3-(3-Phenoxybenzyl)amino-β-carboline: A novel antitumor drug targeting α-tubulin

Article abstract of DOI:10.1016/j.bmcl.2011.06.061

3-(3-Phenoxybenzyl)amino-β-carboline 2h showed extremely-high activity; the IC₅₀ value was 0.074 μM. To verify 2h-induced cell death types, we observed the chromatin condensation, the DNA fragmentation and activated caspase-3 using Hoechst 3334

Full text of DOI:10.1016/j.bmcl.2011.06.061

Bioorganic & Medicinal Chemistry Letters 21 (2011) 4784–4787
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
3-(3-Phenoxybenzyl)amino-b-carboline: A novel antitumor drug targeting
a-tubulin
Reiko Ikeda a,b, Masaki Kurosawa a, Takazumi Okabayashi a, Ayako Takei a, Masamichi Yoshiwara a,
Tadashi Kumakura a, Norio Sakai a, Osamu Funatsu c, Akinori Morita d, Masahiko Ikekita c,d
,
Yumi Nakaike a, Takeo Konakahara a,b,e,
⇑
a Department of Pure and Applied Chemistry, Faculty of Science and Technology, Tokyo University of Science (RIKADAI), Noda, Chiba 278-8510, Japan
b Center for Technologies Against Cancer, Tokyo University of Science (RIKADAI), Noda, Chiba 278-8510, Japan
c Genome and Drug Research Center, Tokyo University of Science (RIKADAI), Noda, Chiba 278-8510, Japan
d Department of Applied Biological Science, Faculty of Science and Technology, Tokyo University of Science (RIKADAI), Noda, Chiba 278-8510, Japan
e Research Institute for Science and Technology, Tokyo University of Science (RIKADAI), Noda, Chiba 278-8510, Japan
a r t i c l e i n f o
a b s t r a c t
Article history:
3-(3-Phenoxybenzyl)amino-b-carboline 2h showed extremely-high activity; the IC50 value was 0.074 lM.
Received 18 March 2011
Revised 14 June 2011
Accepted 14 June 2011
Available online 22 June 2011
To verify 2h-induced cell death types, we observed the chromatin condensation, the DNA fragmentation
and activated caspase-3 using Hoechst 33342, agarose electrophoresis and western blot, and suggesting
2h-induced cell death type was apoptosis. Flow cytometry showed that 2h-treated cell was induced
SubG1 cell population after G2/M cell cycle arrest. In addition, using affinity chromatography and
peptide mass fingerprinting, we found that interacting protein with this compound was
protein.
a-tubulin
Keywords:
b-Carboline
Ó 2011 Elsevier Ltd. All rights reserved.
a-Tubulin
Antitumor drug
Apoptosis
Many anticancer drugs exert their cytotoxic effects by interfer-
ing with cell cycle progression via inhibition or activation of cell
cycle regulators, inhibition of DNA replication, induction of DNA
damage, or disruption of mitotic spindle formation.1 In particular,
ever since taxol was discovered, spindle microtubules have be-
come a popular target for development of new anticancer drugs.2,3
Antimicrotubule agents exert their cytotoxic effects during mitosis
by interfering with the exchange of tubulin subunits between the
microtubules and the free tubulin pool.4 These drugs inhibit cell
proliferation either by increasing microtubule polymerization, for
example, taxol, or by promoting microtubule depolymerization,
for example, the vinca alkaloids, and thus impede cell division
and induce cell death at relatively high concentrations.5 Many
plant-derived products are used to combat malignant tumors.
The chemotherapeutic drugs widely used in clinical oncology in-
clude anticancer compounds extracted from plants such as taxol
and etoposide, or derived from plants alkaloids by simple chemical
modification such as the camptothecin (CPT) derivatives topotecan
and irinotecan.
life, human tissues, and body fluids.6 Harman and norharman are
very well known. Most of these b-carboline derivatives are en-
dowed with antitumor, anticancer, or intercalating properties.
Some of these molecules are benzodiazepine receptors, and their
high DNA binding affinity is thought to be partially responsible
for their pharmacological activity.7–10 The interest in clinical appli-
cation of b-carbolines is based on the physiological effects of these
compounds, such as inhibition of cyclin-dependent kinase (CDK),11
IkappaB kinase (IKK)12 and topoisomerase I.9 However, an optimal
structure of b-carboline derivatives remains not to be identified.
Therefore, in the present study, we synthesized a variety of b-carb-
oline derivatives and evaluated their biological activities.
To determine the cytotoxic effects of synthesized b-carboline
derivatives on a human cervical cancer cell line (HeLa S-3), we used
the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide
(MTT) assay (Tables 1–3). Previously, we reported that 3-benzyl-
amino-b-carboline has high-antitumor activity, with an IC50 value
of 0.11 l
M.13 On the other hand, Boursereau et al. reported that
1-amino-b-carboline has significant anticancer and antiparasite
b-Carboline has a planar tricyclic ring structure, and the deriv-
atives are widely distributed in nature including in plants, marine
activities.14
Therefore, 1-benzylamino-b-carboline derivatives 1a–n were
synthesized from benzaldehyde possessing a methyl, chloro, meth-
oxy, phenoxy, or hydroxy group by reductive amination with
1-amino-b-carboline in moderate to good yields as shown in
⇑
Corresponding author.
E-mail address: konaka@rs.noda.tus.ac.jp (T. Konakahara).
0960-894X/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2011.06.061

R. Ikeda et al. / Bioorg. Med. Chem. Lett. 21 (2011) 4784–4787
4785
Table 1
R4
R3
i)
OHC
Yields and antitumor activity of 1-benzylamino-b-carbolines in HeLa S-3 cells
a
Compds
R2
R3
R4
Yield (%)
IC50
(
lM)
R2
1a
1b
1c
1d
1e
1f
1g
1h
1i
1j
1k
1l
1m
1n
H
Me
H
H
Cl
H
H
H
OMe
H
H
H
H
Me
H
H
Cl
H
Cl
H
OMe
H
OPh
H
H
H
H
Me
H
H
Cl
Cl
H
H
OMe
H
OPh
H
66
92
74
68
68
93
67
69
75
77
54
84
90
75
24
13
22
26
19
20
18
20
15
24
22
1
1
6
2
2
4
3
1
1
1
2
R4
R3
N
(1.5 eqiv)
N
H
benzen, reflux, 24 h
HN
N
N
H
ii) NaBH4 (10 eqiv)
MeOH, reflux, 2 h
R2
1a-n
NH2
(54-93%)
1-Amino-β-carboline
Scheme 1.
The compound 1l with an OPh group at R3 showed the highest anti-
tumor activity; the IC50 value of 1l was 2.3 M. However, the activ-
ity of 1l was lower than that of 3-benzylamino-b-carboline (2a),
which had an IC50 value of 0.11 M (Table 2). These results suggest
that the optimal position for introduction of the benzylamino
group is the 3-position of b-carboline nucleolus. Based on this re-
sult, we started to optimize the structure of 3-benzylamino-b-
carboline to obtain the highest antitumor activity.
H
H
OH
2.3 0.1
l
15
19
2
1
H
l
a
Drug toxicity was determined by 3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyl-
tetrazolium bromide (MTT) assay (see Materials and methods). IC50 is drug con-
centration inhibited the 50% cell growth. Data represent the mean S.D. from four
independent experiments performed in triplicate experiments.
3-Benzylamino-b-carboline derivatives 2a–t were analogously
synthesized from 3-amino-b-carboline and benzaldehydes with a
methyl, hydroxyl, phenoxy, benzyloxy, chloro, nitro, trifluoro-
methyl, or phenyl group at R2–R4 (Scheme 2). Yields are summa-
rized in Table 2.
Cytotoxicity of 2a–t was evaluated using the MTT assay (Table
2). In general, compounds substituted at R2 and R3 showed rela-
tively high antitumor activity. In addition, the antitumor activity
of the compounds substitutes at R3 was higher than that of com-
pounds substituted at R4. In other words, substitution at the R4 po-
sition reduced the antitumor activity. It was notable that 2h which
has a phenoxy group, had the highest-activity among the com-
pounds 2a–t; the IC50 value of 2h is 0.074 lM. However, the intro-
duction of a NO2 or CF3 group significantly reduced the antitumor
activity.
Subsequently, in order to determine the influence of poly-sub-
stitution on antitumor activity, we synthesized 3-(dimethoxybe-
Table 2
Yields and antitumor activity of 3-benzylamino-b-carbolines in HeLa S-3 cells
R2
R3
R4
Yield (%)
IC50
(
l
M)
a
Compds
2a
2b
2c
2d
2e
2f
2g
2h
2i
2j
2k
2l
2m
2n
2o
2p
2q
2r
H
Me
H
H
OH
H
H
H
H
H
Cl
H
H
NO2
H
CF3
H
H
Ph
H
H
H
Me
H
H
OH
H
OPh
H
OBn
H
Cl
H
H
NO2
H
CF3
H
H
H
H
Me
H
H
OH
H
OPh
H
H
H
Cl
H
H
H
H
CF3
H
81
82
81
73
79
80
79
85
73
69
89
89
74
71
82
75
88
74
88
74
0.11 0.01
0.30 0.04
0.21 0.01
6.2 0.2
28
2
0.92 0.10
1.7 0.2
0.074 0.007
1.2 0.1
0.38 0.05
1.0 0.1
0.83 0.05
6.1 1.1
14
>30
1
6.7 0.7
3.1 0.4
47 10
3.0 0.4
>30
nyl)amino-,
and
3-(trimethoxybenyl)amino-b-carboline
derivatives 3d–k by the same method used in the synthesis of 2
as shown in Scheme 3 (see Supplementary data) in moderate to
good yields, and evaluated the antitumor activity of these com-
pounds (Table 3). Compared with di-substituted compounds, the
tri-substituted derivatives were obtained in slightly lower yield
due to steric hindrance. In each case, most of the starting material
was recovered together with the desired product. Our efforts to
improve the yield of poly-substituted product were unsuccessful.
In addition, use of excess aldehyde caused the formation of
bis-benzylated compounds.
2s
2t
H
H
Ph
a
Drug toxicity was determined by 3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyl-
tetrazolium bromide (MTT) assay (see Materials and methods). IC50 is drug con-
centration inhibited the 50% cell growth. Data represent the mean S.D. from four
independent experiments performed in triplicate experiments.
Table 3
Antitumor activity of multi-substituted 3-benzylamino-b-carbolines in HeLa S-3 cells
Compds
R2
R3
R4
R5
R6
Yield (%)
IC50
(lM)
a
The 2-, and 3-methoxy derivatives 3a,b have similarly high
antitumor activity, with the IC50 values of 0.80 and 0.79 lM,
respectively. However, 4-methoxy derivative 3c shows signifi-
cantly reduced antitumor activity. These results suggest that the
optimal position for monomethoxy-substitution was either the
position 2 or 3. To determine the synergistic effect of dimethoxy-
substitution on the antitumor activity, the 2,3-, 2,4-, 2,5-, and
3a
3b
3c
3d
3e
3f
3g
3h
3i
MeO
H
H
MeO
MeO
MeO
MeO
H
H
MeO
H
MeO
H
H
H
H
MeO
H
MeO
H
H
MeO
H
MeO
MeO
H
H
H
H
H
MeO
H
H
MeO
H
H
H
H
H
H
H
MeO
H
H
86
73
82
63
78
72
69
74
73
63
66
0.80 0.06
0.79 0.04
20
4
3.2 0.4
4.8 0.3
6.1 0.6
2.3 0.2
>100
1.7 0.4
4.0 0.8
H
MeO
MeO
MeO
MeO
H
MeO
H
3j
3k
H
H
R4
R3
i)
MeO
0.22 0.03
H
N
a
Drug toxicity was determined by 3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyl-
OHC
NH2
tetrazolium bromide (MTT) assay (see Materials and methods). IC50 is drug con-
centration inhibited the 50% cell growth. Data represent the mean S.D. from four
independent experiments performed in triplicate experiments.
R2
N
R2
N
H
CH3COOH (pH=4)
MeOH, rt, 3 h
N
N
H
R3
R4
Scheme
Supplementary data).
1 and Table 1, according to a usual method (see
ii) NaBH3CN
MeOH, reflux, 4-5 h
2a-t
(52-78%)
1-Amino-β-carboline
Next, we evaluated the cytotoxicity on a human cervical cancer
cell line, HeLa S-3, of compounds 1b–n by MTT method (Table 1).
Scheme 2.

4786
R. Ikeda et al. / Bioorg. Med. Chem. Lett. 21 (2011) 4784–4787
R5
R2
R6
R4
R3
i)
H
N
OHC
R6
R5
NH2
N
R2
N
H
CH3COOH (pH=4)
MeOH, rt, 3 h
N
N
H
R3
R4
ii) NaBH3CN
MeOH, reflux, 4-5 h
1-Amino-β-carboline
3a-k
(52-78%)
Scheme 3.
2,6-dimethoxy-substituted derivatives 3d–g were synthesized in
63%, 78%, 72%, and 69% yields, respectively (Table 3). The
antitumor activities of 3a,b are very high. However, the antitumor
activities of 3d–g are lower than those of 3a,b. Among these com-
pounds, introduction of the 2nd methoxy group at the position 3, 4,
5, or 6 of 3a slightly decreased the antitumor activity. The 4-meth-
oxy group greatly decreased the antitumor activity of the corre-
sponding compound (see 3c in Table 3). Especially, introduction
of the 2nd 4-methoxy group to 3b strongly decreases the antitu-
mor activity of the b-carboline derivative 3h. Although the reason
is not clear, 3-(3,4,5-trimethoxybenzyl)amino-b-carboline (3k)
surprisingly shows the highest antitumor activity among the com-
pounds 3a–k, regardless of its 4-methoxy group; the IC50 value of
Figure 2. Effect of 2h on apoptosis induction in HeLa S-3 cells. Cells were treated
with 2h for 48 h. DNA fragmentation was detected by electrophoresis on a 2%
agarose gel. Marker; 100 bp ladder.
3k is 0.22 lM, as shown in Table 3.
Next, we investigated the type of cell death induced. Both chro-
matin condensation and fragmented nuclei are the classic charac-
teristics of apoptosis. Hoechst 33342 staining showed significant
morphological changes in the nuclear chromatin. After treatment
of HeLa S-3 cells 0, 150, 300, 1000 nM of 2h, we observed the nu-
clear chromatin (Fig. 1). Compared with 2h-treated cells, untreated
cells exhibited less intense nuclear staining. However, a large pro-
portion of 2h-treated cells exhibited fragmented nuclei.
To verify that cell death occurred via apoptosis, a DNA fragmen-
tation assay was performed. DNA fragmentation is a well known
biochemical index of cell apoptosis. After exposure to 2h for
48 h, the genomic DNA was extracted and analyzed by electropho-
resis (Fig. 2). DNA fragmentation was detectable in the cells treated
with 2h (150–1000 nM). Together with fluorescence microscopy
analysis, these observations demonstrated that 2h caused apopto-
tic cell death.
Figure 3. Western blot analysis of activated caspase-3 in HeLa S-3 cells. Cells were
treated with 2h for 6, 12, 24 or 48 h. Control, non-treated cell extract; b-actin was
used as internal control.
pase-3 was readily detected in cells treated with 2h for more
than 12 h, and it enhanced with the time-dependent. These
observations are closely connected with results of the chromatin
condensation and DNA fragmentation, suggesting that 2h induced
apoptosis.
To examine the effects on cell cycle progression, the cells were
exposed to 150 nM of 2h for 12, 24, 36, or 48 h. After treating for
12 h, the percentage of average G2/M population at the control
cells was 65.92% (see Table 4). For 24 h, this population shifted
from G2/M phase to the mean apoptotic population sub-G1 phase.
These results suggest that 2h induces apoptosis through G2/M cells
cycle arrest.
Affinity chromatography, which is based on a highly specific
biological interaction, such as between antigen and antibody, en-
zyme and substrate, or receptor and ligand, is used to separate bio-
chemical mixture. In the present study, we sought to identify the
Caspase-3 is well known as the execution factor of apoptosis.
After determining that 2h induced chromatin condensation and
DNA fragmentation, we further investigated the activation of cas-
pase-3 in 2h-treated cells. As shown in Figure 3, activated cas-
Figure 1. HeLa S-3 cells were treated with 2h for 48 h. Apoptosis was detected by Hoechst 33342 staining.

R. Ikeda et al. / Bioorg. Med. Chem. Lett. 21 (2011) 4784–4787
4787
Table 4
enolase was probably caused by blockade of the glycolytic system,
such as inhibition of the citric acid cycle or the electron transfer
system. During preparation for cell division under conditions of en-
ergy deficiency, a variety of cell cycle regulators might interfere
with the onset of the M phase. Therefore, it does not stand to rea-
son that the cell cycle progressed through the anaphase stage of
cell division after DNA replication. Based on the results of the
Effect of 2h on the distribution of cell-cycle phase in HeLa S-3
Treatment time (h)
SubG1 (%)
G1 (%)
S (%)
G2/M (%)
Control
12
24
36
48
1.51
9.72
32.74
43.51
50.14
56.42
8.40
21.18
18.97
18.38
19.62
15.38
15.93
15.62
20.16
22.45
65.92
30.01
19.19
9.52
present study, it is reasonable to suggest that
a-tubulin might
HeLa S-3 cells were incubated in the absence (control) or presence of 2h for 12, 24,
36, or 48 h. Then, the cells were fixed and stained with propidium iodide to analyze
DNA content by flow cytometer.
interacting with 2h. In addition, the flow cytometric analysis dem-
onstrated G2/M-phase arrest of the cell cycle with 2h. Many micro-
tubule-interfering agents bind to b-tubulin and induce arrest of cell
division. Among them, colchicines, vinblastine and taxol have
played major roles in practical uses as well as in biochemical
studies of microtubules functions. Each of these compounds binds
to b-tubulin, however, the protein that interacted with 2h was
H
N
O
a
-tubulin. Therefore, 2h appears to have a different mechanism
of action compared with the colchicines, vinblastine and taxol.
The function of -tubulin in the apoptosis pathway is unclear,
N
N
a
2h'
and antitumor drugs that are currently in use do not target this
protein. For these reasons, the consequences of 2h effects on
CONHCH2NH2
a-tubulin remain unclear. Characterization of the relationship
between -tubulin and apoptosis requires identification of
a
molecular/biological mechanisms.
Acknowledgments
This work was partially supported by a Grant-in-Aid for Scien-
tific Research from MEXT, a matching fund subsidy from MEXT
2004–2006 (No. 16550148); a grant for the ‘High-Tech Research
Center’ Project for Private Universities, a matching fund subsidy
from MEXT, 2000–2004 and 2005–2007; a Grant from the Japan
Private School Promotion Foundation (2008–2009); a Grant-in-
Aid for Scientific Research form MEXT, a matching fund subsidy
from MEXT 2009–2011 (No. 21590025); and a grant for the Devel-
opment of strategic Research Center in Private Universities sup-
ported by MEXT, Center for Technologies against Cancer (CTC),
2009–2013. We thank Sanae Takasugi, Masaki Takizawa, Takuya
Akahane and Taro Nittono (Department of Pure and Applied Chem-
istry, Faculty of Science and Technology, Tokyo University of Sci-
ence (RIKADAI)) for experimental work in the synthesis of 3d–k
and MTT assay of 3h–i.
α
β
Figure 4. SDS–PAGE analysis of the proteins enriched by affinity chromatography.
The ligand (2h) was introduced using amino-linker (upper panel). M, size-marker;
(1) using NHS-activated sepharose 4 Fast Flow as negative control; (2) solubilized
fraction of interacting protein; (3) insolubilized fraction of interacting protein; (4)
whole cell lysate.
proteins that interact with 2h, because it showed the highest anti-
tumor-activity and induced apoptosis. The compound 2h was used
as the ligand on an affinity column, and was immobilized using an
amino-linker (Fig. 4, upper panel). The proteins that interacted
with the probe 2h0 were separated by electrophoresis, and subse-
quently identified using matrix-assisted laser desorption ioniza-
tion/time of flight-mass spectrometry (MALDI/TOF-MS) (Fig. 4,
Supplementary data
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.bmcl.2011.06.061.
lower panel, a–d). The proteins shown in a–d were a-tubulin, eno-
lase, b-actin, ribosomal protein, respectively.
References and notes
Many anticancer drugs are cytotoxic because they interfere
with cell cycle progression by inhibition of activation of cell cycle
regulation, inhibition of DNA replication, induction of DNA dam-
age, or disruption of mitotic spindle formation.1 In particular, since
taxol was discovered, spindle microtubules have become a popular
target for the development of new anticancer drugs.2,3 These drugs
inhibit cell proliferation either by increasing microtubule polymer-
ization, for example, taxol, or by promoting microtubules depoly-
merization, for example, the vinca alkaloids, thus impeding cell
division and inducing cell death at relatively high concentrations.5
These reports indicate that a delay in progression through mitosis
by the induction of aberrant spindle formation might be another
mechanism by which anticancer drugs alter cancer cell growth.
In the present study, results of affinity chromatography and
peptide mass fingerprinting showed that the protein with the
strongest interaction was enolase (Fig. 4). However, inhibition of
1. Stewart, Z. A.; Westfall, M. D.; Pietenpol, J. A. Trends Pharmacol. Sci. 2003, 24,
139.
2. Gibbs, J. B. Science 2000, 287, 1969.
3. Jordan, M. A. Curr. Med. Chem. Anti-Cancer Agents 2002, 2, 1.
4. Wood, K. W.; Cornwell, W. D.; Jackson, J. R. Curr. Opin. Pharmacol. 2001, 1, 370.
5. Dumontet, C.; Sikic, V. I. J. Clin. Oncol. 1999, 17, 1061.
6. Cao, R.; Peng, W.; Wang, Z.; Xu, A. Curr. Med. Chem. 2007, 14, 479.
7. Hayashi, K.; Nagao, M.; Sugimura, T. Nucleic Acids Res. 1977, 4, 3679.
8. Ishida, J.; Wang, H. K.; Bastow, K. F.; Hu, C. Q.; Lee, K. H. Bioorg. Med. Chem. Lett.
1999, 9, 3319.
9. Cao, R.; Peng, W.; Chen, H.; Ma, Y.; Liu, X.; Hou, X.; Guan, H.; Xu, A. Biochem.
Biophys. Res. Commun. 2005, 338, 1557.
10. Zhao, T. J.; Rosenberg, H. C.; Chiu, T. H. Eur. J. Pharmacol. 1996, 306, 61.
11. Garcia, M. D.; Wilson, A. J.; Emmerson, D. P.; Jenkins, P. R. Org. Biomol. Chem.
2006, 4, 4478.
12. Castro, A. C.; Dang, L. C.; Soucy, F. S.; Grenier, L. Bioorg. Med. Chem. Lett. 2003,
13, 2419.
13. Konakahara, T.; Iida, T.; Iwaki, T.; Kumagai, M.; Sakai, N. JP2006321753.
14. Yohan, B.; Iain, C. Bioorg. Med. Chem. Lett. 2004, 14, 5841.