[(2-Phenylindol-3-yl)methylene]propanedinitriles inhibit the growth of breast cancer cells by cell cycle arrest in G2/M phase and apoptosis

Article abstract of DOI:10.1016/j.bmc.2007.07.046

Cell cycle arrest of malignant cells is an important option for cancer treatment. In this study, we modified the structure of antimitotic 2-phenylindole-3-carbaldehydes by condensation with malononitrile. The resulting methylene propanedinitriles inhibited the growth of MDA-MB 231 and MCF-7 breast cancer cells with IC₅₀ values below 100 nM. Though they exhibited similar structure-activity relationships as the aldehydes, they did not inhibit tubulin polymerization but were capable of blocking the cell cycle in G₂/M phase. The cell cycle arrest was accompanied by apoptosis as demonstrated by the activation of caspases 3 and 9. Since the new 2-phenylindole derivatives also inhibited the growth of transplanted MXT mouse mammary tumors, they are interesting candidates for further development.

Full text of DOI:10.1016/j.bmc.2007.07.046

Available online at www.sciencedirect.com
Bioorganic & Medicinal Chemistry 15 (2007) 7368–7379
[(2-Phenylindol-3-yl)methylene]propanedinitriles inhibit
the growth of breast cancer cells by cell cycle arrest
in G2/M phase and apoptosis
a
a
a
a
´
Michaela Pojarova, Doris Kaufmann, Robert Gastpar, Tsuyuki Nishino,
Przemyslaw Reszka,b Patrick J. Bednarskib and Erwin von Angerera,*
aInstitut fu¨r Pharmazie, Universita¨t Regensburg, D-93040 Regensburg, Germany
bInstitut fu¨r Pharmazie, Universita¨t Greifswald, F.-L.-Jahnstr. 17, D-17487 Greifswald, Germany
Received 24 May 2007; revised 16 July 2007; accepted 25 July 2007
Available online 21 August 2007
Abstract—Cell cycle arrest of malignant cells is an important option for cancer treatment. In this study, we modified the structure of
antimitotic 2-phenylindole-3-carbaldehydes by condensation with malononitrile. The resulting methylene propanedinitriles inhibited
the growth of MDA-MB 231 and MCF-7 breast cancer cells with IC50 values below 100 nM. Though they exhibited similar struc-
ture–activity relationships as the aldehydes, they did not inhibit tubulin polymerization but were capable of blocking the cell cycle in
G2/M phase. The cell cycle arrest was accompanied by apoptosis as demonstrated by the activation of caspases 3 and 9. Since the
new 2-phenylindole derivatives also inhibited the growth of transplanted MXT mouse mammary tumors, they are interesting can-
didates for further development.
ꢀ 2007 Elsevier Ltd. All rights reserved.
1. Introduction
ever, were stable and displayed activities somewhat
lower than those of the parent aldehydes.1 These results
prompted us to convert the aldehyde into other stable
derivatives. In this study, we describe the syntheses
and biological characterization of the Knoevenagel con-
densation products of the aldehydes with malononitrile.
In previous studies,1,2 we found that 2-phenylindole-3-
carbaldehydes 1 (Fig. 1) with appropriate substituents
in the aromatic rings strongly inhibited the growth of
breast cancer cells. Investigations on the mode of action
revealed that tubulin is the primary target of these
agents.2 They bind to the colchicine binding site and pre-
vent the polymerization of the a/b-tubulin dimers to
functional microtubules which are required for vital
functions of the cell including mitosis. Though these
compounds were active in the nanomolar range, their
in vivo activity was disappointing probably due to the
presence of the metabolically unstable aldehyde func-
tion. Chemical modifications of this functional group
should render these molecules stable under physiological
conditions. One of first modifications performed com-
prised the conversion of the aldehyde function into
methyl imines, e.g. 2a, and oximes, e.g. 2b (Fig. 1).
The methyl imines proved to be simply prodrugs that re-
leased the aldehydes after hydrolysis.1 The oximes, how-
The main objective of this study was the search for
compounds which strongly inhibit tumor growth by a
blockade of the cell cycle. As shown for the 2-phenylin-
dole-3-carbaldehydes 1 and other antimitotic agents, the
inhibition of tubulin polymerization is accompanied by
a cell cycle arrest in the G2/M phase.1 This blockade
of the normal cell cycle renders the cell susceptible to
apoptosis. Cell cycle intervention has emerged as an
interesting new strategy to combat cancer.3,4 The first ap-
proved drugs which inhibit tumor growth by cell cycle ar-
rest in G2/M phase were the vinca alkaloids which inhibit
tubulin polymerization.5 More recently, natural products
such as the taxanes have been discovered which block the
cell cycle in the same phase by stabilizing the microtu-
bules.6 Other attractive targets are enzymes controlling
the cell cycle such as cyclin-dependent kinases.7,8 A large
number of heterocycle-based compounds have been
found to inhibit some of these kinases, especially cdk4
Keywords: Phenylindoles; Breast cancer; Cell cycle arrest; Apoptosis.
*
Corresponding author. Tel.: +49 (0) 941 9434821; fax +49 (0) 941
9434820; e-mail: erwin.von-angerer@chemie.uni-regensburg.de
0968-0896/$ - see front matter ꢀ 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2007.07.046

´
M. Pojarova et al. / Bioorg. Med. Chem. 15 (2007) 7368–7379
7369
H
H
X
O
CH2OH
R1
R2
Bu
R3
OMe
N
H
N
N
H
N
N
H
H
H
Indole-3-carbinol
1a-s
2a X = NMe
2b X = NOH
3
R1 = H, alkyl, F, OCH3
R2 = H, Cl, F, OCH3
R3 = H, alkyl, OCH3, CF3
H
N
CN
Cl
NC
O
H
O
R1
R2
HO
H2NO2S
R3
N
H
MeO
NH
N
NH
S
N
H
H
O
Cl
O
O
N
H
6a-s
4
Indirubin
5
R1 = H, alkyl, F, OCH3
R2 = H, Cl, F, OCH3
R3 = H, alkyl, OCH3, CF3
Figure 1. Examples for indole-derived agents that inhibit cell cycle progression through different mechanisms.
and cdk6. Since we are mainly interested in indole deriv-
atives, a selection of those agents containing this struc-
tural element are presented in Figure 1.
elaborated.1 We decided to select the same substituents
because we anticipated a similar mode of action for
the propanedinitriles 6. The conversion of the aldehydes
1 to the corresponding methylene propanedinitriles 6
was readily accomplished by the reaction with malono-
nitriles (Fig. 2). This structural modification should
not alter the physico-chemical properties of the indole
core much and would lead only to minor steric effects.
We also have synthesized some analogues of the methy-
lene propanedinitriles in which one of the nitrile groups
had been replaced by hydrogen, carboxamide, or car-
boxylate groups, but all these modifications abolished
or strongly decreased the activity of the indole deriva-
tives. Thus, the data for these compounds are not
presented.
The simplest example is indole-3-carbinol, a natural
constituent of crucifer plants. It induces the arrest of
the cell cycle in G1 phase and inhibits the growth of
breast cancer cells.9,10 Its major acid-catalyzed reaction
product 3.30-diindolylmethane (3) also interferes with
the cell cycle progression of breast cancer cells.11–13
Other indole-based inhibitors of cyclin-dependent ki-
nases are indirubin and derivatives.14–17 Cell cycle con-
trol in G1 phase can also be achieved with the indolyl
sulfonamide E7070 (4),18,19 whereas the diindolylmeth-
ylene derivative 5 blocks the cell cycle in G2/M phase.20
In the present study, we converted the 2-phenylindole-3-
carbaldehydes 1 which block the cell cycle in G2/M
phase by inhibition of tubulin polymerization to (2-
2.2. Anti-proliferative activity
All of the compounds synthesized were first evaluated
for anti-proliferative activity using hormone-indepen-
dent human MDA-MB 231 breast cancer cells in a
microplate assay. As one of the reference compounds
the unsubstituted [(2-phenylindol-3-yl)methylene]prop-
anedinitrile (6a) was used. It strongly inhibits the growth
of these cells with an IC50 value of 0.43 lM (Table 1).
Based on previous investigations with methoxy-substi-
tuted 2-phenylindole-3-carbaldehydes, a series of com-
pounds with a methoxy group in the para-position of
the phenyl ring and a variety of lipophilic substituents
in positions 5 and/or 6 of the indole including halogens
and alkyl groups of variable length were tested. All 2-(4-
methoxyphenyl)indole derivatives strongly inhibited the
growth of MDA-MB 231 cells with IC50 values below
1 lM (Table 1). The potency of the unsubstituted indole
6a was significantly increased when substituents had
been introduced that add a substantial amount of lipo-
philicity to the parent molecule as exemplified by the
5-alkyl derivatives 6l–n. Elongation of the alkyl chain
from methyl to pentyl decreased the IC50 value from
phenylindol-3-yl)methylenepropanedinitriles
6
and
determined their anti-proliferative activities in MDA-
MB 231 and MCF-7 breast cancer cells. The effects on
the cell cycle were analyzed by flow cytometry. Since a
cell cycle arrest may lead to an apoptotic cell death,
markers of apoptosis such as the activation of caspases
3 and 9 were determined for two representative exam-
ples. A major goal of this study was the discovery of po-
tent antitumor agents that retain their activity in vivo.
Thus, one of the compounds synthesized was evaluated
in the MXT mouse mammary tumor model.
2. Results and discussion
2.1. Chemistry
In previous investigations the most favorable substitu-
tion pattern of the starting 2-phenylindole-3-carbalde-
hydes 1 in respect to antitumor activity had been

´
M. Pojarova et al. / Bioorg. Med. Chem. 15 (2007) 7368–7379
7370
CN
H
NC
O
H
R1
R2
EtOH, piperidine,
reflux, 2 h
R1
R2
CN
CN
R3
R3
H2C
+
N
N
H
H
4a-s
6a-s
R1 = H, alkyl, F, OCH3
R2 = H, Cl, F, OCH3
R3 = H, alkyl, OCH3, CF3
Figure 2. Synthesis of [(2-phenylindol-3-yl)methylene]propanedinitriles 6a–s.
Table 1. Anti-proliferative effects of [(2-phenylindol-3-yl)methy-
lene]propanedinitriles 6a–s on human MDA-MB 231 and MCF-7
breast cancer cells
120
100
80
6j
6h
6l
CN
6n
6m
Vin.
60
NC
H
40
R1
20
R3
0
R2
Compound R1
N
1.00E-09
1.00E-08
1.00E-07
1.00E-06
1.00E-05
H
Concentration [mol/L]
R2
R3
MDA-MB 231a MCF-7b
IC50 (lM)
IC50 (lM)
Figure 3. Inhibitory effects of [[5-alkyl-2-(4-methoxyphenyl)indol-3-
yl]methylene]propanedinitriles 6 with variable alkyl chain length on the
growth of MDA-MB 231 breast cancer cells. Inhibition of cell growth
was determined after incubation for 4 days and subsequent crystal
violet staining of viable cells. Vincristine (Vin) was used as reference
drug.
6a
6b
H
H
H
H
H
0.43
0.19
0.77
0.34
0.30
0.31
0.22
0.25
0.20
0.20
0.19
0.31
0.013
0.076
0.025
0.10
0.20
0.15
0.10
0.068
n.d.c
H
OMe 0.72
OMe 0.59
6c
6d
OMe
H
OMe OMe 0.26
6e
F
H
F
OMe 0.40
OMe 0.28
6f
6g
H
H
Me
OMe Me
0.18
6h
6i
H
Cl
H
H
H
H
H
H
H
H
H
H
OMe 0.28
OMe 0.075
OMe 0.083
OMe 0.21
OMe 0.026
OMe 0.042
OMe 0.046
6-methoxy-2-(4-methoxyphenyl)indole derivative 6d
dropped by one order of magnitude when the ether func-
tion in 6-position was cleaved (data not shown).
Me
6j
6k
n-Pr
iPr
6l
6m
n-Bu
n-Pentyl
n-Hexyl
n-Bu
n-Bu
n-Bu
n-Pentyl
n-Hexyl
All [(2-phenylindol-3-yl)methylene]propanedinitriles 6
were also tested for anti-proliferative activity in MCF-
7 breast cancer cells (Table 1). Since similar activities
were observed in both cell lines an involvement of the
estrogen receptor, present in MCF-7 cells, can be ruled
out. The differences in activity were usually within the
statistical errors of both assays (ꢀ20%) except for a
small number of compounds whose IC50 values differed
by a factor of 2 or more (6a, 6i, 6j, 6l, 6o–q, 6s). There
was no general preference in sensitivity between these
two cell lines though a slight tendency to higher IC50
values in MCF-7 cells was noticed.
6n
6o
Me
Et
0.065
0.076
6p
6q
CF3 0.056
CF3 0.078
CF3 0.15
0.0045
6r
6s
Vincristine
a Inhibition of cell growth determined after incubation for 4 days and
subsequent crystal violet staining of viable cells. Mean values of two
independent experiments with 16-24 replicates, SD are generally less
than 25%.
b Analogous experiment as described for MDA-MB 231 cells with one
exception: The incubation period was 5 days.
c Not determined.
The comparison of the inhibitory effects of the methy-
lene propanedinitriles 6 with those of the corresponding
2-phenylindole-3-carbaldehydes 1 revealed that the lat-
ter are more active by a factor of 3–10.
0.28 to 0.042 lM (Fig. 3). A similar observation was
made with 2-phenylindoles furnished with a methyl,
ethyl, or trifluoromethyl group in para-position of the
phenyl ring (6o–s). The IC50 values are only one order
of magnitude higher than that of the established antimi-
totic drug vincristine. When the methoxy group in this
ring was shifted to the 3-position the activity decreased
from 0.26 to 0.97 lM (IC50 values). A polar character
of the substituents in the indole moiety was detrimental
to the anti-proliferative potency. The IC50 value of the
2.3. Cell cycle arrest in G2/M-phase
Previous studies with 2-phenylindole-3-carbaldehydes 1
have shown that these aldehydes exert their cytotoxic ef-
fects through inhibition of tubulin polymerization and
cell cycle arrest in G2/M-phase. Thus it was of interest
to study the effect of the [(2-phenylindol-3-yl)methy-
lene]propanedinitriles 6 on the cell cycle. In this assay,

´
M. Pojarova et al. / Bioorg. Med. Chem. 15 (2007) 7368–7379
7371
the cell cycle dependent DNA content of MDA-MB 231
2.4. Inhibition of tubulin polymerization
cells was determined by flow cytometry using propidium
iodide in permeabilized cells. First, one of the most ac-
tive methylene propanedinitriles (6n) was studied for
its effect on the cell cycle in comparison to the corre-
sponding aldehyde 1n and vincristine. Treatment of
the cells with any of these three agents led to a decrease
of the peak for cells in the G1/G0 phase and the parallel
increase of the number of cells in G2/M phase due to the
blockade of the cell cycle in G2/M phase (Fig. 4). This
change in the cell cycle distribution was accompanied
by the appearance of significant quantity of cells with
a DNA content lower than that in G1/G0 cells (sub-G1
fraction). The comparison revealed that 6n exerted the
same effect as 1n but required a tenfold higher concen-
tration (0.05 vs 0.5 lM). Vincristine was even more
active and blocked the cell cycle at a 10-nM concentra-
tion (Fig. 4).
One of the possible explanations for the cell cycle arrest
is the inhibition of tubulin polymerization to functional
microtubules as it is observed for antimitotic agents such
as colchicine and combretastatin A-4. Therefore, two
methylene propanedinitriles (6c and 6f) and the corre-
sponding 2-phenylindole-3-carbaldehydes 1c and 1f
were tested for their inhibitory activity on the polymer-
ization of tubulin isolated from calf brains. For compar-
ison the aldehydes 1a and 1b and colchicine were
included in this assay. Cytotoxicities were determined
in two breast cancer cell lines which differ by their estro-
gen receptor status (ER- MDA-MB 231, ER+ MCF-7)
and the vulva carcinoma cell line A-431 which is charac-
terized by high levels of EGF receptors.
The progression of tubulin polymerization was mea-
sured turbidimetrically at 350 nm in cuvettes over a per-
iod of 20 min after the temperature had been raised from
2 to 37 ꢁC. At a concentration of 40 lM, colchicine and
the aldehydes 1b, 1c, and 1f suppressed tubulin polymer-
ization, whereas both methylene propanedinitriles were
inactive (Table 2). The data on cytotoxicity determined
in three different cell lines revealed that the inhibition
of tumor cell growth by the derivatives 6c and 6f does
not depend on the blockade of tubulin polymerization.
In this respect the methylene propanedinitriles 6c and
6f resemble the unsubstituted aldehyde 1a, which also
inhibits cell growth without affecting tubulin polymeri-
Five more derivatives with different cytostatic potencies
were examined for their effects on the cell cycle progres-
sion in increasing concentrations. Already at very low
concentrations a significant sub-G1 fraction appeared
probably due to apoptotic processes in cells of the G1
(G0) pool. Sometimes this phenomenon became evident
before cell cycle arrest occurs (6f in Fig. 5a, 6p in
Fig. 5c). According to the reduced number of cells in
G1/G0 phase, this peak decreases at higher concentra-
tions. Usually blockade of the cell cycle is observed at con-
centrations five- to tenfold higher than the IC50 values.
Control
Vin.; 0.01 µM
Vin.; 0.1 µM
b
c
a
d
0
0
FL3-H
1023
FL3-H
0
1023
1023
FL3-H
1n; 0.01 µM
1n; 0.05 µM
1n; 0.1 µM
0
0
1023
0
1023
1023
FL3-H
FL3-H
FL3-H
6n; 0.5 µM
6n; 1.0 µM
6n; 0.1 µM
0
1023
0
FL3-H
FL3-H
1023
1023
0
FL3-H
Fluorescence channels
Figure 4. Flow cytometry analysis of cell cycle. MDA-MB 231 breast cancer cells were exposed to compounds 1n, 6n, and vincristine in various
concentrations for 24 h. The DNA content was quantified by the standard propidium iodide procedure, as described in the Experimental Section.
Cells are assigned to those in G0/G1- (label b), S- (label c), and G2/M-phase (label d), and to sub-G1 cells (label a), respectively, according to their
DNA content.

´
M. Pojarova et al. / Bioorg. Med. Chem. 15 (2007) 7368–7379
7372
80
70
60
50
40
30
20
10
0
caspase 3 and 9 substrates was used as biochemical mar-
sub-G1
G1/G0
S
ker of apoptosis. Two representative examples of the
methylene propanedinitriles (6m and 6q) were tested
for their ability to increase caspase activity within 24 h
in HL-60 human leukemia cells. Both compounds in-
duce caspase activity at a 1-lM concentration with
somewhat different effects on the cleavage of the caspase
3 and 9 substrates (Fig. 6). The induction of caspase 9-
like activity is much stronger than that observed for the
reference antitumor agent etoposide, even when etopo-
side is applied at an eightfold higher concentration.
Strongly enhanced caspase activities of 6m and 6q were
also observed at 24 h at a concentration of 0.3 lM (data
not shown). The indole derivatives also inhibit the
growth of the HL-60 cells after a 48-h treatment with
IC50 values of 41 6.5 nM (6m) and 70 11 nM (6q),
respectively, as determined by the MTT assay.22 In this
assay, the clinically used anticancer agent etoposide dis-
played an IC50 value of 315 72 nM.
G2/M
ctrl
6f
6f
6f
6f
6f 6o 6o 6o
0.01 0.05 0.1 0.5 1.0 0.1 0.5 1.0 µM
80
sub-G1
G1/G0
S
70
60
50
40
30
20
10
0
G2/M
Although these studies provide strong evidence that
apoptosis is the mechanism of cell death for the title
compounds, necrosis as the competing action cannot
be ruled out. Therefore, we stained HL-60 cells that
were used for the induction of caspases with trypan
blue, a procedure that is applied to detect necrotic
cells23 although the uptake of trypan blue can occur
in both apoptotic and necrotic cells.24 Staining of the
HL-60 cells with trypan blue after a 24-h treatment per-
iod with etoposide (8 lM), 6m (1 lM), and 6q (1 lM)
showed the occurrence of many cells that have taken
up the dye though roughly the same number of cells re-
mained unstained. Untreated control cells showed
hardly any uptake of the dye. Since staining of cells
with trypan blue does not allow a clear distinction to
be made between necrosis and apoptosis, other meth-
ods are required to quantify the number of cells under-
going apoptosis such as counting the cells with nuclear
condensation23 or the addition of cycloheximide which
blocks the synthesis of proteins required for apoptotic
processes.24 Nevertheless, the pictures obtained by light
microscopy of HL-60 cells treated with 1.0 lM of 6m or
6q showed similar morphologies to those treated with
8.0 lM etoposide after 24 h (see Supplementary
material).
ctrl Vin Vin 6h 6h 6h 6n 6n 6n
0.01 0.1 0.5 1.0 5.0 0.1 0.5 1.0 µM
80
70
60
50
40
30
20
10
0
sub-G1
G1/G0
S
G2/M
ctrl Vin 6p
6p
6p
6q 6q
6q
0.1 0.05 0.1 0.5 0.05 0.1 0.5 µM
Figure 5. Cell cycle distribution of MDA-MB 231 cells treated for 24 h
with [(2-phenylindol-3-yl)methylene]propanedinitrile 6f, 6o (panel a),
6h, 6n (panel b), and 6p, 6q (panel c) in various concentrations.
Vincristine (Vin) was used as reference drug. Percentages of sub-G1
cells and cells in G1/G0-, S-, and G2/M-phase are shown. Data refer to
a representative experiment out of two.
zation at relevant concentrations. These results make an
alternative mode of action for the latter compounds
likely.
2.6. Inhibition of MXT mouse mammary tumors
2.5. Activation of caspases
Preliminary studies of the aldehydes 1 in a MXT mouse
mammary tumor model revealed that the antitumor
activity observed in vitro is lost in vivo. This lack of
in vivo activity can be rationalized by the presence of
a metabolically labile aldehyde function. Thus, the alde-
hydes were converted to derivatives with a modified
functional group. Since the indole derivatives 6 have re-
tained their in vitro activities, a representative com-
pound of this series (6f) was evaluated in vivo. As test
model the hormone-independent transplantable MXT
mouse mammary tumor was used. Treatment was
started 24 h after transplantation and lasted two weeks.
At the end of treatment the tumor size of treated ani-
mals was compared with that of untreated control
animals.
The inhibition of cell growth by the indole derivatives 6
is accompanied by the occurrence of a substantial sub-
G1 fraction in the cell cycle analysis. This effect is prob-
ably due to apoptotic processes induced in G1 and/or G2
phase. This observation is in accord with microscopic
images of the cell nucleus taken after treatment with 6f
(0.1 lM). They show blackberry-like structures and a
fragmentation pattern of the nuclei typical for apoptosis
(image not shown). The apoptotic path to cell death is
characterized by the release of cytochrome c from the
mitochondria, the induction of various caspase enzymes,
condensation of the chromosomes, and the fragmenta-
tion of nuclear DNA.21 In this study, the cleavage of

´
M. Pojarova et al. / Bioorg. Med. Chem. 15 (2007) 7368–7379
7373
Table 2. Effects of 2-phenylindole-3-carbaldehydes 1a-c and 1f, and of [(2-phenylindol-3-yl)methylene]propanedinitriles 6c and 6f on the growth of
human cancer cells and tubulin polymerization
X
H
R1
R2
R3
N
H
Compound
R1
R2
R3
X
MDA-MB 231a IC50 (lM)
MCF-7b IC50 (lM)
A-431c IC50 (lM)
TPd % T/C
1a
1b
H
H
H
H
H
F
H
O
0.42 0.07
0.47 0.02
0.26 0.03
0.59 0.06
0.047 0.006
0.28 0.07
0.03 0.01
0.65 0.08
0.18 0.04
0.18 0.07
0.27 0.05
0.043 0.009
0.22 0.09
n.t.e
0.76 0.07
0.21 0.06
0.20 0.12
0.34 0.06
0.18 0.09
0.19 0.04
n.t.e
98
3
7
4
4
5
5
6
5
H
OMe
OMe
OMe
OMe
OMe
O
1c
6c
OMe
OMe
H
O
7
C(CN)2
O
C(CN)2
100
8
1f
6f
Colchicine
H
F
99
1
a Inhibition of cell growth determined after incubation for 4 days and subsequent crystal violet staining of viable cells. Mean values of four
independent experiments with 16–24 replicates SD.
b The incubation period was 5 days.
c The incubation period was 3 days.
d Tubulin polymerization in the presence of inhibitor (40 lM); controls without inhibitor = 100%.
e Not tested.
800
600
400
200
0
mean
median
1800
1500
1200
900
600
300
0
500
400
300
200
100
0
t = 0 h 6m
6q Etop.
t = 0 h 6m
6q Etop.
Figure 6. Effects of 6m (1.0 lM), 6q (1.0 lM), and etoposide (8 lM) on
caspase 3-like (panel a) and caspase 9-like (panel b) activities of HL-60
cells after a 24-h exposure. Controls show caspase activities at the
beginning of treatment. Bars represent means of three independent
experiments SD. All increases are statistically significant (a, p < 0.02;
b, p < 0.001).
Control
Cisplatin
2 mg/kg
6f
6f
10 mg/kg 30 mg/kg
Figure 7. Inhibitory effects of indole 6f and cisplatin on the growth of
transplanted MXT mouse mammary tumors after two weeks of
treatment. Each point of the scatter diagram represents the tumor area
of one animal. The stated doses were administered subcutaneously
three times per week.
The methylene propanedinitrile 6f showed a significant,
dose-dependent inhibition of tumor growth (Fig. 7). The
effect was comparable to that of doxorubicin.25 The ef-
fect of the reference drug cisplatin was not reached in
this experiment because cisplatin was administered in a
sub-toxic dose (2 mg/kg) which led to a 4.5% reduction
of body weight, whereas the animals showed no sign
of toxicity when treated with 10 or 30 mg of 6f per kilo-
gram body weight (Table 3).
Table 3. Effect of 6f and cisplatin on the growth of hormone-resistant
MXT mouse mammary tumors
Compound Dosea Tumor areab % T/Cc Change of body
(mg)
(mm2)
weightd (%)
—
6f
—
10
30
2
294 75
218 62
164 52
+4.5
+3.6
+2.3
ꢁ4.5
74e
56f
3f
2.7. Discussion
6f
Cisplatin
8
8
The aim of this study was the chemical modification of
the antimitotic 2-phenylindole-3-carbaldehydes 1 to
derivatives with in vivo activity. Since the only reactive
group in 1 is the aldehyde function, its conversion to a
methylene propanedinitrile structure should render the
molecule stable to metabolic transformations that re-
duce the bioactivity of the parent structure. The results
a Dose/kg body weight, administered three times per week.
b Means SD after two weeks of treatment.
c Tumor area of treated animals/tumor area of control animals · 100.
d Difference of mean body weights between the first and the last day of
treatment.
e Significant inhibition of tumor growth (p < 0.05).
f Significant inhibition of tumor growth (p < 0.01).

´
M. Pojarova et al. / Bioorg. Med. Chem. 15 (2007) 7368–7379
7374
of this study showed that the carbonyl oxygen in the
2-phenylindole-3-carbaldehydes can be replaced by
the dicyanomethylene group without abolishing the
anti-proliferative activity on cancer cells. It was also
interesting to note that the different substituents in the
methylene propanedinitriles 6 had a similar influence
on the antitumor activity as in the aldehyde series.
The similarity of the structure–activity relationships
indicates a common mode of action for both aldehydes
1 and the corresponding methylene propanedinitriles 6,
but when we investigated the mode of action of the
methylene propanenitriles 6 we found that they must
act through a different mechanism. The aldehydes 1
were shown to inhibit tubulin polymerization and this
action was assumed to be responsible for the strong
antimitotic effects of the aldehydes. The fact that the
propanedinitriles and aldehydes share the same struc-
ture–activity profile, and the significantly higher activity
of most the aldehydes can be rationalized by a common
mode of action for both types of indole derivatives, but
must encompass an additional effect of the aldehydes on
tubulin polymerization. This assumption is supported
by the data for the unsubstituted aldehyde 1a and its
40-methoxy derivative 1b. Both show similar cytostatic
activities but only 1b inhibits tubulin polymerization.
Obviously, the methoxy group in para-position of the
phenyl ring plays a crucial role in the interaction with
tubulin. When the methoxy group is located in the 6-po-
sition of the indole rather than in para-position of the
phenyl ring, no inhibition of tubulin polymerization
was observed.2
An important aspect of the application of anti-prolifer-
ative agents as anticancer drugs is their in vivo activity.
One of the compounds of this study was used to treat
mice bearing transplanted, hormone-independent
MXT mammary tumors. A significant inhibition of tu-
mor growth was achieved without obvious adverse ef-
fects. Though these data are derived from a single
experiment and are preliminary in character they can
be considered as proof of principle. These indole deriv-
atives exhibit in vivo activity which remains to be
optimized.
3. Conclusions
A series of 2-phenylindole-3-carbaldehydes 1, which
have previously been characterized as antimitotic agents
of high potency, were converted to the corresponding
methylene propanedinitriles 6 to render them stable to
possible metabolic transformations of the aldehyde
function. The in vitro assays on anti-proliferative activ-
ity in various cell lines revealed that the potencies of
these 2-phenylindole derivatives decreased only slightly
upon this structural modification. The increase of the
IC50 values in human MDA-MB 231 and MCF-7 breast
cancer cells was generally less than the factor of 5.
Though the methylene propanedinitriles share the same
structure–activity relationship with the aldehydes, their
molecular target(s), however, might be different because
they do not interfere with the polymerization of tubulin,
the dominant molecular action of the 2-phenylindole-3-
carbaldehydes. Similar as the aldehydes, the methylene
propanedinitriles block the cell cycle in G2/M phase
and drive the tumor cells into apoptosis as demonstrated
by the strong increase of caspase 3 and 9 activities. The
question whether the necrotic cell death also plays a role
cannot yet be clearly answered. An important aspect of
this study was the in vivo activity of this class of antimi-
totic agents. Thus, the derivative 6f was evaluated
in vivo for antitumor activity. It significantly inhibited
the growth of transplanted MXT mouse mammary tu-
mors at doses between 10 and 30 mg/kg body weight.
This in vivo activity was not paralleled by the corre-
sponding aldehydes. The biological characteristics of
these new compounds make them interesting candidates
for further development as anticancer agents.
The precise mechanism by which the methylene propan-
edinitriles 6 inhibit cell growth remains unclear. These
agents block the cell cycle in G2/M phase and drive
the cells into an apoptotic cell death. This process is
probably mediated by the activation of caspases such
as caspases 3 and 9. Both enzyme activities are strongly
stimulated at concentrations 4- to 20-fold higher than
IC50 and much lower than those required for the anti-
cancer drug etoposide. These findings are in accord with
the observed fragmentation of nuclear DNA at submi-
cromolar concentrations. From the data obtained, the
growth inhibitory effect of the indoles 6 can be rational-
ized as the result of an apoptotic cell death following the
cell cycle arrest in G2/M phase. However, the reason for
the cell cycle blockade is not evident. An interaction
with tubulin as found for the corresponding aldehydes
can be excluded. Another target we considered was the
EGF receptor associated PTK because the structure of
the methylene propanedinitriles 6 resembles some tyr-
phostins.26 However, an inhibition of EGF receptor
phosphorylation which would cause a cell cycle arrest
in the late G1 phase27 was not observed. Obviously,
the tyrphostins require free phenolic hydroxyl groups
for their action on EGF receptors. We also checked
some other protein tyrosine kinases from human placen-
tal tissue, but found no activity. Whether CDK or cyc-
lins are involved in the observed cell cycle blockade
remains to be investigated. The chemical structure of
the indole-based methylene propanedinitriles, however,
exhibits hardly any similarities with known inhibitors
of CDKs.3
4. Experimental
4.1. General methods
Melting points were determined on a Buchi 510 appara-
¨
tus and are uncorrected. 1H NMR spectra were recorded
in d6-DMSO on AC-250, AVANCE300 and
AVANCE400 spectrometers (Bruker) with TMS as inter-
nal standard and were in accord with the assigned struc-
tures. Mass spectra were obtained with a MAT-311A
spectrometer (Varian). Purity of all compounds was
checked by TLC. Elemental analyses were performed
by the Mikroanalytisches Laboratorium, University of
Regensburg. The syntheses of the starting aldehydes
1a–d,2 1f,2 1e, and 1g–s1 have been described previously.

´
M. Pojarova et al. / Bioorg. Med. Chem. 15 (2007) 7368–7379
7375
4.2. Preparation of [(2-phenylindol-3-yl)methylene]prop-
anedinitriles 6
Anal. for C19H12FN3O; calcd C, 71.92; H, 3.81, N,
13.24; found: C, 72.53; H, 4.89; N, 12.32.
To a suspension of the 2-phenylindole-3-carbaldehyde 1
(50 mmol) and malonic acid dinitrile (55 mmol) in EtOH
(5 mL) were added 2–3 drops of piperidine. The mixture
was then heated under reflux for 2 h. After cooling H2O
(50 mL) was added slowly. The crystalline precipitate
was separated by filtration and purified by recrystalliza-
tion from EtOH or by column chromatography (SiO2)
with CH2Cl2/EtOAc mixtures as the eluent.
4.2.7. [[6-Methoxy-2-(4-methylphenyl)indol-3-yl]methy-
lene]propanedinitrile (6g). Yellow crystals (47% yield),
mp 258 ꢁC (EtOH). 1H NMR d 2.45 (s, 3H, –CH3);
3.87 (s, 3H, OCH3); 7.14 (dd, 3J = 8 Hz, 4J = 2 Hz,
1H, indole-H5); 7.43 (d, 4J = 2 Hz, 1H, indole-H7);
3
7.20, 7.60 (AA0BB0, J = 9 Hz, 4H, phenyl-H); 7.90 (s,
2H, vinyl-H and indole-H4); 12.86 (s, 1H, N-H). Anal.
for C20H15N3O; calcd C, 76.66; H, 4.82; N, 13.41;
found: C, 76.57; H, 4.92; N, 12,62.
4.2.1. [(2-Phenylindol-3-yl)methylene]propanedinitrile (6a).
1
Yellow crystals (16% yield), mp 217–219. H NMR d
4.2.8. [[2-(4-Methoxyphenyl)-5-methylindol-3-yl]methy-
lene]propanedinitrile (6h). Yellow solid (71% yield), mp
7.20–7.74 (m, 9H, ArH); 7.93 (s, 1H, vinyl-H); 7.98–
8.14 (m, 1H, indole-H4). Anal. for C18H11N3; calcd C,
80.28; H, 4.12; N, 15.6; found: C, 79.42; H, 3.99; N,
16.03.
1
260 ꢁC (EtOH). H NMR d 2.43 (s, 3H, –CH3); 3.84
(s, 3H, OCH3); 6.94–7.01 (m 2H, indole-H7,6); 7.45,
7.53 (AA0BB0, 3J = 9 Hz, 4H, phenyl-H); 7.86 (s, 1H, vi-
nyl-H); 8.00 (s, 1H, indole-H4); 12.86 (s, br, 1H, N-H).
Anal. for C20H15N3O; calcd C, 76.66; H, 4.82; N,
13.41; found: C, 76.04; H, 4.68; N, 13.22.
4.2.2. [[2-(4-Methoxyphenyl)-indol-3-yl]methylene]prop-
anedinitrile (6b). Yellow crystals (85% yield), mp 210 ꢁC
(EtOH). 1H NMR d 3.87 (s, 3H, OCH3); 7.20, 7.61
(AA0BB0, 3J = 9.0 Hz, 4H, phenyl-H); 7.25–7.39 (m,
2H, indole-H); 7.51–7.56 (m, 1H, indole-H); 7.93 (s,
1H, vinyl-H); 8.02–8.09 (m, 1H, indole H); 12.93 (s,
1H, N-H). Anal. for C19H13N3O; calcd C, 76.24; H,
4.38; N, 12.04; found: C, 76.08; H, 5.18; N, 12.44.
4.2.9. [[6-Chloro-2-(4-methoxyphenyl)-5-methylindol-3-
yl]methylene]propanedinitrile (6i). Yellow solid (47%
1
yield), mp 234 ꢁC (EtOH). H NMR d (ppm) = 2.45 (s,
3H, –CH3); 3.87 (s, 3H, OCH3); 7.20, 7.60 (AA0BB0,
3J = 9 Hz, 4H, phenyl-H); 7.54 (s, 1H, indole-H4); 7.96
(s, 1H, indole-H7); 8.00 (s, 1H, vinyl-H); 12.31 (s, br,
1H, N-H). Anal. for C20H14ClN3O; calcd C, 69.07; H,
4.06; N, 12.08; found: C, 68.21; H, 4.42; N, 11.63.
4.2.3. [[5-Methoxy-2-(4-methoxyphenyl)-indol-3-yl]meth-
ylene]propanedinitrile (6c). Yellow crystals (89% yield),
1
mp 257 ꢁC (EtOH). H NMR d 3.84 (s, 3H, OCH3);
3
4
3.88 (s, 3H, OCH3); 6.97 (dd, J = 8.8 Hz, J = 2.4 Hz,
1H, indole-H6); 7.20, 7.59 (AA0BB0, 3J = 8.8 Hz, 4H,
phenyl-H); 7.44 (d, 3J = 8.8 Hz, 1H, indole-H7); 7.61
4.2.10. [[2-(4-Methoxyphenyl)-5-n-propylindol-3-yl]meth-
ylene]propanedinitrile (6j). Yellow crystals (46% yield),
1
3
mp 180 ꢁC (EtOH). H NMR d 0.91 (t, J = 7 Hz, 3H,
–CH2–CH3); 1.67 (sext, 3J = 7 Hz, 2H, –CH2–CH2–
CH3); 2.69 (t, 3J = 7 Hz, 2H, –CH2–CH2–); 3.87 (s,
3H, OCH3); 7.17–7.23 (m, 1H, indole-H6); 7.21, 7.60
(d, J = 2.4 Hz, 1H, indole-H4); 7.88 (s, 1H, vinyl-H);
4
12.86 (s, 1H, N-H). Anal. for C20H15N3O2; calcd C,
72.94; H, 4.59; N, 12.76; found: C, 72.87; H, 4.67; N,
12.40.
3
3
(AA0BB0, J = 9 Hz, 4H, phenyl-H); 7.44 (d, J = 8 Hz,
1H, indole-H7); 7.88 (d, 4J = 2 Hz, 1H, indole-H4);
7.92 (s, 1H, vinyl-H); 12.83 (s, 1H, N-H). Anal. for
C22H19N3O; calcd C, 77.40; H, 5.61; N, 12.31; found:
C, 76.78; H, 5.46; N, 11.98.
4.2.4. [[6-Methoxy-2-(4-methoxyphenyl)-indol-3-yl]meth-
ylene]propanedinitrile (6d). Yellow solid (96% yield), mp
1
250 ꢁC. H NMR d 3.84 (s, 3H, OCH3); 3.88 (s, 3H,
OCH3); 6.93 (dd, 3J = 8.8 Hz, 4J = 2.4 Hz, 1H, in-
dole-H5); 7.09 (d, 3J = 2.4 Hz, 1H, indole-H7); 7.20,
7.58 (AA0BB0, 3J = 8.8 Hz, 4H, phenyl-H); 7.81 (s,
1H, vinyl-H); 7.97 (d, 3J = 8.8 Hz, 1H, indole-H4),
12.78 (s, 1H, N-H). Anal. for C20H15N3O2; calcd C,
72.94; H, 4.59; N, 12.76; found: C, 72.67; H, 4.82;
N, 12.51.
4.2.11. [[6-Isopropyl-2-(4-methoxyphenyl)indol-3-yl]methy-
lene]propanedinitrile (6k). Yellow solid (33% yield), mp
145 ꢁC (EtOH). 1H NMR d 1.27 (d, 3J = 7 Hz, 6H,
3
–CH-(CH3)2); 3.01 (sept, J = 7 Hz, 1H, –CH-(CH3)2);
3.86 (s, 3H, OCH3); 7.20, 7.58 (AA0BB0, 3J = 9 Hz,
4H, phenyl-H); 7.18–7.26 (m, 1H, indole-H6); 7.43 (d,
3J = 8 Hz, 1H, indole-H7); 7.87 (s, 1H, indole-H4); 7.99
(s,1H, vinyl-H); 12.86 (s, br, 1H, N-H). Anal. for
C22H19N3O; calcd C, 77.40; H, 5.61; N, 12.31; found:
C, 76.51; H, 5.65; N, 11.57.
4.2.5. [[5-Fluoro-2-(4-methoxyphenyl)-indol-3-yl]methy-
lene]propanedinitrile (6e). Yellow crystals (20% yield),
1
mp 225 ꢁC (EtOH). H NMR d 3.86 (s, 3H, OCH3);
7.15–7.84 (m, 3H, indole-H); 7.20, 7.60 (AA0BB0,
3J = 9.0 Hz, 4H, phenyl-H); 7.98 (s, 1H, vinyl-H);
13.00 (s, 1H, N-H). Anal. for C19H12FN3O; calcd C,
71.92; H, 3.81, N, 13.24; found: 71.92; H, 5.14; N, 12.06.
4.2.12. [[5-n-Butyl-2-(4-methoxyphenyl)indol-3-yl]methy-
lene]propanedinitrile (6l). Yellow solid (78% yield), mp
154 ꢁC (EtOH). 1H NMR d 0.90 (t, 3J = 7 Hz, 3H,
–CH2–CH3); 1.33 (sext, 3J = 7 Hz, 2H, –CH2–CH2–
CH3); 1.64 (quin, 3J = 7 Hz, 2H, –CH2–CH2–CH2–);
2.71 (t, 3J = 7 Hz, 2H, –CH2–CH2–); 3.87 (s, 3H,
4.2.6. [[6-Fluoro-2-(4-methoxyphenyl)-indol-3-yl]methy-
lene]propanedinitrile (6f). Orange crystals (20% yield),
1
3
mp 228-232 ꢁC. H NMR d 3.72 (s, 3H, OCH3); 6.84–
OCH3); 7.20, 7.59 (AA0BB0, J = 9 Hz, 4H, phenyl-H);
3
3
8.02 (m, 3H, ArH); 7.02, 7.43 (AA0BB0, J = 9 Hz, 4H,
7.19 (d, J = 8 Hz, 1H, indole-H6); 7.52 (d, J3 = 8 Hz,
ArH); 7.78 (s, 1H, vinyl-H); 7.90 (m, 1H, indole-H4).
1H, indole-H7); 7.88 (s, 1H, indole-H4); 7.93 (s, 1H,

´
M. Pojarova et al. / Bioorg. Med. Chem. 15 (2007) 7368–7379
7376
3
vinyl-H); 12.82 (s, 1H, N-H). Anal. for C23H21N3O;
calcd C, 77.72; H, 5.95; N, 11.82; found: C, 77.23; H,
5.66; N, 11.60.
2H, J = 8 Hz, phenyl-H); 8.16 (s, 1H, vinyl-H); 13.05
(s, br, 1H, N-H). Anal. for C23H18F3N3; calcd C,
70.22; H, 4.61; N, 10.68; found: C, 69.81; H. 4.67; N,
9.84.
4.2.13. [[2-(4-Methoxyphenyl)-5-n-pentylindol-3-yl]meth-
ylene]propanedinitrile (6m). Yellow solid (75% yield), mp
158 ꢁC (EtOH). 1H NMR d 0.86 (t, 3J = 7 Hz, 3H,
–CH2–CH3); 1.26 (m, 4H, –CH2–(CH2)2– CH3); 1.64
(m, 2H, –CH2–CH2–CH2–); 2.70 (t, 3J = 7 Hz, 2H,
4.2.18. [[5-n-Pentyl-2-[4-(trifluoromethyl)phenyl]indol-3-
yl]methylene]propanedinitrile (6r). Yellow solid (50%
yield), mp 195–196 ꢁC (EtOH). 1H NMR d 0.84 (t,
3H, 3J = 7 Hz, –CH2–CH3); 1.25 (m, 4H, –CH2–
(CH2)2– CH2–); 1.64 (m, 2H, –CH2–CH2–CH2–); 2.69
(t, 2H, 3J = 7 Hz, –CH2–CH2–); 7.20 (dd, 1H,
3J = 8 Hz, 4J = 2Hz, indole-H6); 7.46 (d, 1H,
3J = 8 Hz, indole-H7); 7.84, 7.97 (d, 4H, 3J = 8 Hz,
AA0BB0, phenyl-H); 7.88 (s, 1H, indole-H4); 8.13 (s,
1H, vinyl-H); 13.04 (s, br, 1H, N-H); MS: m/z
(%) = 407 (56, [M]Å+); 350 (100, [M]Å+ꢁC4H9). Anal.
for C24H20F3N3; calcd C, 70.75; H, 4.95; N, 10.31;
found: C, 70.46; H, 5.10; N, 10.26.
3
–CH2–CH2–); 3.87 (s, 3H, OCH3); 7.18 (d, J = 8 Hz,
1H, indole-H6); 7.60, 7.20 (AA0BB0, 3J = 9 Hz, 4H, phe-
3
nyl-H); 7.43 (d, J = 8 Hz, 1H, indole-H7); 7.88 (s, 1H,
Indol-H4); 7.93 (s, 1H, vinyl-H); 12.87 (s, br, 1H,
N-H). Anal. for C24H23N3O; calcd C, 78.02; H, 6.27;
N, 11.37; found: C, 77.57; H, 6.11; N, 11.12.
4.2.14. [[5-n-Hexyl-2-(4-methoxyphenyl)indol-3-yl]methy-
lene]propanedinitrile (6n). Yellow solid (78% yield), mp
131 ꢁC (EtOH). 1H NMR d 0.85 (t, 3J = 7 Hz, 3H,
–(CH2–CH3)); 1.28 (m, 6H, –CH2–(CH2)3– CH3); 1.64
(m, 2H, –CH2–CH2–CH2–), 2.70 (t, 3J = 7 Hz, 2H,
4.2.19. [[5-n-Hexyl-2-[4-(trifluoromethyl)phenyl]indol-3-
yl]methylene]propanedinitrile (6s). Orange crystals (52%
4
1
–CH2–CH2–); 3.87 (s, 3H, OCH3); 7.16 (d, J = 1 Hz,
yield), mp 177–178 ꢁC (EtOH). HNMR d 0.85 (t, 3H,
1H, indole-H6); 7.59, 7.20 (AA0BB0, 3J = 9 Hz, 4H, phe-
3J = 7 Hz, –CH2–CH3); 1.25 (m, 6H, –CH2–(CH2)3–
CH3); 1.64 (m, 2H, –CH2–CH2–CH2–); 2.71 (t, 2H,
3J = 7Hz, –CH2–CH2–); 7.23 (dd, 1H, 3J = 8 Hz,
3
nyl-H); 7.44 (d, J = 8 Hz, 1H, indole-H7); 7.88 (s, 1H,
indole-H4); 7.93 (s, 1H, vinyl-H); 12.85 (s, br, 1H,
N-H). Anal. for C25H25N3O; calcd C, 78.30; H, 6.57;
N, 10.96; found: C, 78.11; H, 6.41; N, 10.83.
3
4J = 2 Hz, indole-H6); 7.49 (d, 1H, J = 8 Hz, indole-
H7); 7.89 (s, 1H, indole-H4); 7.87, 7.99 (d, 4H,
3J = 8 Hz, AA0BB0, phenyl-H); 8.15 (s, 1H, vinyl-H);
13.06 (s, br, 1H, N-H); MS: m/z (%) = 421 (62,
[M]Å+); 350 (100, [M]Å+ꢁC5H11). Anal. for
C25H22F3N3; calcd C, 71.25; H, 5.26; N, 9.97; found:
C, 70.73; H, 4.79; N, 9.22.
4.2.15. [[5-n-Butyl-2-(4-methylphenyl)indol-3-yl]methy-
lene]propanedinitrile (6o). Orange crystals (30% yield),
mp 203–204 ꢁC (EtOH). 1H NMR d 0.90 (t, 3H,
3J = 7 Hz, –CH2–CH3); 1.32 (m, 2H, –CH2–CH2–
CH3); 1.63 (m, 2H, –CH2–CH2–CH2–); 2.42 (s, 3H, –
3
CH3); 2.71 (t, 2H, J = 7 Hz, –CH2–CH2–); 7.19 (dd,
4.3. Materials and reagents for bioassays
1H, 3J = 8 Hz, 4J = 2Hz, indole-H6); 7.45 (d, 3H,
3J = 8Hz, indole-H7 and phenyl-H); 7.54 (d, 2H,
3J = 8 Hz, phenyl-H); 7.91 (s, 1H, indole-H4); 7.92 (s,
1H, vinyl-H); 12.90 (s, br, 1H, N-H). Anal. for
C23H21N3; calcd C, 81.38; H, 6.24; N, 12.38; found: C,
81.01; H, 6.06; N, 12.20.
Drugs and biochemicals were obtained from Sigma
(Deisenhofen, Germany) except where noted. All cell
lines used (MDA-MB 231 and MCF-7 breast cancer
cells, A-431 vulva carcinoma cells) were of human origin
and were obtained from the American Type Culture col-
lection (ATCC, Rockville, MD, U.S.A.). Female
B6D2F1 mice were bought from Charles-River (Sulzfeld,
Germany). The MXT mammary tumors were gener-
ously provided by Dr. A. E. Bogden, Laboratory of
Experimental Oncology, Worcester, Mass, USA.
4.2.16. [[5-n-Butyl-2-(4-ethylphenyl)indol-3-yl]methy-
lene]propanedinitrile (6p). Yellow powder (68% yield),
mp 164–165 (EtOH). 1H NMR
d 0.90 (t, 3H,
3J = 7 Hz, –CH2–CH3); 1.25 (t, 3H, J = 7 Hz, –CH2–
CH3); 1.33 (m, 2H, –CH2–CH2– CH3); 1.63 (m, 2H,
–CH2–CH2–CH2–); 2.72 (m, 4H, –CH2–CH3 and
–CH2–CH2–); 7.19 (dd, 1H, 3J = 8 Hz, 4J = 2 Hz,
indole-H6); 7.46 (m, 3H, indole-H7 and phenyl-H);
7.56 (d, 2H, phenyl-H); 7.92 (s, 2H, vinyl-H and in-
dole-H4); 12.89 (s, br, 1H, N-H); MS: m/z (%) = 353
(100, [M]Å+); 310 (84, [M]Å+ꢁC3H7). Anal. for
C24H23N3 Æ 1/2H2O; calcd C, 79.51; H, 6.62; N 11.59;
found: C, 79.81; H, 6.45; N, 11.35.
3
4.4. Determination of anti-proliferative activity
Hormone-independent human MDA-MB 231 breast
cancer cells were grown in McCoy-5a medium, supple-
mented with L-glutamine (73 mg/L), gentamycin sulfate
(50 mg/L), NaHCO3 (2.2 g/L), and 5% sterilized fetal
calf serum (FCS). At the start of the experiment, the cell
suspension was transferred to 96-well microplates
(100 lL/well). After the cells had grown for 2–3 days
in a humidified incubator with 5% CO2 at 37 ꢁC, med-
ium was replaced by one containing the test compounds
(200 lL/well). Control wells (16/plate) contained 0.1% of
DMF that was used for the preparation of stock solu-
tions. Initial cell density was determined by addition
of glutaric dialdehyde (1% in PBS; 100 lL/well) instead
of test compound. After incubation for about 4 days,
the medium was removed and 100 lL of glutaric dialde-
4.2.17. [[5-n-Butyl-2-[4-(trifluoromethyl)phenyl]indol-3-
yl]methylene]propanedinitrile (6q). Orange powder (64%
yield), mp 212–213 ꢁC (EtOH). 1H NMR d 0.90 (t,
3
3H, J = 7 Hz, –CH2–CH3); 1.33 (m, 2H, –CH2–CH2–
CH3); 1.67 (m, 2H, –CH2–CH2–CH2–); 2.75 (t, 2H,
3J = 7 Hz, –CH2–CH2–); 7.23 (dd, 1H, 3J = 8 Hz,
4J = 2Hz, indole-H6); 7.49 (d, 1H, 3J = 8 Hz, indole-
H7); 7.88 (m, 3H, phenyl-H and indole-H4); 7.99 (d,

´
M. Pojarova et al. / Bioorg. Med. Chem. 15 (2007) 7368–7379
7377
hyde in PBS (1%) was added for fixation. After 15 min,
the solution of aldehyde was decanted. Cells were
stained by treating them for 25 min with 100 lL of an
aqueous solution of crystal violet (0.02%). After decant-
ing, cells were washed several times with water to re-
move adherent dye. After addition of 100 lL EtOH
(70%), plates were gently shaken for 2 h. Optical density
of each well was measured in a microplate autoreader
EL 309 (Bio-tek) at 578 nm.
transferred to centrifugation tubes and carefully under-
layered with a pre-warmed (37 ꢁC) sucrose solution
(10% in PEM buffer solution containing 1 mM GTP, ap-
prox. 10% of the transferred volume). After centrifuga-
tion at 37 ꢁC for 45 min (20,000g), the pellets were
weighed and suspended in ice-cold PEM buffer solution
(3 mL/g) and homogenized in a Teflon-in-glass potter.
After standing in ice for 30 min, the suspension was cen-
trifuged at 2 ꢁC for 30 min (40,000g). The supernatant
was separated and adjusted to 1 mM GTP. By incuba-
tion at 37 ꢁC for 15 min tubulin was polymerized once
again. After centrifugation at 37 ꢁC for 30 min microtu-
bules were obtained as shiny gel-like pellet. The yields
ranged from 2 to 6 g per brain. Aliquots were frozen
in liquid nitrogen and stored at ꢁ70 ꢁC. Purity was
checked by polyacrylamide gel electrophoresis.
For hormone-sensitive MCF-7 human breast cancer
cells a similar procedure to that described for MDA-
MB 231 cells was applied with alterations: cells were
grown in EMEM supplemented with sodium pyruvate
(110 mg/L), gentamycin sulfate (50 mg/L), NaHCO3
(2.2 g/L), phenol red and 10% FCS. An analogous pro-
tocol was applied for A-431 vulva carcinoma cells with
two modifications. The concentration of FCS was re-
duced to 5% and the incubation period was 4 days.
4.6.2. Temperature-induced tubulin polymerization. The
pellet of frozen microtubules was warmed to 37 ꢁC in
a water bath. After addition of the 20-fold volume of
ice-cold PEMG buffer it was homogenized. Depolymer-
ization was completed by keeping the mixture at 0 ꢁC
for 30 min, followed by centrifugation at 2 ꢁC (30 min;
30,000g) to remove insoluble protein. Each reaction
tube contained 0.46 mL of the supernatant and 20 lL
of the DMSO solution of the test compound in varying
concentrations. Reaction mixture were preincubated at
37 ꢁC for 15 min and chilled on ice followed by addition
of 20 lL of a 25 mM GTP solution in PEMG buffer to
each tube. Reaction mixtures was transferred to cuvettes
of a UV spectrophotometer connected to two different
temperature controller. First, the temperature inside
the cuvettes was held at 2 ꢁC. The cuvette holder was
then switched to the second temperature controller at
37 ꢁC and the absorption was measured over a period
of 20 min at 350 nm. Absorption at the start of the reac-
tion was used as baseline. Three independent experi-
ments were performed for the determination of IC50
values. Each experiment had two control reaction mix-
tures; their mean value was defined as 100% and their
turbidity readings were generally within 10% of each
other.
For the determination of the IC50 values in the HL-60
cell line, the MTT assay was used as previously
described.22
4.5. Flow cytometry
MDA-MB 231 cells were grown to 70–80% confluence
on the bottom of a 75 cm2 culture flask. Test substances
were dissolved in DMF and diluted to the required con-
centrations. The content of DMF in medium and in con-
trols after 1000 times dilution was set to 0.1%. For each
concentration one culture flask was used. Cells were ex-
posed to test substances for 24 h. This incubation time
was necessary to obtain a sufficient number of cells
(5 · 106 cells). After incubation the cells were trypsini-
zed, centrifuged (250g) in an excess of serum containing
medium for 10 min, and washed with PBS.
The pellets of cells were resuspended in 1 mL of PBS and
9 mL of 70% EtOH (ice-cold). EtOH and PBS were
removed by centrifugation and the cells were washed with
PBS. After the addition of 0.5 mL of PBS and 0.5 mL of
DNA extraction buffer, the cells were transferred into
Eppendorf cups (2 mL), incubated for 5 min at room tem-
perature, and centrifuged again. The cells are stained with
1 mL of propidium iodide solution for at least 30 min at
room temperature and then analyzed by flow cytometry
using a FACS-CaliburTM. For analysis, data files for four
parameters were collected for 8,000–15,000 events from
each sample. The flow rate was adjusted to ca. 300 cells/
s for appropriate accuracy. Data were analyzed by the
WinMDI 2.8 software.
4.7. Caspase induction studies28
HL-60 human acute myeloid leukemia cells (DSMZ)
were maintained in logarithmic growth in RPMI 1640
supplemented with 10% FCS, 30 mg/L of penicillin G
and 40 mg/L of streptomycin sulfate. Apoptosis was in-
duced by treatment of cells (1 · 105 cells/mL) with the
indole derivatives or with etoposide as the reference
drug. Cells were incubated in a humid atmosphere of
5% CO2 at 37 ꢁC and the cell lysates were prepared after
24 h of treatment.
4.6. Tubulin polymerization assay
4.6.1. Isolation and purification of calf brain tubulin. The
cortex of one or two fresh calf brains in ice-cold PEM
buffer (1 mL/g tissue, +16 mg DTE/100 mL buffer solu-
tion) was homogenized in portions. After centrifugation
(90 min; 20,000g) at 2–4 ꢁC, the supernatant was care-
fully decanted. The concentrations of GTP and ATP
were adjusted to 0.1 and 2.5 mM, respectively. After
stirring gently at 37 ꢁC for 30 min, the solution was
For the preparation of the cell-free extracts, all steps
were performed at 4 ꢁC. Cells were pelleted at 500g for
4 min, washed with PBS, once again pelleted at 500g
for 3 min and resuspended in hypotonic Hepes buffer
(10 mM Hepes, pH 7.0; 5 mM 1,4-dithiothreitol
(DTT); 2 mM Na–EDTA; 0.1% (w/v) CHAPS). After
a 20-min incubation on ice, lysates were centrifuged at
10,000g for 10 min. The supernatant was removed while

´
M. Pojarova et al. / Bioorg. Med. Chem. 15 (2007) 7368–7379
7378
taking care to avoid the pellet. Fifty microliter aliquots
containing 2–4 lg of cytosolic protein (estimated by the
Bradford method) were pre-incubated with 10 lM of
appropriate protease inhibitor (benzyloxycarbonyl-
Asp-Glu-Val-Asp-chloromethylketone and acetyl-Leu-
Glu-His-Asp-chloromethylketone, both from Bachem.
Heidelberg, Germany, for caspases 3 and 9, respectively)
or DMSO (assay without inhibitor) for 5 min at 21 ꢁC
and then diluted to 198 lL with the assay buffer
(50 mM Hepes; 5 mM DTT; 2 mM Na–EDTA; 0.1%
(w/v) CHAPS). The samples were frozen at ꢁ32 ꢁC.
All experiments were performed within one week of ex-
tract preparation.
after transplantation. Compound 6f was suspended in
polyethylene glycol 400/1.8% saline (1:1); the reference
drug cisplatin was dissolved in 0.9% saline. Both agents
were administered subcutaneously on Monday, Wednes-
day, and Friday. After two weeks of treatment, the tu-
mor area was determined by transdermal calliper
measurements of two perpendicular axes, one across
the largest diameter. The change of body weight be-
tween start and end of therapy was recorded in order
to detect obvious toxicity.
Acknowledgments
The authors thank Dr. Doris Marko, University of Kar-
lsruhe, for helpful discussions and Renate Liebl, Univer-
sity of Regensburg, for excellent technical assistance.
Caspase 3 and 9 activities were assessed by monitoring
cleavage of fluorochrome tagged synthetic substrates
Acetyl-Asp-Met-Gln-Asp-7-amino-4-methyl-coumarin
(Ac-DMQD-AMC) and Acetyl-Leu-Glu-His-Asp–ami-
no-4-trifluoromethyl-coumarin (Ac-LEHD-AFC) (Ba-
chem), respectively, by HPLC. After thawing of
lysates, 2 lL of appropriate substrate was added (final
concentration of 150 and 200 lM for Ac-DMQD-
AMC and Ac-LEHD-AFC, respectively). The samples
were incubated for 30 min at 30 ꢁC and the specific
products quantified by RP-HPLC with a fluorescence
detector (L-7485, Merck). Fluorescence was measured
by using an excitation wavelength of 380 nm or
400 nm and emission wavelength of 460 or 505 nm for
released AMC or AFC, respectively. Control experi-
ments (data not shown) confirmed that the release of
AMC or AFC was linear for at least 60 min at the con-
ditions specified.
Supplementary data
Supplementary data associated with this article can be
found, in the online version, at doi:10.1016/j.bmc.
2007.07.046.
References and notes
1. Kaufmann, D.; Pojarova, M.; Vogel, S.; Liebl, R.;
Gastpar, R.; Gross, D.; Nishino, T.; Pfaller, T.; von
Angerer, E. Bioorg. Med. Chem. 2007, 15, 5122–5136.
2. Gastpar, R.; Goldbrunner, M.; Marko, D.; von Angerer,
E. J. Med. Chem. 1998, 41, 4965–4972.
3. Buolamwini, J. K. Curr. Pharm. Des. 2000, 6, 379–392.
4. Webster, K. R. Chem. Res. Toxicol. 2000, 13, 940–943.
5. von Angerer, E. Exp. Opin. Ther. Patents 1999, 9, 1069–
1081.
6. von Angerer, E. Curr. Opin. Drug Discovery Dev. 2000, 3,
575–584.
7. Benson, C.; Kaye, S.; Workman, P.; Garrett, M.; Walton,
M.; de Bono, J. Br. J. Cancer 2005, 92, 7–12.
8. Sielecki, T. M.; Boylan, J. F.; Benfield, P. A.; Trainor, G.
L. J. Med. Chem. 2000, 43, 1–18.
9. Garcia, H. H.; Brar, G. A.; Nguyen, D. H. H.; Bjeldanes,
L. F.; Firestone, G. L. J. Biol. Chem. 2005, 280, 8756–
8764.
To estimate a specific caspase activity, measurements of
blank (substrate only), negative control (substrate + cell
lysate + inhibitor) and positive controls (substrate + cell
lysate) were performed. From each result obtained as
fluorescence units (peak area) blank values were sub-
tracted. Standards containing 0–15 pmol of AMC and
0–30 pmol of AFC were utilized to determine the
amount of fluorochrome released. The results are pre-
sented as amount of AMC or AFC release (pmol/mg/
min), which can be blocked by selective protease
inhibitor.
10. Cover, C. M.; Hsieh, S. J.; Tran, S. H.; Hallden, G.; Kim,
G. S.; Bjeldanes, L. F.; Firestone, G. L. J. Biol. Chem.
1998, 273, 3838–3847.
4.8. Microscopy of HL-60 cells
11. Hong, C.; Firestone, G. L.; Bjeldanes, L. F. Biochem.
Pharmacol. 2002, 63, 1085–1097.
12. Hong, C.; Kim, H.; Firestone, G. L.; Bjeldanes, L. F.
Carcinogenesis 2002, 23, 1297–1305.
13. Rahman, K. W.; Li, Y.; Wang, Z.; Sarkar, S. H.; Sarkar,
F. H. Cancer Res. 2006, 66, 4952–4960.
14. Nam, S.; Buettner, R.; Turkson, J.; Kim, D.; Cheng, J. Q.;
Muehlbeyer, S.; Hippe, F.; Vatter, S.; Merz, K. H.;
Eisenbrand, G.; Jove, R. Proc. Natl. Acad. Sci. U.S.A.
2005, 102, 5998–6003.
15. Moon, M. J.; Lee, S. K.; Lee, J. W.; Song, W. K.; Kim, S.
W.; Kim, J. I.; Cho, C.; Choi, S. J.; Kim, Y. C. Bioorg.
Med. Chem. 2006, 14, 237–246.
16. Kim, S. A.; Kim, Y. C.; Kim, S. W.; Lee, S. H.; Min, J. J.;
Ahn, S. G.; Yoon, J. H. Clin. Cancer Res. 2007, 13, 253–
259.
HL-60 cells at a density of 500,000/ml in 10% FCS/
RPMI medium were treated with either 0.1% DMF
(control), 1.0 lM 6m, 1.0 lM 6q, or 8 lM etoposide
for 24 h at 37 ꢁC. Cells were photographed directly at
a 400-fold magnification with a Sony alpha100 digital
camera mounted on a Zeiss Axiovert 25 phase-contrast
microscope.
4.9. Determination of antitumor activity in mice
Hormone resistant MXT mouse mammary tumors were
grown in ovariectomized female B2D2F1 mice before
transplantation. At the start of the experiment, tumor
pieces of 1 mm3 were serially transplanted into 8- to 9-
week-old female B2D2F1 mice. Animals were randomly
assigned to groups of 6–8 and treatment started 24 h
17. Eisenbrand, G.; Hippe, F.; Jakobs, S.; Muehlbeyer, S. J.
Cancer Res. Clin. Oncol. 2004, 130, 627–635.

´
M. Pojarova et al. / Bioorg. Med. Chem. 15 (2007) 7368–7379
7379
18. Supuran, C. T. Exp. Opin. Invest. Drugs 2003, 12, 283–
287.
19. Owa, T.; Yoshino, H.; Okauchi, T.; Yoshimatsu, K.;
Ozawa, Y.; Sugi, N. H.; Nagasu, T.; Koyanagi, N.; Kitoh,
K. J. Med. Chem. 1999, 42, 3789–3799.
23. Suzuki, I.; Kondoh, M.; Nagashima, F.; Fujii, M.;
Asakawa, Y.; Watanabe, Y. Planta Med. 2004, 401–406.
24. Zhang, L.; Mizumoto, K.; Sato, N.; Ogawa, T.; Kusum-
oto, M.; Niiyama, H.; Tanaka, M. Cancer Lett. 1999, 142,
129–137.
20. Andreani, A.; Granaiola, M.; Leoni, A.; Locatelli, A.;
Morigi, R.; Rambaldi, M.; Garaliene, V.; Farruggia, G.;
Masotti, L. Bioorg. Med. Chem. 2004, 12, 1121–1128.
21. Chang, H. Y.; Yang, X. Microbiol. Mol. Biol. Rev. 2000,
64, 821–846.
25. Szepeshazi, K.; Schally, A. V.; Nagy, A. Breast Cancer
Res. Treat. 1999, 56, 267–276.
26. Levitzki, A. Biochem. Pharmacol. 1990, 40, 913–918.
27. Kleinberger-Doron, N.; Shelah, N.; Capone, R.; Gazit,
A.; Levitzki, A. Exp. Cell Res. 1998, 241, 340–351.
28. Saczewski, F.; Reszka, P.; Gdaniec, M.; Grunert, R.;
Bednarski, P. J. J. Med. Chem. 2004, 47, 3438–3449.
22. Bracht, K.; Boubakari; Grunert, R.; Bednarski, P. J. Anti-
¨
Cancer Drugs 2006, 17, 41–51.