QSAR modeling, synthesis and bioassay of diverse leukemia RPMI-8226 cell line active agents

Article abstract of DOI:10.1016/j.ejmech.2010.08.033

A rigorous QSAR modeling procedure employing CODESSA PRO descriptors has been utilized for the prediction of more efficient anti-leukemia agents. Experimental data concerning the effect on leukemia RPMI-8226 cell line tumor growth of 34 compounds (treated

Full text of DOI:10.1016/j.ejmech.2010.08.033

European Journal of Medicinal Chemistry 45 (2010) 5183e5199
Contents lists available at ScienceDirect
European Journal of Medicinal Chemistry
journal homepage: http://www.elsevier.com/locate/ejmech
Original article
QSAR modeling, synthesis and bioassay of diverse leukemia RPMI-8226 cell line
active agents
,
a,
Alan R. Katritzkya , Adel S. Girgis b, Svetoslav Slavova, Srinivasa R. Tala a, Iva Stoyanova-Slavova a
*
a Center for Heterocyclic Compounds, University of Florida, Department of Chemistry, Gainesville, FL 32611-7200, United States
b Pesticide Chemistry Department, National Research Centre, Dokki, 12622 Cairo, Egypt
a r t i c l e i n f o
a b s t r a c t
Article history:
A rigorous QSAR modeling procedure employing CODESSA PRO descriptors has been utilized for the
Received 18 June 2010
Received in revised form
6 August 2010
Accepted 10 August 2010
Available online 19 August 2010
prediction of more efficient anti-leukemia agents. Experimental data concerning the effect on leukemia
RPMI-8226 cell line tumor growth of 34 compounds (treated at a dose of 10 mM) was related to their
chemical structures by a 4-descriptor QSAR model. Four bis(oxy)bis-urea and bis(sulfanediyl)bis-urea
derivatives (4a, 4b, 8, 11a) predicted as active by this model, together with 11b predicted to be of low
activity, were synthesized and screened for anti-tumor activity utilizing 55 different tumor cell lines.
Compounds 8 and 11a showed anti-tumor properties against most of the adopted cell lines with growth
inhibition exceeding 50%. The highly promising preliminary anti-tumor properties of compounds 8 and
11a, were screened at serial dilutions (10ꢀ4e10ꢀ8 mM) for determination of their GI50 and TGI against the
Keywords:
Leukemia RPMI-8226 cell line
Anti-tumor agents
Bis(oxy)bis-urea derivatives
Bis(sulfanediyl)bis-urea derivatives
QSAR
screened human tumor cell lines. Compound 11a (GI50 ¼ 1.55, TGI ¼ 8.68
mM) is more effective than
M) against the target leukemia RPMI-8226 cell line. Compound
compound 8 (GI50 ¼ 58.30, TGI ¼ >100
m
11a also exhibits highly pronounced anti-tumor properties against NCI-H226, NCI-H23 (non-small cell
lung cancer), COLO 205 (colon cancer), SNB-75 (CNS cancer), OVCAR-3, SK-OV-3 (ovarian cancer), A498
(renal cancer) MDA-MB-231/ATCC and MDA-MB-468 (breast cancer) cell lines (GI50 ¼ 1.95, 1.61, 1.38, 1.56,
CODESSA PRO
1.30, 1.98, 1.18, 1.85, 1.08, TGI ¼ 8.35, 6.01, 2.67, 8.59, 4.01, 7.01, 5.62, 6.38, 5.63
mM, respectively). Thus 11a
could be a suitable lead towards the design of broad spectrum anti-tumor active agents targeting various
human tumor cell lines.
Ó 2010 Elsevier Masson SAS. All rights reserved.
1. Introduction
therapeutic agents with novel modes of action and fewer side
effects are still needed [6].
Cancer is a leading cause of death world wide. Despite techno-
logical advances, the index of cancer cure remains low and its
treatment is a challenge. The childhood cancer Leukemia affects
a significant segment of the population, with 26% of all cases and
30% mortality [1]. While the incidence of leukemia has remained
relatively unchanged, its treatment has made progress and since
1950, mortality rates for childhood cancer declined by more than
50%. In 1960, few could survive acute lymphoid “childhood
leukemia” while now 86% of children and teens so diagnosed are
still alive 5 years later [2,3]. The overall five-year (1995e2001)
survival rate for acute myeloid leukemia has also increased from
38% to 65% [4]. Despite the success of clinical trials, new agents and
treatments have limitations related to their side effects and the
development of acquired drug resistance [5]. Active new
A wide range of QSAR1 tools for modeling has been developed
during the past three decades; our group has utilized CODESSA
PRO2 [7] which calculates numerous quantitative descriptors using
information extracted from the molecular structure. CODESSA PRO
has successfully correlated and predicted diverse physicochemical
and biological properties [8e11]. The present study reports the
QSAR modeling of biological data obtained at the National Cancer
Institute, Bethesda, USA using the leukemia RPMI-8226 cell line
according to standard procedures [12e17a]. Based on the QSAR
study we predicted new molecules expected to be bio-active, and
synthesized representative examples. The training dataset adopted
herein the present study includes many types of chemical struc-
tures (Table 1), thus diversity of pharmacological modes of action is
expected. Many of the compounds in the training dataset, espe-
cially the nicotinamide derivatives (compounds 14e24, Table 1) are
1
* Corresponding author. Tel.: þ1 352 392 0554; fax: þ1 352 392 9199.
E-mail address: katritzky@chem.ufl.edu (A.R. Katritzky).
Quantitative StructureeActivity Relationship.
COmprehensive DEscriptors for Structural and Statistical Analysis.
2
0223-5234/$ e see front matter Ó 2010 Elsevier Masson SAS. All rights reserved.
doi:10.1016/j.ejmech.2010.08.033

5184
A.R. Katritzky et al. / European Journal of Medicinal Chemistry 45 (2010) 5183e5199
closely structurally related to potent inducers of apoptosis agents
5. Results and discussion
such as N-(4-methoxy-2-nitrophenyl)pyridine-3-carboxamide
(EC50 ¼ 1.6, 0.88, 2.9
m
M in the caspase activation assay in human
The original dataset of 34 compounds comprises 5 structurally
distinct subsets containing respectively ten (1e10), three (11e13),
eleven (14e24), two (27e28) and six (29e34) compounds, together
with two singletons 25 and 26. Theheterogeneityof the datasetof 34
compounds and the need to select a robust test set representative of
the general population, led us to the following procedure: i) for each
of the three bigger subsets (1e10; 14e24; 29e34) the average anti-
tumor activity was found and the member of each subset charac-
terized with an activity closest to the average (i.e., compounds 1, 17
and 29 respectively) was moved to the test set; ii) to estimate the
ability of the model to predict the activity of structurally diverse
compounds the small subset of three compounds (11e13) and both
the singletons (25 and 26) were also moved to the test set. Thus,
a training set of 26 compounds and an external test set of 8
compounds (approximately 1/4 of the total) were formed,
complying with standard QSAR recommendations [22b].
breast cancer T47D, ZR75-1 and colorectal DLD-1 cell lines,
respectively) and 6-methyl-N-(4-ethoxy-2-nitrophenyl)pyridine-
3-carboxamide (EC50 ¼ 0.082, 0.052, 0.11
mM in the caspase acti-
vation assay in human breast cancer T47D, ZR75-1 and colorectal
DLD-1 cell lines, respectively) [17b]. However, the present work is
focused on the preparation and anti-tumor bioactivity screening of
compounds predicted from a QSAR model and presented as
a medicinal chemistry study targeting the discovery of anti-tumor
active hits.
2. Dataset
Percentage growth of the tumor “leukemia RBMI-8226 cell line”
treated at a dose of 10 mM with the 34 compounds tested was
measured relative to control experiments (Table 1) [12e14,18].
For the next stage, the BMLR stepwise regression algorithm,
using subsets of noncollinear descriptors, generated the best
n-parameter regression equations (n ꢁ 2), based on the highest R2
and F values obtained in the process of calculations. Since, the
correlations are not statistically reliable if the variables are mutu-
ally intercorrelated, the BMLR rejects the simultaneous use of
descriptors with intercorrelation coefficients larger than a certain
user defined threshold value (in this case we took R2 ¼ 0.5) (Fig. 2).
QSAR models with up to 6 descriptors were generated (the
maximum number allowed according to the 5:1 rule of thumb e
the ratio between the data points and the number of descriptors).
We considered the statistical parameters including the square of
the correlation coefficient (R2), the cross-validated correlation
coefficient (R2cv), the Fisher criterion (F) and the predictive R2 (Rt2est).
We then relied on Occam’s parsimony principle to select a model
with 4 descriptors as optimal (see Table 2).
3. Methodology
Compound structures were drawn using ChemDraw Ultra 11.0
[19] and each structure was further optimized by the following
procedure:
i) conformational search analysis utilizing the “random walk”
method as implemented in HyperChem v. 7.5 [20] was used to
identify the lowest energy conformer;
ii) in view of the size and complexity of the anti-leukemia agents
studied, the lowest energy conformation obtained in step “i”
was subjected to a preliminary optimization with molecular
mechanics force field (MMþ);
iii) the AM1 [21] semi-empirical method using an RMS gradient
of 0.05 kcal/mol was then applied to obtain the final geom-
etries of the compounds for export to CODESSA PRO software
as the next step.
N ¼ 26 (training set); R2 ¼ 0.771; Rc2v ¼ 0.602; R2test ¼ 0.599 (for
the test set of 8 compounds described above); F ¼ 17.7; s2 ¼ 0.0376.
Ranges: Observed (0.780; ꢀ2.035). Predicted (0.887; ꢀ2.138).
The descriptors of Table 2 can be arranged based on their level of
significance (“t-criterion”) in the following order: “WPSA-3
Weighted PPSA4 (PPSA3 * TMSA/1000) (MOPAC PC)” > “Maximum
resonance energy for bond CeN” > “Principal moment of inertia
B” > “HA dependent HDSA-1/TMSA (MOPAC PC) (all)”.
The most important descriptor, namely the “WPSA-3 Weighted
PPSA” represents the portion of the total molecular surface area
which is positively charged. The negative regression coefficient sign
of this descriptor suggests that bulky molecules, with the presence
of few electronegative atoms such as O, S, Cl, N will be of reduced
potency as anticancer agents.
More than 500 constitutional, topological, geometrical,
quantum chemical and electrostatic descriptors were calculated
with the CODESSA PRO software package. The LogP descriptor was
also calculated and added to the set of CODESSA descriptors as an
important transportation limiting factor. The optimal descriptor
subset was then selected by the BMLR3 [22a] algorithm based on
the selection of orthogonal descriptor pairs.
4. Multi-linear modeling
The second descriptor in order of significance is the “Maximum
resonance energy for bond CeN”. As expected this descriptor has
largest values for compounds containing a nitrile group (see ID 25,
27 and 28 in Table 1). The negative regression coefficient of this
descriptor implies that the high resonance energy of the CeN bond
lowers the anti-tumor activity of compounds containing a nitrile
group; this is well illustrated by compounds 25, 27 and 28 which
are among the least active agents with inhibition activities of 0.96,
11.91 and 24.86, respectively.
The “Principal moment of inertia B” is related to the optimal size
and shape of the active molecules; molecules of size larger than the
optimal may well experience steric hindrance and thus inability to
bind efficiently to the target.
4.1. Data distribution and transformation of the original property
To improve the distribution of the experimental data, trans-
formations of the percentage inhibition of the tumor relative to
control experiments were performed. The logarithmic trans-
formation was found to produce a data distribution close to
normal (Fig. 1) and was thus preferred to alternative trans-
formations involving reciprocal, logarithmic reciprocal and
squared functions. One compound (ID #25), which is in fact one of
the two singletons in the dataset, disturbed slightly the normal
distribution of the data causing a long left branch of the proba-
bility density function. However, this compound fits well the
general QSAR relationship carried out in Table 2, and was retained
in the model.
3
4
Best Multi-Linear Regression.
Partial Positive Surface Area.

A.R. Katritzky et al. / European Journal of Medicinal Chemistry 45 (2010) 5183e5199
5185
Table 1
Structural formulae of the compounds under study and observed and predicted anti-tumor properties.
ID
Chemical structure
Observed %
growth of tumor
Observed % growth
inhibition of tumor
Log(observed %
growth inhibition
of tumor)
Log(predicted %
growth inhibition
of tumor)
D
(obs. ꢀ pred.)
1a*
38.54
38.04
20.74
10.58
36.59
61.46
61.96
79.26
89.42
63.41
1.789
1.792
1.899
1.951
1.802
2.104
2.138
1.966
1.702
1.750
ꢀ0.315
ꢀ0.345
ꢀ0.067
0.250
2a
3a
4a
5b
0.052
(continued on next page)

5186
A.R. Katritzky et al. / European Journal of Medicinal Chemistry 45 (2010) 5183e5199
Table 1 (continued)
ID Chemical structure
Observed %
growth of tumor
Observed % growth
inhibition of tumor
Log(observed %
growth inhibition
of tumor)
Log(predicted %
growth inhibition
of tumor)
D
(obs. ꢀ pred.)
6b
13.37
66.75
17.73
86.63
33.25
82.27
1.938
1.522
1.915
1.785
1.761
1.795
0.153
7b
ꢀ0.239
8b
0.120
9b
89.11
10.89
1.037
1.164
ꢀ0.127
10b
87.94
12.06
1.081
1.168
ꢀ0.087

A.R. Katritzky et al. / European Journal of Medicinal Chemistry 45 (2010) 5183e5199
5187
Table 1 (continued)
ID
Chemical structure
Observed %
growth of tumor
Observed % growth
inhibition of tumor
Log(observed %
growth inhibition
of tumor)
Log(predicted %
growth inhibition
of tumor)
D
(obs. ꢀ pred.)
11c*
66.79
65.71
70.18
87
33.21
34.29
29.82
13
1.521
1.535
1.475
1.114
1.417
1.416
1.176
1.315
0.105
0.119
0.299
12c*
13c*
14c
ꢀ0.201
15c
76
24
1.380
1.346
0.035
(continued on next page)

5188
Table 1 (continued)
A.R. Katritzky et al. / European Journal of Medicinal Chemistry 45 (2010) 5183e5199
ID
Chemical structure
Observed %
growth of tumor
Observed % growth
inhibition of tumor
Log(observed %
growth inhibition
of tumor)
Log(predicted %
growth inhibition
of tumor)
D
(obs. ꢀ pred.)
16c
49
61
54
29
51
39
46
71
1.708
1.591
1.663
1.851
1.615
1.620
1.544
1.677
0.093
17c*
18c
19c
ꢀ0.029
0.119
0.174
20c
85
15
1.176
1.267
ꢀ0.091

A.R. Katritzky et al. / European Journal of Medicinal Chemistry 45 (2010) 5183e5199
5189
Table 1 (continued)
ID
Chemical structure
Observed %
growth of tumor
Observed % growth
inhibition of tumor
Log(observed %
growth inhibition
of tumor)
Log(predicted %
growth inhibition
of tumor)
D
(obs. ꢀ pred.)
21c
47
53
1.724
1.485
0.239
22c
70
30
1.477
1.606
ꢀ0.129
23c
42
58
1.763
1.812
ꢀ0.049
24c
21
79
1.898
1.642
0.256
25c*
99.04
0.96
ꢀ0.018
0.336
ꢀ0.353
(continued on next page)

5190
Table 1 (continued)
A.R. Katritzky et al. / European Journal of Medicinal Chemistry 45 (2010) 5183e5199
ID
Chemical structure
Observed %
growth of tumor
Observed % growth
inhibition of tumor
Log(observed %
growth inhibition
of tumor)
Log(predicted %
growth inhibition
of tumor)
D
(obs. ꢀ pred.)
26c*
90.89
88.09
75.14
81.67
90.85
9.11
11.91
24.86
18.33
9.15
0.960
1.076
1.396
1.263
0.961
1.078
1.290
1.242
0.871
1.253
ꢀ0.119
ꢀ0.214
0.154
27c
28c
29c*
30c
0.392
ꢀ0.291

A.R. Katritzky et al. / European Journal of Medicinal Chemistry 45 (2010) 5183e5199
5191
Table 1 (continued)
ID
Chemical structure
Observed %
growth of tumor
Observed % growth
inhibition of tumor
Log(observed %
growth inhibition
of tumor)
Log(predicted %
growth inhibition
of tumor)
D
(obs. ꢀ pred.)
31c
85.26
93.97
ꢀ8.45
45.39
14.74
1.169
0.780
2.035
1.737
0.900
0.888
2.007
1.733
0.268
32c
6.03
ꢀ0.108
33c
108.45
0.029
34c
54.61
0.004
The compounds marked with an asterisk are part of the test set.
a
Ref. [14].
Ref. [12].
Ref. [18].
b
c
As the magnitude of the “HA dependent HDSA-1/TMSA (MOPAC
PC) (all)” descriptor increases, so does the observed anti-leukemia
activity. This descriptor is an indicator for the ability of a molecule
to participate in a hydrogen bond formation. Ligandereceptor
interactions are generally donoreacceptor based with no covalent
bonds formed. This is consistent with the positive regression
coefficient of this descriptor.
the proposed QSAR to chance correlations we fitted the model to
randomly reordered activity values and then comparing the
statistical parameters with those obtained for the actual activities.
[23] Twenty such randomizations, produced on average R2 ¼ 0.413
(ranging 0.361e0.472). The substantial difference between the
actual R2 of 0.771 and the averaged R2 from the scrambling
procedure indicates the stability of the model.
As emphasized above, the limited size of our dataset requires
extensive internal and external validation to prove the stability and
to estimate the “true” predictive power of the model. In addition to
crossvalidation and external validation we therefore also applied
a scrambling procedure to the model. To examine the sensitivity of
6. Chemistry
Based on the model of Table 2, we predicted the activity of some
30 bis-urea containing-compounds similar in structure to the most

5192
A.R. Katritzky et al. / European Journal of Medicinal Chemistry 45 (2010) 5183e5199
Table 2
Best 4-descriptor QSAR model.
ID
X
DX
t
Descriptor name
0
1
5.117
ꢀ0.02310
0.4471
0.002899
11.45
ꢀ7.970
Intercept
WPSA-3 Weighted PPSA
(PPSA3*TMSA/1000) (MOPAC PC)
Maximum resonance energy for
bond C-N
2
ꢀ0.09639
0.01437
ꢀ6.709
3
4
ꢀ353.7
56.32
0.8610
ꢀ6.281
Principal moment of inertia B
HA dependent HDSA-1/TMSA
(MOPAC PC) (all)
3.834
4.453
Alkylation of 2-aminothiophenol 9 with 1,4-dibromobutane in
refluxing ethanol gave diamine dihydrochloride 10 (Scheme 3). Bis-
ureas 11a,b (68e75%) were obtained by reaction of isocyanates and
diamine dihydrochloride 10 in dry THF.
7. Anti-tumor activity screening
Fig. 1. Probability density function of the logarithmic percentage tumor growth inhi-
bition for the studied set of 34 compounds.
Compounds (4a,b; 8; 11a,b) were screened for anti-tumor
activity at a dose of 10 mM utilizing human tumor cell lines, rep-
resenting leukemia, melanoma and cancers of the lung, colon,
brain, ovary, breast, prostate and kidney. Adriamycin was used as
a reference standard according to the previously reported stan-
dard procedure. [12e17] The results obtained (Table 3) represent
percentage growth of the tumor cell lines treated with compounds
under investigation relative to control cell experiments. The data
obtained revealed that compounds 8 and 11a showed promising
anti-tumor properties against most of the adopted cell lines,
whereas compounds 4a,b and 11b revealed little significant effect
active one (compound #33 in Table 1). We selected five of these (4a,
4b, 8, 11a and 11b) in order to provide a measure of the predictive
power of our model: four of these were selected from among these
predicted as most active and one of low predicted activity was also
picked. The predicted percentages of tumor growth inhibition
were: 4a e 109.7, 4b e 61.6, 8 e 94.2, 11a e 59.7 (high activity) and
11b e 36.7 (low activity). As described below compounds 4a, 4b, 8,
11a, and 11b were synthesized and their anti-tumor activity
screened through the Developmental Therapeutics Program of
National Cancer Institute, Bethesda, USA.
Schemes 1e3 outline the routes used for synthesis of target
compounds 4a, 4b, 8, 11a, and 11b. Reaction of the potassium salts
of acetamidophenols 1 and 5 with 1,3-dibromopropane in refluxing
DMF afforded diacetamide derivatives 2 and 6 respectively. Treat-
ment of 2 and 6 with conc. HCl in refluxing ethanol gave the dia-
mino analogues 3 and 7 respectively; which on reaction with the
appropriate isocyanate in dry tetrahydrofuran at room temperature
yielded the bis-ureas 4a,b and 8 in 89e93% yield (Schemes 1 and 2).
(considering >50% inhibition at 10 mM as noticeable activity).
Highly active compounds 8 and 11a were screened at serial dilu-
tions (10ꢀ4e10ꢀ8 mM) to determine GI50 (the concentration
resulting in a 50% growth inhibition of the tumor compared with
the control experiments) and TGI (the concentration resulting in
a 100% growth inhibition of the tumor compared with the control
experiments). The data obtained (Table 4) indicate that, compound
11a is more effective (GI50 ¼ 1.55, TGI ¼ 8.68
mM) than 8
M) against the targeted leukemia
(GI50 ¼ 58.30, TGI ¼ >100
m
RPMI-8226 cell line. Compound 8 however, reveals highly prom-
ising anti-tumor properties against SR (leukemia), SNB-75 (CNS
cancer) and MALME-3M (melanoma) cell lines (GI50 ¼ 0.197, 1.48,
0.26, TGI ¼ 4.01, 5.34, 5.31
mM, respectively), while compound 11
exhibits highly pronounced anti-tumor properties against NCI-
H226, NCI-H23 (non-small cell lung cancer), COLO 205 (colon
cancer), SNB-75 (CNS cancer), OVCAR-3, SK-OV-3 (ovarian cancer),
A498 (renal cancer) MDA-MB-231/ATCC and MDA-MB-468 (breast
cancer) cell lines (GI50 ¼ 1.95, 1.61, 1.38, 1.56, 1.30, 1.98, 1.18, 1.85,
1.08, TGI ¼ 8.35, 6.01, 2.67, 8.59, 4.01, 7.01, 5.62, 6.38, 5.63
mM,
respectively). It seems likely that a combination of a sulfnyl
function with a chlorophenylurea residue could be a very suitable
choice for designing broad spectrum anti-tumor active agents
targeting various human tumor cell lines (Figs. 3 and 4).
8. Conclusions
In conclusion, the use of CODESSA PRO software provided
a robust QSAR model describing the bioactivity of 34 compounds
tested against the “leukemia RBMI-8226 cell line”. Based on the
above model, we synthesized five novel bis-urea containing-
compounds, four predicted to be active and one less active. Initial
anti-tumor activity results at a dose of 10 mM utilizing 55 different
tumor cell lines, indicated that 11a and 8 possess promising
Fig. 2. Predicted vs experimental Log(percentage of tumor growth inhibition).
activity. Further screening at serial dilutions (10ꢀ4e10ꢀ8 mM)

A.R. Katritzky et al. / European Journal of Medicinal Chemistry 45 (2010) 5183e5199
5193
HO
O
O
1) KOH, EtOH
NH
2) Br(CH2)3Br, DMF
HN
NH
H3C
O
O
CH3
H3C
O
2
1
1) EtOH, HCl
2) NaOH
O
O
O
O
RNCO
HN
NH
Et3N, THF
H3N
NH3 Cl
Cl
O
NH
R
HN
O
4
R
3
4a, R = 4-ClC6H4 (92% yield)
Scheme 1.
showed compound 11a to have GI50 ¼ 1.55, TGI ¼ 8.68
m
M, and 8 to
s ¼ broad singlet, m ¼ multiplet), coupling constants (J values) are
expressed in Hz. Elemental analyses were performed on a Carlo
Erba EA-1108 instrument. Anhydrous THF was obtained by distil-
lation immediately prior to use, from sodium/benzophenone ketyl.
Purity of compounds was determined by elemental analyses; purity
of the target compounds was ꢁ95%.
have GI50 ¼ 58.30, TGI ¼ >100 M against the targeted leukemia
m
RPMI-8226 cell line. Compound 11a exhibits highly pronounced
anti-tumor properties against NCI-H226, NCI-H23 (non-small cell
lung cancer), COLO 205 (colon cancer), SNB-75 (CNS cancer),
OVCAR-3, SK-OV-3 (ovarian cancer), A498 (renal cancer) MDA-MB-
231/ATCC and MDA-MB-468 (breast cancer) cell lines, suggesting
that 11a could be a lead for developing broad spectrum anti-tumor
active agents.
9.1. Synthesis of diacetamides (2, 6) (general procedure)
The appropriate acetamidophenol 1, 5 (10 mmol) was dissolved
in absolute ethanol containing KOH (10 mmol). The solvent was
evaporated under reduced pressure, the residual material dissolved
in DMF (10 mL) and 1,3-dibromopropane (5 mmol) was added. The
mixture was heated under reflux for the appropriate time. The
solvent was evaporated under reduced pressure and the residue
was poured in water (100 mL). The separated solid was collected,
washed with water and crystallized from a suitable solvent
affording the corresponding 2, 6.
9. Experimental section
Melting points were determined using a capillary melting point
apparatus equipped with a digital thermometer and are uncor-
rected. 1H (300 MHz) and 13C (75 MHz) NMR spectra were recorded
in DMSO-d6 (with tetramethylsilane as the internal standard),
unless otherwise stated. Data are reported as follows: chemical
shift, multiplicity (s ¼ singlet, d ¼ doublet, t ¼ triplet, q ¼ quartet, br
O
O
OH
1) KOH, EtOH
2) Br(CH2)3Br, DMF
NH
NH
HN
H3C
O
O
CH3
H3C
O
6
5
1) EtOH, HCl
2) NaOH
O
O
O
O
NCO
MeO
NH
HN
Et3N, THF
Cl
NH3 Cl
H3N
HN
O
O
NH
7
OMe
OMe
8, 93%
Scheme 2.

5194
A.R. Katritzky et al. / European Journal of Medicinal Chemistry 45 (2010) 5183e5199
H3N
Cl
SH
1) Na, EtOH
S
Br
+
S
Br
2) MeOH, HCl
NH2
NH
3 Cl
9
10
R
O
NH
HN
S
S
RNCO, THF, Et3N
NH
11
11a, R = 4-ClC6H4 (68% yield)
11b, R = 4-MeOC6H4 (75% yield)
HN
R
O
Scheme 3.
9.2. N,N0-{4,40-[Propane-1,3-diyl(oxy)]bis(4,1-phenylene)}
diacetamide (2)
(OCH2), 115.9, 116.6, 140.2, 152.3 (arom. C). Anal. Calcd for
C15H18N2O2: C, 69.74; H, 7.02; N, 10.84. Found: C, 69.96; H, 7.18; N,
10.82.
Reaction time 1 h; colorless microcrystals from methanol; mp
178e180 ꢂC, lit. [24] mp 191e192 ꢂC; yield 44%. 1H NMR (DMSO-d6)
d
9.6. 2,20-[Propane-1,3-diylbis(oxy)]dianiline dihydrochloride (7)
1.99 (s, 6H, 2 CH3), 2.12 (quintet, J ¼ 6.3 Hz, 2H, OCH2CH2), 4.06 (t,
J ¼ 6.3 Hz, 4H, 2 OCH2), 6.87 (d, J ¼ 8.7 Hz, 4H, arom. H), 7.46 (d,
Reaction time 5 h; colorless microcrystals after washing with
J ¼ 9.0 Hz, 4H, arom. H), 9.78 (s, 2H, 2 NH); 13C NMR (DMSO-d6)
diethyl ether; mp 256e258 ꢂC, lit. [26] mp 306e308 ꢂC; yield 97%.
1H NMR (DMSO-d6)
d
2.25 (quintet, J ¼ 5.85 Hz, 2H, CH2), 4.35 (t,
d
23.8 (CH3), 28.7 (OCH2CH2), 64.3 (OCH2), 114.4, 120.5, 132.6, 154.2
(arom. C), 167.7 (CO). Anal. Calcd for C19H22N2O4: C, 66.65; H, 6.48;
N, 8.18. Found: C, 66.40; H, 6.55; N, 8.12.
J ¼ 6.0 Hz, 4H, 2 OCH2), 7.01 (t, J ¼ 7.8 Hz, 2H, arom. H), 7.25 (d,
J ¼ 8.1 Hz, 2H, arom. H), 7.35 (t, J ¼ 7.8 Hz, 2H, arom. H), 7.48 (d,
J ¼ 7.5 Hz, 2H, arom. H), 10.15 (br s, 4H, 2 NH2); 13C NMR (DMSO-d6)
9.3. N,N0-{2,20-[Propane-1,3-diyl(oxy)]bis(2,1-phenylene)}
d 28.3 (CH2), 65.1 (OCH2), 113.2, 120.7, 120.9, 123.8, 129.0, 151.5
diacetamide (6)
(arom. C).
9.7. Synthesis of 2,20-[1,4-butanediylbis(thio)]bisbenzenamine
dihydrochloride (10)
Reaction time 2 h; colorless microcrystals from ethanol; mp
192e194 ꢂC, lit. [25] mp 193.5e194.5 ꢂC; yield 52%. 1H NMR (CDCl3)
d
2.11 (s, 6H, 2 COCH3), 2.30e2.42 (m, 2H, CH2), 4.26 (t, J ¼ 6.0 Hz,
4H, 2 OCH2), 6.90e7.06 (m, 6H, arom. H), 7.73 (br s, 2H, 2 NH), 8.34
A mixture of 2-aminothiophenol 9 (1.08 mL, 10 mmol) and 1,4-
dibromobutane (0.6 mL, 5 mmol) in absolute ethanol containing
sodium metal (0.23 g, 10 mmol) was heated under reflux for 2 h.
After cooling, the reaction mixture was poured into ice-cold water
(200 mL) and extracted with diethyl ether. The organic layer was
washed with saturated sodium carbonate solution then with water
and dried over anhydrous sodium sulfate, then evaporated to
dryness under reduced pressure. The residue was dissolved in
methanol (20 mL) and conc. HCl (37%) was added dropwise. The
separated colorless crystals were collected and washed with diethyl
(d, J ¼ 7.5 Hz, 2H, arom. H); 13C NMR (CDCl3)
d 25.0 (CH3), 29.2
(CH2), 66.1 (OCH2), 111.6, 120.5, 121.9, 124.0, 128.2, 147.0 (arom. C),
168.4 (CO).
9.4. General procedure for the synthesis of dianilines (3,7)
A solution of the appropriate 2, 6 (1.5 mmol) in absolute ethanol
(15 mL) containing conc. HCl (4 mL) was heated under reflux for the
appropriate time. The reaction mixture was poured into ice-cold
water (200 mL) and made alkaline with NaOH solution (50%). The
separated solid was collected washed with water and crystallized
from a suitable solvent affording the corresponding 3. In the case of
6 the dihydrochloride salt, which separated under reflux conditions
was collected, washed with diethyl ether, air dried and used
without further purification.
ether
affording
2,20-[1,4-butanediylbis(thio)]bisbenzenamine
dihydrochloride (10) in pure form, which was used without further
purification (1.625 g, 86 % yield), mp 245e247 ꢂC (lit. [27] mp
253e255 ꢂC).
9.8. General procedure for reaction of 3,7 and 10 with isocyanates
9.5. 4,40-[Propane-1,3-diylbis(oxy)]dianiline (3)
A solution of the appropriate 3, 7 or 10 (2.5 mmol) with the
corresponding isocyanate (5 mmol) in dry tetrahydrofuran (10 mL)
was stirred at room temperature (25 ꢂC) for the suitable time. In the
case of 7 and 10, triethylamine (5 mmol) was added to the reaction
mixture. The separated solid was collected, washed with THF to
afford 4a,b. In the case of 8, 11a the separated solid was washed
with water and then with hot methanol to give colorless
Reaction time 6 h; colorless microcrystals from benzene; mp
104e106 ꢂC, lit. [24] mp 119e121 ꢂC; yield 65%. 1H NMR (CDCl3)
d
2.18 (quintet, J ¼ 6.3 Hz, 2H, OCH2CH2), 3.41 (br s, 4H, 2 NH2), 4.07
(t, J ¼ 6.3 Hz, 4H, 2 OCH2), 6.63 (d, J ¼ 8.7 Hz, 4H, arom. H), 6.75 (d,
J ¼ 8.7 Hz, 4H, arom. H); 13C NMR (CDCl3)
d 29.8 (OCH2CH2), 65.5

A.R. Katritzky et al. / European Journal of Medicinal Chemistry 45 (2010) 5183e5199
5195
Table 4
GI50 and TGI values (
Table 3
mM) of 8 and 11a utilizing human tumor cell lines.
Anti-tumor properties of the tested compounds at a dose of 10
tumor cell lines.
mM utilizing human
Panel/Cell line
Compound 8
Compound 11a
GI50
TGI
GI50
TGI
Panel/cell line
Percentage growth of tumor cell lines treated with the
tested compounds
Leukemia
CCRF-CEM
HL-60(TB)
K-562
MOLT-4
RPMI-8226
SR
14.4
26.7
31.7
17.7
58.3
0.2
78.9
68.0
>100
36.6
>100
4.0
2.5
2.4
1.3
2.4
1.5
2.23
>100
4a
4b
8
11a
11b
7.2
Leukemia
CCRF-CEM
HL-60(TB)
MOLT-4
SR
>100
>100
157
105
118
88.7
NT
NT
90.4
95.5
90.4
NT
L7.12
18.8
-5.8
L39.7
77
L14.9
L63.4
-6.5
L32.1
L4
75.6
120
128
143
NT
8.7
6.5
Non-small cell lung cancer
A549/ATCC
EKVX
HOP-62
HOP-92
NCI-H226
NCI-H23
NCI-H322M
NCI-H460
NCI-H522
RPMI-8226
>100
10.5
3.3
>100
>100
9.4
8.6
>100
>100
>100
>100
>100
2.6
2.4
3.4
2.0
1.9
1.6
3.7
1.4
1.4
>100
>100
>100
Non-small cell lung cancer
A549/ATCC
EKVX
HOP-62
HOP-92
NCI-H226
NCI-H23
NCI-H322M
NCI-H460
NCI-H522
107
95.0
96.2
181
123
96.5
102
103
111
86.7
95.7
81.5
L80.0
2.2
38.5
NT
87.1
2.8
82.4
17.7
63.7
NT
-5.5
12.1
L45.9
L0.98
-7.8
16.2
10.2
NT
83.1
2.7
7.1
8.3
6.0
84.6
68.1
85.2
88.6
90.0
89.2
56.2
>100
12.4
>100
>100
>100
121
>100
12.8
>100
94.5
90.2
96.8
98.3
Colon cancer
COLO 205
HCC-2998
HCT-116
HCT-15
HT29
KM12
SW-620
91.6
L50.8
14.3
NT
>100
NT
1.4
2.67
0.7
1.9
2.6
1.7
2.5
2.7
>100
4.8
>100
>100
>100
>100
Colon cancer
COLO 205
HCT-116
HCT-15
HT29
KM12
166
94.8
117
114
110
104
124
83.3
118
107
117
118
18.6
L36.8
65.5
54.9
23.8
NT
110
72.7
99.9
115
89.8
104
5.3
>100
>100
>100
>100
>100
L14.3
16.8
10.5
6.3
>100
>100
>100
>100
SW-620
0.7
13.6
CNS cancer
SF-268
SF-295
SF-539
SNB-19
SNB-75
U251
CNS cancer
SF-268
SF-295
SF-539
SNB-19
SNB-75
U251
7.3
29.9
3.4
4.3
1.5
>100
>100
>100
26.1
5.3
2.8
0.6
2.1
1.9
1.6
2.7
>100
ND
7.3
>100
8.6
125
103
71.7
86.7
94.5
101
50.7
49.0
L29.6
47.6
L34.8
L50.9
25.5
L14.0
0.7
23.3
10.4
104
89.4
86.2
86.3
81.3
96.2
97.3
91.8
108
92.2
103
3.7
28.8
>100
89.8
15.9
Melanoma
LOX IMVI
MALME-3M
M14
MDA-MB-435
SK-MEL-2
SK-MEL-28
SK-MEL-5
UACC-257
UACC-62
Melanoma
LOX IMVI
MALME-3M
M14
MDA-MB-435
SK-MEL-2
SK-MEL-28
SK-MEL-5
UACC-62
>100
0.3
45.7
>100
>100
>100
>100
>100
>100
>100
5.3
>100
>100
>100
>100
>100
>100
>100
2.5
2.5
2.4
2.6
2.9
2.5
2.1
3.2
NT
>100
98.1
101
96.7
110
118
111
100
99.3
93.3
95.5
109
108
204
23.7
55.1
61.4
41.6
65.8
45.7
48.7
3.9
5.5
3.4
14.9
6.1
L51.0
14.8
5.5
92.9
108
88.7
91.5
89.4
103
85.0
66.7
9.4
>100
>100
>100
74.1
16.4
>100
99.5
109
98.9
16.5
NT
Ovarian cancer
IGROV1
OVCAR-3
OVCAR-4
OVCAR-5
OVCAR-8
NCI/ADR-RES
SK-OV-3
Ovarian cancer
IGROV1
85.1
112
110
94.6
64.1
99.8
90.2
116
109
4.9
L11.4
L67.2
46.0
L65.3
59.8
L51.7
-8.7
17.8
12.6
NT
49.9
101
>100
>100
>100
>100
8.9
>100
>100
>100
>100
>100
>100
>100
3.90
1.3
2.1
2.9
3.3
2.2
2.0
>100
OVCAR-3
OVCAR-4
OVCAR-5
OVCAR-8
NCI/ADR-RES
SK-OV-3
4.0
89.0
85.6
NT
80.9
102
>100
>100
>100
ND
88.4
70.1
94.4
88.4
-0.1
NT
>100
5.0
122
7.9
7.01
Renal cancer
786-0
A498
ACHN
SN12C
TK-10
Renal cancer
786-0
A498
93.7
107
106
105
129
86.9
L31.5
L9.3
46.3
L21.3
22.0
13.5
L28.0
11.4
6.2
16.1
93.8
105
94.0
99.8
93.1
72.1
ND
>100
72.9
>100
>100
>100
>100
2.8
1.2
2.1
1.6
3.4
2.6
2.8
2.4
>100
5.6
9.7
>100
>100
>100
>100
>100
144
100
87.3
123
71.4
16.0
>100
>100
3.1
>100
4.0
ACHN
CAKI-1
RXF 393
SN12C
TK-10
UO-31
68.8
L20.7
L33.4
49.7
>100
Prostate cancer
PC-3
DU-145
UO-31
>100
93.0
121
108
107
L36.5
39.4
NT
11.2
NT
104
Prostate cancer
PC-3
DU-145
>100
>100
>100
>100
1.7
2.0
>100
>100
Breast cancer
MCF7
MDA-MB-231/ATCC
HS 578T
BT-549
T-47D
103
111
85.4
102
106
114
55.0
28.2
34.0
46.4
81.2
9.2
4.9
4.9
-4.8
5.5
NT
L10.7
96.
88.1
84.2
83.6
90.1
76.0
93.3
88.2
88.1
83.6
91.1
Breast cancer
MCF7
MDA-MB-231/ATCC
HS 578T
BT-549
T-47D
>100
8.8
>100
1.5
>100
NT
>100
>100
>100
29.2
>100
1.7
1.8
1.8
2.7
0.6
1.1
13.5
6.4
14.1
>100
MDA-MB-468
22.6
5.7
5.6
NT ¼ Not tested.
MDA-MB-468
NT
NT ¼ Not Tested, ND ¼ Not Determined.

5196
A.R. Katritzky et al. / European Journal of Medicinal Chemistry 45 (2010) 5183e5199
Fig. 3. Dose response curves of 1,10-{2,20[propane-1,3-diylbis(oxy)]bis(2,1-phenylene)}bis[3-(4-methoxyphenyl)urea] (8).

A.R. Katritzky et al. / European Journal of Medicinal Chemistry 45 (2010) 5183e5199
5197
Fig. 4. Dose response curves of 1,10-{2,20-[butane-1,4-diylbis(sulfanediyl)]bis(2,1-phenylene)}bis[3-(4-chlorophenyl)urea] (11a).

5198
A.R. Katritzky et al. / European Journal of Medicinal Chemistry 45 (2010) 5183e5199
microcystals, which require no further purification. In the case of
11b, however, after washing the reaction mixture with water, it was
crystallized from n-butanol.
2H, 2 NH); 13C NMR (DMSO-d6)
d 27.5 (CH2), 33.5 (SCH2), 55.1
(OCH3), 114.0, 119.9, 120.3, 122.4, 122.8, 128.1, 132.7, 133.2, 139.9,
152.5 (arom. C), 154.5 (CO). Anal. Calcd for C32H34N4O4S2: C, 63.76;
H, 5.69; N, 9.29. Found: C, 63.38; H, 5.96; N, 8.93.
9.9. 1,10-{4,40-[Propane-1,3-diylbis(oxy)]bis(4,1-phenylene)}bis[3-
(4-chlorophenyl)urea] (4a)
9.14. Anti-tumor activity screening
Reaction time 3 h; colorless microcrystals after washing with
Anti-tumor activity screening for compounds (4a,b; 8, 11a,b) at
a dose of 10 mM utilizing 55 different human tumor cell lines,
tetrahydrofuran; mp 311e313 ꢂC, yield 92%. 1H NMR (DMSO-d6)
d
2.13 (quintet, J ¼ 5.85 Hz, 2H, OCH2CH2), 4.08 (t, J ¼ 6.0 Hz, 4H, 2
representing leukemia, melanoma and cancers of the lung, colon,
brain, ovary, breast, prostate and kidney was carried out using
adriamycin as a reference standard according to standard proce-
dure [12e17]. The human tumor cell lines of the cancer screening
panel are grown in RPMI-1640 medium containing 5% fetal bovine
OCH2), 6.89 (d, J ¼ 8.7 Hz, 4H, arom. H), 7.28e7.36 (m, 8H, arom.),
7.47 (d, J ¼ 8.7 Hz, 4H, arom. H), 8.51 (s, 2H, 2 NH), 8.73 (s, 2H, 2
NH); 13C NMR (DMSO-d6)
d 28.8 (OCH2CH2), 64.4 (OCH2), 114.7,
119.6, 120.1, 125.1, 128.6, 132.6, 138.9, 152.6 (arom. C), 153.8 (CO).
Anal. Calcd for C29H26Cl2N4O4: C, 61.60; H, 4.63; N, 9.91. Found: C,
61.57; H, 4.52; N, 9.76.
serum and 2 mM
L-glutamine. For a typical screening experiment,
cells are inoculated in 96-well-microtiter plates in 100
mL at plating
densities ranging from 5000 to 40,000 cells/well depending on the
doubling time of individual cell lines. After cell inoculation, the
microtiter plates are incubated at 37 ꢂC, 5% CO2, 95% air and 100%
relative humidity for 24 h prior to addition of test compounds. After
24 h, two plates of each cell lines are fixed in situ with trichloro-
acetic acid (TCA), to represent a measurement of the cell population
for each cell line at the time of test compound addition (time zero,
Tz). Tested compounds were solubilized in dimethyl sulfoxide at
400-fold the desired final maximum test concentration and stored
frozen prior to use. At the time of test compound addition, an
aliquot of frozen concentrate was thawed and diluted to twice the
desired final maximum test concentration with complete medium
9.10. 1,10-{4,40[Propane-1,3-diylbis(oxy)]bis(4,1-phenylene)}bis[3-
(4-methoxyphenyl)urea] (4b)
Reaction time 6 h; colorless microcrystals after washing with
tetrahydrofuran; mp 307e309 ꢂC; yield 89%. 1H NMR (DMSO-d6)
d
2.13 (quintet, J ¼ 6.15 Hz, 2H, OCH2CH2), 3.71 (s, 6H, 2 OCH3), 4.07
(t, J ¼ 6.15 Hz, 4H, 2 OCH2), 6.86 (d, J ¼ 8.7 Hz, 4H, arom. H), 6.88 (d,
J ¼ 7.8 Hz, 4H, arom. H), 7.35 (d, J ¼ 9.0 Hz, 8H, arom. H), 8.38 (s, 4H,
4 NH); 13C NMR (DMSO-d6)
d 28.8 (OCH2CH2), 55.1 (OCH3), 64.4
(OCH2), 114.0, 114.6, 119.9, 133.0, 133.1, 152.9, 153.6 (arom. C), 154.3
(CO). Anal. Calcd for C31H32N4O6: C, 66.89; H, 5.79; N, 10.07. Found:
C, 66.94; H, 5.79; N, 9.99.
containing 50
mg/mL gentamicin. Aliquots of 100 mL of the tested
compound dilutions were added to the appropriate microtiter wells
9.11. 1,10-{2,20[Propane-1,3-diylbis(oxy)]bis(2,1-phenylene)}bis[3-
(4-methoxyphenyl)urea] (8)
already containing 100
concentrations.
mL of medium, to give in the required final
Following the test compound addition, the plates were incu-
bated for an additional 48 h at 37 ꢂC, 5% CO2, 95% air and 100%
relative humidity. For adherent cells, the assay was terminated by
the addition of cold TCA. Cells were fixed in situ by the gentle
Reaction time 24 h; colorless microcrystals after washing with
methanol; mp 234e236 ꢂC; yield 93%. 1H NMR (DMSO-d6)
d 2.34
(quintet, J ¼ 6.0 Hz, 2H, CH2), 3.71 (s, 6H, 2 OCH3), 4.31 (t, J ¼ 6.0 Hz,
4H, 2 OCH2), 6.85e6.92 (m, 8 H, arom. H), 7.02e7.06 (m, 2H, arom.
H), 7.36 (d, J ¼ 8.4 Hz, 4H, arom. H), 7.98 (s, 2H, 2 NH), 8.09e8.12 (m,
addition of 50 mL of cold 50% (w/v) TCA (final concentration, 10%
TCA) and incubated for 60 min at 4 ꢂC. The supernatant was dis-
2H, arom. H), 9.19 (s, 2H, arom. H); 13C NMR (DMSO-d6)
d
28.6 (CH2),
carded, and the plates were washed five times with tap water and
55.2 (OCH3), 65.1 (OCH2), 111.7, 114.1, 118.7, 120.0, 120.6, 121.8, 129.0,
132.8, 146.8, 152.6 (arom. C), 154.5 (CO). Anal. Calcd for C31H32N4O6:
C, 66.89; H, 5.79; N, 10.07. Found: C, 66.52, H, 5.88, N, 9.82.
air dried. Sulforhodamine B (SRB) solution (100 mL) at 0.4% (w/v) in
1% acetic acid was added to each well, and plates were incubated
for 10 min at room temperature. After staining, unbound dye was
removed by washing five times with 1% acetic acid and the plates
were air dried. Bound stain was subsequently solubilized with
10 mM trizma base, and the absorbance was read on an automated
plate at 515 nm. For suspension cells, the methodology was the
same except that the assay was terminated by fixing settled cells at
9.12. 1,10-{2,20-[Butane-1,4-diylbis(sulfanediyl)]bis(2,1-phenylene)}
bis[3-(4-chlorophenyl)urea] (11a)
Reaction time 48 h; colorless microcrystals; mp 210e212 ꢂC;
yield 68%. 1H NMR (DMSO-d6)
d
1.62 (br s, 4H, 2 CH2), 2.82 (br s,
the bottom of the wells by gently adding 50
mL of 80% TCA (final
4H, 2 SCH2), 6.98 (t, J ¼ 7.5 Hz, 2 H, arom. H), 7.25 (t, J ¼ 7.8 Hz, 2
H, arom. H), 7.32 (d, J ¼ 8.7 Hz, 4 H, arom. H), 7.39 (d, J ¼ 7.8 Hz,
2H, arom. H), 7.49 (d, J ¼ 8.7 Hz, 4H, arom. H), 8.02 (d, J ¼ 8.1 Hz,
2H, arom. H), 8.27 (s, 2H, 2 NH), 9.63 (s, 2H, 2 NH); 13C NMR
concentration, 16% TCA). Table represents the observed
percentage growth of each cell line treated with a test compound
relative to control cell line experiments.
3
Due to the preliminarily nature of the observed anti-tumor
properties of compounds 8 and 11a, we screened the activity at
serial dilutions (10ꢀ4e10ꢀ8 mM), adopting the described procedure,
and using the seven absorbance measurements [time zero (Tz),
control growth (C) and test growth in the presence of the tested
compound at the five concentration levels (Ti)], to calculate the
percentage growth at each of the test compound concentration
levels.
(DMSO-d6)
d 27.5 (CH2), 33.4 (SCH2), 119.6, 120.6, 122.9, 123.4,
125.4, 128.1, 128.7, 133.0, 138.7, 139.4 (arom. C), 152.3 (CO). Anal.
Calcd for C30H28Cl2N4O2S2: C, 58.91; H, 4.61; N, 9.16. Found: C,
58.59; H, 4.59; N, 8.98.
9.13. 1,10-{2,20-[Butane-1,4-diylbis(sulfanediyl)]bis(2,1-phenylene)}
bis[3-(4-methoxyphenyl)urea] (11b)
Percentage growth inhibition is calculated as [(Ti ꢀ Tz)/
(C ꢀ Tz)] ꢃ 100 for concentrations in which Ti ꢁ Tz, and by [(Ti ꢀ Tz)/
Tz] ꢃ 100 for concentrations for which Ti < Tz.
Reaction time 24 h; mp 202e203 ꢂC; yield 75%. 1H NMR (DMSO-
d 1.62 (br s, 4H, 2 CH2), 2.81 (br s, 4H, 2 SCH2), 3.71 (s, 6H, 2
OCH3), 6.87 (d, J ¼ 9.0 Hz, 4H, arom. H), 6.96 (dt, J ¼ 1.2, 7.5 Hz, 2 H,
arom H), 7.24 (dt, J ¼ 1.2, 7.5 Hz, 2 H, arom. H), 7.35e7.41 (m, 6 H,
arom. H), 8.05 (d, J ¼ 8.1 Hz, 2H, arom. H), 8.17 (s, 2H, 2 NH), 9.33 (s,
d6)
Growth inhibition of 50% (GI50) is calculated from [(Ti ꢀ Tz)/
(C ꢀ Tz)] ꢃ 100 ¼ 50, which is the test compound concentration
resulting in a 50% reduction in the net protein increase (as

A.R. Katritzky et al. / European Journal of Medicinal Chemistry 45 (2010) 5183e5199
5199
[13] A.S. Girgis, Regioselective synthesis of dispiro[1H-indene-2,30-pyrrolidine-
200,300-[3H]indole-1,200 (100H)-diones of potential anti-tumor properties, Eur. J.
Med. Chem. 44 (2009) 91e100.
measured by SRB staining) in control cells during the drug incu-
bation (Table 4).
[14] A.S. Girgis, Regioselective synthesis and stereochemical structure of anti-
tumor active dispiro[3H-indole-3,20pyrrolidine-30,300-piperidine]-2(1H),400-
diones, Eur. J. Med. Chem. 44 (2009) 1257e1264.
Acknowledgment
[15] 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, Feasibility of drug screening
with panels of human tumor cell lines using a microculture tetrazolium assay,
Cancer Res. 48 (1988) 589e601.
[16] M.R. Grever, S.A. Schepartz, B.A. Chabner, The National Cancer Institute:
cancer drug discovery and development program, Semin. Oncol. 19 (1992)
622e638.
[17] (a) M.R. Boyd, K.D. Paull, Some practical considerations and applications of the
national cancer institute in vitro anticancer drug discovery screen, Drug Dev.
Res. 34 (1995) 91e109;
We thank the National Cancer Institute, Bethesda, USA for anti-
tumor activity screening through the Developmental Therapeutics
Program. This study was supported by research grant of U.S.eEgypt
Joint Board on Scientific and Technological Cooperation, grant no.
MAN 10-007-002. We thank Dr. C.D. Hall for helpful discussions and
a Reviewer for very helpful insightful comments and constructive
criticism.
(b) S.X. Cai, B. Nguyen, S. Jia, J. Herich, J. Guastella, S. Reddy, B. Tseng, J. Drewe,
S. Kasibhatla, Discovery of substituted N-phenyl nicotinamides as potent
References
inducers of apoptosis using
a cell- and caspase-based high throughput
screening assay, J. Med. Chem. 46 (2003) 2474e2481.
[18] Unpublished data, Anti-tumor activity of these compounds screened through
the Developmental Therapeutics Program of National Cancer Institute,
Bethesda, USA and observed anti-tumor activity properties of these 24
compounds are given Table 1.
[19] ChemDraw Ultra v.11, CambridgeSoft Corporation, Cambridge, MA. www.
cambridgesoft.com
[20] Hyperchem, v. 7.5, Hypercube Inc., Gainesville, FL. www.hyper.com
[21] M.J.S. Dewar, E.G. Zoebisch, E.F. Healy, J.J.P. Stewart, AM1: a new general
purpose quantum mechanical molecular model, J. Am. Chem. Soc. 107 (1985)
3902e3909.
[1] Canadian Cancer Statistics 2005. Canadian Cancer Society e National Cancer
Institute of Canada, Toronto, Canada, 2005.
[2] Canadian Cancer Statistics 2002. National Cancer Institute of Canada, Toronto,
Canada, 2002.
[3] E.N. MacKay, A.H. Sellers, A statistical survey of leukemia in Ontario and at the
Ontario cancer foundation clinics, 1938e1958, Can. Med. Assoc. J. 96 (1967)
1626e1635.
[4] American Cancer Society, Cancer Facts and Figures 2006. American Cancer
Society, Atlanta, 2006.
[5] J. Robert, C. Jarry, Multidrug resistance reversal agents, J. Med. Chem. 46
(2003) 4805e4817.
[6] S. Hassan, S. Dhar, M. Sandstrom, D. Arsenau, M. Budnikova, I. Lokot,
N. Lobanov, M.O. Karlsson, R. Larsson, E. Lindhagen, Cytotoxic activity of a new
paclitaxel formulation, Pacliex, in vitro and in vivo, Cancer Chemother.
Pharmacol. 44 (2005) 47e54.
[22] (a) M. Karelson, Molecular Descriptors in QSAR/QSPR. Wiley Interscience,
2000;
(b) T. Ghafourian, M. Barzegar-Jalali, N. Hakimiha, M.T.D. Cronin, Quantitative
structureepharmacokinetic relationship modelling: apparent volume of
distribution, J. Pharm. Pharmacol. 56 (2004) 339e350.
[7] CODESSA PRO Software. University of Florida, 2002.www.codessa-pro.com.
[8] A.R. Katritzky, S.H. Slavov, D.A. Dobchev, M. Karelson, QSAR modeling of the
[23] L. Eriksson, J. Jaworska, A.P. Worth, M.T.D. Cronin, R.M. McDowell,
P. Gramatica, Methods for reliability and uncertainty assessment and for
applicability evaluations of classification- and regression-based QSARs,
Environ. Health Perspect. 111 (2003) 1361e1375.
[24] P.A. Henderson, O. Niemeyer, C.T. Imrie, Methylene-linked liquid crystal
dimmers, Liq. Cryst. 28 (2001) 463e472.
[25] W.A. Jacobs, M. Heidelberger, On nitro- and aminophenoxyacetic acids, J. Am.
Chem. Soc. 39 (1917) 2188e2224.
[26] J.N. Ashley, R.F. Collins, M. Davis, N.E. Sirett, The chemotherapy of schistoso-
miasis, Part I. Derivatives and Analogues of au-Di-(p-aminophenoxy)alkanes,
J. Chem. Soc. (1958) 3298e3313.
antifungal activity against Candida albicans for
a diverse set of organic
compounds, Bioorg. Med. Chem. 16 (2008) 7055e7069.
[9] A.R. Katritzky, Z. Wang, S. Slavov, M. Tsikolia, D. Dobchev, N.G. Akhmedov,
C.D. Hall, U.R. Bernier, G.G. Clark, K.J. Linthicum, Synthesis and bioassay of
improved mosquito repellents predicted from chemical structure, Proc. Natl.
Acad. Sci. 105 (2008) 7359e7364.
[10] S.C. Basak, D. Mills, Prediction of partitioning properties for environmental
pollutants using mathematical structural descriptors, Arkivoc ii (2005) 60e76.
[11] A. Thakur, QSAR study on benzenesulfonamide dissociation constant pKa:
physicochemical approach using surface tension, Arkivoc xiv (2005) 49e58.
[12] A.S. Girgis, H.M. Hosni, F.F. Barsoum, Novel synthesis of nicotinamide deriv-
atives of cytotoxic properties, Bioorg. Med. Chem. 14 (2006) 4466e4476.
[27] A.A. Abbas, A.S. Girgis, A convenient synthesis of thiamacrocyclic dilactams,
Heteroatom. Chem. 18 (2007) 249e254.