Bioorg. Med. Chem. 2009, 17, 1731.
[7] G. V. Shah, A. Muralidharan, S. Thomas, M. Gokulgandhi, M. Mudit, M. Khanfar, K. El Sayed, Mol.
Cancer Ther. 2009, 8, 509.
[8] G. L. Patrick, ꢄAn Introduction to Medicinal Chemistryꢀ, 3rd edn., Oxford University Press Inc.,
New York, 2005.
[9] K. R. Acharya, E. D. Sturrock, J. K. Riodan, M. R. W. Ehlers, Nat. Rev. Drug Discovery 2003, 2, 891.
[10] A. A. Patchett, E. Harris, E. W. Tristram, M. J. Wyvratt, M. T. Wu, D. Taub, E. R. Peterson, T. J.
Ikeler, J. ten Broeke, L. G. Payne, D. L. Ondeyka, E. D. Thorsett, W. J. Greenlee, N. S. Lohr, R. D.
Hoffsommer, H. Joshua, W. V. Ruyle, J. W. Rothrock, S. D. Aster, A. L. Maycock, F. M. Robinson,
R. Hirschmann, C. S. Sweet, E. H. Ulm, D. M. Gross, T. C. Vassil, C. A. Stone, Nature 1980, 288, 280.
[11] D. J. Abraham, ꢄBurgerꢀs Medicinal Chemistry and Drug Discoveryꢀ, 6th edn., John Wiley and Sons,
New York, 2003.
[12] A. P. Feinberg, I. Creese, S. H. Snyder, Proc. Natl. Acad. Sci. U.S.A. 1976, 73, 4215.
[13] C. R. Ganelllin, S. M. Roberts, ꢄMedicinal Chemistry – The Role of Organic Chemistry in Drug
Researchꢀ, 2nd edn., Academic press, New York, 1994.
[14] R. B. Silverman, ꢄThe Organic Chemistry of Drug Design and Drug Actionꢀ, 2nd edn., Elsevier
Academic Press, MA, 2004.
[15] B. W. Richardson, Med. Times Gaz. 1869, 2, 703.
[16] A. B. H. Funcke, M. J. E. Ernsting, R. F. Rekker, W. T. Nauta, Arzneim. Forsch. 1953, 3, 503.
[17] A. K. Ghose, J. J. Wendoloski, Persp. Drug Discovery Design 1998, 9–11, 253.
[18] G.-Z. Ao, Y.-H. Zhang, H. Ji, G. Deng, Yaoxue Xuebao 2004, 39, 803.
[19] C. Syldatk, V. Lehmensiek, G. Ulrichs, U. Bilitewski, K. Krohn, H. Hçke, F. Wagner, Biotechnol.
Lett. 1992, 14, 99.
[20] Y. Luo, L. Ma, H. Zheng, L. Chen, R. Li, C. He, S. Yang, X. Ye, Z. Chen, Z. Li, Y. Gao, J. Han, G. He,
L. Yang, Y. Wei, J. Med. Chem. 2010, 53, 273.

CHEMISTRY & BIODIVERSITY – Vol. 8 (2011)
1485
[21] A. I. Khodair, A. A. El-Barbary, Y. A. Abbas, D. R. Imam, Phosphorus, Sulfur Silicon Relat. Elem.
2001, 170, 261.
[22] P. Kitipichai, R. La Peruta, G. M. Korenowski, G. E. Wnek, J. Polym. Sci., Part A: Polym. Chem.
1993, 31, 1365.
[23] A. K. Sen, S. Mukhopadhyay, Indian J. Chem., Sect. B 1983, 22, 939.
[24] B. Malhotra, A. K. Dwivedi, V. Hora, I. M. Hora, J. Indian Chem. Soc. 1990, 67, 74.
[25] J. Tanaka, K. Nakayasu, Patent FR 2585701 A1, 1987.
[26] K. R. Bharucha, V. Pavilanis, D. Ajdukovic, H. M. Schrenk, Patent DE 2329745, 1974.
[27] G. Billek, Monatsh. Chem. 1961, 92, 352.
[28] W. A. Cetenko, D. T. Connor, R. J. Sorenson, P. C. Unangst, S. R. Stabler, Patent EP 343643, 1989.
[29] S. Ostergaard, H. B. Olsen, N. C. Kaarsholm, P. Madsen, P. Jakobsen, S. Ludvigsen, G. Schluckebier,
D. B. Steensgaard, A. K. Petersen, Patent WO 2004080480, 2004.
[30] M. Lamiri, K. Bougrin, B. Daou, M. Soufiaoui, E. Nicolas, E. Giralt, Synth. Commun. 2006, 36, 1575.
[31] L. Weiss, Cancer Metastasis Rev. 2000, 19, 193.
[32] I. J. Fidler, Nat. Rev. Cancer 2003, 3, 453.
[33] M. G. Lampugnani, Methods Mol. Biol. 1999, 96, 177.
[34] K. V. Lu, K. A. Jong, A. K. Rajasekaran, T. F. Cloughesy, P. S. Mischel, Lab. Investig. 2004, 84, 8.
[35] B. Herren, K. J. Garton, S. Coats, D. F. Bowen-Pope, R. Ross, E. W. Raines, Exp. Cell Res. 2001, 271,
152.
[36] C. Huang, Z. Rajfur, C. Borchers, M. D. Schaller, K. Jacobson, Nature 2003, 424, 219.
[37] K. T. Mc Henry, S. V. Ankala, A. K. Ghosh, G. Fenteany, ChemBioChem 2002, 3, 1105.
[38] A. Vogt, K. E. Pestell, B. W. Day, J. S. Lazo, P. Wipf, Mol. Cancer Ther. 2002, 1, 885.
[39] C. Klein-Soyer, J. P. Cazenave, J. M. Herbert, J. P. Maffrand, J. Cell Physiol. 1994, 160, 316.
[40] E. G. Fischer, A. Stingl, C. J. Kirkpatrick, J. Immunol. Methods 1990, 128, 235.
[41] D. Philp, T. Huff, Y. S. Gho, E. Hannappel, H. K. Kleinman, FASEB J. 2003 17, 2103.
[42] E.-J. Moon, Y. M. Lee, O.-H. Lee, M.-J. Lee, S.-K. Lee, M.-H. Chung, Y.-I. Park, C.-K. Sung, J.-S.
Choi, K.-W. Kim, Angiogenesis 1999, 3, 117.
[43] P. C. Ma, T. Kijima, G. Maulik, E. A. Fox, M. Sattler, J. D. Griffin, B. E. Johnson, R. Salgia, Cancer
Res. 2003, 63, 6272.
[44] R. S. Chung, P. A. Adlard, J. Dittmann, J. C. Vickers, M. I. Chuah, A. K. West, J. Neurochem. 2004,
88, 454.
[45] L. W. Morgan, A. Wieslander, M. Davies, T. Horiuchi, Y. Ohta, M. J. Beavis, K. J. Craig, J. D.
Williams, N. Topley, Kidney Int. 2003, 64, 1854.
[46] Y. Saga, H. Mizukami, Y. Takei, K. Ozawa, M. Suzuki, Int. J. Oncol. 2003, 23, 1109.
[47] Y. C. Martin, M. G. Bures, E. A. Danaher, J. DeLazzer, I. Lico, P. A. Pavlik, J. Comput.-Aided Mol.
Des. 1993, 7, 83.
[48] R. D. Clark, J. Chem. Inf. Comput. Sci. 1997, 37, 1181.
[49] D. P. Marriott, I. G. Dougall, P. Meghani, Y.-J. Liu, D. R. Flower, J. Med. Chem. 1999, 42, 3210.
[50] G. Spadoni, C. Balsamini, G. Diamantini, B. Di Giacomo, G. Tarzia, M. Mor, P. V. Plazzi, S. Rivara,
V. Lucini, R. Nonno, M. Pannacci, F. Fraschini, B. M. Stankov, J. Med. Chem. 1997, 40, 1990.
[51] H.-J. Kim, H. Choo, Y. S. Cho, K. T. I. No, A. N. Pae, Bioorg. Med. Chem. 2008, 16, 636.
[52] D. S. Patel, P. V. Bharatam, J. Comput.-Aided Mol. Des. 2006, 20, 55.
[53] M. Khanfar, B. A. Asal, M. Mudit, A. Kaddoumi, K. A. El Sayed, Bioorg. Med. Chem. 2009, 17,
6032.
[54] I. V. Tetko, Drug Discovery Today 2005, 10, 1497.
[55] G. Szileighl, Arzneim.-Forsch. 1984, 35, 1260.
Received June 18, 2010