Phenylmethylene hydantoins as prostate cancer invasion and migration inhibitors. CoMFA approach and QSAR analysis

Article abstract of DOI:10.1016/j.ejmech.2010.08.066

Prostrate cancer constitutes the second leading cause of cancer deaths in men in United States. In the process of discovery of new antiproliferative and anti-metastatic agents against prostate cancer, marine-derived phenylmethylene hydantoin (PMH) derivatives were identified with activity level range between 50 and 200 μM. 3D-QSAR CoMFA model was used in virtual screening of commercially available derivatives of PMH. PMH derivatives with manifold increase in anti-migratory and anti-invasive activities were discovered using wound-healing and Cultrex invasion assays. Benzene ring replacement with other heterocyclic rings did not significantly improve the methylene hydantoins activities. Multivariate analysis performed on the whole series of methylene hydantoins, which further supported the findings of CoMFA model. Predictive QSAR model with conventional r² and cross-validated coefficient (q ²) values up to 0.982 and 0.803 were established. The molecular volume (MV) and the log P were identified as critical parameters for methylene hydatoins migration inhibitory activity. PMH is a novel anti-metastatic lead class with potential therapeutic activity against prostate cancer.

Full text of DOI:10.1016/j.ejmech.2010.08.066

European Journal of Medicinal Chemistry 45 (2010) 5397e5405
Contents lists available at ScienceDirect
European Journal of Medicinal Chemistry
journal homepage: http://www.elsevier.com/locate/ejmech
Original article
Phenylmethylene hydantoins as prostate cancer invasion and migration
inhibitors. CoMFA approach and QSAR analysis
*
Mohammad A. Khanfar, Khalid A. El Sayed
Department of Basic Pharmaceutical Sciences, College of Pharmacy, University of Louisiana at Monroe, 1800 Bienville Drive, Monroe, Louisiana 71201, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
Prostrate cancer constitutes the second leading cause of cancer deaths in men in United States. In the
process of discovery of new antiproliferative and anti-metastatic agents against prostate cancer, marine-
derived phenylmethylene hydantoin (PMH) derivatives were identified with activity level range between
Received 2 July 2010
Received in revised form
27 August 2010
Accepted 27 August 2010
Available online 8 September 2010
50 and 200 mM. 3D-QSAR CoMFA model was used in virtual screening of commercially available deriv-
atives of PMH. PMH derivatives with manifold increase in anti-migratory and anti-invasive activities
were discovered using wound-healing and Cultrex invasion assays. Benzene ring replacement with other
heterocyclic rings did not significantly improve the methylene hydantoins activities. Multivariate anal-
ysis performed on the whole series of methylene hydantoins, which further supported the findings of
CoMFA model. Predictive QSAR model with conventional r2 and cross-validated coefficient (q2) values up
to 0.982 and 0.803 were established. The molecular volume (MV) and the log P were identified as critical
parameters for methylene hydatoins migration inhibitory activity. PMH is a novel anti-metastatic lead
class with potential therapeutic activity against prostate cancer.
Keywords:
Phenylmethylene Hydantoins
Anti-invasion
Prostate cancer
CoMFA
QSAR
Ó 2010 Elsevier Masson SAS. All rights reserved.
1. Introduction
platelet aggregation inhibition, anti-arrhythmic and antihyperten-
sive, anti-diabetic, neuroprotective, HDL/cholesterol modulating,
Prostate cancer remains the leading cause of men’s mortality
worldwide [1]. Despite important progress in the early diagnosis of
prostate neoplasias through the measurement of prostate-specific
antigen levels, 10% of newly diagnosed patients have some
evidence of locally advanced prostate cancer and 5% already have
distant metastasis at the time of diagnoses [2,3]. Postmortem
analysis of deaths attributed to prostate cancer reveals that most
subjects have evidence of metastatic disease [4]. Some treatment
alternatives exist with limited curative potential in the case of
locally advanced prostate cancer [5e7]. In contrast, patients with
evidence of distant metastasis have very poor prognosis and no
curative treatment exists [8]. Prostate cancer deaths are a result of
metastatic disease and treatment of such metastatic disease is one
of the major therapeutic challenges. Thus, there is urgent need to
develop novel therapeutic approaches and strategies to control
metastatic prostate cancer.
antiviral, anticonvulsant, and growth hormone secretagogue [9,10].
On the other hand, much interest has been given to their cancer
chemotherapeutic activity. Hydantoins were shown to inhibit MMPs
[11] and were patented in 1987 for their tyrosine kinase inhibitory
and antiallergic activities [12]. 2-[4-[(5-Oxo-2-thioxo-4-imidazoli-
dinylidene)-methyl]phenoxy] propionic acid and analogs bind the
ERK1 or ERK2 kinase, inhibit phosphorylation, and slow cell prolif-
eration in vitro [13]. Peptide hydantoin analogs were patented for
their ability to inhibit thrombocyte aggregation, metastasis, and
osteoclast binding to bone surfaces [14]. A
b-carboline hydantoinwas
found to have phosphodiesterase 5 and colon tumor cell growth
inhibition activity [15]. 5-Benzylidene hydantoins inhibited A549
lung cancer cell line proliferation via dual mechanism; inhibiting
EGFR autophosphorylation and increasing p53 level [16]. Azaspiro
bicyclic hydantoin derivatives have anti-tumor and anti-angiogenic
activity against ovarian and osteosarcoma cells through down regu-
lation of the secretion of VEGF [17]. Substituted hydantoin derivatives
have also been patented as MEK1 and MEK2 inhibitors, which are
usefultargets inthe treatmentofcellproliferativedisordersincluding
cancer [18]. Fluorinated spirohydantoin derivative induced cytotox-
icity in chronic myelogenous leukemia by inhibiting the cell growth
via interfering with DNA replication [19]. Hydantoin derivatives were
found to have androgen receptor modulation, which can be used for
treating prostate cancer and other androgen receptor-related
Over the past few years, hydantoin derivatives have attracted
a great attention due to their diverse biological activities. These
include fungicidal, herbicidal, anti-inflammatory, anti-HIV, analgesic,
cannabinoid receptor-1, 5HT, purine P-2X receptor antagonism,
* Corresponding author. Tel.: þ1 318 342 1725; fax: þ1 318 342 1737.
E-mail address: elsayed@ulm.edu (K.A. El Sayed).
0223-5234/$ e see front matter Ó 2010 Elsevier Masson SAS. All rights reserved.
doi:10.1016/j.ejmech.2010.08.066

5398
M.A. Khanfar, K.A. El Sayed / European Journal of Medicinal Chemistry 45 (2010) 5397e5405
diseases [20]. Phenylemthylene hydantoins (PMH) were reported to
inhibit GSK-3 , potential target for the treatment of cancer, Alz-
vibration bands at a higher frequency region (1660e1675 cmꢀ1
)
b
characteristic for Z-oriented D5,6 bonds, which are distinguishable
from those expected for the E isomers (1630e1640 cmꢀ1) [9]. The
chemical shift of the most diagnostic olefinic proton H-6 was
downfield shifted in the 1H NMR spectra (dH 6.40e7.00), which
further confirmed the Z-orientation of the D5,6 system. The expec-
ted chemical shift of H-6 in the E isomers should be 6.20e6.30 ppm.
The downfield shift of H-6 in the Z isomer is attributed to the
anisotropic effect of the spatially nearby C-4 carbonyl group [9].
heimer’s disease, bipolar disorders, and type-2 diabetes [21].
Natural products have proven to be the most reliable source of
new anticancer entities [22,23]. Nearly 63% of anticancer drugs
introduced over the last 25 years are natural products or can be
traced back to a natural products origin [24]. A recent review lists
79 natural products or natural product analogs that entered clinical
trial as anticancer agents between 2005 and 2007 [25]. Ethanolic
extract of the abundant shallow water Red Sea sponge Hemimycale
arabica inhibited the proliferation and invasion of the highly
metastatic human prostate cancer PC-3M cell line [26,27]. Liquid
chromatography of this extract afforded the known (Z)-5-(4-
hydroxybenzylidene)-hydantoin (1) [27]. This natural PMH (1) and
the related synthetic (Z)-5-(4-(ethylthio)benzylidene)-hydantoin
(2) showed potent in-vitro anti-growth and anti-invasive proper-
ties against PC-3M prostate cancer cells in MTT and spheroid
disaggregation assays [26,27]. PMHs 1 and 2 also showed anti-
invasive activity in orthotopic xenograft of PC-3 M cells in nude
mice model [26]. They decreased orthotopic tumor growth and
inhibited the formation of tumor micrometastases in distant organs
without apparent cytotoxic effects at the test doses [23]. PMH 2 also
showed a potent activity in LPB-Tag mice transgenic model [26].
PMH 2 remarkably reduced growth of primary tumors and their
metastasis in reproductive organs, decreased morbidity and
increased survival of the LPB-Tag mice [26]. Recently, 1 and 2 were
found to decrease total CD44 and CD44 v7e10 expression, which
could partly explain their anti-metastatic activity [21].
Previous study from our laboratory has investigated the influ-
ence of substitutions in the phenyl ring of the PMH on the prostate
cancer migration inhibitory activity [27]. Using regioselective and
cost-effective condensation reaction of hydantoin and substituted
benzaldehyde, several diverse derivatives of PMH molecule that
exhibit different degrees of activity were structurally modified [27].
On the basis of this investigation, predictive 3D-QSAR (CoMFA)
model with conventional r2 and cross-validated coefficient (q2)
values up to 0.910 and 0.651 were established [27]. The main
objective of this study is to further discover and optimize new PMH
derivatives with enhanced anti-migratory and anti-invasive activ-
ities against prostate cancer cells. To accomplish this aim, three
strategies were applied. Firstly, CoMFA-based virtual screening of
in-house database of virtually sketched PMHs using all benzalde-
hydes that are commercially available in SigmaeAldrich and VWR
repositories was conducted. Since all structural modifications in the
first report were focused on the substitutions of the phenyl ring of
PMH, the second applied strategy was the investigation of the
influence of replacing the phenyl ring with other heterocyclic rings.
Finally, predictive quantitative structure-activity relationship
(QSAR) was derived based on the relevant physiochemical
descriptors of the compounds that would be of value in guiding
lead PMH design. The partial least square (PLS) method was applied
to find relationships between the in vitro activity and the physi-
ochemical data.
2.2. Virtual screening and biological evaluation
3D-QSAR model of 35 PMH derivatives was generated and
validated using Comparative Molecular Field Analysis (CoMFA)
[27]. Predictive 3D-QSAR model with conventional r2 and cross-
validated coefficient (q2) values up to 0.910 and 0.651, respectively,
were established [27]. CoMFA steric and electrostatic fields from
PLS analysis were generated [27]. In this analysis both steric and
electrostatic fields contribute to the QSAR equation by 70.6% and
29.4%, respectively [27]. This suggested that variation in biological
activity of inhibitors is dominated by differences in steric (van der
Waals) interactions with the hydantoins active site.
Virtual screening (VS) has been a fruitful strategy for lead
identification for several decades. Its use has substantially
increased in recent years, especially with the advent of online
databases, ending the discovery of new hits out of millions of
virtual structures. VS offers means to systematically select a small
number of compounds, which can then be tested in lower-
throughput assays, increasing the likelihood of discovery and
identification of compounds with desired activities as compared to
the more-expensive and time-consuming physical high-
throughput screening. The study herein applied ligand-based
virtual screening using the validated CoMFA model to design and
discover new PMH derivatives with enhanced anti-migratory and
anti-invasive activities (Fig. 1). All commercially available
substituted benzaldehydes in VWR and SigmaeAldrich repositories
were collected and the corresponding PMH derivatives were
virtually sketched using SYBYL 8.0. 636 virtual PMH derivatives
were subjected to CoMFA (PLS) model and predicted cutoff IC50
value of 50
derivatives were found to have calculated IC50 less than 50
only 15 were readily available for synthesis (Table 1). All
compounds were first tested at 100 M in wound-healing assay
mM was used as selection criteria. Eighteen PMH
mM, but
m
against the metastatic prostate cancer (PC-3) and compounds with
percentage inhibition ꢁ50% were further tested to determine their
IC50 values (Table 1). Five compounds (3, 4, 7, 9, and 10) exhibited
potent anti-migratory activity, including compound 10 with 5-folds
better activity compared to the lead PMH 2. Compound 10
(IC50 ¼ 8.1
doseeresponse manner in wound-healing assay (Fig. 2) without
significant cytotoxicity up to 200 M as assessed by the MTT assay.
mM) showed a strong inhibition of PC-3 migration in
m
Working consistently with CoMFA model, the common feature
between the five most active compounds is the presence of bulky
aliphatic and aromatic substitution at the para-position of the
phenyl ring although they have different electronic properties.
2. Results and discussion
2.1. Chemistry
O
NH
PMHs were synthesized using Scheme 1 [9]. This Scheme
includes base-catalyzed condensation of hydantoin with
substituted benzaldehydes [28]. Geometrical isomerism (E/Z
isomers) was possible due to the restricted rotation around the
exocyclic D5,6 double bond of the PMHs, however this method
regioselectively afforded the Z-geometry as confirmed by the
spectral data. The IR spectra of PMHs showed the stretching
O
HN
H
N
O
O
NaHCO3/Aq. EtOH
H
O
Ethanolamine
Reflux
HN
R
R
Scheme 1. General synthetic scheme of phenylmethylene hydantoins.

M.A. Khanfar, K.A. El Sayed / European Journal of Medicinal Chemistry 45 (2010) 5397e5405
5399
Table 1
Structures of PMHs and their biological activities in wound-healing assay against
PC-3 cells.
O
NH
HN
O
R1
R2
R3
R5
R4
Cmpd R1
R2
R3
R4
R5
% Inhibition IC50
a
at 100 mM
1
2
H
H
H
H
OH
SEt
H
H
H
H
34
127.9 ꢂ 8.1
42.9 ꢂ 4.7
42.9
(at 50
76.3
77.5
36.3
m
M)
3
4
5
H
H
H
H
Br
OCH2(4-F-Ph)
OEt
H
OEt
H
H
H
H
23.8 ꢂ 3.3
35.7 ꢂ 1.5
86 ꢂ 10.1
OMe OCH2CO-
pyrrolidine
6
7
8
OPr
H
H
H
H
H
OPr
H
OMe
H
H
H
H
68.4
91.8
33.2
77.1 ꢂ 4.0
19.7 ꢂ 2.9
94 ꢂ 11.7
OCH2COPh
OCH2CO-
mopholine
OCH2CCH
9
H
H
H
H
Cl
OMe
Br
H
H
H
H
H
H
H
H
H
H
H
H
H
H
88.8
97.4
66.7
75.9
80.9
23
21.8 ꢂ 3.5
8.1 ꢂ 1.4
Fig. 1. Outlines of the virtual screening methodology that was applied to discover new
active PMHs.
10
11
12
13
14
15
16
17
OMe OPr
H
H
H
H
OCH2CH2CH(CH3)2
pyrrolidine
SMe
OPr
OtBu
46.1 ꢂ 8.0
48.7 ꢂ 6.1
44.7 ꢂ 9.3
110 ꢂ 12.3
91 ꢂ 7.8
H
OMe
OMe Cl
OMe Cl
H
Bioisosteric replacement of the phenyl ring of PMH with other
heterocyclic ring was attempted (Fig. 3). Excluding compound 21,
replacement of the phenyl with other heteroaromatic functionality
did not improve the anti-migratory activity relative to the lead 2.
Bioisosteric replacement of the thioethane moiety in 2 with
a thiophene in 21 showed a good anti-migratory activity in wound-
39
OPr
OtBu
OMe 19
H 52.7
187 ꢂ 17.9
60.8 ꢂ 4.9
H
a
The values are the mean ꢂ SD. Each experiment was conducted in triplicate.
healing assay (IC50 ¼ 32.5
mM) and equipotent activity to the lead 2
(IC50 ¼ 42.9 M).
m
compound 10 may be required to enhance its potency to nanomolar
activity, while maintaining its low cytotoxicity.
Anti-invasive activities of active compounds in wound-healing
assay (2e4, 7, 9, 10, 12, 13, 21) were measured using Trevigen’s
CultrexÒ 96 well Basement Membrane Extract (BME) cell invasion
assay kit against the highly metastatic PC-3 cells [28,29]. This assay
employs a simplified Boyden Chamber design with a polyethylene
Overall activities profile of PMHs showed that long bulky
substituents with various aliphatic or aromatic groups at the para-
position of the benzene ring are favorable with manifold increase in
anti-migratory and anti-invasive activities. This finding is confirmed
by QSAR analysis, as described later. Alternatively, electron-donating
substitution on one meta-position and electron-withdrawing on the
other meta-position with limited steric properties are associated
with activity improvement. However, ortho-substituent is not
tolerable, which dramatically decreases the activity.
terephthalate membrane (8 mm). Detection of cell invasion is
quantified using calcein-AM [28,29]. Cells internalize calcein-AM
and intracellular esterases cleave the acetomethylester moiety to
generate free calcein. Free calcein fluoresces brightly, and this
fluorescence may be used to quantify the number of cells that have
invaded across BME [28,29]. Anti-invasive activities of 2e4, 7, 9, 10,
12, 13, and 21 at 50 mM doses are shown in Fig. 4. Comparable to
wound-healing assay, PMH 10 was 5-folds more active than the hit
PMH 2. Other active PHMs also showed variable range of activities.
Interestingly, PMH 21 was 3-folds more potent than 2, though it was
equipotent in wound-healing assay, suggesting multiple targets for
this compound. Active compounds were then tested for their cyto-
toxic activity using MTT assay (Fig. 5). Excluding 13, PMHs did not
display any significant cytotoxicity against PC-3 over 72 h at the
maximum tested dose (200 mM). This shows a clear difference
between the cytotoxic and anti-invasive activity levels. Different
derivatives of PMH were tested in vivo for their anticonvulsant [9]
2.3. QSAR analysis
Multivariate analysis was carried out on all the methylene
hydantoins which were characterized by 14 physicochemical
descriptors representing various size, electronic, and lipophilic
properties (Supporting Information Table 1). The lipophilicity of the
methylene hydantoins was described by theoretically-determined
log P (clog P) values and number of hydrophobic center (HC). The
size parameters of methylene hydantoins were described by
molecular volume (MV), molecular area (MA), bond counts (BC),
molar refractivity (MR), and polar surface area (PSA). The electronic
properties of hydantoins were described by the dipole moment
(DM), highest occupied molecular orbital (HOMO), lowest unoc-
cupied molecular orbital (LUMO), hydrogen bond acceptor (HBA),
hydrogen bond donor (HBD), and ionization potential (IP). All
physiochemical parameters were theoretically calculated using the
SYBYL 8.0 spreadsheet autofill module [30]. The difference in 13C
and GSK-3
b inhibitory [21] activity using maximal electroshock
seizure and liver-glycogen level, respectively. These compounds
were well tolerated with no reported fatalities. These in-vivo and in-
vitro data presented herein provide a preliminary indication of
excellent safety and tolerability of PMHs as active entity. The high
activity and lowcytotoxicity of 10 could provide a basis for designing
new analogs with wide therapeutic window. Future modification of

5400
M.A. Khanfar, K.A. El Sayed / European Journal of Medicinal Chemistry 45 (2010) 5397e5405
Fig. 2. Doseeresponse activity of 10 in wound-healing assay against PC-3 cell line. Magnification ¼ 100ꢃ.
chemical shift
(Dd), an additional electronic parameter, was
QSAR study [31]. The partial least squares projection to latent
experimentally determined by NMR studies.
Principal component analysis (PCA) was used to evaluate the
molecular diversity of the methylene hydantoins selected for the
structures (PLS), which is a regression extension of PCA, was later
used to quantify and predict the relationship between the activity
and physicochemical descriptors [31]. It was imperative to evaluate
O
O
O
O
H
NH
O
HN
HN
N
N
HN
NH
O
N
HN
O
NH
N
NH
19
N
S
HN
O
21
18
20
O
O
O
O
O
NH
O
O
N
NH
HN
NH
HN
NH
HN
HN
N
H
O
NH
N
O
O
O
25
24
22
23
O
O
O
HO
I
HN
HN
HO
HN
NH
NH
NH
N
N
N
H
H
O
O
O
28
27
26
Fig. 3. Structural modification of PMH using different heterocycles.

M.A. Khanfar, K.A. El Sayed / European Journal of Medicinal Chemistry 45 (2010) 5397e5405
5401
Table 2
120
100
80
60
40
20
0
Experimental and predicted pIC50 of the training and test sets.
a
b
Compound
Actual pIC50
Pred. pIC50
Sets
1
2
3
4
5
6
7
8
ꢀ2.106
ꢀ1.632
ꢀ1.376
ꢀ1.552
ꢀ1.934
ꢀ1.887
ꢀ1.294
ꢀ1.973
ꢀ1.338
ꢀ0.908
ꢀ1.663
ꢀ1.687
ꢀ1.650
ꢀ2.041
ꢀ1.959
ꢀ2.271
ꢀ1.783
ꢀ2.251
ꢀ1.978
ꢀ1.696
ꢀ1.511
ꢀ2.189
ꢀ1.937
ꢀ1.777
ꢀ1.734
ꢀ1.996
ꢀ2.299
ꢀ1.913
ꢀ2.321
ꢀ1.638
ꢀ1.22
Training
Training
Training
Training
Training
Test
ꢀ1.495
ꢀ1.953
ꢀ1.991
ꢀ1.639
ꢀ1.900
ꢀ1.403
ꢀ0.884
ꢀ1.292
ꢀ1.682
ꢀ1.865
ꢀ2.002
ꢀ1.909
ꢀ2.666
ꢀ1.753
ꢀ2.281
ꢀ1.689
ꢀ1.463
ꢀ1.599
ꢀ1.902
ꢀ1.931
ꢀ1.573
ꢀ1.881
ꢀ2.045
ꢀ2.175
ꢀ1.908
Test
Training
Training
Training
Test
Training
Training
Training
Training
Test
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
DMSO
2
3
4
7
9
10
12
13
21
50 µM
Fig. 4. Anti-invasive activities of PMHs against PC-3 cells using CultrexÔ assay kit.
Error bars indicate the SD of n ¼ 3/concentration.
Training
Training
Test
Training
Training
Training
Training
Training
Training
Training
Training
Training
the predictivity of the generated QSAR models. The molecules were
divided into training and test sets. Selection of the training set and
the test set compounds was done by considering the fact that test
set compounds represent a range of biological activity similar to
that of the training set. Thus, the test set was a true representative
of the training set. This was achieved by arbitrarily setting aside five
compounds as a test set with a regularly distributed biological data.
The mean (standard deviation) of the activity of the training and
test sets was 77.92 (50.12) and 85.02 (63.93), respectively. Table 2
shows the actual and predicted activity for both training and test
sets. Twenty-three hydantoins were used to build these QSAR
models. The IC50 (concentration causing 50% inhibitory effect on
a
pIC50 (-logIC50) values were experimentally tested in the wound-healing assay
in PC-3 cells.
b
Predicted pIC50 from the QSAR model 3.
less important ones from the analysis to improve the predictive
power of the model. The variable importance projection (VIP) plot
is a measure of the relative importance of variables included in the
analysis with respect to their contribution to the activity. Descrip-
tors with a VIP value greater than 1 were considered to have above-
average influence on the activity [31].
the PC-3 migration) values were converted to pIC50 (ꢀlog IC50
)
values and used as dependent variables in the analysis.
PLS analysis of all of the compounds in the training set and using
all the 14 descriptors resulted in a QSAR model with a moderate
cross-validated correlation coefficient (q2) value of 0.474 and
conventional correlation coefficient (r2) value of 0.849 (model 1,
Table 3). Earlier reports stated that models with q2 greater than 0.5
were considered to have predictive capability better than chance,
and with higher q2, higher reliability could be attained [31]. Hence,
the model was further refined to improve q2 and in turn its
predictive power. On the basis of the relative importance, out of 14
physicochemical descriptors, some might not significantly
contribute to the activity, and hence, it was decided to exclude the
Inspection of the VIP plot showed little contribution from many
descriptors. HOMO, LUMO, IP, HBA, PSA, DM, and 13C chemical shift
difference were found to be less important because their VIP plot
values were less than 1 (Fig. 6). The exclusion of these descriptors
improved the model and accounted for 95% and predicted 76% of
the anti-migratory activity of the methylene hydantoins (model 2,
Table 3). Examination of the residuals from the LOO cross-validated
predictions indicated that compound 27 might be an outlier.
Omission of this compound resulted in an increase in the q2 value to
0.803 for the remaining 22 compounds (model 3, Table 3). The
outlier status of 27 could stem from the presence of an electron-
withdrawing bulky iodine atom attached to the pyrrole ring. One
would expect a higher potency for 27 than it exhibited based on the
significant importance of MV to the activity. This analysis predicted
the activities of the compounds of test set with predictive corre-
lation coefficient (r2Pred) of 0.574. Fig. 7 shows the plot of the pre-
dicted versus observed activities of the training and test sets.
120
100
80
60
40
20
0
Table 3
Summary of PLS models developed for PMHs by the QSAR analysis.
Model
description
No. of
q2
r2
nb
No. of
descriptors
RMSEEc
componentsa
Model 1
Model 2
Model 3
4
3
3
0.474
0.757
0.803
0.849
0.954
0.982
23
23
22
14
7
7
0.052
0.008
0.005
DMSO
2
3
4
7
9
10
12
13
21
200 µM
a
Number of significant components obtained in the PLS analysis.
Number of compounds used in the analysis.
Root-mean-square error of estimation in the analysis (predicted vs observed).
b
c
Fig. 5. Cytotoxic activity of PMHs at 200
indicate the SD of n ¼ 3/concentration.
m
M against PC-3 using MTT assay. . Error bars

5402
M.A. Khanfar, K.A. El Sayed / European Journal of Medicinal Chemistry 45 (2010) 5397e5405
-3
-2.5
-2
-1.5
-1
-0.5
-0.5
-1
-1.5
-2
-2.5
Actual pIC50
Fig. 6. VIP Plot from PLS analysis of data in model 1 (Table 3).
Fig. 7. Actual versus calculated pIC50 of PMHs as predicted from model 3 (-, training
set, :, test set).
A careful inspection of VIP plot and descriptor’s correlation
coefficients in model 3 revealed the importance of both size and
lipophilic parameters. The descriptors were found in the order of
increasing importance: MA < MV < BC < clog P. The MV, BC, and
clog P were directly related to the activity, while MA was inversely
related (Eq. (3) in the Supporting Information). All other electronic
parameters such as HOMO, LUMO, DM were less contributing
according to model 1 and VIP plot. Based on the VIP and model 3
(Fig. 6 and Supporting Information Eq. (3)), the most influential two
descriptors were found to be clog P and MV. This is expected
because many of the active methylene hydantoins have lipophilic
and bulky substituents especially at the para-position of the ben-
zylidene moiety (e.g., hydantoins 3, 7, 9, and 10). These findings are
strongly correlated with the CoMFA model, where the steric
contribution to activity variation was 70.6% [27].
performed on a Shimadzu HPLC system (Columbia, MD) using
Agilent 5
m
C18 Column (150 ꢃ 4.6 mm id; Agilent, Santa Clara, CA),
and isocratic elution (CH3CNeH2O, 75:25) with UV detection set at
305 and 220 nm to verify the purity of each tested compound. A
purity of >98% has been established for all compounds. TLC anal-
yses were carried out on precoated Si gel 60 F254 500
using MeOH-CH2Cl2 (2:8) as a developing solvent.
mm TLC plates,
4.2. Preparation of phenylmethylene hydantoins
Hydantoin (0.1 g) was dissolved in 1 mL H2O while heating at
70 ꢄC in oil bath with continuous stirring (Scheme 1) [9]. The pH
was adjusted to 7.0 using saturated NaHCO3 solution after complete
dissolution. The temperature was then raised to 90 ꢄC after the
addition of 0.09 mL ethanolamine. Equimolar quantity of the cor-
responding aldehyde solution in 1e2 mL EtOH was then added
drop-wise with continuous stirring. The reaction was kept under
reflux for approximately 5e8 h. The reaction was monitored by TLC
every hour till a yellow or white precipitate was formed. After
complete depletion of the starting aldehyde, the mixture was
cooled and the precipitate was filtered and washed with EtOHeH2O
(1:5) before recrystallization from EtOH.
3. Conclusions
In this study, we present a successful example of employing
CoMFA-based virtual screening to identify new PMH derivatives
with enhanced anti-migratory and anti-invasive activities. A set of
15 PMHs were selected from the virtual 635 compounds for bio-
logical testing. New PMHs with manifold increase in the activity
were identified with modest efforts and expenses by screening only
15 PMHs, highlighting an important feature of the virtual screening
approach. QSAR study using multivariate analysis gave a significant
PLS model that could account for 98% and predict 80% of their anti-
migratory activity. The model demonstrated the significance
contribution of the size and lipophilicity on the activity of methy-
lene hydantoins. PMH is a novel anti-metastatic lead class that can
be easily, regioselectively, and cost-effectively synthesized. These
advantages bode well for the emergence of future PMHs as
potential therapeutics for the control and prevention of metastatic
prostate cancer.
4.2.1. (Z)-5-(3-methoxy-4-(2-oxo-2-(pyrrolidin-1-yl)ethoxy)
benzylidene)imidazolidine-2,4-dione (5)
Compound 5 was prepared according to the abovementioned
procedure to afford 0.244 g, brown powder (70.7% yield). 1H NMR
(DMSO-d6, 400 MHz)
d 1.76 (m, 2H), 1.88 (m, 2H), 3.31 (m, 2H),
3.46 (m, 2H), 3.85 (s, 3H), 4.75 (s, 3H), 6.39 (s, 1H), 6.85 (d, 1H,
J ¼ 9.2), 7.12 (d, 1H, J ¼ 9.2), 7.14 (s, 1H), 10.81 (brs, 2H). 13C NMR
(DMSO-d6, 100 MHz)
d 24.1 (CH2), 26.7 (CH2), 45.3 (CH2), 46.2
(CH2), 56.3 (CH3), 67.1 (CH2), 109.7 (CH), 113.3 (CH), 113.7 (CH),
123.3 (CH), 126.6 (qC), 126.8 (qC), 148,7 (qC), 149.4 (qC), 156.4 (qC),
166.0 (qC), 166.2 (qC); HREIMS m/z 344.1263 [MꢀH]þ (calcd for
C17H18N3O5, 344.1252).
4. Experimental section
4.1. General experimental procedures
4.2.2. (Z)-5-(2,4-dipropoxybenzylidene)imidazolidine-2,4-dione (6)
Nearly 0.273 g of 5 was prepared as a brown powder (89.8%
The 1H and 13C NMR spectra were recorded in d6-DMSO, using
TMS as an internal standard, on a JEOL Eclipse NMR spectrometer
(Tokyo, Japan) operating at 400 MHz for 1H and 100 MHz for 13C.
The HREIMS experiments were conducted in Louisiana State
University on a 6200-TOF LCMS (Agilent, Santa Clara, CA) equipped
with a multimode source (mixed source that can ionize the samples
alternatively by ESI and APCI). Analytical HPLC analyses were
yield). 1H NMR (DMSO-d6, 400 MHz)
d 0.98 (m, 6H), 2.51 (m, 4H),
3.97 (m, 4H), 6.53 (dd, 1H, J ¼ 2.2, 8.8), 6.57 (d, 2H, J ¼ 2.2), 6.66 (s,
1H), 7.54 (d, 1H, J ¼ 8.8), 10.21 (brs, 1H), 11.11 (brs, 1H). 13C NMR
(DMSO-d6, 100 MHz) d 11.0 (CH3), 11.1 (CH3), 22.5 (CH2), 22.6 (CH2),
69.7 (CH2), 70.1 (CH2), 99.9 (CH), 103.7 (CH), 106.6 (CH), 114.8
(qC), 126.5 (qC), 130.7 (CH), 156.0 (qC), 158.6 (qC), 161.1 (qC), 166.2
(qC). HREIMS m/z 303.1355 [MꢀH]þ (calcd for C16H19N2O4, 303.1345).

M.A. Khanfar, K.A. El Sayed / European Journal of Medicinal Chemistry 45 (2010) 5397e5405
5403
4.2.3. (Z)-5-(3-methoxy-4-(2-oxo-2-phenylethoxy)benzylidene)
imidazolidine-2,4-dione (7)
4.2.9. (Z)-5-(3-chloro-2,6-dimethoxy-4-propoxybenzylidene)
imidazolidine-2,4-dione (16)
Compound 16 was prepared to afford 0.172 g, orange powder
(50.6% yield). 1H NMR (DMSO-d6, 400 MHz)
Compound 7 was prepared to afford 0.202 g, yellow powder
(57% yield). 1H NMR (DMSO-d6, 400 MHz)
d
3.87 (s, 3H), 5.60 (s,
d
0.97 (t, 3H, J ¼ 7.2),
2H), 6.37 (s, 1H), 6.86 (d, 1H, J ¼ 8.8), 7.09 (dd, 1H, J ¼ 1.8, 8.4), 7.13
(d, 1H, J ¼ 2.2), 7.51 (m, 1H), 7.56 (dd, 1H, J ¼ 7.7, 7.7), 7.67 (d, 1H,
J ¼ 7.3), 7.80 (m, 1H), 7.99 (d, 1H, J ¼ 7.0),. 13C NMR (DMSO-d6,
1.73 (m, 2H), 3.88 (s, 3H), 4.01 (t, 2H, J ¼ 6.6), 6.16 (s, 1H), 6.82 (s,
1H), 10.65 (brs, 1H), 11.23 (brs, 1H). 13C NMR (DMSO-d6, 100 MHz)
d
10.5 (CH3), 22.8 (CH2), 55.9 (CH3), 60.1 (CH3), 74.3 (CH2), 96.3 (CH),
100 MHz)
d
56.3 (CH3), 70.9 (CH2), 109.8 (CH), 112.8 (CH), 113.0
100.0 (qC), 102.2 (qC), 122.1 (CH), 124.8 (qC), 152.9 (qC), 154.4 (qC),
155.1 (qC), 157.2 (qC) 166.3 (qC); HREIMS m/z 341.1008 [MþH]þ
(calcd for C15H18ClN2O5, 341.0904).
(CH), 123.2 (CH), 126.7 (qC), 126.9 (qC), 128.4 (CH), 129.5 (CH),
134.5 (qC), 134.7 (CH), 148.4 (qC), 149.4 (qC), 156.5 (qC), 166.4 (qC),
195.1 (qC). HREIMS m/z 351.1000 [MꢀH]þ (calcd for C19H15N2O5,
351.0986).
4.2.10. (Z)-5-(4-tert-butoxybenzylidene)imidazolidine-2,4-dione
(17)
4.2.4. (Z)-5-(3-methoxy-4-(2-morpholino-2-oxoethoxy)
benzylidene)imidazolidine-2,4-dione (8)
Compound 17 was prepared to afford 0.227 g, white powder
(83.4% yield). 1H NMR (DMSO-d6, 400 MHz)
d
1.37 (s, 9H), 6.53 (s,
1H), 7.05 (d, 2H, J ¼ 8.4), 7.46 (d, 2H, J ¼ 8.4), 10.53 (brs, 1H), 11.03
(brs,1H). 13C NMR (DMSO-d6, 100 MHz)
27.9 (CH3), 78.9 (qC), 110.1
Compound 8 was prepared to afford 0.243 g, white powder
(66.7% yield). 1H NMR (DMSO-d6, 400 MHz)
d
3.45 (dd, 4H, J ¼ 4.2,
d
11.9), 3.57 (dd, 4H, J ¼ 4.2, 18.3), 3.85 (s, 3H), 4.87 (s, 2H), 6.38 (s,
(CH), 123.7 (CH), 127.7 (qC), 128.2 (qC), 129.9 (CH), 156.2 (qC), 166.1
1H), 6.88 (d, 1H, J ¼ 8.8), 7.14 (m, 1H), 7.15 (m, 1H), 10.82 (brs, 1H) .
(qC); HREIMS m/z 259.1084 [MꢀH]þ (calcd for C14H15N2O3, 259.1088).
13C NMR (DMSO-d6, 100 MHz)
d 40.3 (CH2), 56.2 (CH3), 60.2 (CH2),
68.2 (CH2), 110.5 (CH), 113.5 (CH), 117.5 (CH), 126.2 (CH), 126.3 (qC),
133.4 (qC), 148.3 (qC), 148.7 (qC), 153.2 (qC), 166.4 (qC) 167.6 (qC).
HREIMS m/z 360.1206 [MꢀH]þ (calcd for C17H18N3O6, 360.1201).
4.2.11. (Z)-5-((4-methyl-1H-imidazol-5-yl)methylene)
imidazolidine-2,4-dione (19)
Compound 19 was prepared to afford 0.118 g, yellow powder
(61.5% yield). 1H NMR (DMSO-d6, 400 MHz)
d
2.29 (s, 3H), 6.39
(s, 1H), 7.76 (s, 1H), 9.59 (brs, 1H), 11.03 (brs, 1H). 13C NMR (DMSO-
d6, 100 MHz) 9.6 (CH3), 101.0 (CH), 126.0 (qC), 129.6 (qC), 132.7
4.2.5. (Z)-5-(3-bromo-5-methoxy-4-propoxybenzylidene)
imidazolidine-2,4-dione (10)
d
PMH 10 was prepared to afford 0.164 g, yellow powder (46.3%
(qC), 134.6 (CH), 154.2 (qC), 165.4 (qC); HREIMS m/z 193.0729
yield). 1H NMR (DMSO-d6, 400 MHz)
d
0.99 (t, 3H, J ¼ 7.3), 1.70 (m,
2H), 3.88 (s, 3H), 6.35 (s,1H), 7.20 (s,1H), 7.15 (s,1H),10.99 (brs,1H) .
13C NMR (DMSO-d6, 100 MHz)
11.1 (CH3), 23.6 (CH2), 56.8 (CH3),
[MþH]þ (calcd for C8H9N4O2, 193.0720).
d
4.2.12. (Z)-5-((3H-pyrrolo[2,3-b]pyridin-3-yl)methylene)
imidazolidine-2,4-dione (20)
74.9 (CH2), 107.5 (CH), 114.0 (CH), 117.9 (qC), 125.5 (CH), 128.6 (qC),
130.7 (qC), 145.5 (qC), 153.9 (qC), 156.4 (qC), 166.0 (qC). HREIMS m/z
355.0252 [MþH]þ (calcd for C14H16BrN2O4, 355.0293).
Compound 20 was prepared to afford 0.135 g, yellow powder
(59.2% yield). 1H NMR (DMSO-d6, 400 MHz)
d 6.73 (s, 1H), 7.16 (dd,
1H, J ¼ 4.8, 8.1), 8.24 (s, 1H), 8.25 (m, 1H), 8.28 (d, 1H, J ¼ 4.8), 8.31
4.2.6. (Z)-5-(4-(isopentyloxy)benzylidene)imidazolidine-2,4-dione
(11)
(s,1H), 10.24 (brs,1H),11.07 (brs, 1H). 13C NMR (DMSO-d6, 100 MHz)
d
79.7 (qC), 101.5 (CH), 107.8 (qC), 117.0 (CH), 119.7 (qC), 125.0 (qC),
Compound 11 was prepared to afford 0.240 g, white powder
127.5 (CH), 144.3 (CH), 148.9 (qC), 155.9 (qC), 165.9 (qC); HREIMS m/
(87.6% yield). 1H NMR (DMSO-d6, 400 MHz)
d
0.94 (d, 6H, J ¼ 6.6),
z 252.0750 [M]þ (calcd for C12H11N3O2Na, 252.0743).
1.61 (m, 2H), 1.78 (m, 1H), 4.03 (t, 2H, J ¼ 6.6), 6.38 (s, 1H), 6.95 (d,
2H, J ¼ 8.8), 7.57 (d, 2H, J ¼ 8.8), 10.50 (brs, 1H), 11.18 (brs, 1H). 13C
4.2.13. (Z)-5-((6-(thiophen-2-yl)pyridin-2-yl)methylene)
imidazolidine-2,4-dione (21)
NMR (DMSO-d6, 100 MHz)
d 24.1 (CH3), 24.9 (CH), 38.2 (CH2), 65.0
(CH2),108.7 (CH),126.1 (CH),127.8 (qC),129.9 (qC), 130.4 (CH), 155.3
(qC),156.9 (qC),166.0 (qC). HREIMS m/z 273.1244 [MꢀH]þ (calcd for
C15H17N2O3, 273.1239).
Compound 21 was prepared to afford 0.164 g, brown powder
(53.9% yield). 1H NMR (DMSO-d6, 400 MHz)
d 6.50 (s, 1H), 7.21
(d, 1H, J ¼ 3.6), 7.49 (dd, 1H, J ¼ 3.6, 4.8), 7.72 (d, 1H, J ¼ 5.2), 7.84 (s,
3H), 8.31 (s, 1H), 9.99 (brs, 1H), 11.02 (brs, 1H). 13C NMR (DMSO-d6,
4.2.7. (Z)-5-(2-methoxy-4-propoxybenzylidene)imidazolidine-2,4-
dione (14)
100 MHz) d 104.8 (CH), 118.1 (CH), 124.9 (CH), 126.7 (CH), 129.0
(CH), 129.5 (CH), 132.6 (qC), 138.9 (CH), 144.3 (qC), 151.8 (qC), 153.9
(qC), 154.9 (qC), 165.6 (qC); HREIMS m/z 270.0336 [MꢀH]þ (calcd
for C13H8N3O2S, 270.0343).
Compound 14 was prepared to afford 0.211 g, yellow powder
(76.5% yield). 1H NMR (DMSO-d6, 400 MHz)
d
0.99 (t, 3H, J ¼ 7.0),
1.73 (m, 2H), 3.87 (s, 3H), 4.04 (t, 2H, J ¼ 6.4), 6.29 (s, 1H), 6.55 (m,
1H), 6.56 (m, 1H), 8.02 (d, 1H, J ¼ 8.8), 10.76 (brs, 1H), 11.17 (brs, 1H).
4.2.14. (Z)-5-(benzo[d][1,3]dioxol-4-ylmethylene)imidazolidine-
2,4-dione (22)
13C NMR (DMSO-d6, 100 MHz)
d 10.2 (CH3), 22.9 (CH2), 56.8 (CH3),
70.3 (CH2), 103.2 (CH), 106.3 (CH), 107.7 (qC), 123.2 (CH), 124.7 (qC),
131.7 (CH),144.2 (qC),153.8 (qC),156.2 (qC),166.1 (qC); HREIMS m/z
277.1109 [MþH]þ (calcd for C14H17N2O4, 277.1188).
Compound 22 was prepared to afford 1.67 g, white powder (72%
yield). 1H NMR (DMSO-d6, 400 MHz)
d
6.12 (s, 2H), 6.31 (s, 1H), 6.87
(dd, 1H, J ¼ 1.5, 7.7), 6.90 (dd, 1H, J ¼ 8.0, 7.7), 8.12 (dd, 1H, J ¼ 8.0,
1.5) 10.22 (s, 1H), 11.30 (s, 1H). 13C NMR (DMSO-d6, 100 MHz)
99.8
d
4.2.8. (Z)-5-(4-tert-butoxy-3-chloro-2-methoxybenzylidene)
imidazolidine-2,4-dione (15)
(CH2),114.7 (CH),116.2 (qC),120.8 (CH),121.9 (CH),122.4 (CH),124.7
(qC), 147.2 (qC), 147.1 (qC), 158.7 (qC), 162.5 (qC); HREIMS m/z
255.0380 [MþNa]þ (calcd for C11H8N2O4Na, 255.0376).
Compound 15 was prepared to afford 0.131 g, white powder
(40.4% yield). 1H NMR (DMSO-d6, 400 MHz)
d
1.41 (s, 9H), 3.83 (s,
3H), 6.72 (s, 1H), 6.84 (d, 1H, J ¼ 8.8), 7.89 (d, 1H, J ¼ 8.8), 10.55 (brs,
1H), 11.06 (brs, 1H). 13C NMR (DMSO-d6, 100 MHz)
27.7 (CH3), 59.8
4.2.15. (Z)-5-(indolin-7-ylmethylene)imidazolidine-2,4-dione (23)
Compound 23 was prepared to afford 0.102 g, white powder
d
(CH3), 77.8 (qC), 104.4 (CH), 108.0 (qC), 108.2 (qC), 122.5 (CH), 124.1
(qC), 128.7 (CH), 150.2 (qC), 154.8 (qC), 156.2 (qC), 166.0 (qC);
HREIMS m/z 325.0889 [MþH]þ (calcd for C15H18ClN2O4, 325.0955).
(44.5% yield). 1H NMR (DMSO-d6, 400 MHz)
d 2.50 (m, 2H), 4.39 (t,
2H, J ¼ 8.8), 6.59 (s, 1H), 7.13 (dd, 1H, J ¼ 7.7, 7.3), 7.25 (d, 1H, J ¼ 7.3),
8.43 (s, 1H), 8.52 (s, 1H), 10.72 (brs, 1H), 11.02 (brs, 1H). 13C NMR

5404
M.A. Khanfar, K.A. El Sayed / European Journal of Medicinal Chemistry 45 (2010) 5397e5405
(DMSO-d6, 100 MHz)
d
27.4 (CH2), 47.9 (CH2), 114.0 (CH), 117.4 (qC),
cells in serum free RPMI medium were added per well to the top
chamber containing the tested PMH (50 M). About 150 L of RPMI
medium was then added to the lower chamber containing 10% FBS
and penicillin/streptomycin and using fibronectin (1 L/mL) and N-
122.3 (CH), 122.9 (CH), 124.1(CH), 130.8 (qC), 132.3 (qC), 135.2 (qC),
156.5 (qC), 156.9 (qC); HREIMS m/z 252.0750 [MþNa]þ (calcd for
C12H11N3O2Na, 252.0743).
m
m
m
formyl-met-leu-phe (10 nM) as chemoattractants. Cells were
allowed to migrate to lower chamber at 37 ꢄC in CO2 incubator. After
24 h, top and bottom chambers were aspirated and washed with
4.2.16. (Z)-5-((6-methoxypyridin-2-yl)methylene)imidazolidine-
2,4-dione (24)
Compound 24 was prepared to afford 0.161 g, yellow powder
washing buffer supplemented with the kit. About 100 mL of Cell
(73.5% yield). 1H NMR (DMSO-d6, 400 MHz)
d
3.99 (s, 3H), 6.44 (s,
Dissociation Solution/Calcein-AM solution was added to the bottom
chamber and incubated at 37 ꢄC in CO2 incubator for 1 h. The cells
internalize calcein-AM, and the intracellular esterases cleaved the
AM moiety to generate free calcein. Fluorescence of the samples was
determined at 485 nm excitation, 520 nm emission, using ELISA
plate reader (BioTek, VT). The number of cells that have invaded
through the BME coat was calculated using a standard curve.
1H), 6.75 (d, 1H, J ¼ 8.0), 7.22 (d, 1H, J ¼ 7.3), 7.73 (dd, 1H, J ¼ 7.3,
8.0), 9.62 (brs, 1H), 11.47 (brs, 1H). 13C NMR (DMSO-d6, 100 MHz)
d
53.8 (CH3), 106.1 (CH), 110.7 (CH), 119.8 (CH), 131.5 (qC), 140.5
(CH), 151.7(qC), 154.9 (qC), 163.7 (qC), 165.6 (qC); HREIMS m/z
218.0563 [MꢀH]þ (calcd for C10H8N3O3, 218.0566).
4.2.17. (Z)-5-((1H-imidazol-5-yl)methylene)imidazolidine-2,4-
dione (25)
4.6. Wound-healing assay
Compound 25 was prepared to afford 0.091 g, yellow powder
(51.1% yield). 1H NMR (DMSO-d6, 400 MHz)
d
6.41 (s, 1H), 7.59
(s, 1H), 7.86 (s, 1H), 9.68 (brs, 1H), 11.17 (brs, 1H). 13C NMR (DMSO-
d6, 100 MHz) 102.2 (CH), 123.1 (qC), 131.1 (qC), 135.5 (CH), 136.7
Cells were plated onto sterile 24-well plates and allowed to
recover for a confluent cell monolayer formed in each well (>90%
confluence). Wounds were then inflicted to each cell monolayer
d
(qC), 154.2 (qC), 165.7 (qC), 165.6 (qC); HREIMS m/z 179.0551
[MþH]þ (calcd for C7H6N4O2, 179.0579).
using a sterile 200 mL pipette tip. Media were removed, cell
monolayers were washed once with PBS, and then fresh media
containing test compounds were added to each well. Test
compounds were prepared in DMSO at different concentrations
and added to the plates, each in triplicate using DMSO as a vehicle
control. The incubation was carried out for 24 h under serum-
starved conditions, after which media was removed and cells were
fixed and stained using Diff-Quick staining (Dade Behring Diag-
nostics). The number of cells migrated on the scratched wound
were counted under the microscope in three or more randomly
selected fields (magnification: 400ꢃ). Final results are expressed as
mean ꢂ SEM per 400ꢃ field. All treatments, including the controls,
were documented photographically. IC50 for each compound was
calculated using nonlinear regression (curve fit) of log concentra-
tion versus the number of cells/well implemented using GraphPad
Prism 5.0 (GraphPad Software, La Jolla, CA).
4.2.18. (Z)-5-((4-iodo-1H-pyrrol-2-yl)methylene)imidazolidine-
2,4-dione (27)
Compound 27 was prepared to afford 0.253 g, white powder
(83.8% yield). 1H NMR (DMSO-d6, 400 MHz)
d
6.46 (s, 1H), 6.69 (s,
1H), 748 (s, 1H), 10.45 (bs, 1H), 10.92 (bs, 1H). 13C NMR (DMSO-d6,
100 MHz) 108.8 (CH), 109.2(CH), 111.2 (qC), 119.0 (CH), 129.2.5
d
(qC), 131.7(qC), 154.7 (qC), 165.2 (qC); HREIMS m/Hz 303.95710
[MþH]þ (calcd for C8H6IN3O2, 303.9583).
4.3. Cell culture
Prostate cancer cell line PC-3 was purchased from ATCC (Man-
assas, VA). The cell line was grown in 10% fetal bovine serum (FBS)
and RPMI 1640 (GIBCO-Invitrogen, NY) supplemented with
2 mmol/L glutamine, 100
mg/mL penicillin G, and 100
mg/mL
4.7. Determination of the 13C NMR chemical Shift at the
(C5)
b-carbon
streptomycin, at 37 ꢄC under 5% CO2.
4.4. MTT proliferation assay
The 13C NMR chemical shift of the
the electronic influence of the substituents at the phenyl ring. The
difference in the chemical shift Dd was given by (dH dR), where dR
is the carbon chemical shift of the substituted phenylmethylene
hydantoin and dH is that of the unsubstituted compound. The Dd
value is known to be sensitive to the electronic influence of the
substituents on the aryl moieties [34,35].
b-carbon was used to identify
The growth of PC-3 cancer cell lines was measured using MTT kit
(TACSÔ, TREVIGENÒ, Inc.) [32,33]. Exponentially growing cells
were plated in a 96 well plate at a density of 8 ꢃ 103 cells per well,
and allowed to attach for 24 h. Complete growth medium was then
ꢀ
replaced with 100
trogen, NY) containing 200
culture continued at 37 ꢄC under 5% CO2. After 72 h of culture, the
cells were treated with MTT solution (10
L/well) at 37 ꢄC for 4 h.
The color reaction was stopped by the addition of solubilization/
stop solution (100
L/well), and the incubation at 37 ꢄC continued
mL of RPMI of serum free medium (GIBCO-Invi-
m
M of the specific tested PMH and
4.8. Molecular modeling
m
Three-dimensional structure building and all modeling were
performed using the SYBYL program package [30], version 8.0,
installed on DELL desktop workstations equipped with a dual
2.0 GHz IntelÒ XeonÒ processor running the Red Hat Enterprise
Linux (version 5) operating system. Energy minimizations were
performed using the Tripos force field with a distance-dependent
dielectric and the Powell conjugate gradient algorithm with
a convergence criterion of 0.01 kcal/(mol A) [36]. Partial atomic
charges were calculated using the semiempirical program MOPAC
6.0 and applying the AM1 [37] before calculating their physico-
chemical properties. Size, electronic, and lipophilic properties were
theoretically calculated using the SYBYL 8.0 spreadsheet autofill
module. The multivariate analyses were performed with JMP 8.0
(SAS Institute Inc.) using the default settings.
m
to completely dissolve the formazan product. Absorbance of the
samples was determined at 570 nm with an ELISA plate reader
(BioTek, VT). The number of cells per well was calculated against
a standard curve prepared by plating various concentration of cells,
as determined by hemocytometer, at the start of each experiment.
4.5. CultrexÒ cell invasion assay
Anti-invasive activities were measured using Trevigen’s CultrexÒ
Cell Invasion Assay as previously described [28,29]. About 50
basement membrane extract (BME) coat was added per well. After
incubation for 4 h at 37 ꢄC in a 5% CO2, 50,000 cells/50
L of PC-3
mL of
m

M.A. Khanfar, K.A. El Sayed / European Journal of Medicinal Chemistry 45 (2010) 5397e5405
5405
[16] V. Zuliani, C. Carmi, M. Rivara, M. Fantini, A. Lodola, F. Vacondio, F. Bordi,
P.V. Plazzi, A. Cavazzoni, M. Galetti, R.R. Alfieri, P.G. Petronini, M. Mor, Eur. J.
Med. Chem. 44 (2009) 3471e3479.
[17] C.S. Basappa, Ananda Kumar, S.N. Swamy, K. Sugahara, K.S. Rangappa, Bioorg.
Med. Chem. 17 (2009) 4928e4934.
Appendix. Supplementary material
Supplementary data associated with this article can be found in
the online version, at doi:10.1016/j.ejmech.2010.08.066.
[18] S. Chen, N.J.S. Huby, N. Kong, J.A. Moliterni, O.J. Morales, WO 2009021887 A1
20090219, PCT Int. Appl. (2009).
References
[19] C.V. Kavitha, M. Nambiar, C.S. Ananda Kumar, B. Choudhary, K. Muniyappa,
K.S. Rangappa, S.C. Raghavan, Biochem. Pharm. 77 (2009) 348e363.
[20] M.E. Jung, C.L. Sawyers, S. Ouk, C. Tran, J. Wongvipat, WO 2007126765 A2
20071108, PCT Int. Appl. (2007).
[21] (a) M.A. Khanfar, B. AbuAsal, M. Mudit, A. Kaddoumi, K.A. El Sayed, Bioorg.
Med. Chem. 17 (2009) 6032e6039;
(b) K. Yang, Y. Tang, K.A. Iczkowski, Am. J. Transl. Res. 2(2010) 88e94.
[22] K.V. Sashidhara, K.N. White, P. Crews, J. Nat. Prod. 72 (2009) 588e603.
[23] K.H. Lee, J. Nat. Prod. 73 (2010) 500e516.
[24] D.J. Newman, G.M. Cragg, J. Nat. Prod. 70 (2007) 461e477.
[25] N. Fleshner, A.R. Zlotta, Cancer 110 (2007) 1889e1899.
[26] G.V. Shah, A. Muralidharan, S. Thomas, M. Gokulgandhi, M. Mudit, M. Khanfar,
K.A. El Sayed, Mol. Cancer Ther. 8 (2009) 509e520.
[27] M. Mudit, M. Khanfar, A. Muralidharan, S. Thomas, G.V. Shah, R.W.M. van
Soest, K.A. El Sayed, Bioorg. Med. Chem. 17 (2008) 1731e1738.
[28] K.P. Tamilarasan, G.K. Kolluru, M. Rajaram, M. Indhumathy, R. Saranya,
S. Chatterjee, BMC Cell Biol. 7 (2006) 17e30.
[29] P.R. Borghesani, J.M. Peyrin, R. Klein, J. Rubin, A.R. Carter, P.M. Schwartz,
A. Luster, G. Corfas, R.A. Segal, Development 6 (2002) 1435e1442.
[30] Tripos Associates, SYBYL Molecular Modeling Software, Version 8.0. Tripos
Associates, St. Louis, MO, 2007. http://www.tripos.com.
[31] S. Wold, C. Albano, W.J. Dunn III, U. Edlund, K. Esbensen, P. Geladi, S. Hellberg,
E. Johanson, W. Lindberg, M. Sjostrom, NATO Sci. Ser. C 138 (1984) 17e95.
[32] T. Mosmann, J. Immunol. Methods 65 (1983) 55e63.
[33] M.C. Alley, D.A. Scudiero, A. Monks, M.L. Hursey, M.J. Czerwinski, D.L. Fine,
B.J. Abbott, J.G. Mayo, R.H. Shoemaker, M.R. Boyd, Cancer Res. 48 (1988) 589e601.
[34] G.N. Wernly Chung, J.M. Mayer, A. Tsantili-Kakoulidou, B. Testa, Int. J. Pharm.
63 (1990) 129e134.
[1] A. Jemal, R. Rebecca Siegel, E. Ward, Y. Hao, J. Xu, M. Thun, Cancer J. Clin. 59
(2009) 225e249.
[2] I.M. Thompson, P.J. Goodman, C.M. Tangen, M.S. Lucia, G.J. Miller, L.G. Ford,
M.M. Lieber, R.D. Cespedes, J.N. Atkins, S.M. Lippman, S.M. Carlin, A. Ryan,
C.M. Szczepanek, J.J. Crowley, C.A. Coltman Jr., New Engl. J. Med. 349 (2003)
215e224.
[3] T. Makinen, T.L. Tammela, M. Hakama, U.H. Stenman, S. Rannikko, J. Aro,
H. Juusela, L. Maattanen, A. Auvinen, Clin. Cancer Res. 9 (2003) 2435e2439.
[4] L. Bubendorf, A. Schopfer, U. Wagner, G. Sauter, H. Moch, N. Willi, T.C. Gasser,
M.J. Mihatsch, Hum. Pathol. 31 (2000) 578e583.
[5] M. Bolla, L. Collette, L. Blank, P. Warde, J.B. Dubois, R.O. Mirimanoff, G. Storme,
J. Bernier, A. Kuten, C. Sternberg, J. Mattelaer, J. Lopez Torecilla, J.R. Pfeffer,
C. Lino Cutajar, A. Zurlo, M. Pierart, Lancet 360 (2002) 103e106.
[6] A.V. D’Amico, J. Manola, M. Loffredo, A.A. Renshaw, A. DellaCroce,
P.W. Kantoff, J. Am. Med. Assoc. 292 (2004) 821e827.
[7] D.R. Shepard, D. Raghavan, Nat. Rev. Clin. Oncol. 7 (2010) 13e21.
[8] J.C. Cheville, D. Tindall, C. Boelter, R. Jenkins, C.M. Lohse, V.S. Pankratz,
T.J. Sebo, B. Davis, M.L. Blute, Cancer 95 (2002) 1028e1036.
[9] J.C. Thenmozhiyal, P.T. Wong, W.K. Chui, J. Med. Chem. 47 (2004) 1527e1535.
[10] M. Meuesel, M. Gutschow, Org. Prep. & Procedures Int. 36 (2004) 391e443.
[11] B. Gabos, M. Munck Af Rosenschoeld, L. Ripa, WO 2009007747 A2 20090115,
PCT Int. Appl. (2009).
[12] L. Naohiro, S. Tadayoshi, K. Ikuo, Y. Katsuji, A. Yutaka, Y. Toshiaki, Jpn. Kokai
Tokkyo Koho (1987) JP 62029570 A 19870207.
[13] P. Shapiro, A.D. Mackerell, WO 2006105237 A2 20061005, PCT Int. Appl.
(2006).
[35] X. Rong, M.L. Go, Chem. Pharm. Bull. 45 (1997) 476e481.
[36] M. Clark, R.D. Cramer III, N. van Opdenbosch, Comput. Chem. 10 (1989)
982e1012.
[14] W. Koenig, G. Zoller M. Just, B. Jablonka, DE 4126277 A1 19930211, Ger. Offen.
(1993).
[15] A.H. Abadi, B.D. Gary, H.N. Tinsley, G.A. Piazza, M. Abdel-Halim, Eur. J. Med.
Chem. 45 (2010) 1278e1286.
[37] J.J. Stewart, J. Comput. Aided Mol. Des. 4 (1990) 1e105.