1-Aryl-2-dimethylaminomethyl-2-propen-1-one hydrochlorides and related adducts: A quest for selective cytotoxicity for malignant cells

Article abstract of DOI:10.1016/j.bmc.2008.03.060

The primary objective of this study was to discover one or more clusters of compounds which are not equitoxic but display cytoselectivity toward different malignant cells. Furthermore a most important consideration is that such molecules should also displ

Full text of DOI:10.1016/j.bmc.2008.03.060

Available online at www.sciencedirect.com
Bioorganic & Medicinal Chemistry 16 (2008) 5747–5753
1-Aryl-2-dimethylaminomethyl-2-propen-1-one hydrochlorides
and related adducts: A quest for selective cytotoxicity
for malignant cells
Hari N. Pati,a Umashankar Das,a Masami Kawase,b Hiroshi Sakagami,c Jan Balzarini,d
Erik De Clercqd and Jonathan R. Dimmocka,*
aCollege of Pharmacy and Nutrition, University of Saskatchewan, 110 Science Place, Saskatoon, Saskatchewan, Canada S7N 5C9
bFaculty of Pharmaceutical Sciences, Matsuyama University, 4-2 Bunkyo-Cho, Matsuyama, Ehime 790-8578, Japan
cDepartment of Diagnostic and Therapeutic Sciences, Meikai University School of Dentistry, Saitama 350-0238, Japan
dRega Institute for Medical Research, Katholieke Universiteit Leuven, Minderbroedersstraat 10, B-3000, Belgium
Received 13 February 2008; revised 21 March 2008; accepted 24 March 2008
Available online 27 March 2008
Abstract—The primary objective of this study was to discover one or more clusters of compounds which are not equitoxic but dis-
play cytoselectivity toward different malignant cells. Furthermore a most important consideration is that such molecules should also
display greater cytotoxic potencies to tumors than normal tissues. Two series of compounds are described which meet these criteria,
namely the 1-aryl-2-dimethylaminomethyl-2-propen-1-one hydrochlorides 1a–e and 1-aryl-3-dimethylamino-2-hydroxymethyl-1-
propanone hydrochlorides 2a–e. A number of these compounds possess marked cytotoxic potencies (IC50 and CC50 values within
the 10À6 and 10À7 molar range) which are greater than these of the reference drug melphalan. Statistical analyses demonstrated that
cytotoxic potencies are influenced by the size of the aryl substituents in series 1 and to some extent by the electronic properties of the
aryl groups in series 2. The mode of action of a representative compound 1e in HL-60 cells included inducing apoptosis and acti-
vation of caspases À3, À8, and À9.
ꢀ 2008 Elsevier Ltd. All rights reserved.
1. Introduction
The reasons for preparing series 1 and the related ad-
ducts 2 and 3 included the following general consider-
ations. First, conjugated styryl ketones are designed as
thiol alkylators having little or no capacity to interact
with the amino or hydroxy groups of cellular constitu-
ents.1–3 Consequently these molecules may be devoid of
the genotoxic properties of a number of currently avail-
able anticancer drugs4 since thiol groups are absent in
nucleic acids. This concept of thiol-specificity represents
a markedly different approach in the design of putative
anticancer drugs than is often followed. Second, this no-
vel design may lead to cytotoxics which are not cross-
resistant to contemporary anticancer medication. Sup-
port for this contention comes from the observation that
several drug resistant cell lines were free from cross-resis-
tance to a series of Mannich bases of conjugated styryl
ketones.5 Third, since enones may undergo indiscrimi-
nate thiol alkylation prior to reaching a target site, a pro-
drug approach was incorporated into the project.
One of the major problems involved in treating different
cancers with drugs is the lack of their demonstrating
appreciably greater toxicity to malignancies than normal
cells. Hence novel groups of molecules are required
which are not general biocidal agents displaying equi-
toxicity but vary considerably in their potencies toward
different cells. In particular, these compounds should
provide unequivocal evidence of being more cytotoxic
to neoplastic cells rather than normal tissues. The objec-
tive of this report is to disclose that the 1-aryl-2-dimeth-
ylaminomethyl-2-propen-1-one hydrochlorides 1 and
the related adducts 2 are novel clusters of compounds
with selective toxicity to tumors.
Keywords: 1-Aryl-3-dimethylamino-1-propanones; Cytotoxicity; Apo-
ptosis; Caspases.
*
The specific reasons for the design of the compounds in
series 1 were to create a series of molecules having the
Corresponding author. Tel.: +1 306 966 6331; fax: +1 306 966
6377; e-mail: jr.dimmock@usask.ca
0968-0896/$ - see front matter ꢀ 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2008.03.060

5748
H. N. Pati et al. / Bioorg. Med. Chem. 16 (2008) 5747–5753
results of the Phase II investigation in which the selective
locus of thiol attack
toxicity of various compounds to neoplastic cells is
clearly demonstrated as well as an indication of the
means whereby bioactivity is mediated (Fig. 2).
β
H
R
H
CH3
α
N
H
Cl
CH3
2. Bioevaluations
O
All of the compounds in series 1–3 were evaluated
against transformed human CEM and Molt 4/C8 T-lym-
phocytes. In addition, these compounds were assessed
against murine L1210 leukemic cells. These data are pre-
sented in Table 1. Six representative compounds 1a, b, e,
2b, e, and 3 were evaluated against approximately 54 hu-
man tumor cell lines representing different types of can-
cers. The cytotoxicity against these neoplasms, and in
particular toward leukemic cell lines, is summarized in
Table 2. The cytotoxic potencies of 1a–e, 2a–e, and 3
were also assessed using the following human neoplastic
cells lines, namely HL-60 promyelocytic leukemic cells
and HSC-2 and HSC-4 squamous cell carcinomas. In
addition, evaluation toward non-malignant human
HGF gingival fibroblasts, HPLF periodontal ligament
fibroblasts and HPC pulp cells was undertaken. These
results are portrayed in Table 3. A representative com-
pound 1e caused apoptosis in HL-60 cells as indicated
1
Figure 1. Design of the thiol alkylator 1.
following structural features, namely (1) a conjugated
enone group attached to an aryl ring, (2) a sterically
unhindered b carbon atom where interactions with thi-
ols take place, and (3) an amino group located close to
the a,b-unsaturated keto pharmacophore. The reasons
for the last molecular component are as follows. First,
protonation of the basic center leading to a quadrivalent
nitrogen atom will increase the electrophilicity to thiols
as illustrated in Figure 1. Second, on occasions the
extracellular pH around tumor cells may be low6,7 and
the pH of neoplasms may be lower than in normal tis-
sues.8 Since the ratio of ions to free base is dependent
on the pH of the medium,9 the percentage of ions can
be greater, close to or in the malignant cells which
may lead to preferential toxicity to neoplasms. Third,
the amine hydrochloride portion of 1–3 confers drug-
likeness and increases water solubility.
Table 1. Evaluation of series 1–3 against human CEM and Molt 4/C8
T-lymphocytes and murine L1210 leukemic cells
Compound
IC50 (lM)a
SIb
IC50 (lM)a
In the light of these considerations, a twofold strategy
was adopted, namely Phase I (an initial examination
to ascertain whether some preliminary data warranted
proceeding to Phase II) and Phase II (a detailed investi-
gation of whether the compounds display cytoselectivity
and a probing of the mode of action of these novel cyto-
toxins). The Phase I study has been completed.10 The
compounds were prepared as follows. Condensation of
the appropriate arylethanone with dimethylamine
hydrochloride and excess of formaldehyde led to the for-
mation of 1a–e. In the presence of water, the Mannich
bases 1a–e were converted into the corresponding ami-
noalcohols 2a–e while reaction of 1c with 2-mercap-
toethanol gave 3. The IC50 values of the compounds in
the series 1 and 2 toward human WiDr colon cancer
cells were in the low micromolar range. On the other
hand, 3 has very low potency (IC50 = 311 lM) and fur-
ther analogs in series 3 were not prepared. A stability
study revealed that 2c but not 3 reverted to 1c in solu-
tion. The purpose of the current report is to reveal the
CEM
Molt 4/C8
L1210
1a
1b
65.6 33.8
32.6 7.0
7.91 0.53
11.4 1.9
10.6 8.5
1.42 0.07
33.1 19.5
12.2 0.0
8.22 2.28
1.15 0.03
>500
146 91
2.2
1.1
1.1
3.1
2.0
4.1
1.1
3.0
2.6
4.3
ꢀ1.0
1.3
52.2 15.8
26.5 21.7
8.61 1.85
11.1 0.9
5.08 4.52
1.65 0.10
22.0 18.5
42.6 1.7
6.55 4.29
1.57 0.05
>500
30.2 19.0
8.95 0.20
35.5 7.4
21.3 14.0
5.84 1.89
30.9 18.6
36.8 2.6
21.0 13.0
4.90 2.19
>500
1c
1d
1e
2a
2b
2c
2d
2e
3
Melphalanc
2.47 0.03
3.24 0.79
2.13 0.03
a The IC50 value is the concentration of compound required to inhibit
the growth of the cells by 50%.
b The letters SI refer to the selectivity index of each compound which is
the ratio of the highest to lowest IC50 values of each compound
toward the two T-lymphocytes.
c The data for melphalan is taken from Ref. 23 copyright (2006) with
permission of Elsevier.
HO
HO
O2N
S
R
H
H
R
CH3
N
CH3
N
CH3
N
HCl
HCl
HCl
CH3
CH3
CH3
O
O
O
1
2
3
Figure 2. Structures of series 1–3. The aryl substituents in series 1 and 2 are as follows, namely (a) R = H; (b) R = Cl; (c) R = NO2; (d) R = CH3: (e)
R = OCH3.

H. N. Pati et al. / Bioorg. Med. Chem. 16 (2008) 5747–5753
5749
Table 2. Evaluation of 1a, b, e, 2b, e, 3, and melphalan against a panel of human tumor cell linesa
Compound
All cell lines
Leukemic cell lines, IC50 (lM)b
SId
Average GI50 (lM)b
SIc
67.6
CCRF-CEM
K562
RPMI-8226
SR
0.55
3.98
14.5
0.63
Average
1a
1b
7.94
>20.4
>10.2
4.68
0.40
3.31
0.37
0.32
—
2.19
3.47
1.05
0.33
0.49
>100
43.7
1.91
—
1.26
3.59
9.33
0.43
10.3
>84.5
29.5
6.3
>30.2
>269
251
>5.7
>1.1
10.9
0.4
1e
2b
21.4
—
2e
3
Melphalan
4.57
>97.7
26.9
2512
>2.63
118
20.0
38.0
66.1
—
>100
1.86
>100
6.17
ꢀ1.2
0.9
a The compounds were evaluated against several groups of human tumors including leukemia.
b The GI50 and IC50 values refer to the quantities of compounds required to inhibit 50% of the growth of the cells as explained in the text.
c The selectivity index (SI) figures are the differences between the highest and lowest GI50 values.
d The selectivity index (SI) figures refer to the quotients of the average GI50 value of all the cell lines and the average IC50 figure for leukemic cells.
Table 3. Evaluation of 1a–e, 2a–e, 3, and melphalan against normal and malignant cells
Compound
Normal cells, CC50 (lM)a
Tumor cells, CC50 (lM)a
HGF
HPC
HPLF
ave
HL-60
SIb
HSC-2
SIb
HSC-4
SIb
1a
1b
26
6.1
31
52
36
12
3.0
29
1.2
1.4
3.1
2.1
1.2
1.0
1.3
3.1
1.1
1.9
2.4
>5.7
33
1.1
1.1
1.3
0.9
1.2
0.8
1.0
2.3
1.2
0.9
1.5
>2.5
6.0
30
9.2
26
7.1
25
<3.1
1.6
>2.3
16
5.1
8.2
3.4
5.9
5.2
6.0
73
6.4
19
1c
1d
19.3
6.1
6.9
5.1
5.5
5.4
227
5.7
5.3
400
>200
8.1
9.2
5.2
12
7.0
7.0
5.0
7.7
225
6.8
4.7
397
>200
0.45
<3.4
0.41
<3.1
35
16
7.6
6.1
6.5
8.0
99
1e
2a
6.8
4.4
>2.1
12
2b
2c
5.8
>2.5
6.4
271
4.9
176
9.8
4.9
400
>200
2d
2e
<3.1
0.46
353
6.0
>2.2
10
6.3
2.5
165
35
5.8
5.2
264
81
4.0
390
>200
3
1.1
>33
Melphalanc
a The CC50 value is the concentration of the compound required to kill 50% of the cells. The figures quoted are the average of two independent
determinations which differed by less than 5%. The highest concentrations employed were 400 lM except solubility considerations precluded the use
of more than 200 lM of melphalan.
b The letters SI refer to the selectivity index which was obtained by dividing the average CC50 value for the three normal cells by the CC50 figure of
each malignant cell line.
c The data for melphalan was taken from Ref. 21 copyright (2007) with permission of Elsevier.
in Figure 3. In addition, 1e activated caspases-3, À8 and
À9 in HL-60 cells and caspase-3 in HSC-2 carcinomas
which are presented in Figure 3.
Hammett sigma, Hansch pi, and molar refractivity
(MR) constants of the aryl group. Negative correlations
were noted between the MR values of the aryl substitu-
ents and the IC50 figures in the CEM, Molt4/C8 and
L1210 bioassays (p < 0.05).Thus in developing series 1,
groups with greater size than the aryl substituents em-
ployed in 1a–e should be used. A similar analysis was
undertaken with 2a–e which revealed a positive correla-
tion between the r constants and the IC50 values in the
L1210 screen. This observation reveals that in the future
the placement of electron releasing groups in the aryl
ring in analogs of 2a–e should lead to compounds with
greater cytotoxic properties. No other correlations were
noted (p > 0.05).
3. Discussion
All of the compounds 1–3 were evaluated toward CEM
and Molt 4/C8 T-lymphocytes in order to discern the
ability of these compounds to inhibit the growth of
transformed cells of human origin. In addition, many
anticancer drugs are effective toward murine L1210
cells11 and therefore this bioassay was employed with
a view to detecting promising lead molecules. The data
are presented in Table 1. In particular, the potencies
of 2a and 2e having IC50 values of <2 lM in two-thirds
of the screens as well as 1c and 2d with IC50 figures be-
low 10 lM in most of the assays are noteworthy.
In view of the important contribution of steric factors to
cytotoxic potencies in series 1, consideration was given
to the possibility that the topography of the aryl ring
may also affect the magnitude of the IC50 values.
Accordingly the torsion angles (h) between the aryl rings
and the adjacent carbonyl group were calculated. In the
case of 1a–e, the h figures are À73.1, À69.5, À59.5,
À64.4, and À67.4, respectively. Linear and semilogarith-
mic plots between these figures and the IC50 values
The different cytotoxic potencies of 1a–e may have been
caused by the variation in the electronic, hydrophobic
and steric properties of the aryl substituents. Accord-
ingly linear and semilogarithmic plots were made be-
tween the IC50 values of 1a–e in each screen and the

5750
H. N. Pati et al. / Bioorg. Med. Chem. 16 (2008) 5747–5753
Figure 3. (A) Evaluation of the effects of 1e in HL-60 and HSC-2 cells (5 · 105 cells) on DNA fragmentation after incubation for 6 h with the
concentration indicated (in lM). M is the molecular weight marker of DNA. The arrow indicates a large DNA fragment. Reproducible results were
obtained when repeating the experiment. (B) The effect of 1e on the activation of caspases in neoplastic cells (4 · 104 cells) after incubation for 4 h.
Each value is the mean of two or three experiments.
revealed a negative correlation in the CEM test
(p < 0.05). Hence, the insertion of bulky substituents in
the ortho and meta positions of the aryl rings when cre-
ating further analogs of series 1 may lead to compounds
with increased potencies. The h values of 2a–e are 17.5,
24.3, 26.8, 30.7, and 15.7, respectively, but no correla-
tions between these figures and cytotoxic potencies were
noted (p > 0.05) confirming that no evidence was gener-
ated that steric factors contribute significantly to bioac-
tivity in series 2.
0.6, and 1.3 times the potencies of melphalan, respec-
tively, while the relevant figures for 2e are 2.2, 0.7, and
1.4, respectively. Clearly both of these aminoalcohols
are lead molecules which should be developed further.
Other compounds with 25% or more of the potencies
of melphalan are 1c (in all three screens), 1e (L1210 as-
say), and 2d (CEM and L1210 tests).
The biodata in Table 1, and also Tables 2 and 3 vide in-
fra, confirm the lack of cytotoxic potency of 3. However,
the concept of activated soft compounds has been pro-
posed which refers to the combination of a bioactive
molecule to an inert carrier moiety.12 Thus the attach-
ment of a pharmacophore to 3 such as esterification of
a cytotoxic acid via the hydroxy group of 3 may be a
worthwhile venture in the future.
The question of whether masking the enone structure of
1a–e by hydration yielding 2a–e led to greater cytotoxic
potencies was addressed by comparing the IC50 values of
compounds bearing the same aryl substituents in each
screen. Thus 1a was compared with 2a in the CEM as-
say, then in the Molt 4/C8 test and finally in the
L1210 screen and so forth. Standard derivations were ta-
ken into account. The results indicated that 2a > 1a and
1c > 2c in all three bioassays while 2e > 1e when consid-
ering CEM and Molt 4/C8 T-lymphocytes. In the
remaining cases equal potencies were observed. Hence
both the nature of the aryl ring and the general structure
of the series of the compounds need to be considered
when developing these compounds.
An important feature of antineoplastic agents is their
ability to display preferential cytotoxicity toward malig-
nant cells rather than normal tissues. Such a property
would be displayed by compounds which are not general
biocidal agents and therefore exert different potencies in
bioassays. Thus a comparison between the cytotoxicity
of the compounds toward transformed human CEM
and Molt 4/C8 cells was undertaken. The selectivity in-
dex (SI) values of each compound were calculated using
the ratio of the higher and lower IC50 figures and these
data are presented in Table 1. Compounds in which the
difference in IC50 values were more than doubled
(SI > 2) were noted, namely 1a, d, and all members of
A comparison was made between the IC50 values of the
most potent compounds with the alkylating agent mel-
phalan which is used in cancer chemotherapy. In the
CEM, Molt 4/C8 and L1210 assays, 2a possesses 1.7,

H. N. Pati et al. / Bioorg. Med. Chem. 16 (2008) 5747–5753
5751
series 2 except 2b. This result compares favourably with
melphalan with a SI figure of 1.3.
The SI values are clearly dependent on the cell line under
consideration. Thus the percentage of compounds which
possess SI figures greater than 2 is 100, 30, and 10,
respectively, when considering HL-60, HSC-2, and
HSC-4 cells, respectively. Of particular note are the SI
values of 10 or greater displayed by 1c, d, and 2a, e when
considering the greater toxicity toward HL-60 neo-
plasms than normal cells. The CC50 values against the
malignant cells vary considerably and in general the rel-
ative sensitivities to each chemical are HL-60 > HSC-
2 > HSC-4. This observation demonstrates further the
selective toxicity of these compounds which may reveal
a disparity in the lethal effects to normal and neoplastic
cells.
Compounds 1a, b, e, 2b, e, and 3 were evaluated
against approximately 54 human tumor cell lines
which were derived from the following groups of neo-
plasms namely leukemia, melanoma and non-small cell
lung, colon, central nervous system, ovarian, renal and
prostate cancers; with the exception of 1a, all com-
pounds were also evaluated using various malignant
breast cells.13 In this determination the quantity of
compound required to inhibit the growth of the cells
by 50% is computed. However in those cases where
50% inhibition of growth was not achieved at the
highest concentration, namely 100 lM, the figure of
100 lM was used in determining the average inhibi-
tory concentration. Hence, the term GI50 rather than
IC50 is utilized.
The final segment of this study was directed to obtaining
some understanding of the way in which cytotoxicity
was caused and why potencies to various malignant cells
differed. Many cytotoxic agents cause apoptosis,15
which can be due to activation of caspases. Treatment
of HL-60 cells by a representative compound 1e revealed
that internucleosomal DNA was induced using 0.5, 1,
and 2 lM of this Mannich base (Fig. 3A). At higher
concentrations of 4, 8, and 16 lM, this phenomenon
did not occur and only a smear pattern of DNA frag-
mentation was observed. On the other hand, there was
no unequivocal evidence of DNA fragmentation using
the same concentrations of 1e with HSC-2 cells. Figure
3B indicates that activation of caspases À3, À8, and
À9 occurred in HL-60 cells by 1e but this process was
virtually absent in HSC-2 cells. Thus 1e caused apopto-
tic cell death in HL-60 cells but the death of HSC-2 cells
is by alternative mechanisms. These results suggest that
some of the compounds in series 1 and 2 cause apopto-
sis. In addition, a compound may cause toxicity by dif-
ferent mechanisms depending on the cell line. This
phenomenon may be a contributing factor to the differ-
ent SI values observed in this project which further en-
hances the potential of these compounds for further
development.
The biodata are presented in Table 2. The average IC50
figures reveal that the compounds in series 1 and 2 are
potent cytotoxins in general. Of particular interest are
the two compounds in series 2 with GI50 figures of less
than 5 lM. Examination of the mean graphs14 revealed
that most of the compounds displayed a greater toxicity
to leukemic cells than other cell lines. With the exception
of 3, the potencies toward various leukemic cells are
noteworthy; specifically 2b has submicromolar IC50 val-
ues and 1a, b possess IC50 figures of less than 5 lM. All
of the compounds in series 1 and 2 demonstrated greater
antileukemic properties than melphalan. These evalua-
tions confirm the cytotoxic properties of series 1 and
2. In regard to selective toxicity for leukemic cells,
impressive SI values were obtained for 1a, 1b and in par-
ticular 2b.
This preferential toxicity for certain neoplastic cells
which have been observed may translate into a display
for selective toxicity to malignant rather than normal
cells. This possibility was addressed in the third bioas-
say involving both normal and malignant cells. All of
the compounds in series 1–3 were evaluated against
HGF, HPC, and HPLF normal cell lines as well as to-
ward HL-60, HSC-2, and HSC-4 neoplasms. These
data are presented in Table 3. The results confirm that
the compounds in series 1 and 2 are potent cytotoxins
toward malignant cells whereby 77% of the CC50 val-
ues are less than 10 lM. The HL-60 cells are the most
sensitive to 1a–e and 2a–e and in the case of 1d, 2a,
and 2e, the CC50 figures are submicromolar toward
this cell line. The CC50 values of most of the com-
pounds against HL-60, HSC-2, and HSC-4 cells are
lower than melphalan. Linear and semilogarithmic
plots were undertaken between the CC50 figures gener-
ated in the HL-60, HSC-2 and HSC-4 assays with the
r, p, and MR constants of the aryl substituents and
the h values of the compounds in series 1 and 2. Cyto-
toxic potencies were negatively correlated with the
MR constants in the HSC-2 screen while in the case
of series 2, a positive correlation was noted between
the r values in all three cell lines. These results are
similar to the observations made in reviewing the bio-
data in Table 1.
4. Conclusions
The biodata presented in Table 1 reveal that the amino-
alcohols 2a and 2e are clearly lead molecules in terms of
both cytotoxic potencies and SI values. When evalua-
tions of representative compounds against a number of
human tumor cell lines took place, 2b emerged as a note-
worthy candidate for future development in terms of po-
tency and selective toxicity especially to leukemic cells
(Table 2). An important feature of most of these com-
pounds in series 1 and 2 is their lethal effects toward pro-
myelocytic leukemic HL-60 cells as the figures in Table 3
reveal. In particular 1d, 2a, and 2e have submicromolar
CC50 values. Of considerable importance is the observa-
tion of the selective toxicity for HL-60 cells displayed by
various molecules, especially 1c, d and 2a, e with SI fig-
ures of 10 or more. A representative compound 1e
caused apoptosis in HL-60 but not HSC-2 cells. In addi-
tion, while 1e activated caspases À3, À8, and À9 in HL-
60 cells, this effect was virtually absent in HSC-2 cells.
This observation of different mechanisms of action is

5752
H. N. Pati et al. / Bioorg. Med. Chem. 16 (2008) 5747–5753
mandatory when compounds are to be developed which
are tumor-specific and spare normal tissues. This study
has demonstrated the need to develop these prototypic
molecules in series 1 and 2. In particular the very favour-
able properties of 2a and 2e both in terms of cytotoxic
potencies and selective toxicity to different cells should
be noted.
5.2.3. Evaluation of the compounds in series 1–3 against
normal and malignant human tumor cell lines. The meth-
odology used in assessing the cytotoxicity of various
compounds to the normal HGF, HPC, and normal
HPLF cell lines as well as neoplastic HL-60, HSC-2,
and HSC-4 cells has been described previously20 and re-
cently summarized.21
5.2.4. Evaluation of the ability of 1e to cause apoptosis
and activate caspases in HSC-2 and HL-60 cells. The
methodology of evaluating whether apoptosis and acti-
vation of caspases À3, À8, and À9 occurs have been pre-
viously described22 and summarized.21
5. Experimental protocols
5.1. Chemistry
5.1.1. Synthesis of compounds. The compounds in series
1–3 were prepared by the methodologies described
previously.10
Acknowledgments
5.1.2. Molecular modeling. Molecular modeling used a
BioMedCache programme.16
The authors thank the Canadian Institutes of Health
Research for an operating grant to J.R. Dimmock and
the Ministry of Education, Science, Sports and Culture
of Japan for a Grant-in-Aid (No. 14370607) to H. Saka-
gami. The Flemish Fonds voor Wetenschappelijk Ond-
erzoek provided funds to J. Balzarini and E. De
Clercq who thank Mrs. L. van Berckelaer for undertak-
ing the CEM, Molt 4/C8, and L1210 assays. The U.S.
National Cancer Institute kindly generated the biodata
which are presented in Table 2.
5.1.3. Statistical analyses. The Hammett sigma, Hansch
pi, and molar refractivity constants were obtained from
the literature.17 The linear and semilogarithmic plots
were made using a commercial statistics package.18
The significant p values (<0.05) that are generated from
the data presented in Table 1 when the IC50 values of the
compounds in series 1 were plotted against various
physicochemical parameters are as follows [bioassay,
physical constant, linear(l) or semilogarithmic(sl) plot
in parentheses]: 0.023 (CEM, MR, l), 0.005 (Molt4/C8,
MR, l), 0.025 (Molt4/C8, MR, sl), 0.015 (L1210, MR,
l), 0.046 (L1210, MR, sl), 0.040 (CEM, h, sl). In the case
of series 2, the relevant figure is as follows: 0.009 (L1210,
r, l).
References and notes
1. Pati, H. N.; Das, U.; Sharma, R. K.; Dimmock, J. R.
Mini-Rev. Med. Chem. 2007, 7, 131–139.
2. Mutus, B.; Wagner, J. D.; Talpas, C. J.; Dimmock, J. R.;
Phillips, O. A.; Reid, R. S. Anal. Biochem. 1989, 177, 237–
243.
3. Dimmock, J. R.; Raghavan, S. K.; Logan, B. M.; Bigam,
G. E. Eur. J. Med. Chem. 1983, 18, 248–254.
4. Okey, A. B.; Harper, P. A. In Principles of Medical
Pharmacology; Kalant, H., Grant, D. M., Mitchell, J.,
Eds., 7th Ed.; Elsevier: Canada, Toronto, 2007; p 902.
5. Dimmock, J. R.; Kumar, P.; Quail, J. W.; Pugazhenthi,
U.; Yang, J.; Chen, M.; Reid, R. S.; Allen, T. M.; Kao, G.
Y.; Cole, S. P. C.; Batist, G.; Balzarini, J.; De Clercq, E.
Eur. J. Med. Chem. 1995, 30, 209–217.
The related p value (<0.05) obtained from the CC50 fig-
ures of the compounds in series 1 evaluated against HL-
60, HSC-2, and HSC-4 cells is as follows: 0.045 (HSC-2,
MR, l). In the case of series 2, the relevant figures are as
follows: 0.035 (HL-60, r, l), 0.035 (HSC-2, r, l), 0.019
(HSC-2, r, sl), 0.036 (HSC-4, r, l), 0.017 (HSC-4, r, sl).
5.2. Bioassays
6. Ojugo, A. S. E.; McSheehy, P. M. J.; McIntyre, D. J. O.;
McCoy, C.; Stubbs, M.; Leach, M. O.; Judson, I. R.;
Griffiths, J. R. NMR Biomed. 1999, 12, 495–504.
7. Stubbs, M.; McSheey, P. M. J.; Griffiths, J. R.; Bashford,
C. L. Mol. Med. Today 2000, 6, 15–19.
8. Van den Berg, A. P.; Wike-Hooley, J. L.; Van den Berg-
Blok, A. E.; Van der Zee, J.; Reinhold, H. S. Eur. J.
Cancer Clin. Oncol. 1982, 18, 457–462.
5.2.1. Evaluation of 1a–e, 2a–e, and 3 against transformed
CEM and Molt 4/C8 T-lymphocytes and murine L1210
leukemic cells. The methodology for these bioassays
using CEM, Molt 4/C8 and L1210 cells has been de-
scribed previously.19 In brief, different concentrations
of each compound were incubated at 37 ꢁC for 72 h
(CEM and Molt 4/C8 T-lymphocytes) or 48 h (L1210
cells).
9. Albert, A. In Selective Toxicity, 7th ed.; Chapman and
Hall: London, 1985, pp. 642–643.
10. Pati, H. N.; Das, U.; Ramirez-Erosa, I. J.; Dunlop, D. M.;
Hickie, R. A.; Dimmock, J. R. Chem. Pharm. Bull. 2007,
55, 511–515.
11. Suffness, M.; Douros, J. In Methods in Cancer Research;
De Vita, V. T., Jr., Busch, H., Eds.; Academic Press, New
York, 1979; Vol. XVI, Part A, p. 84.
12. Bodor, N. Adv. Drug Res. 1984, 13, 255–331.
13. Boyd, M. R.; Paull, K. D. Drug Dev. Res. 1995, 34, 91–
109.
14. Grever, M. R.; Schepartz, S. A.; Chabner, B. A. Semin.
Oncol. 1992, 19, 622–638.
5.2.2. Examination of 1a, b, e, 2b, e, and 3 against
approximately 54 human tumor cell lines. A literature pro-
cedure was utilized when assaying various compounds
against a number of human tumor cell lines.13 The con-
centrations of the compounds were 10À8–10À4 M except
in the case of melphalan quantities of 10À7.6–10À3.6 M
were employed. The number of cell lines whose growth
was not inhibited by 50% at the maximum concentration
of 10À4 M was not achieved/ total number of cell lines
examined were 5/53 (1b), 4/49 (1e), and 54/55 (3).

H. N. Pati et al. / Bioorg. Med. Chem. 16 (2008) 5747–5753
5753
15. Tsurusawa, M.; Saeki, K.; Fujimoto, T. Int. J. Hematol.
1997, 66, 79–88.
16. BioMedCache 6.1 Windows, BioMedCache, Fujitsu
America, Inc., 2003.
Shirataki, Y.; Sakagami, H.; Satoh, K.; Nakashima, H.;
´
´ ´
Molnar, A.; Spengler, G.; Gyemant, N.; Ugocsai, K.;
Molnar, J. Phytother. Res. 2004, 18, 212–223.
´
21. Das, U.; Kawase, M.; Sakagami, H.; Ideo, A.; Shimada,
´
´
J.; Molnar, J.; Barath, Z.; Bata, Z.; Dimmock, J. R.
Bioorg. Med. Chem. 2007, 15, 3373–3380.
17. Hansch, C.; Leo, A. J. Substituent Constants for Correla-
tion Analysis in Chemistry and Biology; John Wiley and
Sons: New York, 1979, p. 49.
18. Statistical Package for Social Sciences, SPSS for Windows,
Standard Version, release 13.0, Chicago: SPSS Inc., 2004.
19. Baraldi, P. B.; Nunez, M. Del. C.; Tabrizo, M. A.; De
22. Takeuchi, R.; Hoshijima, H.; Onuki, N.; Nagasaka, H.;
Chowdbury, S. A.; Kawase, M.; Sakagami, H. Anticancer
Res. 2005, 25, 4037–4042.
23. Das, U.; Gul, H. I.; Alcorn, J.; Shrivastav, A.; George, T.;
Sharma, R. K.; Nienaber, R. K.; De Clercq, E.; Balzarini,
J.; Kawase, M.; Kan, N.; Tanaka, T.; Tani, S.; Werbovetz,
K. A.; Yakovich, A. J.; Manavathu, E. K.; Stables, J. P.;
Dimmock, J. R. Eur. J. Med. Chem. 2006, 41, 577–585.
´
Clercq, E.; Balzarini, J.; Bermejo, J.; Estevez, F.; Romagn-
oli, R. J. Med. Chem. 2004, 47, 2877–2886.
20. Motohashi, N.; Wakabayashi, H.; Kurihara, T.; Fukushi-
ma, H.; Yamada, T.; Kawase, M.; Sohara, Y.; Tani, S.;