New tricks for an Old natural product: Discovery of highly potent evodiamine derivatives as novel antitumor agents by systemic structure-activity relationship analysis and biological evaluations

Article abstract of DOI:10.1021/jm300605m

Evodiamine is a quinazolinocarboline alkaloid isolated from the fruits of traditional Chinese herb Evodiae fructus. Previously, we identified N13-substituted evodiamine derivatives as potent topoisomerase I inhibitors by structure-based virtual screening and lead optimization. Herein, a library of novel evodiamine derivatives bearing various substitutions or modified scaffold were synthesized. Among them, a number of evodiamine derivatives showed substantial increase of the antitumor activity, with GI₅₀ values lower than 3 nM. Moreover, these highly potent compounds can effectively induce the apoptosis of A549 cells. Interestingly, further computational target prediction calculations in combination with biological assays confirmed that the evodiamine derivatives acted by dual inhibition of topoisomerases I and II. Moreover, several hydroxyl derivatives, such as 10-hydroxyl evodiamine (10j) and 3-amino-10-hydroxyl evodiamine (18g), also showed good in vivo antitumor efficacy and low toxicity at the dose of 1 mg/kg or 2 mg/kg. They represent promising candidates for the development of novel antitumor agents.

Full text of DOI:10.1021/jm300605m

Article
pubs.acs.org/jmc
New Tricks for an Old Natural Product: Discovery of Highly Potent
Evodiamine Derivatives as Novel Antitumor Agents by Systemic
Structure−Activity Relationship Analysis and Biological Evaluations
Guoqiang Dong,† Shengzheng Wang,† Zhenyuan Miao, Jianzhong Yao, Yongqiang Zhang, Zizhao Guo,
Wannian Zhang,* and Chunquan Sheng*
Department of Medicinal Chemistry, School of Pharmacy, Second Military Medical University, 325 Guohe Road, Shanghai 200433,
People’s Republic of China
S
* Supporting Information
ABSTRACT: Evodiamine is a quinazolinocarboline alkaloid isolated from the fruits of traditional Chinese herb Evodiae fructus.
Previously, we identified N13-substituted evodiamine derivatives as potent topoisomerase I inhibitors by structure-based virtual
screening and lead optimization. Herein, a library of novel evodiamine derivatives bearing various substitutions or modified scaffold
were synthesized. Among them, a number of evodiamine derivatives showed substantial increase of the antitumor activity, with
GI50 values lower than 3 nM. Moreover, these highly potent compounds can effectively induce the apoptosis of A549 cells.
Interestingly, further computational target prediction calculations in combination with biological assays confirmed that the
evodiamine derivatives acted by dual inhibition of topoisomerases I and II. Moreover, several hydroxyl derivatives, such as
10-hydroxyl evodiamine (10j) and 3-amino-10-hydroxyl evodiamine (18g), also showed good in vivo antitumor efficacy and low
toxicity at the dose of 1 mg/kg or 2 mg/kg. They represent promising candidates for the development of novel antitumor agents.
discovery of new therapeutic agents and also define novel drug
targets.3 One of the most significant examples is the discovery
and development of camptothecin (CPT, 1)-based antitumor
INTRODUCTION
■
Natural products have long been an important source of drugs or
leads.1 In the case of antitumor agents, natural products or their
synthetic derivatives still comprise more than 50% of the drugs
that are used for cancer chemotherapy.2 Natural products often
serve as biologically prevalidated drugs or leads that can lead to
Received: May 2, 2012
© XXXX American Chemical Society
A
dx.doi.org/10.1021/jm300605m | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Figure 1. Chemical structures of camptothecin and evodiamine derivatives.
agents.4 Compound 1 is isolated from the Chinese tree
Camptotheca acuminata in 1966.5 Nearly 10 years later, its
unique mode of action was found to be selective inhibition
of DNA topoisomerase I (Top1).6 Structural modifications of
natural product 1 have generated two effective antitumor agents,
namely topotecan (TPT)7 and irinotecan (IRT),8 and a number
of new drug candidates (e.g., rubitecan,9 lurtotecan,10 and
exatecan11). This successful story highlights the importance of
medicinal chemistry and pharmacological efforts on leads from
natural products.
Evodiae fructus (Chinese name: Wu-Chu-Yu) is a traditional
Chinese herb that has been widely used for gastrointestinal
disorders, amenorrhea, headache, and postpartum hemorrhage
for thousands of years.12,13 Evodiamine (2, Figure 1), a quinazo-
linocarboline alkaloid, is the major bioactive constituent isolated
from the fruits of Evodiae fructus. Compound 2 has shown a wide
range of biological activities including anti-inflammatory,14−16
antiobesity,17−20 and antitumor21 effects. In particular, the cyto-
toxicity of compound 2 on various human cancer cell lines has
been extensively studied. Compound 2 has been reported
to inhibit the proliferation of a wide variety of tumor cells by
inducing their apoptosis.22−25 It has been recently reported that
various molecular mechanism involved in compound 2-induced
apoptosis including caspase-dependent26−28 and -independent29
pathways, sphingomyelin pathway,30 PI3K/Akt/caspase, and
Fas-L/NF-κB signaling pathways.31
Although the antiproliferative and apoptotic effects of
evodimaine have been well studied, it is unsuitable to subject
compound 2 for direct clinical development because it exhibits
antitumor activity only at micromolar concentrations.21 The
antitumor potency and physicochemical properties of compound
2 remain to be significantly improved. Interestingly, compound 2
has several attractive features as an antitumor lead. First, it shows
broad-spectrum antitumor activities. Besides its antiproliferative
and apoptotic effects, compound 2 suppresses the invasion and
migration of human colon carcinoma cells and melanoma cells to
the lung.32,33 Second, the toxicity of compound 2 is relatively low.
It was reported that compound 2 had no toxic effects against
normal peripheral blood mononuclear cells,22 and its LD50 value
in mice was 77.79 mg/kg.34 Third, compound 2 is a drug-like
molecule that meets the criteria of Lipinski’s “rule of five”35 and
also meets the rules for lead-like or fragment-like molecules.36
The relatively low molecular weight of compound 2 (MW = 303)
makes it suitable for further optimization into drug-like
candidates because the MW value is often increased during the
process of lead-to-candidate.37,38 In addition, compound 2 has
good synthetic accessibility that provides a good opportunity to
efficiently generating the library of derivatives.
In our previous investigations, compound 2 was identified as a
novel Top1 inhibitor by structure-based virtual screening.45 It
shares a unique “L type” conformation in the active site of Top1.
Only the indole ring of compound 2 intercalates at the DNA
cleavage site and forms a hydrogen bond with Arg364. On the
basis of the binding mode, preliminary optimization of compound 2
was performed by design and synthesis of a series of N-substituted
analogues. Compared to compound 2, the 4-chloro benzoyl
derivative 3, the most active one, showed significantly improved
Top1 inhibitory activity, antitumor spectrum, and cytotoxicity.45
Inspired by these encouraging results, the present inves-
tigation will provide systemic structure−activity relationship
(SAR) studies of compound 2. As a result, a series of highly active
derivatives were successfully identified. They showed excellent
antiproliferative activities at nanomolar concentrations with a
broad spectrum. Particularly, their good in vivo antitumor
potency were confirmed in two xenograft models. Moreover, in
silico target identification in combination with biological assays
indicated that these derivatives acted by dual inhibition of Top1
and Top2.
CHEMISTRY
■
The synthesis of the target compounds were based on the
coupling of ring ABC (substituted 3,4-dihydro-β-carboline) with
ring DE (substituted N-methyl isatoic anhydride). Ring ABC
(6 and 12) were prepared by reacting substituted tryptamine 4
with ethyl formate and followed by intramolecular ring closure in
the presence of POCl3. Starting from substituted 2-amino-
benzoic acid, it was reacted with triphosgene to give isatoic
anhydride, which was methylated by CH3I to afford ring DE
(9 and 13). Then, intermediate 6 (or 12) was treated with
compound 13 (or 9) in CH2Cl2 at room temperature to give
A-ring derivatives 10a−g (Scheme 1), E-ring derivatives 14a−o
(Scheme 2), and A, E-ring derivatives 17a−e (Scheme 3) with
moderate to high yields. Hydroxyl substituted derivatives, such as
10j, 11e, 14p, 14q, and 18a−e, were obtained by demethylation
of the corresponding methoxyl derivatives in the presence of
BBr3 and CH2Cl2 at low temperature. Subsequently, the hydroxyl
group could be alkylated by treating with various haloalkanes
and K2CO3 to afford compounds 10k−p. Nitro-substituted
derivatives 10h and 10i were synthesized by direct nitration of
compound 2. Reduction of the nitro group of compounds 14j
and 17d using 10% Pd/C and H2 afforded the amino derivatives
14r and 18f. Compound 14r was further alkylated or acylated to
give compounds 15a−d (Scheme 2).
In the presence of NaH and DMF, several derivatives, such as
10, 14f, 14i, 19, 21, and 28, were treated with 4-chlorobenzoyl
chloride to afford the N13-substituted compounds 11a−d, 11e,
16a−b, 20, 22, and 29. Treatment of compound 2 or compound
10g with Lawesson’s reagent afforded the thiocarboyl derivatives
19 and 23. Reduction of D-ring amide group by LiAlH4 yielded
compound 21 (Scheme 4). In the presence of 2-iodoxybenzoic
acid (IBX) and DMSO, rutaecarpine analogues 26a−b were
Despite the promising properties of compound 2 as a good
starting point for the development of novel antitumor agents, its
medicinal chemistry and molecular targets are poorly under-
stood. Most of the synthetic and medicinal chemistry efforts were
focused on rutaecarpine, a N14 analogue of compound 2.39−44
B
dx.doi.org/10.1021/jm300605m | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
a
Scheme 1
a
Reagents and conditions: (a) ethyl formate, reflux, 6 h, yield 100%; (b) POCl3, CH2Cl2, 0 °C, 6 h, yield 85.4−93.6%; (c) triphosgene, THF, reflux,
4 h, yield 95%; (d) CH3I, KOH, DMF, 2 h, yield 87.4%; (e) CH2Cl2, rt, 12 h, yield 52.6−71.6%; (f) KNO3, H2SO4, 0 °C, 12 h, yield 58.1−91.3%;
(g) K2CO3, EtOH, reflux, 6 h, yield 76.3−88.8%; (h) 4-chlorobenzoyl chloride, NaH, DMF, 80 °C, 12 h, yield 28.9−52.1%; (i) BBr3, CH2Cl2,
−78 °C, 6 h, yield 58.3−63.2%.
synthesized by the oxidation of demethyl derivatives 25a−b.
Coupling of carboline 12 with 2-hydroxybenzoyl chloride (27)
afforded the oxa- derivative 28, which was subsequently acylated
to compound 29.
RESULTS AND DISCUSSION
■
A-Ring Modified Derivatives. Previously, the effects of
N13 substitutions of compound 2 on the biological activities
have been investigated.45 Herein, SAR studies are mainly focused
on introducing various substitutions on compound 2 or
modifying its scaffold. First, a series of C10 and C12 substituted
derivatives were synthesized. The growth inhibitory activities
toward human cancer cell lines A549 (lung cancer), MDA-
MB-435 (breast cancer), and HCT116 (colon cancer) were
determined using MTT assay.45 Compound 1, TPT, and IRT
were used as reference drugs. It can be seen that the type and
position of the substitutions play an important role for the
antiproliferative potency (Table 1). The 10-chloro derivative 10c
was inactive toward all the three cancer cell lines. When the
chlorine atom of compound 10c was replaced by other halogen
atoms (i.e., fluorine, bromine, and iodine), improved antitumor
activity was observed. The iodine atom (compound 10e) seems
Two isomers of compound 18a were synthesized by
ruthenium(II)-catalyzed asymmetric transfer hydrogenation
(Scheme 5).46 The β-carboline intermediate 35 was prepared
by reacting 4-(benzyloxy)aniline with ethyl 2-oxopiperidine-3-
carboxylate, followed by treating with HCOOH. In the presence
of POCl3, β-carboline 35 was condensed with 5-fluoro-2-
(methylamino)benzoate in dry THF to afford the dehydro
derivative 37 in 50% yield. Using Noyori’s procedure,47
asymmetric reduction of 37 catalyzed by RuCl[(S,S)-Tsdpen]-
(p-cymene) or RuCl[(R,R)-Tsdpen](p-cymene) gave (S)-38
and (R)-38, respectively. Finally, after removal of the protection
group, compounds (S)-18a and (R)-18a were obtained in
90% ee.
C
dx.doi.org/10.1021/jm300605m | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
a
Scheme 2
a
Reagents and conditions: (a) CH2Cl2, 45 °C, 6 h, yield 13−87%; (b) BBr3, CH2Cl2, −20 °C, 5 h, yield 50−52%; (c) 10% Pd/C, H2, DMF, rt, 12 h,
yield 73%; (d) 1,3-dibromopropane or 3-bromopropyne, K2CO3, DMSO, 60 °C, 7 h, yield 27−32%; (e) Ac2O or cyclopropanecarbonyl chloride,
Et3N, CH2Cl2, rt, 2 h, yield 73−82%; (f) 4-chlorobenzoyl chloride, NaH, DMF, 80 °C, 8 h, yield 31−86%.
a
Scheme 3
a
Reagents and conditions: (a) CH2Cl2, 45 °C, 6 h, yield 45−91%; (b) BBr3, CH2Cl2, −20 °C, 5 h, yield 31−63%; (c) 10% Pd/C, H2, DMF, rt, 12 h,
yield 79%; (d) BBr3, CH2Cl2, −20 °C, 5 h, yield 31%.
to be the most favorable at C10. Compound 10e showed
higher antiproliferative activity (GI50 range: 0.26−11.39 μM)
than reference drug 2 and IRT, but it was less potent than
compound 1 and TPT. On the other hand, the movement
of the chlorine atom from C10 (compound 10c) to C12
(compound 10a) led to the substantial increase of the
antitumor activity. Compound 10a showed excellent anti-
proliferative activity against MDA-MB-435 and HCT116
cancer cell lines with GI50 value lower than 0.003 μM, which
was more active than all the reference drugs. The introduc-
tion of the nitro or methyl group on the C10 position of
compound 2 resulted in two inactive derivatives (10f and 10h).
D
dx.doi.org/10.1021/jm300605m | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
a
Scheme 4
a
Reagents and conditions: (a) Lawesson’s reagent, PhMe, reflux, 6 h, 47.9−65.6%; (b) LiAlH4, THF, rt, 12 h, yield 58.6%; (c) 4-chlorobenzoyl
chloride, NaH, DMF, 80 °C,12 h, yield 69.5%; (d) 4-chlorobenzoyl chloride, NaH, DMF, 80 °C,12 h, yield 56.9%; (e) CH2Cl2, rt, 6 h, 78.5−84.1%;
(f) IBX, DMSO, rt, 12 h, yield 86.7−89.9%; (g) CH2Cl2, rt, 12 h, yield 57.6%; h, 4-chlorobenzoyl chloride, NaH, DMF, 0 °C, 12 h, yield 57.2%.
The 10-nitro derivative 10h can be activated by adding another
nitro group on the C12 position. The dinitro derivative 10i
showed good activity against the MDA-MB-435 cell line (GI50 =
0.16 μM), but it was also inactive toward the other two cancer cell
lines.
In our previous studies, the 4-chloro-benzoyl group was found
to be favorable for the N13 substitution.45 In an attempt to
investigate the synergistic effects of the A-ring and B-ring indole
NH substitutions, C10 and N13 disubstituted derivatives 11a−e
were designed and synthesized. For some inactive or weakly
active derivatives, such as compounds 10c, 10d, 10f and 10h, the
addition of a 4-chloro-benzoyl group on the N13 atom resulted
in improved antiproliferative activity and antitumor spectrum.
For example, the 10-methyl derivative 10f was inactive, whereas
its corresponding C10 and N13 disubstituted derivative 11c
showed good antiproliferative activity against the MDA-MB-435
(GI50 = 1.14 μM) and HCT116 (GI50 = 11.4 μM) cell lines. For
the highly active compound 10j, the corresponding analogue 11e
showed comparable activity against the A549 cell line, but its
potency against the other two cell lines was decreased.
E-Ring Modified Derivatives. 1-Methyl derivative (14a)
showed potent in vitro antitumor activity, with GI50 values
ranging from 0.011 to 0.23 μM (Table 2). When the C1 methyl
group was moved to C3 (compound 14h) or C4 (compound
14m), the antiproliferative activity was decreased obviously.
Introducing the substitutions on the C2 position led to the loss of
the activity because two C2-substituted derivatives, 14b and 14c,
The most potent compound in this series is the 10-hydroxyl
derivative 10j, whose GI50 value was lower than 0.003 μM for all
the three cancer cell lines. In contrast, the replacement of the C10
hydroxyl group by a methoxyl group (compound 10g) led to
significant decrease or loss of the antitumor activity. To further
validate the importance of the hydroxyl group, a series of ether
or ester derivatives 10k−p were designed and synthesized. The
in vitro antitumor assay revealed that all the ether derivatives
showed decreased antiproliferative activity. Particularly, they
almost lost the activity against the A549 cancer cell line. For the
SAR of the substitutions on the hydroxyl group, the oxirane
methyl group (compound 10m) was relatively better than
the ethyl (compound 10k), propyl (compound 10l), and allyl
(compound 10n) group. The benzyl substituted analogy 10p was
nearly inactive. In contrast, the ester derivative 10p was more
active than the ether analogues, especially for the activity against
the HCT116 cell line.
E
dx.doi.org/10.1021/jm300605m | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
a
Scheme 5
a
Reagents and conditions: (a) NaNO2, hydrochloric acid, H2O, 0−5 °C, 2 h; (b) KOH, H2O, rt, 12 h; (c) rt, 10 h; (d) HCOOH, reflux, 0.5 h, yield
59%; (e) POCl3, THF, reflux, 3 days, yield 50%; (f) Cat. A, HCOOH:Et3N (5:2), 0 °C, 8 h, yield 87%; (g) Cat. B, HCOOH:Et3N (5:2), 0 °C, 8 h,
yield 83%; (h) 10% Pd/C, H2, DMF, rt, 12 h, yield 85−86%
were both inactive. In contrast, most of the C3-substituted
derivatives, except for 14o and 14j, showed good antitumor
activity. Among them, 3-fluoro (14d) and 3-chloro (14e)
derivatives were the most active compounds in this series, with
GI50 value lower than 0.003 μM for all the three cancer cell lines.
For the other 3-halogen derivatives, the 3-bromo derivative 14f
showed moderate activity, whereas the 3-iodine derivative 14g
was almost inactive. When the C3 position was substituted by the
methyl (compound 14h) or methoxyl (compound 14i) group,
good antiproliferative activity was also observed for them, but
they were less potent than the 3-fluoro or 3-chloro derivatives. As
compared to the 3-substituted analogues, significantly decreased
activity was observed for the corresponding 4-substituted
derivatives (compounds 14k−n).
14n by a hydroxyl group (compounds 14p and 14q) led to
significantly improved antitumor activity against the MDA-MB-
435 and HCT116 cell lines. Particularly, the GI50 value of the
3-hydroxyl derivatives (14p) for the MDA-MB-435 cell line was
lower than 0.003 μM. The 3-nitro derivative 14j was inactive,
but its reduced product 14r showed potent activity against the
MDA-MB-435 (GI50 = 0.15 μM) and HCT116 (GI50 = 8.92 μM)
cell lines. When the amine group of compound 14r was
dialkylated, the corresponding compounds 15a and 15b showed
increased activity against the HCT116 cell line. However, two
amide derivatives 15c and 15d were only weakly active. Notably,
all the amine and amide analogues were inactive again the A549
cell line. Compounds 16a and 16b were synthesized to evaluate
the synergistic effects of the C3 and N13 substitutions.
Unfortunately, the results were the reverse of the A-ring SAR.
The additional 4-chloro-benzoyl group on the N13 position had
negative effects on the antiproliferative activity.
Adding another methoxyl group on the C2 position of com-
pound 14i resulted in the decreased activity (compound 14o).
Interestingly, the replacement of the methoxyl group of 14i and
F
dx.doi.org/10.1021/jm300605m | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
introduction of a C3-fluoro, chloro, or hydroxyl group on the
E-ring also led to substantial increase of the antiproliferative
activity. In an attempt to investigate their synergistic effects
with E-ring, disubstituted analogues 17a−e and 18a−g were
synthesized by combing the favorable substitutions of the
A- and E- ring. It can be seen that all the hydroxyl derivatives
are more active than their corresponding methoxyl analogues
(Table 3). The results were consistent with A- and E-ring SAR,
Table 1. In Vitro Antitumor Activity of A-Ring Modified
Evodiamine Derivatives (GI50, μM)
compd
A549
MDA-MB-435
HCT116
10a
10b
10c
10d
10e
10f
10g
10h
10i
10j
10k
10l
10m
10n
10o
10p
11a
11b
11c
11d
11e
1
2.55 0.17
6.85 0.41
>200
<0.003
27.35 1.25
>200
<0.003
8.56 0.63
>200
>200
37.57
>200
11.39 1.05
>200
0.26 0.03
>200
0.78 0.02
>200
Table 3. In Vitro Antitumor Activity of A-Ring and E-Ring
Disubstituted Evodiamine Derivatives (GI50, μM)
>200
8.97 0.52
>200
>200
>200
>200
compd
A549
MDA-MB-435
HCT116
>200
0.16 0.01
<0.003
>200
<0.003
<0.003
17a
17b
17c
17d
17e
18a
18b
18c
18d
18e
18f
200 13.68
13 1.03
>200
<0.003
2.0 0.06
9.8 0.67
>200
0.78 0.05
0.004 0.000
0.033 0.001
0.15 0.02
42 3.02
>200
5.93 0.23
3.07 0.10
0.61 0.03
4.39 0.25
41.25
19.17 1.32
8.05 0.74
2.98 0.35
13.96 1.28
>200
>200
48.88 2.33
>200
>200
98 5.87
<0.003
21 1.29
>200
<0.003
<0.003
1.9 0.04
>200
0.08 0.001
72.72 5.22
38.73 4.02
1.14 0.05
11.83 0.93
<0.002
0.01 0.002
>200
0.12 0.005
0.75 0.04
1.6 0.08
0.23 0.02
62 5.25
4.0 0.13
0.005 0.000
0.003 0.000
0.66 0.04
0.013 0.001
12 1.03
0.005 0.000
0.007 0.000
0.07 0.001
0.009 0.000
6.6 0.50
22.54 1.62
>200
19.65 0.86
11.40 0.98
70.99 6.11
0.05 0.002
0.083 0.001
100 7.23
0.40 0.02
7.1 0.25
87.20 5.20
26.58 1.09
0.17 0.005
100 9.60
3.96 0.64
18g
0.076 0.006
0.007 0.000
0.033 0.003
20.20 1.65
0.034 0.001
12.2 1.94
2
which further highlighted the importance of hydroxyl group for
the antitumor activity. In addition, excellent antiproliferative
activity was retained after the combination of A-ring and E-ring
substitutions. Highly active compounds 18a−c, 18e, and 18g
were more potent than the reference drugs. For example, the GI50
value of the 3-fluoro-10-hydroxyl derivative (18a) was lower than
0.003 μM against all the three cancer cell lines. Compounds 18b,
18c, 18e, and 18g were more selective to the MDA-MB-435 and
HCT116 cell lines with GI50 values in low nanomolar range.
More importantly, after introducing one or two hydrophilic
groups (i.e., OH and NH2) on the scaffold of compound 2, the
water solubility can be significantly improved, which is very
helpful for their in vivo antitumor efficacy.
D-Ring Modified Derivatives. Unlike the structural
optimization of A-ring and E-ring, the investigation of D-ring
SAR was mainly focused on the modification of the scaffold.
First, the C5 carbonyl group of compound 2 was replaced by a
thiocarbonyl group (compound 19) or reduced to the methylene
group (compound 21). As compared to compound 2, the
thiocarbonyl derivative 19 showed improved antiproliferative
activity against all the three cancer cell lines (GI50 range: 12.24−
90.73 μM). To validate the positive effect of the thiocarbonyl
group, compound 23 was synthesized and evaluated. Similarly,
the thiocarbonyl derivative 23 was more active than the cor-
responding carbonyl analogue 10g. On the other hand, the
antitumor activity of the reduced product 21 was similar to that
of compound 2. Unfortunately, the addition of the 4-chloro-
benzoyl group on the N13 position of compounds 19 and 21 led
to nearly loss of the activity. The SAR results are reverse to those
of compound 2, indicating that there might be different SARs
when the scaffold of compound 2 was changed. Second, the
SAR of the N14 methyl group was investigated. Two N14
demethyl derivatives, 25a and 25b, were synthesized, which
showed decreased activities. When the free amine group of
compounds 25a−b was oxidized to the enamine group (26a−b),
namely changing the scaffold of compound 2 to rutaecarpine,48
their antiproliferative activity was still lower than the corresponding
a
TPT
b
IRT
5.0 0.05
a
b
TPT = topotecan. IRT = irinotecan.
Table 2. In Vitro Antitumor Activity of E-Ring Modified
Evodiamine Derivatives (GI50, μM)
compd
A549
MDA-MB-435
HCT116
14a
14b
14c
14d
14e
14f
0.23 0.015
>200
0.084 0.002
>200
0.011 0.001
>200
>200
>200
>200
<0.003
<0.003
<0.003
<0.003
<0.003
<0.003
14.49 1.06
>200
1.50 0.08
2.94 0.09
0.32 0.01
0.29 0.02
>200
14.07 1.04
>200
14g
14h
14i
32.17 2.89
0.94 0.01
>200
0.61 0.03
1.25 0.05
>200
14j
14k
14l
200 15.36
5.65 0.25
14.30 1.03
128.11 15.20
>200
42.60 5.03
1.43 0.10
20.20 1.62
22.26 2.01
8.97 0.88
<0.003
>200
4.56 0.18
13.14 0.68
19.71 1.32
>200
14m
14n
14o
14p
14q
14r
15a
15b
15c
15d
16a
16b
>200
0.013 0.000
2.14 0.24
8.92 0.95
3.85 0.29
0.21 0.01
70.50 6.00
186.57 23.65
39.71 0.33
88.68 5.84
62.41 5.65
>200
0.24 0.02
0.15 0.02
0.20 0.01
2.03 0.12
59.15 4.23
>200
>200
>200
>200
>200
>200
8.72 0.55
184.39 12.29
>200
Synergistic Effects of A-Ring and E-Ring Modifications.
The results of A-ring SAR indicated that the C3 hydroxyl group is
important for the antitumor activity. On the other hand, the
G
dx.doi.org/10.1021/jm300605m | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
the S-isomer was about 580-fold more potent than the R-isomer.
However, there was reverse trend for the CNE cell line (human
thyroid carcinoma) and (R)-18a showed slightly higher activity
than the S-isomer.
Table 4. In Vitro Antitumor Activity of D-Ring Modified
Evodiamine Derivatives (GI50, μM)
compd
A549
MDA-MB-435
HCT116
19
90.73 8.09
>200
12.24 1.26
>200
46.49 3.88
>200
Summary of the SARs. Summary of the SARs is depicted in
Figure 2. First, the introduction of the proper substitution on
the scaffold of compound 2 can led to substantial improvement
of the antitumor activity. Positions C1, C3, C10, and C12 are
favorable to be substituted. Fluorine is favored at the C3 position,
while chlorine is suitable for both C3 and C10 position. It should
be noted that the free hydroxyl group is important for the
excellent antiproliferative potency. Second, the modification of
D-ring scaffold is tolerated. The transformation of the C5
carbonyl to the thiocarbonyl group or changing the N14-methyl
group to an oxygen atom has positive effects on the antitumor
activity. Third, there are synergistic effects for the C3 and C10
substitutions. Compounds 18a−c and 18g, bearing one or two
hydrophilic groups, not only showed excellent antitumor activity
but also possess good water solubility, which were subjected to
the evaluation of in vivo antitumor efficacy.
Antitumor Spectrum of the Selected Derivatives. From
the above results, a number of derivatives of compound 2 were
found to be highly potent against A549, HCT116, and MDA-
MB-435 cell lines. To evaluate their antiproliferative effects in a
broader manner, the activity of compounds 10a, 14d, and 18a
were tested against a variety of human cancer cell lines including
U20S (osteosarcoma), PANC-1 (prostate carcinoma), HepG2
(liver cancer), PC-3 (pancreatic cancer), HL-60 (acute
myeloblastic leukemia), and Saos-2 (osteogenic sarcoma). As
depicted in Table 6, the tested compounds showed broad anti-
tumor spectrum. They are particularly effective against the
HL-60 cell line with GI50 values lower than 3 nM. They also
showed moderate to good antiproliferative activity against the
PANC-1, HepG2, and PC-3 cell lines. Notably, compound 18a
showed remarkably higher activity against the PANC-1 cell
line (GI50 < 3 nM) than compound 1 (GI50 = 6.29 μM). With the
exception of compound 14d, the other compounds including the
20
21
>200
16.17 1.24
95.95 2.66
3.69 0.23
18.50 0.96
>200
>200
22
>200
>200
23
106.17 9.67
60.12 3.22
>200
21.38 1.08
26.71 1.77
>200
25a
25b
26a
26b
28
>200
>200
137.7 10.05
20.0 1.76
3.03 0.15
>200
47.2 3.02
2.40 0.12
>200
15.9 1.44
3.13 0.20
>200
29
derivatives of compound 2 (Table 4). Third, the N14-methyl
group of compound 2 was replaced by the oxygen atom
to afford a new scaffold (oxa-evodiamine). Interestingly,
compound 28 was far more potent than compound 2 with GI50
values in the range of 2.4−3.13 μM. However, its N13 substituted
derivative 29 was inactive. The result further validated the fact
that different D-ring scaffolds resulted in different SAR for the
N13 position.
Effect of C13b Chirality on the Antitumor Activity.
Compound 2 and its derivatives have an asymmetric center at the
C13b position (Figure 1). The effect of C13b chirality on the
antitumor activity was investigated for the first time. Both R- and
S- isomers of compound 18a, a highly active derivative, were
synthesized by asymmetric transfer hydrogenation. To obtain
robust results, their antiproliferative activity was evaluated for
five kinds of human cancer cell lines. As shown in Table 5, the
C13b chirality showed distinct effects toward different cell lines.
For the A549 and HCT116 cell lines, (S)-18a was slightly more
active than the R-isomer. Significant difference of the activity
was observed for the ZR-75−30 (human breast cancer) and
KYSE (human esophageal squamous cell carcinoma) cell lines;
Table 5. In Vitro Antitumor Activity of the Isomers of Compound 18a (GI50, μM)
isomers
A549
HCT116
ZR-75-30
CNE
KYSE-150
(R)-18a
(S)-18a
0.0046 0.000
0.0037 0.000
0.011 0.000
0.0036 0.000
1.75 0.09
5.88 0.42
8.11 0.79
1.23 0.10
0.003 0.000
0.003 0.000
Figure 2. Summary of the structure−activity relationships of the evodiamine derivatives.
H
dx.doi.org/10.1021/jm300605m | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Table 6. In Vitro Antitumor Spectrum of Selected Evodiamine Derivatives (GI50, μM)
compd
U20S
PANC-1
HepG2
PC-3
HL-60
Saos-2
10a
14d
18a
1
>296
129 10.23
>296
0.12 0.01
1.70 0.09
<0.003
2.31 0.13
188 12.55
23.8 1.65
0.22 0.006
>296
34.2 2.85
92.3 6.77
>287
<0.003
<0.003
<0.003
<0.003
>296
205 16.02
>296
>287
6.29 0.50
>287
Figure 3. (A) The effect of evodiamine derivatives on human cancer cell morphology. A549 cells were treated with DMSO and 5 μM of compounds 10j
and 18g for 48 h, and representative photographs were captured under a light microscope. (B) Evodiamine derivatives-induced cell apoptosis. A549 cells
were treated with DMSO and 5 μM of compounds 10j and 18g for 48 h. Apoptosis was examined by flow cytometry (n = 3). Representative photographs
from three independent experiments were displayed.
reference drug 1 were inactive against the U20S and Saos-2 cell
lines.
same conditions. Moreover, the effects of several active derivatives
on the induction of apoptosis in A549 cells were evaluated by
flow cytometry. The percentage of apoptotic cells in the control
group and DMSO group at 48 h was 0.47% and 0.43%,
respectively. After treating with 5 μM of compounds for 48 h,
the percentage of apoptotic cells for compounds 10j and 18g was
13.07% and 14.48%, respectively, suggesting that they showed
remarkable apoptotic effect.
Top1 Inhibitory Activity and the Binding Modes.
Previously, Top1 was identified as a target of compound 2 and
its derivatives. They act by stabilizing a covalent Top1−DNA
complex called the cleavable complex.45 Herein, Top1-mediated
DNA cleavage assays with purified Top1 were used to investigate
The Effect of Selected Derivatives on Apoptosis in
A549 Cells. The effect of selected derivatives of compound 2,
namely compounds 10j and 18g, on the induction of apop-
tosis was also evaluated by cell morphology (Figure 3A) and
fluorescence-activated cell sorting (Figure 3B). The in vitro
morphological changes of the cells treated with selected derivatives
were examined. Visual assessment of the immunostaining
suggested that the majority of A549 cells became small, round,
and floating after thetreatment with 5 μM of various derivatives for
48 h. In contrast, the untreated control group and DMSO group
displayed a normal shape and well-ordered cytoskeleton under the
I
dx.doi.org/10.1021/jm300605m | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Figure 4. Top1 and Top2 inhibitory activity of the evodiamine derivatives. (A) Inhibition of Top1 relaxation activity at 100 μM. Lane 1, supercoiled
plasmid DNA; lane 2, DNA + Top1; lane 3, DNA + Top1 + evodiamine; lane 4, DNA + Top1 + compound 1; lanes 5−17, DNA + Top1 + evodiamine
derivatives (14a, 14q, 14d, 14e, 14p, 14r, 18a, 10a, 18e, 18b, 10j, 18g, and 18c, respectively). (B) Inhibition of Top1 relaxation activity at 50 μM. The
lanes are the same as the assay at 100 μM. (C) Inhibition of Top2 relaxation activity at 50 μM. Lane 1, supercoiled plasmid DNA; lane 2, DNA + Top2;
lane 3, DNA + Top2 + etoposide; lane 4, DNA + Top2 + evodiamine; lanes 5−17, DNA + Top2 + evodiamine derivatives.
the inhibitory activity of representative compounds with high
activity (i.e., compounds 10a, 10j, 14a, 14d, 14e, 14p, 14q, 14r,
18a, 18b, 18c, 18e, and 18g). Top1 and supercoilded DNA
pBR322 were incubated in the presence of the test compounds,
whose ability to stabilize the cleavable complex were evaluated by
the appearance of short DNA fragments. As shown in Figure 4A,
all the tested compounds were found to be active against Top1-
mediated relaxation of supercoiled DNA at the concentration
of 100 μM. All of them showed higher inhibitory activity than
compound 2, which was only active at the concentration of
500 μM. We further tested the Top1 inhibitory activity at the
concentration of 50 μM, and four compounds, namely 10a, 18b,
10e, and 18g, retained inhibitory activity (Figure 4B). Notably,
the tested compounds were less active than compound 1, but
they showed comparable or better antiproliferative activities,
suggesting that there might be additional targets for them.
Next, the binding modes of compounds 10j, 18e, 18g, and 18c
were investigated by molecular docking and molecular dynamics
(MD) simulations. All of these compounds bear one or two
hydroxyl group on ring A or ring E, and thus the function of
these hydroxyl groups can be investigated. GOLD 5.0.149 and
Desmond,50 with previously established protocols,45 were used
for computer modeling. As depicted in Figure 5, the binding
poses of the derivatives were similar to that of compound 2. Their
indole ring intercalated at the site of DNA cleavage and formed
base-stacking interactions with both the −1 (upstream) and
+1 (downstream) base pairs. However, unlike compound 2, its
derivatives formed different hydrogen bonding interactions with
the active site of Top1. Previously, the indole NH of compound 2
was found to form a hydrogen bond with Arg364.45 For the
hydroxyl derivatives, their indole NH was deviated from Arg364.
Instead, the oxygen atom of A-ring hydroxyl group formed
hydrogen bonding interaction with Arg364. Notably, the D-ring
carbonyl oxygen atom formed an additional hydrogen bond with
Asn352. However, no direct interaction between E-ring hydroxyl
or amine group and Top1 was observed.
Discovery of Evodiamine Derivatives as Dual Top1/
Top2 Inhibitors by in Silico Target Identification. The
results of Top1 assay indicated that the antiproliferative activity
of the tested derivatives did not correlate well with their Top1
inhibitory. It is inferred that there might be additional targets for
them. To validate the hypothesis, ChemMapper,51,52 a free web
server for computational target identification, was used to predict
the potential targets for the highly potent compound 14d.
ChemMapper is based on the concept that compounds sharing
high 3D similarities may have relatively similar target association
profile. ChemMapper performs the 3D similarity searching, ranking,
and superposition by combining the strength of molecular shape
superposition and chemical feature matching. After searching
the MDDR database,53 a list of potential targets was identified (see
Table S1 in Supporting Information). Interestingly, topoisomerase
II (Top2), a well-known antitumor target, was found to be ranked
the fourth in the list. Top2 exists in two isoforms, α (alpha) and
β (beta), and Top2α is commonly used for assaying.54 Thus, the
Top2α inhibitory activity of compound 14d was determined in a
relaxation assay.55 As shown in Figure 4C, compound 14d was
proven to be a potent inhibitor of the catalytic activity of Top2.
Particularly, it was more active than the well-known Top2α inhibitor
etoposide (for relative Top2α inhibition rate, see Table S2 in
Supporting Information). Moreover, compound 14d was observed
to be competitive binding with ATP and the increase of the
ATP concentration resulted in the loss of the inhibitory activity
of compound 14d. Inspired by the results, other highly potent
compounds were also assayed for their Top2α inhibitory activity
(Figure 4C). To our delight, most of the tested compounds showed
moderate to good Top2α inhibitory activity, except for compounds
10a, 14e, and 14p. Besides compound 14d, derivatives 14q, 18a,
18b, 10j, 18f, and 18c also showed better activity than etoposide.
The above results indicated that compound 2 and its derivatives are
dual inhibitors of Top1 and Top2. As compared to compound 2,
the significant increase of the antitumor activities for its derivatives
J
dx.doi.org/10.1021/jm300605m | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Figure 5. Schematic representation of the proposed binding mode for compounds 10j (A), 18e (B), 18g (C), and 18c (D) in the Top1−DNA complex.
The figure was generated using PyMol (http://www.pymol.org/).
Figure 6. Schematic representation of the proposed binding mode for compounds 10j (A), 18e (B), 18g (C), and 18c (D) in the ATP-binding domain
of Top2α. The figure was generated using PyMol (http://www.pymol.org/).
may be due to their improved inhibitory activity toward both Top1
and Top2.
surrounding residues. For compound 18e, its C4 hydroxyl group
formed two hydrogen bonds with Asp94 and Asn150,
respectively. For compounds 18g and 18c, hydrogen bonding
interaction was observed between its C3 amine or hydroxyl
group with Asn150 and Thr159. The docking model was
consistent with the fact that E-ring hydroxyl or amine was
important for the activity.
DNA Binding Properties. Interaction of the compounds
with DNA induces both hypochromic and bathochromic shifts
in the absorption spectrum.56 The interaction of compounds 10j,
To investigate the interactions between the tested compounds
and Top2α, compounds 10j, 18e, 18g, and 18c were subjected
to molecular docking and molecular simulations. They were
observed to be well-fitted into the ATP-binding domain of
Top2α (Figure 6). The A-ring C10 hydroxyl group formed a
hydrogen bond with the backbone amide of Ile88. Interestingly,
the E-ring hydroxyl or amine group that did not directly interact
with Top1 formed hydrogen bonding interactions with the
K
dx.doi.org/10.1021/jm300605m | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Figure 7. UV/vis absorption spectra of compounds 10j, 18c, 18e, and 18g (50 μM) alone (free drug lines) or with increasing concentrations of ct-DNA
from 20 to 100 μM, respectively.
18c, 18e, and 18g with calf thymus DNA (ct-DNA) were
examined by absorbance titration experiments (Figure 7). As
DNA concentration was increased, both hypochromic and
bathochromic shifts were not observed in these experiments. It
was concluded that these compounds could not bind to the
DNA. To further prove our conclusion, Top1 unwinding assay
and Hoechst 33258 displacement experiments were performed
to test whether these compounds could intercalate into DNA or
bind to the minor groove of DNA. A negatively supercoiled
pBR322 DNA was relaxed by an excess amount of Top1 in the
present of the test compounds. Then, the compounds were
removed by phenol extraction. If the compound is an inter-
calator, such as ethidium bromide, the intercalation of the
compound between DNA base pairs induces a positively
supercoiled DNA.57 As shown in Figure 8, the intercalator
ethidium bromide induced a supercoiled DNA following Top1
treatment. However, DNA remained relaxed in the present of 10j,
18c, 18e, and 18g, indicating that they were not DNA intercalators.
The fluorescence of the groove binder Hoechst 33258 increases by
140-fold upon DNA binding.58,59 So the displacement of Hoechst
33258 could be monitored by a decrease in the fluorescence
intension.60,61 As the test compounds concentration was increased,
the fluorescence intension remained unchanged, indicating that
compounds 10j, 18c, 18e, and 18g failed to bind to the minor
groove of DNA (Supporting Information).
Mechanism of Top1 and Top2 Inhibition. Both Top1 and
Top2 inhibitors were classified into two categories, poisons and
catalytic inhibitors. Top1 poison, such as CPT, could stabilize
the Top1−DNA cleavage complex and increase the intensity of
nicked band in the EtBr containing gel. Compared with CPT,
compounds 10j, 18c, 18e, and 18g could also increase the
intensity of nicked band (Figure 9A), indicating these compounds
were Top1 poisons. Top2 poisons, including etoposide and
doxorubicin, act by stabilizing Top2−DNA cleavage complex and
lead to the formation of a new DNA band corresponding to linear
plasmid DNA. In contrast to etoposide, no linear plasmid DNA
was detected in the case of test compounds (10j, 18c, 18e, and
18g, Figure 9B). This result suggested that these compounds did
not stabilize the Top2−DNA cleavage complex. It was concluded
that these compounds acted as Top2 catalytic inhibitors rather
than Top2 poisons.
Inhibition of the Tumor Growth of A549 and HCT116
Tumor Xenografts in Nude Mice. To investigate the in vivo
effects of the synthesized compounds on tumor growth, two
xenograft models were prepared by inoculating human lung
cancer A549 or human colon cancer HCT116 cells on the nude
L
dx.doi.org/10.1021/jm300605m | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Figure 8. Investigation of the intercalating ability of the evodiamine derivatives based on Top1 unwinding assay. Lane 1, supercoiled plasmid DNA
(pBR322); lane 2, DNA + Top1; lane 3, DNA + Top1 + EtBr (1 μg/mL); lanes 4−7, DNA + Top1 + evodiamine derivatives (10j, 18c, 18e, and 18g at
100 μM, respectively). The position of nicked DNA (Nck), relaxed DNA (Rel), and supercoiled pBR322 (SC) was indicated.
Figure 9. Top1 and Top2 DNA cleavage gels of the evodiamine derivatives. (A) Top1-mediated DNA cleavage assay gel of test compounds using
supercoiled pUC19. Lane 1, supercoiled plasmid DNA (pUC19); lane 2, DNA + Top1; lane 3, DNA + Top1 + compound 1 (100 μM); lanes 4−7, DNA +
Top1 + evodiamine derivatives (10j, 18c, 18e, and 18g at 100 μM, respectively). (B) Top2-mediated DNA cleavage assay gel of test compounds using
supercoiled pBR322. Lane 1, supercoiled plasmid DNA (pBR322); lane 2, DNA + Top2; lane 3, DNA + Top2 + etoposide (100 μM); lanes 4−7, DNA +
Top2 + evodiamine derivatives (10j, 18c, 18e, and 18g at 100 μM, respectively). The position of nicked DNA (Nck), liner DNA (Lin), relaxed DNA (Rel),
and supercoiled DNA (SC) was indicated.
mice. Because the in vivo antitumor potency usually does not
correlate well with their in vitro antiproliferative activities,
representative compounds were selected to evaluate their
potency in the two xenograft models. The compounds for in
vivo testing were chosen according to their chemical structures
and antitumor activity. Initially, compounds 10j and 14d were
tested in the A549 xenograft model at the dose of 1 mg/kg.
As shown in Figure 10A, intraperitoneal (ip) injection of the
two compounds for 13 days significantly inhibited tumor growth
(p < 0.05). Next, the dose of compounds 14e and 18a were
doubled (2 mg/kg), but no improvement of the tumor growth
inhibition was observed (Figure 10B). Meanwhile, all the com-
pounds were found to be well tolerated during the test and
showed no significant loss of body weight, suggesting that the
toxicity of them is low. Among them, the 10-hydroxyl derivative
10j showed the best in vivo potency. In the HCT116 xenograft
model, all the tested compounds (i.e., 10a, 10j, 18a, 18b, 18c,
and 18g) also showed significant inhibitory effects on the tumor
growth at the dose of 2 mg/kg (Figure 10C). Except for
compounds 10a, 18b, and 18c, body weight of the nude mice was
not significantly affected during the test (see Figure S1 in
Supporting Information). As compared to compound 10j, its
3-chloro derivative 18b showed increased in vivo potency,
whereas the potency of its 3-fluoro derivative 18a was decreased.
The highest in vivo antitumor potency was observed for 3-amino-
10-hydroxyl evodimine (18g).
The Requirements and Advantages of Dual Top1/Top2
Inhibitors. Topoisomerases are validated antitumor targets that
act as complex molecular engines to modulate DNA topology
and maintain chromosome superstructure and integrity.62 Top1
or Top2 inhibitors show outstanding antitumor efficacy, which
form the basis of many cancer chemotherapeutic protocols.
Despite decades of study, the investigation of new biological
functions of topoisomerases as therapeutic targets and also the
development of novel inhibitors remains highly active.62−64 The
available evidence suggests that the therapeutic potential of
topoisomerase is far from exhausted.64 For many forms of
malignant tumors, potent cytotoxic agents, such as topoisomer-
ase inhibitors, are frequently used as the only remedy to limit
rapid tumor growth, progression, and metastasis. Moreover,
topoisomerase inhibitors can be essentially used in combination
with targeted therapeutic agents, such as kinase inhibitors,
especially for the treatment of advanced solid tumors. However,
current Top1 or Top2 inhibitors have two major limitations:
toxicity and drug resistance. In this scenario, the development of
dual Top1/Top2 inhibitors represents a promising approach to
overcome the drawbacks of single inhibitors.65,66 Dual Top1/
Top2 inhibitors have several advantages as novel antitumor
agents. Given that both Top1 and Top2 are well-qualified anti-
tumor targets and haveoverlapping functions inDNA metabolism,
targeting both Top1 and Top2 could increase overall antitumor
activity with reduced toxic side effects.67 On the other hand, the
M
dx.doi.org/10.1021/jm300605m | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
resistance to Top1 or Top2 inhibitors is an important problem
that remains to be solved.63 It was reported that the emergence
of resistance to Top1 inhibitors is often accompanied by a
concomitant rise in the level of Top2 expression and vice versa.67
In this regard, a single compound able to inhibit both Top1 and
Top2 could be helpful to prevent mechanism-based drug
resistance. Despite the high interest in such compounds, there
are relatively few classes of dual inhibitors.68 The present study
discovered evodiamine derivatives as a novel type of dual Top1/
Top2 inhibitors. Interestingly, they showed excellent in vitro and
in vivo antiproliferative activity with low toxicity. The results
preliminary validated the advantages of dual Top1/Top2
inhibitors and the designed compounds represent high-quality
antitumor leads for further optimization.
CONCLUSION
■
Compound 2 represents an attractive lead structure for the
development of novel antitumor agents. Inspired by our previous
findings, the SAR of compound 2 was investigated systemically.
The results highlighted the importance of hydroxyl group on the
A-ring and E-ring. Further molecular modeling and mechanism
studies confirmed that the derivatives acted by dual inhibition
of Top1 and Top2. The hydroxyl group on the scaffold was
important for the antitumor activity because it could form
important hydrogen bonding interactions with both Top1 and
Top2. Furthermore, the introduction of the hydroxyl group is
also helpful to improve the water solubility. As a result, a series of
highly potent antitumor molecules were identified which showed
promising features as antitumor candidates. They showed
excellent antiproliferative activity, with their GI50 values lower
than 3 nM. Moreover, they can effectively induce the apoptosis of
cancer cells in A549 cells. Importantly, compounds 10j, 18b, 18c,
and 18g showed good in vivo antitumor efficacy and low toxicity
at the dose of 2 mg/kg. They are worthy of further evaluation as
potential chemotherapeutic agents. On the other hand, besides
Top1 and Top2, there also might be additional targets for the
synthesized compounds, which remained to be further identified.
Further SAR and target identification studies are in progress.
EXPERIMENTAL SECTION
■
Chemistry. General Methods. 1H NMR and 13C NMR spectra were
recorded on Bruker AVANCE300, AVANCE500, or AVANCE600
spectrometer (Bruker Company, Germany), using TMS as an internal
standard and CDCl3 or DMSO-d6 as solvents. Chemical shift are given
in ppm (δ). Elemental analyses were performed with a MOD-1106
instrument and were consistent with theoretical values within 0.4%.
The mass spectra were recorded on an Esquire 3000 LC-MS mass
spectrometer. Silica gel thin-layer chromatography was performed on
precoated plates GF-254 (Qingdao Haiyang Chemical, China). Purity of
the compounds was analyzed by HPLC (Agilent Technologies 1260
Infinity) using 40:60 MeOH/H2O as the mobile phase with a flow rate
of 0.8 mL/min on a C18 column (Aglilent 20RBA × SB-C18, 5 μm,
4.6 mm × 150 mm). All compounds exhibited greater than 95% purity.
All solvents and reagents were analytically pure, and no further purifica-
tion was needed. All starting materials were commercially available.
12-Chloroevodiamine (10a). A solution of 8-chloro-4,9-dihydro-
3H-pyrido[3,4-b] indole69 (6a, 0.30 g, 1.5 mmol) and N-methyl isatoic
anhydride70 (9, 0.26 g, 1.5 mmol) in CH2Cl2 (5 mL) was stirred
overnight at room temperature. Then target compound 10a was
precipitated from this solution and washed by CH2Cl2 to give a white
solid (0.29 g, yield 58.2%). 1H NMR (CDCl3, 500 MHz) δ: 2.52 (s, 3H,
N14-CH3), 2.96−2.97 (m, 2H, C8-H2), 3.27−3.29 (m, 1H, C7-H), 4.87−
4.90 (m, 1H, C7-H), 5.94 (s, 1H, C13b-H), 7.12 (t, J = 7.8 Hz, 1H,
C10-H), 7.18 (d, J = 8.0 Hz, 1H, C9-H), 7.24−7.27 (m, 2H), 7.49−7.51
(m, 2H), 8.12 (d, J = 7.8 Hz, 1H, C4-H), 8.47 (s, 1H, N13-H). 13C NMR
Figure 10. Antitumor efficacy of evodiamine derivatives in the xenograft
models. (A) The efficacy of compounds 10j and 14d in the A549
xenograft model at the dose of 1 mg/kg. (B) The efficacy of compounds
14e and 18a in the A549 xenograft model at the dose of 2 mg/kg. (C)
The efficacy of compounds 10a, 10j, 18a, 18b, 18c, and 18g in the
HCT116 xenograft model at the dose of 2 mg/kg. Data are presented
as the mean SEM; n = 6 nude mice per group: (*) P < 0.05, (**)
P < 0.01, versus control group, determined with Student’s t test.
N
dx.doi.org/10.1021/jm300605m | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
(MeOD, 75 MHz) δ: 19.8, 29.2, 39.4, 68.5, 113.6, 117.0, 118.4, 120.0,
121.1, 121.7, 122.2, 122.7, 126.4, 127.7, 129.1, 133.2, 134.2, 150.9, 165.4.
MS (ESI, positive) m/z calcd for C19H17ClN3O (M + H): 338.81; found
338.61. Anal. (C19H16ClN3O) C, H, N. HPLC purity: 98.3%. The
synthetic method for target compounds 10b−10g was similar to that of
compound 10a.
123.9, 126.7, 129.0 (2C), 131.8, 133.1, 153.7, 164.7. MS (ESI, positive)
m/z calcd for C21H22N3O2 (M + H): 348.42; found 348.53. Anal.
(C21H21N3O2) C, H, N. HPLC purity: 99.2%. Compounds 10l−10o
were synthesized by a similar procedure.
1-Methylevodiamine (14a). A solution of 3,4-dihydro-β-carboline
12 (0.05 g, 0.3 mmol) and 1,8-dimethyl-1H-benzo[d][1,3]oxazine-2,4-
dione (0.06 g, 0.3 mmol) in CH2Cl2 (10 mL) was stirred at 45 °C for 6 h
and then cooled to room temperature. The solvent was concentrated in
vacuo, and the crude product was purified by column chromatography
(hexane/EtOAc = 5:1) to give compound 14a (0.08 g, yield 80%) as a
10-Nitroevodiamine (10h). A solution of KNO3 (0.67 g, 6.6 mmol)
and compound 2 (2 g, 6.6 mmol) in concentrated sulfuric acid (20 mL)
was stirred at 0 °C for 12 h. Then the solution was poured to ice−water
mixture (100 mL), and a yellow solid was precipitated from the solution.
The crude product was collected and purified by silica gel column
chromatography (hexane/EtOAc = 3:1) to give 10h as yellow solid
(2.10 g, yield 91.3%). 1H NMR (DMSO-d6, 500 MHz) δ: 2.89 (s, 3H,
N14-CH3), 2. 94−2.98 (m, 2H, C8-2H), 3.21−3.26 (m, 1H, C7-H),
4.63−4.67 (m, 1H, C7-H), 6.20 (s, 1H, C13b-H), 6.99 (t, J = 7.5 Hz, 1H,
C3-H), 7.07 (d, J = 7.9 Hz, 1H, C1-H), 7.48−7.53 (m, 2H), 7.80 (d, J =
7.5 Hz, 1H, C4-H), 8.02 (dd, J = 9.0 Hz, 2.2 Hz, 1H, C11-H), 8.51 (s, 1H,
C9-H), 11.87 (s, 1H, N13-H). 13C NMR (DMSO-d6, 150 MHz) δ: 19.2,
36.8, 40.7, 69.5, 112.1, 114.2, 115.5, 117.2, 117.6, 119.1, 120.5, 125.3,
128.0, 133.6, 134.9, 139.6, 140.6, 148.5, 164.1. MS (ESI, positive) m/z
calcd for C19H17N4O3 (M + H): 349.36; found 349.91. Anal.
(C19H16N4O3) C, H, N. HPLC purity: 98.1%.
1
white solid. H NMR (DMSO-d6, 500 MHz) δ: 2.18 (s, 3H), 2.42
(s, 3H), 2.82 (m, 1H), 2.93 (m, 1H), 3.19 (m, 1H), 4.25 (m, 1H), 5.94
(s, 1H), 7.04−7.80 (m, 7H), 11.21 (s, 1H). 13C NMR (DMSO-d6,
150 MHz) δ: 14.3, 16.9, 20.2, 39.3, 69.3, 112.1, 112.1, 118.8, 119.2,
122.4, 124.9, 126.2, 126.3, 127.9, 128.6, 132.7, 134.8, 137.4, 149.5, 164.1.
MS (ESI, negative) m/z calcd for C20H18N3O (M − H): 316.38; found
316.97. Anal. (C20H19N3O) C, H, N. HPLC purity: 98.7%. Compounds
14b−14i and 14m−14o were synthesized by a similar procedure.
3-Hydroxyevodiamine (14p). To a stirred solution of 14i (0.1 g,
0.3 mmol) in CH2Cl2 (10 mL), a solution of BBr3 (0.06 mL, 0.6 mmol)
in CH2Cl2 (5 mL) was added over a period of 2−3 min at −20 °C. The
reaction mixture was stirred under nitrogen atmosphere for 5 h, and then
5 mL of methanol was added. The reaction mixture was treated with
saturated sodium bicarbonate (50 mL) and extracted with CH2Cl2 (3 ×
50 mL). The combined organic layers were dried over MgSO4 and
filtered. The solvent was removed under reduced pressure, and the
residue was purified by silica gel column chromatography (CH2Cl2/
MeOH = 100:1) to give compound 14p as a white solid (0.05 g, yield
50%). 1H NMR (DMSO-d6, 300 MHz) δ: 2.36 (s, 3H), 2.79−2.82 (m,
2H), 3.13−3.16 (m, 1H), 4.62−4.66 (m, 1H), 5.95 (s, 1H), 6.93−7.13
(m, 5H), 7.28 (d, J = 2.7 Hz, 1H), 7.36 (d, J = 8.1 Hz, 1H), 7.50 (d,
J = 8.1 Hz, 1H), 9.45 (s, 1H), 11.25 (s, 1H). 13C NMR (DMSO-d6,
150 MHz) δ: 19.8, 36.9, 39.3, 68.9, 111.5, 111.59, 113.1, 118.3, 118.8,
120.8, 121.8, 123.5, 123.8, 125.7, 129.3, 136.8, 142.6, 153.4, 163.6. MS
(ESI, negative) m/z calcd for C19H16N3O2 (M − H): 318.35; found
318.82. Anal. (C19H17N3O2) C, H, N. HPLC purity: 98.3%. The
synthetic method for compound 14q was similar to the synthesis of
compound 14p.
10,12-Dinitroevodiamine (10i). A solution of KNO3 (1.33 g,
13.2 mmol) and compound 2 (2 g, 6.6 mmol) in concentrated sulfuric
acid (20 mL) was stirred in an ice bath for 12 h. After that, the solution
was poured into ice−water mixture (100 mL) and a yellow solid was
precipitated from this solution. The crude product was collected and
purified by silica gel column chromatography (hexane/EtOAc = 3:1) to
1
give 10i as yellow solid (1.51 g, yield 58.1%). H NMR (DMSO-d6,
500 MHz) δ: 2.62 (s, 3H, N14-CH3), 2.90 (m, 2H, C8-H2), 3.10−3.18
(m, 1H, C7-H), 4.65−4.71 (m, 1H, C7-H), 6.05 (s, 1H, C13b-H), 7.11 (t,
J = 7.5 Hz, 1H, C3-H), 7.22 (d, J = 8.1 Hz, 1H, C1-H), 7.54 (t, J = 7.8 Hz,
1H, C2-H), 7.88 (d, J = 7.5 Hz, 1H, C4-H), 8.83 (s, 1H, C11-H), 9.05 (s,
1H, C9-H), 12.65 (s, 1H, N13-H). 13C NMR (DMSO-d6, 150 MHz) δ:
19.7, 35.9, 39.4, 67.7, 114.3, 117.2, 119.5, 121.1, 121.7, 122.0, 128.1,
129.7, 131.2, 131.8, 133.1, 135.7, 139.4, 149.9, 163.6. MS (ESI, negative)
m/z calcd for C19H14N5O5 (M − H): 392.34; found 392.48. Anal.
(C19H15N5O5) C, H, N. HPLC purity: 97.6%.
10-Hydroxyevodiamine (10j). To a stirred solution of compound
10g (0.5 g, 1.5 mmol) in CH2Cl2 (30 mL), BBr3 (0.5 mL) was added in
one portion and stirred for 6 h at −78 °C. After the addition of methanol
(0.4 mL) and CH2Cl2 (50 mL), the solution was washed with saturated
NaHCO3 aqueous solution (60 mL) and brine (40 mL) and dried over
Na2SO4, and the solvent was removed under reduced pressure. The
residue was purified by silica gel column chromatography (hexane/
EtOAc = 2:1) to afford 10j (0.30 g, yield 63.2%) as white solid. 1H NMR
(DMSO-d6, 500 MHz) δ: 2.65−2.68 (m, 1H, C8-H), 2.80−2.89 (m, 1H,
C8-H), 2.89 (s, 3H, N14-CH3), 3.15−3.19 (m, 1H, C7-H), 4.58−4.62 (m,
1H, C7-H), 6.08 (s, 1H, C13b-H), 6.61−6.62 (d, 1H, J = 8.5 Hz, C11-H),
6.75 (s, 1H, C9-H), 6.94 (t, J = 7.5 Hz, 1H, C3-H), 7.02 (d, J = 8.2 Hz,
1H, C1-H), 7.14 (d, J = 8.5 Hz, 1H, C12-H), 7.46 (t, J = 7.5 Hz, 1H,
C2-H), 7.78 (d, J = 7.8 Hz, 1H, C4-H), 8.68 (s, 1H, C10-OH), 10.70 (s,
1H, N13-H). 13C NMR (DMSO-d6, 75 MHz) δ: 20.0, 31.1, 36.8, 70.4,
102.6, 111.5, 112.5, 112.5, 117.5, 119.5, 120.5, 127.1, 128.4, 131.4, 131.6,
133.9, 149.1, 151.1, 164.7. MS (ESI, positive) m/z calcd for C19H18N3O2
(M + H): 320.37; found 320.14. Anal. (C19H17N3O2) C, H, N. HPLC
purity: 98.6%. Compound 11e was synthesized by a similar procedure.
10-Ethoxyevodiamine (10k). A solution of compound 10j (0.32 g,
1 mmol), K2CO3 (0.16 g, 1.2 mmol), and bromoethane (0.13 g,
1.2 mmol) in ethanol (20 mL) was refluxed for 6 h. Then, the solvent
was removed under reduced pressure, and the residue was purified by
silica gel column chromatography (hexane/EtOAc = 3:1) to afford 10k
(0.28 g, 79.9%) as a yellow solid. 1H NMR (CDCl3, 500 MHz) δ: 1.45 (t,
J = 7.0 Hz, 3H, C10-OCH2CH3), 2.51 (s, 3H, N14-CH3), 2.92 (m, 2H,
C8-H2), 3.27 (m, 1H, C7-H), 4.10 (q, J = 7.0 Hz, 2H, C10-OCH2CH3),
4.85 (m, 1H, C7-H), 5.90 (s, 1H, C13b-H), 6.92 (d, J = 8.8 Hz, 1H,
C11-H), 7.02 (d, J = 2.0 Hz, 1H, C9-H), 7.14 (d, J = 7.7 Hz, 1H, C1-H),
7.23 (t, J = 7.5 Hz,1H, C3-H), 7.29 (d, J = 8.8 Hz, 1H, C12-H), 7.49 (t,
1H, C2-H), 8.10−8.12 (m, 2H). 13C NMR (CDCl3, 75 MHz) δ: 15.0,
20.2, 37.2, 39.6, 64.2, 68.9, 101.8, 112.0, 113.4, 113.8, 122.1 (2C), 123.6,
4-Hydroxyevodiamine (14q). White solid (0.054 g, yield 52%). 1H
NMR (DMSO-d6, 300 MHz) δ: 2.79−2.80 (m, 1H), 2.93−2.95 (m,
1H), 2.99 (s, 3H), 3.20−3.23 (m, 1H), 4.57−4.60 (m, 1H), 6.24 (s, 1H),
6.28 (d, J = 8.4 Hz, 1H), 6.38 (d, J = 8.4 Hz, 1H), 6.99 (t, J = 7.5 Hz, 1H),
7.10 (t, J = 7.5 Hz, 1H), 7.27−7.36 (m, 2H), 7.45 (d, J = 7.5 Hz, 1H),
11.00 (s, 1H), 12.51 (s, 1H). 13C NMR (acetone-d6, 150 MHz) δ: 21.0,
38.1, 42.0, 71.9, 105.3, 108.9, 110.4, 113.0 113.6, 119.8, 120.7, 123.7,
127.9, 131.6, 136.6, 138.4, 151.2, 163.7, 170.9. MS (ESI, negative) m/z
calcd for C19H16N3O2 (M − H): 318.35; found 318.94. Anal.
(C19H17N3O2) C, H, N. HPLC purity: 97.6%.
3-Aminoevodiamine (14r). 10% Pd/C (0.01 g) was added to a
stirring solution of 14j (0.12 g, 0.34 mmol) in DMF (10 mL), and the
reaction mixture was stirred under hydrogen atmosphere for 12 h at
room temperature. The mixture was diluted with water (50 mL) and
then extracted with EtOAc (3 × 50 mL). The combined organic layers
were washed with saturated sodium chloride solution (3 × 50 mL), dried
over anhydrous Na2SO4, and concentrated under reduced pressure. The
residue was purified by silica gel column chromatography (gradient
CH2Cl2/MeOH = 100:1 to 100:2) to give compound 14r as a yellow
solid (0.08 g, yield 73%). 1H NMR (DMSO, 500 MHz) δ: 2.27 (s, 3H),
2.77−2.79 (m, 1H), 2.88−2.92 (m, 1H), 3.11−3.13 (m, 1H), 4.63−4.67
(m, 1H), 5.10 (s, 2H), 5.90 (s, 1H), 6.78 (dd, J = 8.5 Hz, 2.5 Hz, 1H),
6.94 (d, J = 8.5 Hz, 1H), 7.01 (t, J = 7.5 Hz, 1H), 7.10−7.14 (m, 2H),
7.36 (d, J = 8.0 Hz, 1H), 7.51 (d, J = 8.0 Hz, 1H), 11.26 (s, 1H). 13C
NMR (DMSO-d6, 75 MHz): δ 20.11, 36.98, 39.23, 69.53, 111.75,
112.05, 112.54, 118.55, 119.18, 120.27, 122.21, 123.91, 124.34, 126.21,
129.42, 137.28, 141.46, 145.26, 164.87. MS (ESI, positive) m/z calcd for
C19H19N4O (M + H): 319.38; found 319.93. Anal. (C19H18N4O) C, H,
N. HPLC purity: 98.9%. The synthetic method for target compounds
18f was similar to the synthesis of compound 14r.
3-(Azetidin-1-yl)evodiamine (15a). To a stirred mixture of com-
pound 14r (0.1 g, 0.31 mmol), potassium carbonate (0.08 g, 0.62 mmol),
O
dx.doi.org/10.1021/jm300605m | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
and DMSO (5 mL), 1,3-dibromopropane (0.25 g, 1.24 mmol) was
added dropwise within 1 h at room temperature. Then the mixture was
heated to 60 °C and stirred for 7 h. After cooling, the mixture was diluted
with water (50 mL) and extracted with EtOAc (3 × 50 mL). The organic
layer was dried over Na2SO4, and the solvent was evaporated in vacuo.
The crude product was purified by silica gel column chromatog-
raphy (CH2Cl2/MeOH = 100:2) to give compound 15a as a red solid.
C19H19N4O2 (M + H): 335.38; found 335.85. Anal. (C19H18N4O2) C,
H, N. HPLC purity: 98.5%.
5-Thioxoevodiamine (19). To a stirred solution of compound 2
(0.3 g, 1 mmol) in toluene (30 mL), Lawesson’s reagent (0.61 g,
1.5 mmol) was added and stirred for 6 h at 110 °C. Then the solvent was
removed under reduced pressure. The residue was purified by silica gel
column chromatography (hexane/EtOAc = 4:1) to afford 19 as a yellow
solid (0.21 g, yield 65.6%). 1H NMR (CDCl3, 500 MHz) δ: 2.45 (s, 3H),
3.03−3.09 (m, 2H), 3.81−3.90 (m, 1H), 5.56−5.62 (m, 1H), 5.75 (s,
1H), 7.10 (d, J = 8.1 Hz, 1H), 7.16−7.23 (m, 2H), 7.29 (d, J = 7.5 Hz,
1H), 7.42−7.49 (m, 2H), 7.63 (d, J = 7.8 Hz, 1H), 8.30 (s, 1H), 8.56 (d,
J = 8.1 Hz, 1H). 13C NMR (CDCl3, 75 MHz) δ: 19.6, 35.3, 47.5, 69.7,
111.3, 112.5, 115.4, 118.3, 119.4, 120.0, 121.7, 122.5, 126.4, 127.8, 132.4,
132.8, 136.6, 145.5, 190.7. MS (ESI, positive) m/z calcd for C19H18N3S
(M + H): 320.43; found 320.90. Anal. (C19H17N3S) C, H, N. HPLC
purity: 99.5%. Compounds 23 was synthesized by a similar procedure.
5-(Methylene)evodiamine (21). To a stirred solution of compound 2
(0.3 g, 1 mmol) in THF (30 mL), LiAlH4 (0.11 g, 3 mmol) was added
and the mixture was stirred for 12 h at room temperature. After the
addition of H2O (0.2 mL), the solvent was removed under reduced
pressure. The residue was purified by silica gel column chromatography
(hexane/EtOAc = 4:1) to afford 21 as a pale solid (0.17 g, yield 58.6%).
1H NMR (500 MHz, CDCl3) δ: 2.66 (s, 3H), 2.81(m, 2H), 3.05 (m,
1H), 3.34 (m, 1H), 3.86 (d, J = 13.1 Hz, 1H), 4.07 (d, J = 14.1 Hz, 1H),
4.82 (s, 1H), 6.95 (d, J = 7.2 Hz, 1H), 6.99−7.07 (m, 2H), 7.10−7.23
(m, 3H), 7.36 (d, J = 7.8 Hz, 1H), 7.55 (d, J = 7.5 Hz, 1H), 8.24 (s, 1H).
13C NMR (CDCl3, 75 MHz) δ: 20.6, 39.1, 50.1, 55.7, 73.6, 110.7 (2C),
1
(0.03 g, yield 27%). H NMR (DMSO-d6, 300 MHz) δ: 2.28 (t, J =
7.2 Hz, 2H), 2.35 (s, 3H), 2.80−2.83 (m, 2H), 3.10−3.13 (m, 1H), 3.79
(t, J = 7.2 Hz, 4H), 4.62−4.67 (m, 1H), 5.94 (s, 1H), 6.64 (dd, J =
8.4 Hz, 2.7 Hz, 1H), 6.88 (d, J = 2.7 Hz, 1H), 6.98−7.12 (m, 3H), 7.35
(d, J = 7.8 Hz, 1H), 7.49 (d, J = 7.8 Hz, 1H), 11.26 (s, 1H). 13C NMR
(DMSO-d6, 150 MHz) δ: 13.5, 16.5, 18.7, 19.8, 37.1, 52.1, 69.0, 109.1,
111.5, 111.6, 116.8, 118.3, 118.8, 121.8, 123.1, 123.4, 125.7, 129.4,
136.8, 141.2, 148.7, 163.8. MS (ESI, positive) m/z calcd for C22H23N4O
(M + H): 359.44; found 359.76. Anal. (C22H22N4O) C, H, N. HPLC
purity: 98.4%.
3-Acetamidoevodiamine (15c). Compound 14r (0.1 g, 0.3 mmol)
and triethylamine (0.04 mL, 0.3 mmol) were dissolved in CH2Cl2
(15 mL). Acetic anhydride (0.05 g, 0.47 mmol) was added dropwise
within 30 min, and the mixture was stirred at room temperature for 2 h.
The precipitate was collected by filtration and washed with alcohol
1
to give 15c (0.09 g, yield 82%) as a red solid. H NMR (DMSO-d6,
300 MHz) δ: 2.04 (s, 3H), 2.67 (s, 3H), 2.83−2.85 (m, 2H), 3.19−3.23
(m, 1H), 4.59−4.63 (m, 1H), 5.93 (s, 1H), 7.00 (t, J = 7.8 Hz, 1H),
7.08−7.13 (m, 2H), 7.35 (d, J = 7.8 Hz, 1H), 7.48 (d, J = 7.8 Hz, 1H),
7.73 (dd, J = 9.0 Hz, 2.7 Hz, 1H), 8.04 (d, J = 2.7 Hz, 1H), 10.23 (s, 1H),
11.27 (s, 1H). 13C NMR (DMSO-d6, 75 MHz) δ: 19.6, 23.8, 36.5, 55.0,
69.2, 111.6, 111.6, 118.2, 118.3, 118.8, 120.1, 121.2, 121.9, 124.5, 125.8,
129.8, 133.8, 136.7, 145.1, 163.8, 168.0. MS (ESI, positive) m/z calcd for
C21H21N4O2 (M + H): 361.42; found 361.96. Anal. (C21H20N4O2) C, H,
N. HPLC purity: 97.6%.
3-Fluoro-10-methoxyevodiamine (17a). Reaction of 6-methoxy-
3,4-dihydro-β-carboline 6g (0.3 g, 1.5 mmol) and 1-methyl-6-fluoro-
1H-benzo[d][1,3]oxazine-2,4-dione (0.2 g, 1.0 mmol) as described for
the synthesis of 14a, followed by purification using silica gel column
chromatography (hexane/EtOAc = 3:1) gave 17a (0.25 g, yield 71%) as
a white solid. 1H NMR (DMSO-d6, 300 MHz) δ: 2.86 (s, 3H), 3.03−
3.06 (m, 2H), 3.38−3.40 (m, 1H), 3.95 (s, 3H), 4.79−4.83 (m, 1H),
6.26 (s, 1H), 6.95 (dd, J = 8.7 Hz, 2.4 Hz, 1H), 7.19 (d, J = 2.4 Hz, 1H),
7.35−7.39 (m, 1H), 7.44 (d, J = 9.0 Hz, 1H), 7.54−7.61 (m, 1H), 7.72
(d, J = 9.0 Hz, 3.3 Hz, 1H), 11.21 (s, 1H). 13C NMR (DMSO-d6,
75 MHz) δ: 19.6, 36.6, 55.3, 69.3, 100.1, 111.4, 112.1, 112.3, 113.2,
113.5, 120.4, 121.6, 122.0, 126.1, 130.2, 131.7, 146.0, 153.4, 163.0. MS
(ESI, positive) m/z calcd for C20H19FN3O2 (M + H): 352.38; found
352.92. Anal. (C20H18FN3O2) C, H, N. HPLC purity: 99.2%. The
synthetic method for target compounds 17b−17e was similar to the
synthesis of compound 17a.
3-Fluoro-10-hydroxyevodiamine (18a). Reaction of 17a (0.1 g,
0.28 mmol) and BBr3 (0.06 mL, 0.6 mmol) as described for the synthesis
of 14p, followed by purification using silica gel column chromatography
(CH2Cl2/MeOH = 100:1) gave 18a as a white solid (0.05 g, yield 52%).
1H NMR (DMSO-d6, 600 MHz) δ: 2.67 (s, 3H), 2.72−2.73 (m, 1H),
2.75−2.81 (m, 1H), 3.14−3.18 (m, 1H), 4.56−4.59 (m, 1H), 6.02 (s,
1H), 6.29 (dd, J = 9.0 Hz, 2.4 Hz, 1H), 6.76 (d, J = 2.4 Hz, 1H), 7.13−
7.15 (m, 2H), 7.33−7.36 (m, 1H), 7.51 (dd, J = 9.0 Hz, 2.4 Hz, 1H), 8.67
(s, 1H), 10.79 (s, 1H). 13C NMR (DMSO-d6, 150 MHz) δ: 20.0, 37.0,
40.9, 69.8, 102.7, 111.2, 112.5, 112.7, 113.7, 113.9, 120.9, 121.0, 121.8,
127.0, 130.5, 131.5, 146.4, 151.2, 163.5. ESI-MS MS (ESI, negative)
m/z calcd for C19H15FN3O2 (M − H): 336.34; found 336.84. Anal.
(C19H16FN3O2) C, H, N. HPLC purity: 98.7%. Compounds 18b−18e
and 18g were synthesized by a similar procedure.
3-Amino-10-hydroxyevodiamine (18g). Yellow solid (0.03 g, yield
31%). 1H NMR (DMSO-d6, 600 MHz) δ: 2.27 (s, 3H), 2.49−2.51 (m,
2H), 2.68−2.70 (m, 1H), 5.09 (s, 2H), 5.84 (s, 1H), 6.62 (dd, J = 8.4 Hz,
2.4 Hz, 1H), 6.76 (dd, J = 8.4 Hz, 2.4 Hz, 1H), 6.79 (d, J = 2.4 Hz, 1H),
6.93 (d, J = 2.4 Hz, 1H), 7.12−7.15 (m, 2H), 8.67 (s, 1H), 10.90 (s, 1H).
13C NMR (DMSO-d6, 75 MHz) δ: 19.9, 37.0, 48.6, 69.0, 102.3,
110.6, 111.7, 111.9, 112.0, 119.4, 123.4, 123.9, 126.4, 129.8, 131.3,
140.3, 145.1, 150.6, 164.0. MS (ESI, positive) m/z calcd for
111.4, 118.1 (2C), 119.1 (2C), 121.5, 121.8, 126.4, 126.7, 130.4, 136.0,
148.0. MS (ESI, positive) m/z calcd for C19H20N3 (M + H): 290.38;
found 290.53. Anal. (C19H19N3) C, H, N. HPLC purity: 99.1%.
3-Fluoro-10-methoxy-7,8,13b,14-tetrahydroindolo[2′,3′:3,4]-
pyrido[2,1-b]quinazolin-5(13H)-one (25a). A solution of intermediate
6g (1 g, 5.0 mmol) and 24 (0.90 g, 5.0 mmol) in CH2Cl2 (30 mL) was
stirred for 6 h at room temperature. Then, the target compound 25a was
1
precipitated from the solvent as a pale solid (1.42 g, yield 84.1%). H
NMR (CDCl3, 600 MHz) δ: 2.90−2.92 (m, 2H), 3.15−3.20 (m, 1H),
3.87 (s, 3H), 4.98−5.02 (m, 1H), 5.87 (s, 1H), 6.91 (dd, J = 9.0 Hz,
2.4 Hz, 1H), 6.98 (dd, J = 8.4 Hz, 4.2 Hz, 1H), 7.00 (d, J = 2.4 Hz, 1H),
7.14 (t, J = 8.4 Hz, 1H), 7.28 (d, J = 8.4 Hz, 1H), 7.79 (dd, J = 8.4 Hz,
3.0 Hz, 1H), 8.11 (s, 1H). 13C NMR (DMSO-d6, 150 MHz) δ: 20.0,
38.7, 55.3, 63.8, 100.3, 109.2, 111.9, 112.3, 113.3, 117.2, 120.7, 126.1,
131.0, 131.3, 143.7, 153.4, 154.8, 156.3, 162.7. MS (ESI, positive) m/z
calcd for C19H17FN3O2 (M + H): 338.36; found 338.21. Anal.
(C19H16FN3O2) C, H, N. HPLC purity: 98.8%. Compound 25b was
synthesized by a similar procedure.
3-Fluoro-10-methoxyrutecarpine (26a). To a solution of com-
pound 25a (0.50 g, 1.5 mmol) in DMSO (20 mL), IBX (0.41 g,
1.5 mmol) was added and stirred for 2 h at room temperature. Then
H2O (40 mL) was added and extracted with EtOAc (3 × 30 mL). The
combined organic phases were dried over Na2SO4, and the solvent was
removed under reduced pressure. The residue was purified by silica gel
column chromatography (hexane/EtOAc = 3:1) to afford 26a as a pale
solid (0.44 g, yield 86.7%). 1H NMR (CDCl3, 600 MHz) δ: 3.22 (t, J =
7.2 Hz, 2H), 3.88 (s, 3H), 4.57 (t, J = 7.2 Hz, 2H), 7.01 (s, 1H), 7.03 (d,
J = 9.6 Hz, 1H), 7.36 (d, J = 8.4 Hz, 1H), 7.45 (t, J = 9.6 Hz, 1H), 7.70 (s,
1H), 7.94 (dd, J = 8.4 Hz, 3.0 Hz, 1H), 9.02 (s, 1H). 13C NMR (DMSO-
d6, 75 MHz) δ: 19.4, 31.1, 55.8, 100.9, 111.6, 113.9, 116.4, 117.9, 122.1,
123.2, 125.5, 127.7, 129.6, 134.4, 144.8, 145.3, 154.3, 159.2, 160.5. MS
(ESI, positive) m/z calcd for C19H15FN3O2 (M + H): 336.34; found
336.96. HPLC purity: 98.2%. The synthetic method for compound 26b
was similar to the synthesis of compound 26a.
7,8,13,13b-Tetrahydro-5H-benzo[5′,6′][1,3]oxazino[3′,2′:1,2]-
pyrido[3,4-b]indol-5-one (28). To a solution of 2-hydroxybenzoic
acid (1 g, 7.2 mmol) in CH2Cl2 (50 mL), sulfurous dichloride (1.71 g,
14.4 mmol) was added and stirred for 1 h at 40 °C. Then the solvent and
excess sulfurous dichloride were removed under reduced pressure. The
residue was added to a solution of intermediate 12 (1.02 g, 6 mmol) in
CH2Cl2 (30 mL) and stirred for 12 h at room temperature After reaction,
the solvent was removed under reduced pressure and the residue was
purified by silica gel column chromatography (hexane/EtOAc = 4:1) to
P
dx.doi.org/10.1021/jm300605m | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
1
afford 28 as a white solid (1.01 g, yield 57.6%). H NMR (DMSO-d6,
500 MHz) δ: 2.93−3.02 (m, 2H), 3.21−3.22 (m, 1H), 4.71−4.74 (m, 1H),
6.68 (s, 1H), 7.07 (t, J = 7.7 Hz, 1H), 7.13−7.27 (m, 3H), 7.42 (d, J =
8.2 Hz, 1H), 7.57−7.60 (m, 2H), 7.91 (d, J = 7.6 Hz, 1H), 11.52 (s, 1H).
13C NMR (CDCl3, 75 MHz) δ: 14.1, 39.1, 81.2, 111.6, 113.7, 116.3,
118.6, 119.3, 120.3, 123.0, 123.7, 125.9, 127.1, 128.8, 134.2, 137.2, 156.7,
163.0. MS (ESI, positive) m/z calcd for C18H15N2O2 (M + H): 291.32;
found 291.43. Anal. (C18H14N2O2) C, H, N. HPLC purity: 98.6%.
6-Benzyloxy-2,3,4,9-tetrahydro-1H-β-carbolin-1-one (35). 4-
(Benzyloxy)aniline (1.4 g, 5.8 mmol) was dissolved in 3 M hydrochloric
acid (10 mL). A solution of sodium nitrite (0.5 g, 8.7 mmol) in water
(5 mL) was added dropwise within 30 min. After stirring for 2 h under
the temperature lower than 5 °C, the solution of ethyl 2-oxopiperidine-
3-carboxylate (1.0 g, 5.8 mmol) and KOH (0.4 g, 6.9 mmol) in water
(20 mL), which had been stirred for 12 h at room temperature, was
added under the ice−water bath condition. Then the reaction mixture
was further stirred for 10 h at room temperature. The brown precipitate
was filtered and washed with water to give the crude product 34, which
was dissolved in 98% HCOOH (15 mL). The mixture was heated to
110 °C for 0.5 h and then cooled to room temperature. Next, the
reaction was poured into cold water to afford a red mixture that was
extracted three times with EtOAc. The organic layer was dried over
Na2SO4, and the solvent was removed under reduced pressure. The
residue was purified by flash chromatography (CH2Cl2/MeOH =
100:2) to give the title compound 35 as a brown solid (1.1 g, yield 59%).
1H NMR (DMSO-d6, 300 MHz) δ: 2.86 (t, J = 6.6 Hz, 2H), 3.47 (t, J =
6.6 Hz, 2H), 5.09 (s, 2H), 6.93 (m, 1H), 7.16 (d, J = 1.8 Hz, 1H), 7.25−
7.47 (m, 6H), 7.53 (s, 1H), 11.45 (s, 1H). MS (ESI, positive) m/z calcd
for C18H17N2O2 (M + H): 293.33; found 293.38. Anal. (C18H16N2O2)
C, H, N. HPLC purity: 99.1%.
(S)-3-Fluoro-10-hydroxyevodiamine (S-18a). A solution of (S)-38
(30 mg, 0.07 mmol) and 10% Pd/C (3 mg) in EtOAc (10 mL) was
stirred under hydrogen atmosphere at room temperature for 12 h. The
mixture was filtered, and the solvent was evaporated under reduced
pressure. The residue was purified by column chromatography (hexane/
EtOAc = 3:1) to give compound (S)-18a as a white solid (20 mg, yield
86%). 1H NMR (DMSO-d6, 500 MHz) δ: 2.68 (s, 3H), 2.72−2.74 (m,
2H), 2.76−2.83 (m, 1H), 4.58−4.61 (m, 1H), 6.04 (s, 1H), 6.62 (dd, J =
8.6 Hz, 2.1 Hz, 1H), 6.77 (s, 1H), 7.14−7.17 (m, 2H), 7.37 (d, J =
3.0 Hz), 7.52 (dd, J = 8.8 Hz, 2.9 Hz, 1H), 8.69 (s, 1H), 10.81 (s, 1H).
MS (ESI, negative) m/z calcd for C19H15FN3O2 (M − H): 336.34;
found 336.84. HPLC (Daicel Chiralpak AD, i-PrOH/hexane = 30/70,
flow rate = 0.8 mL/min, λ = 254 nm): tmajor = 7.09 min, tminor = 10.09 min,
ee = 90%. [α]D22 = +341° (c = 2.6 mg/mL, acetone).
(R)-3-Fluoro-10-hydroxyevodiamine (R-18a). The synthetic meth-
od for compound (R)-18a was similar to that of compound (S)-18a.
White solid, yield 85%. HPLC (Daicel Chiralpak AD, i-PrOH/hexane =
30/70, flow rate = 0.8 mL/min, λ = 254 nm): tmajor = 9.04 min, tminor
=
7.14 min, ee = 90%. [α]D22 = −374° (c = 2.0 mg/mL, acetone).
Top1 Inhibitory Activity Assay. DNA relaxation assays were
employed according to the procedure described in previous studies.45
The reaction mixture contained 35 mM Tris-HCl (pH 8.0), 72 mM KCl,
5 mM MgCl2, 5 mM dithiothreitol, 5 mM spermidine, 0.1% bovine
serum albumin (BSA), pBR322 plasmid DNA (0.25 μg), the indicated
drug concentrations (1% DMSO), and 1 unit of Top1 (TaKaRa
Biotechnology Co., Ltd., Dalian) in a final volume of 20 μL. Reaction
mixtures were incubated for 15 min at 37 °C and stopped by addition of
2 μL of 10× loading buffer (0.9% sodium dodecyl sulfate (SDS), 0.05%
bromophenol blue, and 50% glycerol). Electrophoresis was carried out
in a 0.8% agarose gel in TAE (Tris-acetate-EDTA) at 8 V/cm for 1 h.
Gels were stained with ethidium bromide (0.5 μg/mL) for 60 min. The
DNA band was visualized over UV light and photographed with Gel Doc
Ez imager (Bio-Rad Laboratories Ltd.).
3-Fluoro-10-benzyloxydehydroevodiamine (37). To a stirring
solution of 35 (0.3 g, 1 mmol) in dry THF (20 mL) was added
POCl3 (0.14 mL, 1.5 mmol) at 60 °C, and the mixture was stirred under
nitrogen atmosphere at the same temperature. After 40 min, methyl
5-fluoro-2-(methylamino)benzoate (0.28 g, 1.5 mmol) was added and
the reaction mixture was stirred under reflux for 3 days. The mixture
was diluted with water (100 mL) and then extracted with EtOAc (3 ×
100 mL). The combined organic layers were washed with saturated
sodium chloride solution (3 × 100 mL), dried over anhydrous Na2SO4,
and concentrated under reduced pressure. The residue was purified by
column chromatography (gradient CH2Cl2/MeOH = 100:2 to 100:5)
Top2 Inhibitory Activity Assay. DNA Top2α inhibitory activity of
the compounds was measured using Topoisomerase II Drug Screening
Kit (TopoGEN, Inc.). The reaction mixture contained 50 mM Tris-HCl
(pH 8.0), 150 mM NaCl, 10 mM MgCl2, 5 mM dithiothreitol, 30 μg/mL
bovine serum albumin (BSA), 2 mM ATP, pBR322 plasmid DNA
(0.25), the indicated drug concentrations (1% DMSO), and 0.75 unit of
Top2α (TopoGEN, Inc.) in a final volume of 20 μL. Reaction mixtures
were incubated for 30 min at 37 °C and stopped by addition of 2 μL
of 10% SDS. After that, 2 μL of 10× gel loading buffer (0.25%
bromophenol blue, 50% glycerol) was added. The reaction products
were analyzed on 1% agarose gel at 8 V/cm for 1 h with TAE (Tris-
acetate-EDTA) as the running buffer. Gels were stained with ethidium
bromide (0.5 μg/mL) for 60 min. The DNA band was visualized over
UV light and photographed with Gel Doc Ez imager (Bio-Rad
Laboratories Ltd.).
Absorbance Titration. Interaction between the investigational
compounds and DNA was evaluated by absorbance titration.71−73
The calf thymus DNA (ct-DNA, purchased from Sigma-Aldrich) was
dissolved in DPBS (2.67 mM KCl, 1.47 mM KH2PO4, 137.93 mM NaCl,
8.06 mM Na2HPO4, pH 7.4). DNA concentration was determined
by UV absorbance at 260 nm using the molar absorption coefficient
13200 M−1/cm in units of DNA bp. Various compounds (50 μM) were
prepared in DPBS with 1% DMSO in the presence or absence of
increasing concentrations of ct-DNA (0−100 μM). Absorption spectra
were recorded in the 220−400 nm spectral range after equilibration at
room temperature for 20 min using Synergy 2 multimode microplate
reader (BioTek Instruments, Inc.).
1
to give compound 37 as a yellow solid (0.23 g, yield 50%). H NMR
(DMSO-d6, 600 MHz) δ: 3.13 (t, J = 7.2 Hz, 2H), 4.01 (s, 3H), 4.45 (t,
J = 7.2 Hz, 2H), 5.11 (s, 2H), 7.03 (dd, J = 9.0 Hz, 2.5 Hz, 1H), 7.06 (dd,
J = 9.4 Hz, 3.0 Hz, 1H), 7.14−7.18 (m, 1H), 7.24 (d, J = 2.4 Hz, 1H),
7.29−7.32 (m, 2H), 7.36−7.39 (m, 3H), 7.44 (d, J = 7.8 Hz, 1H), 7.47
(d, J = 1.4 Hz, 1H), 11.60 (s, 1H).
(S)-3-Fluoro-10-benzyloxyevodiamine (S-38). To a stirred solution
of 37 (50 mg, 0.11 mmol) and RuCl[(S,S)-Tsdpen](p-cymene) (4 mg,
6 mmol %) in DMF (2.5 mL) was added a 5:2 HCO2H−triethylamine
(18 μL). Then the reaction mixture was stirred at 0 °C for 8 h, diluted
with water (15 mL), and extracted with EtOAc (3 × 15 mL). The
organic layer was dried over Na2SO4, and the solvent was evaporated in
vacuo. The residue was purified by column chromatography (hexane/
EtOAc = 4:1) to give compound (S)-38 as a white solid (40 mg, yield
87%). 1H NMR (DMSO-d6, 600 MHz) δ: 2.65 (s, 3H), 2.77−2.83 (m,
1H), 3.14−3.19 (m, 1H), 4.58−4.61 (m, 1H), 5.07 (s, 2H), 6.04 (s, 1H),
6.83 (dd, J = 8.7 Hz, 2.4 Hz, 1H), 7.08 (d, J = 2.4 Hz, 1H), 7.16 (dd, J =
8.9 Hz, 4.5 Hz, 1H), 7.24 (d, J = 8.7 Hz, 1H), 7.29 (t, J = 7.3 Hz, 1H),
7.33−7.38 (m, 3H), 7.45 (d, J = 7.6 Hz, 2H), 7.52 (dd, J = 8.8 Hz, 3.1 Hz,
1H), 10.99 (s, 1H). MS (ESI, positive) m/z calcd for C26H23FN3O2
(M + H): 428.48; found 428.35. Anal. (C26H22FN3O2) C, H, N. HPLC
(Daicel Chiralpak AS, i-PrOH/hexane = 30/70, flow rate = 0.8 mL/min,
λ = 254 nm): tmajor = 15.87 min, tminor = 37.87 min, ee = 99%. [α]D
+333° (c = 3.0 mg/mL, acetone).
(R)-3-Fluoro-10-benzyloxyevodiamine (R-38). The synthetic
method for compound (R)-38 was similar to that of compound (S)-
38. White solid, yield 83%. HPLC (Daicel Chiralpak AS, i-PrOH/
hexane = 30/70, flow rate = 0.8 mL/min, λ = 254 nm): tmajor = 37.87 min,
tminor = 15.87 min, ee = 99%. [α]D22 = −351° (c = 1.3 mg/mL, acetone).
Top1 Unwinding Assay. Unwinding assays was employed to
determine whether test compounds could intercalate into plasmid
DNA.57,61 The compounds (100 μM) or ethidium bromide (1 μg/mL)
were incubated with pBR322 plasmid DNA (0.25 μg) at 37 °C for
15 min in 20 μL of Top1 buffer (35 mM Tris-HCl (pH 8.0), 72 mM KCl,
5 mM MgCl2, 5 mM dithiothreitol, 5 mM spermidine, 0.1% BSA) to
ensure binding equilibrium. Five units of Top1 (TaKaRa Biotechnology
(Dalian) Co., Ltd.) was added to the reaction mixture and incubated at
37 °C for 60 min to ensure relaxation of the DNA. The reaction mixture
was stopped by the sequential addition of 2 μL of 10% SDS and 2 μL of
22
=
Q
dx.doi.org/10.1021/jm300605m | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
proteinase K at 2 mg/mL for 20 min at 50 °C. After extraction with
an equal volume of phenol/chloroform/isoamyl alcohol (25:24:1), the
samples were processed as described in the Top1 relaxation assay.
Top2-Mediated DNA Cleavage Assay. The same Top2 drug
screening kit was used as above, but this time an amount of 8 units of
hTop2α was added.74,75 Proteinase K (200 μg/mL final concentration)
was added, and the samples were incubated at 50 °C for 20 min.
Electrophoresis was run in TAE with 0.5 μg/mL ethidium bromide.
Top1-Mediated DNA Cleavage Assay. DNA cleavage assay was
performed as previously described.76,77 In brief, 20 μL volumes
containing 35 mM Tris-HCl (pH 8.0), 72 mM KCl, 5 mM MgCl2,
5 mM dithiothreitol, 5 mM spermidine, 0.1% BSA, pUC19 plasmid
DNA (0.25 μg), 4 units of Top1 (TaKaRa Biotechnology (Dalian)
Co.,Ltd.), and the indicated drug concentrations (1% DMSO) was
incubated for 15 min at 37 °C and stopped by the addition of 2 μL of
10% SDS. Proteinase K (200 μg/mL final concentration) was added,
and the samples were incubated at 50 °C for 20 min. After that, 2 μL
of 10× gel loading buffer (0.25% bromophenol blue, 50% glycerol)
was added. The samples were analyzed on 1% agarose (containing
0.5 μg/mL ethidium bromide) at 8 V/cm for 1 h with TAE as the
running buffer. The DNA band was visualized over UV light and
photographed with Gel Doc Ez imager (Bio-Rad Laboratories, Ltd.).
In Vitro Cytotoxicity Assay. Cells were plated in 96-well microtiter
plates at a density of 5 × 103/well and incubated in a humidified
atmosphere with 5% CO2 at 37 °C for 24 h. Test compounds were
added onto triplicate wells with different concentrations and 0.1%
DMSO for control. After they had been incubated for 72 h, 20 μL of
MTT (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide)
solution (5 mg/mL) was added to each well and the plate was incubated
for an additional 4 h. The formazan was dissolved in 100 μL of DMSO.
The absorbance (OD) was read on a WellscanMK-2 microplate reader
(Labsystems) at 570 nm. The concentration causing 50% inhibition
of cell growth (GI50) was determined by the Logit method.61,75 All
experiments were performed three times.
suspensions were implanted subcutaneously into the right axilla region of
mice. Treatment was begun when implanted tumors had reached a
volume of about 100−300 mm3 (after 17 days). The animals were
randomized into appropriate groups (six animals/treatment and eight
animals for the control group) and administered by ip injection once on
day 17 after implantation of cells. Tumor volumes were monitored by
caliper measurement of the length and width and calculated using the
formula of TV = 1/2 × a × b2, where a is the tumor length and b is the
width. Tumor volumes and body weights were monitored every 4 days
over the course of treatment. Mice were sacrificed on day 30−33 after
implantation of cells, and tumors were removed and recorded for analysis.
ASSOCIATED CONTENT
* Supporting Information
■
S
List of potential targets for compound 14d, relative inhibition
rate of the tested derivatives for Top1 and Top2α, experimental
protocol and results of the Hoechst 33258 displacement
experiment, structural characterization of target compounds,
and the change of body weight for the nude mice treated with
selected derivatives. This material is available free of charge via
the Internet at http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
■
*For C.Q.S.: phone/fax, 86-21-81871239, E-mail, shengcq@
hotmail.com. For W.N.Z.: phone/fax, 86-21-81871243; E-mail,
zhangwnk@hotmail.com.
Author Contributions
†These two authors contributed equally to this work.
Notes
The authors declare no competing financial interest.
Apoptosis Detection Assay. A549 cells (5 × 105 cells/mL) were
seeded in six-well plates and treated with compounds at a concentration
of 5 μM for 48 h. The cells were then harvested by trypsinization and
washed twice with cold PBS. After centrifugation and removal of the
supernatants, cells were resuspended in 400 μL of 1× binding buffer,
which was then added to 5 μL of annexin V-FITC and incubated at
room temperature for 15 min. After adding 10 μL of PI, the cells were
incubated at room temperature for another 15 min in the dark. The
stained cells were analyzed by a flow cytometer (BD Accuri C6).
Molecular Modeling. Selected compounds were docked into the
active site of Top1 by a similar procedure as described previously.45
The X-ray crystallographic structure of the ATPase domain of human
Top2α (PDB code: 1ZXM78) was applied using the CHARMm force
field (Discovery Studio 3.0 software package79), and the residues were
corrected for physiological pH. Mg ion and two water molecules (927
and 928) were conserved. The binding region was defined as all the
atoms that are 12 Å around the centroid of ADPNP in the crystal
structure. Docking runs were performed using the program GOLD 5.0.1
on a Linux PC, and the standard default settings were utilized with the
following exceptions: (i) maximal ligand flexibility was allowed, (ii) a
default maximum of 10000 genetic operations was performed, and (iii)
water molecules were allowed to toggle “on” during the docking. Early
termination was allowed if the top three solutions were within rmsd
value of 1.5 Å. ChemPLP was used for the scoring of the final poses. MD
calculations were used to validate the binding poses obtained from
molecular docking. Desmond80−83 (version Desmond_Maestro_academic-
2009−02, http://www.deshawresearch.com) was used to carry out
all the simulations, and the detailed protocols were similar to those in
our previous studies.45 The results indicated that the docked con-
figuration and stable conformation after MD calculations were similar,
with rmsd values lower than 2 Å, and important hydrogen bonding
interactions were retained.
ACKNOWLEDGMENTS
■
This work was supported in part by the National Natural Science
Foundation of China (grant 81222044, 30930107), the 863 Hi-
Tech Program of China (grant 2012AA020302), Shanghai
Rising-Star Program (grant 12QH1402600), Science and
Technology Commission of Shanghai (grant 11431920402),
and Shanghai Municipal Health Bureau (grant XYQ2011038).
We gratefully acknowledge technical support from Dr. Xiaofeng
Liu and Prof. Honglin Li (School of Pharmacy, East China
University of Science and Technology, P. R. China) for in silico
target identification.
ABBREVIATIONS USED
■
CPT, camptothecin; Top1, topoisomerase I; Top2, topoisomer-
ase II; TPT, topotecan; IRT, irinotecan; SAR, structure−activity
relationship; IBX, 2-iodoxybenzoic acid; MD, molecular
dynamics
REFERENCES
■
(1) Newman, D. J.; Cragg, G. M.; Snader, K. M. Natural products as
sources of new drugs over the period 1981−2002. J. Nat. Prod. 2003, 66,
1022−1037.
(2) Newman, D. J.; Cragg, G. M. Natural products as sources of new
drugs over the last 25 years. J. Nat. Prod. 2007, 70, 461−477.
(3) Balamurugan, R.; Dekker, F. J.; Waldmann, H. Design of
compound libraries based on natural product scaffolds and protein
structure similarity clustering (PSSC). Mol. Biosyst. 2005, 1, 36−45.
(4) Oberlies, N. H.; Kroll, D. J. Camptothecin and taxol: historic
achievements in natural products research. J. Nat. Prod. 2004, 67, 129−
135.
In Vivo Antitumor Activity. BALB/C nude male mice (certificate
SCXK-2007−0005, weighing 18−20 g) were obtained from the
Shanghai Experimental Animal Center, Chinese Academy of Sciences.
A549 lung cancer cell suspensions or HCT116 colon cancer cell
(5) Wall, M. E.; Wani, M. C.; Cook, C. E.; Palmer, K. H.; McPhail, A.
T.; Sim, G. A. The Isolation and Structure of Camptothecin, a Novel
R
dx.doi.org/10.1021/jm300605m | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Alkaloidal Leukemia and Tumor Inhibitor from Camptotheca
Acuminata. J. Am. Chem. Soc. 1966, 88, 3888−3890.
(6) Hsiang, Y. H.; Hertzberg, R.; Hecht, S.; Liu, L. F. Camptothecin
induces protein-linked DNA breaks via mammalian DNA topoisomer-
ase I. J. Biol. Chem. 1985, 260, 14873−14878.
(7) Kingsbury, W. D.; Boehm, J. C.; Jakas, D. R.; Holden, K. G.; Hecht,
S. M.; Gallagher, G.; Caranfa, M. J.; McCabe, F. L.; Faucette, L. F.;
Johnson, R. K.; et al. Synthesis of water-soluble (aminoalkyl)-
camptothecin analogues: inhibition of topoisomerase I and antitumor
activity. J. Med. Chem. 1991, 34, 98−107.
(8) Garcia-Carbonero, R.; Supko, J. G. Current perspectives on the
clinical experience, pharmacology, and continued development of the
camptothecins. Clin. Cancer Res. 2002, 8, 641−661.
(9) Giovanella, B. C.; Stehlin, J. S.; Hinz, H. R.; Kozielski, A. J.; Harris,
N. J.; Vardeman, D. M. Preclinical evaluation of the anticancer activity
and toxicity of 9-nitro-20(S)-camptothecin (Rubitecan). Int. J. Oncol.
2002, 20, 81−88.
(10) Eckhardt, S. G.; Baker, S. D.; Eckardt, J. R.; Burke, T. G.; Warner,
D. L.; Kuhn, J. G.; Rodriguez, G.; Fields, S.; Thurman, A.; Smith, L.;
Rothenberg, M. L.; White, L.; Wissel, P.; Kunka, R.; DePee, S.;
Littlefield, D.; Burris, H. A.; Von Hoff, D. D.; Rowinsky, E. K.; Phase, I
and pharmacokinetic study of GI147211, a water-soluble camptothecin
analogue, administered for five consecutive days every three weeks. Clin.
Cancer Res. 1998, 4, 595−604.
(11) Minami, H.; Fujii, H.; Igarashi, T.; Itoh, K.; Tamanoi, K.; Oguma,
T.; Sasaki, Y. Phase I and pharmacological study of a new camptothecin
derivative, exatecan mesylate (DX-8951f), infused over 30 minutes
every three weeks. Clin. Cancer Res. 2001, 7, 3056−3064.
(12) Liao, J. F.; Chen, C. F.; Chow, S. Y. Pharmacological studies of
Chinese herbs. (9) Pharmacological effects of Evodiae fructus. J.
Formosan Med. Assoc. 1981, 80, 30−38.
(13) Yu, X.; Wu, D. Z.; Yuan, J. Y.; Zhang, R. R.; Hu, Z. B.
Gastroprotective effect of fructus evodiae water extract on ethanol-
induced gastric lesions in rats. Am. J. Chin. Med. 2006, 34, 1027−1035.
(14) Chiou, W. F.; Sung, Y. J.; Liao, J. F.; Shum, A. Y.; Chen, C. F.
Inhibitory effect of dehydroevodiamine and evodiamine on nitric oxide
production in cultured murine macrophages. J. Nat. Prod. 1997, 60,
708−711.
(15) Ko, H. C.; Wang, Y. H.; Liou, K. T.; Chen, C. M.; Chen, C. H.;
Wang, W. Y.; Chang, S.; Hou, Y. C.; Chen, K. T.; Chen, C. F.; Shen, Y. C.
Anti-inflammatory effects and mechanisms of the ethanol extract of
Evodia rutaecarpa and its bioactive components on neutrophils and
microglial cells. Eur. J. Pharmacol. 2007, 555, 211−217.
(16) Liu, Y. N.; Pan, S. L.; Liao, C. H.; Huang, D. Y.; Guh, J. H.; Peng,
C. Y.; Chang, Y. L.; Teng, C. M. Evodiamine represses hypoxia-induced
inflammatory proteins expression and hypoxia-inducible factor 1alpha
accumulation in RAW264.7. Shock 2009, 32, 263−269.
(17) Kobayashi, Y.; Nakano, Y.; Kizaki, M.; Hoshikuma, K.; Yokoo, Y.;
Kamiya, T. Capsaicin-like anti-obese activities of evodiamine from fruits
of Evodia rutaecarpa, a vanilloid receptor agonist. Planta Med. 2001, 67,
628−633.
(18) Bak, E. J.; Park, H. G.; Kim, J. M.; Yoo, Y. J.; Cha, J. H. Inhibitory
effect of evodiamine alone and in combination with rosiglitazone on in
vitro adipocyte differentiation and in vivo obesity related to diabetes. Int.
J. Obes. 2010, 34, 250−260.
(19) Wang, T.; Wang, Y.; Kontani, Y.; Kobayashi, Y.; Sato, Y.; Mori, N.;
Yamashita, H. Evodiamine improves diet-induced obesity in a
uncoupling protein-1-independent manner: involvement of antiadipo-
genic mechanism and extracellularly regulated kinase/mitogen-activated
protein kinase signaling. Endocrinology 2008, 149, 358−366.
(20) Wang, T.; Wang, Y.; Yamashita, H. Evodiamine inhibits
adipogenesis via the EGFR-PKCalpha-ERK signaling pathway. FEBS
Lett. 2009, 583, 3655−3659.
(23) Kan, S. F.; Huang, W. J.; Lin, L. C.; Wang, P. S. Inhibitory effects
of evodiamine on the growth of human prostate cancer cell line LNCaP.
Int. J. Cancer 2004, 110, 641−651.
(24) Liao, C. H.; Pan, S. L.; Guh, J. H.; Chang, Y. L.; Pai, H. C.; Lin, C.
H.; Teng, C. M. Antitumor mechanism of evodiamine, a constituent
from Chinese herb Evodiae fructus, in human multiple-drug resistant
breast cancer NCI/ADR-RES cells in vitro and in vivo. Carcinogenesis
2005, 26, 968−975.
(25) Yang, Z. G.; Chen, A. Q.; Liu, B. Antiproliferation and apoptosis
induced by evodiamine in human colorectal carcinoma cells (COLO-
205). Chem. Biodivers. 2009, 6, 924−933.
(26) Kan, S. F.; Yu, C. H.; Pu, H. F.; Hsu, J. M.; Chen, M. J.; Wang, P. S.
Anti-proliferative effects of evodiamine on human prostate cancer cell
lines DU145 and PC3. J. Cell. Biochem. 2007, 101, 44−56.
(27) Chen, M. C.; Yu, C. H.; Wang, S. W.; Pu, H. F.; Kan, S. F.; Lin, L.
C.; Chi, C. W.; Ho, L. L.; Lee, C. H.; Wang, P. S. Anti-proliferative effects
of evodiamine on human thyroid cancer cell line ARO. J. Cell. Biochem.
2010, 110, 1495−1503.
(28) Zhang, C.; Fan, X.; Xu, X.; Yang, X.; Wang, X.; Liang, H. P.
Evodiamine induces caspase-dependent apoptosis and S phase arrest in
human colon lovo cells. Anticancer Drugs 2010, 21, 766−776.
(29) Lee, T. J.; Kim, E. J.; Kim, S.; Jung, E. M.; Park, J. W.; Jeong, S. H.;
Park, S. E.; Yoo, Y. H.; Kwon, T. K. Caspase-dependent and caspase-
independent apoptosis induced by evodiamine in human leukemic
U937 cells. Mol. Cancer Ther. 2006, 5, 2398−2407.
(30) Huang, H.; Zhang, Y.; Liu, X.; Li, Z.; Xu, W.; He, S.; Huang, Y.;
Zhang, H. Acid sphingomyelinase contributes to evodiamine-induced
apoptosis in human gastric cancer SGC-7901 cells. DNA Cell Biol. 2011,
30, 407−412.
(31) Wang, C.; Li, S.; Wang, M. W. Evodiamine-induced human
melanoma A375-S2 cell death was mediated by PI3K/Akt/caspase and
Fas-L/NF-kappaB signaling pathways and augmented by ubiquitin−
proteasome inhibition. Toxicol. in Vitro 2010, 24, 898−904.
(32) Ogasawara, M.; Matsubara, T.; Suzuki, H. Inhibitory effects of
evodiamine on in vitro invasion and experimental lung metastasis of
murine colon cancer cells. Biol. Pharm. Bull. 2001, 24, 917−920.
(33) Ogasawara, M.; Matsubara, T.; Suzuki, H. Screening of natural
compounds for inhibitory activity on colon cancer cell migration. Biol.
Pharm. Bull. 2001, 24, 720−723.
(34) Yang, X. W.; Zhang, H.; Li, M.; Du, L. J.; Yang, Z.; Xiao, S. Y.
Studies on the alkaloid constituents of Evodia rutaecarpa (Juss) Benth
var. bodinaieri (Dode) Huang and their acute toxicity in mice. J. Asian
Nat. Prod. Res. 2006, 8, 697−703.
(35) Lipinski, C. A.; Lombardo, F.; Dominy, B. W.; Feeney, P. J.
Experimental and computational approaches to estimate solubility and
permeability in drug discovery and development settings. Adv. Drug
Deliv. Rev. 2001, 46, 3−26.
(36) Hann, M. M.; Oprea, T. I. Pursuing the leadlikeness concept in
pharmaceutical research. Curr. Opin. Chem. Biol. 2004, 8, 255−263.
(37) Oprea, T. I.; Davis, A. M.; Teague, S. J.; Leeson, P. D. Is there a
difference between leads and drugs? A historical perspective. J. Chem. Inf.
Comput. Sci. 2001, 41, 1308−1315.
(38) Manly, C. J.; Chandrasekhar, J.; Ochterski, J. W.; Hammer, J. D.;
Warfield, B. B. Strategies and tactics for optimizing the Hit-to-Lead
process and beyonda computational chemistry perspective. Drug
Discovery Today 2008, 13, 99−109.
(39) Baruah, B.; Dasu, K.; Vaitilingam, B.; Mamnoor, P.; Venkata, P. P.;
Rajagopal, S.; Yeleswarapu, K. R. Synthesis and cytotoxic activity of
novel quinazolino-beta-carboline-5-one derivatives. Bioorg. Med. Chem.
2004, 12, 1991−1994.
(40) Chen, Z.; Hu, G.; Li, D.; Chen, J.; Li, Y.; Zhou, H.; Xie, Y.
Synthesis and vasodilator effects of rutaecarpine analogues which might
be involved transient receptor potential vanilloid subfamily, member 1
(TRPV1). Bioorg. Med. Chem. 2009, 17, 2351−2359.
(41) Decker, M. Novel inhibitors of acetyl- and butyrylcholinesterase
derived from the alkaloids dehydroevodiamine and rutaecarpine. Eur. J.
Med. Chem. 2005, 40, 305−313.
(21) Jiang, J.; Hu, C. Evodiamine: a novel anti-cancer alkaloid from
Evodia rutaecarpa. Molecules 2009, 14, 1852−1859.
(22) Fei, X. F.; Wang, B. X.; Li, T. J.; Tashiro, S.; Minami, M.; Xing, D.
J.; Ikejima, T. Evodiamine, a constituent of Evodiae fructus, induces anti-
proliferating effects in tumor cells. Cancer Sci. 2003, 94, 92−98.
S
dx.doi.org/10.1021/jm300605m | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
(42) Hong, Y. H.; Lee, W. J.; Lee, S. H.; Son, J. K.; Kim, H. L.; Nam, J.
M.; Kwon, Y.; Jahng, Y. Synthesis and biological properties of benzo-
annulated rutaecarpines. Biol. Pharm. Bull. 2010, 33, 1704−1709.
(43) Wang, B.; Mai, Y. C.; Li, Y.; Hou, J. Q.; Huang, S. L.; Ou, T. M.;
Tan, J. H.; An, L. K.; Li, D.; Gu, L. Q.; Huang, Z. S. Synthesis and
evaluation of novel rutaecarpine derivatives and related alkaloids
derivatives as selective acetylcholinesterase inhibitors. Eur. J. Med.
Chem. 2010, 45, 1415−1423.
(44) Pin, F.; Comesse, S.; Daïch, A. Intramolecular N-aza-
amidoalkylation in association with WitkopeWinterfeldt oxidation as
the key step to synthesize Luotonin-A analogues. Tetrahedron 2011, 67,
5564−5571.
(45) Dong, G.; Sheng, C.; Wang, S.; Miao, Z.; Yao, J.; Zhang, W.
Selection of evodiamine as a novel topoisomerase I inhibitor by
structure-based virtual screening and hit optimization of evodiamine
derivatives as antitumor agents. J. Med. Chem. 2010, 53, 7521−7531.
(46) Nakayama, A.; N., K.; M., k.; H., T. Straightforward asymmetric
total synthesis of (+)-evodiamine, a major indole alkloid in herbal
medicine “Wu Zhu Yu”. Heterocycles 2008, 76, 861−865.
(47) Fujii, A.; Hashiguchi, S.; Uematsu, N.; Ikariya, T.; Noyori, R.
Ruthenium(II)-Catalyzed Asymmetric Transfer Hydrogenation of
Ketones Using a Formic Acid−Triethylamine Mixture. J. Am. Chem.
Soc. 1996, 118, 2521−2522.
(48) Lee, S. H.; Son, J. K.; Jeong, B. S.; Jeong, T. C.; Chang, H. W.; Lee,
E. S.; Jahng, Y. Progress in the studies on rutaecarpine. Molecules 2008,
13, 272−300.
(49) Jones, G.; Willett, P.; Glen, R. C.; Leach, A. R.; Taylor, R.
Development and validation of a genetic algorithm for flexible docking.
J. Mol. Biol. 1997, 267, 727−748.
(50) Desmond_Maestro_academic-2009−02; http://www.
deshawresearch.com.
(51) Liu, X.; Jiang, H.; Li, H. SHAFTS: a hybrid approach for 3D
molecular similarity calculation. 1. Method and assessment of virtual
screening. J. Chem. Inf. Model. 2011, 51, 2372−2385.
(52) Lu, W.; Liu, X.; Cao, X.; Xue, M.; Liu, K.; Zhao, Z.; Shen, X.; Jiang,
H.; Xu, Y.; Huang, J.; Li, H. SHAFTS: a hybrid approach for 3D
molecular similarity calculation. 2. Prospective case study in the
discovery of diverse p90 ribosomal S6 protein kinase 2 inhibitors to
suppress cell migration. J. Med. Chem. 2011, 54, 3564−3574.
(53) MDL Drug Data Report; Symyx Technologies, Inc.: Sunnyvale,
CA.
(62) Vos, S. M.; Tretter, E. M.; Schmidt, B. H.; Berger, J. M. All tangled
up: how cells direct, manage and exploit topoisomerase function. Nature
Rev. Mol. Cell Biol. 2011, 12, 827−841.
(63) Pommier, Y. DNA topoisomerase I inhibitors: chemistry, biology,
and interfacial inhibition. Chem. Rev. 2009, 109, 2894−2902.
(64) Bailly, C. Contemporary Challenges in the Design of
Topoisomerase II Inhibitors for Cancer Chemotherapy. Chem. Rev.
2012, 112, 3611−3640.
(65) Denny, W. A.; Baguley, B. C. Dual topoisomerase I/II inhibitors
in cancer therapy. Curr. Top. Med. Chem. 2003, 3, 339−353.
(66) Dallavalle, S.; Gattinoni, S.; Mazzini, S.; Scaglioni, L.; Merlini, L.;
Tinelli, S.; Beretta, G. L.; Zunino, F. Synthesis and cytotoxic activity of a
new series of topoisomerase I inhibitors. Bioorg. Med. Chem. Lett. 2008,
18, 1484−1489.
(67) Salerno, S.; Da Settimo, F.; Taliani, S.; Simorini, F.; La Motta, C.;
Fornaciari, G.; Marini, A. M. Recent advances in the development of
dual topoisomerase I and II inhibitors as anticancer drugs. Curr. Med.
Chem. 2010, 17, 4270−4290.
(68) Sheng, C.; Miao, Z.; Zhang, W. New strategies in the discovery of
novel non-camptothecin topoisomerase I inhibitors. Curr. Med. Chem.
2011, 18, 4389−4409.
(69) Martin, S. F.; Benage, B.; Hunter, J. E. A concise strategy for the
syntheses of indole alkaloids of the heteroyohimboid and corynan-
theioid families. Total syntheses of ( )-tetrahydroalstonine, ( )-cath-
enamine and ( )-geissoschizine. J. Am. Chem. Soc. 1988, 110, 5925−
5927.
(70) D’Souza, A. M.; Spiccia, N.; Basutto, J.; Jokisz, P.; Wong, L. S.;
Meyer, A. G.; Holmes, A. B.; White, J. M.; Ryan, J. H. 1,3-Dipolar
cycloaddition−decarboxylation reactions of an azomethine ylide with
isatoic anhydrides: formation of novel benzodiazepinones. Org. Lett.
2011, 13, 486−489.
(71) Aleksic, M.; Bertosa, B.; Nhili, R.; Uzelac, L.; Jarak, I.; Depauw, S.;
David-Cordonnier, M. H.; Kralj, M.; Tomic, S.; Karminski-Zamola, G.
Novel Substituted Benzothiophene and Thienothiophene Carboxani-
lides and Quinolones: Synthesis, Photochemical Synthesis, DNA-
Binding Properties, Antitumor Evaluation and 3D-Derived QSAR
Analysis. J. Med. Chem. 2012, 55, 5044−5060.
(72) Wang, P.; Leung, C. H.; Ma, D. L.; Lu, W.; Che, C. M.
Organoplatinum(II) complexes with nucleobase motifs as inhibitors of
human topoisomerase II catalytic activity. Chem. Asian J. 2010, 5, 2271−
2280.
(73) Li, X.; Lin, Y.; Wang, Q.; Yuan, Y.; Zhang, H.; Qian, X. The novel
anti-tumor agents of 4-triazol-1,8-naphthalimides: synthesis, cytotox-
icity, DNA intercalation and photocleavage. Eur. J. Med. Chem. 2011, 46,
1274−1279.
(74) Baviskar, A. T.; Madaan, C.; Preet, R.; Mohapatra, P.; Jain, V.;
Agarwal, A.; Guchhait, S. K.; Kundu, C. N.; Banerjee, U. C.; Bharatam, P.
V. N-Fused imidazoles as novel anticancer agents that inhibit catalytic
activity of topoisomerase IIalpha and induce apoptosis in G1/S phase. J.
Med. Chem. 2011, 54, 5013−5030.
(75) Du, L.; Liu, H. C.; Fu, W.; Li, D. H.; Pan, Q. M.; Zhu, T. J.; Geng,
M. Y.; Gu, Q. Q. Unprecedented citrinin trimer tricitinol B functions as a
novel topoisomerase IIalpha inhibitor. J. Med. Chem. 2011, 54, 5796−
5810.
(76) Marshall, K. M.; Holden, J. A.; Koller, A.; Kashman, Y.; Copp, B.
R.; Barrows, L. R. AK37: the first pyridoacridine described capable of
stabilizing the topoisomerase I cleavable complex. Anticancer Drugs
2004, 15, 907−913.
(77) Routier, S.; Peixoto, P.; Merour, J. Y.; Coudert, G.; Dias, N.; Bailly,
C.; Pierre, A.; Leonce, S.; Caignard, D. H. Synthesis and biological
evaluation of novel naphthocarbazoles as potential anticancer agents. J.
Med. Chem. 2005, 48, 1401−1413.
(78) Wei, H.; Ruthenburg, A. J.; Bechis, S. K.; Verdine, G. L.
Nucleotide-dependent domain movement in the ATPase domain of a
human type IIA DNA topoisomerase. J. Biol. Chem. 2005, 280, 37041−
37047.
(54) Nitiss, J. L. Targeting DNA topoisomerase II in cancer
chemotherapy. Nature Rev. Cancer 2009, 9, 338−350.
(55) Jun, K. Y.; Lee, E. Y.; Jung, M. J.; Lee, O. H.; Lee, E. S.; Park Choo,
H. Y.; Na, Y.; Kwon, Y. Synthesis, biological evaluation, and molecular
docking study of 3-(3′-heteroatom substituted-2′-hydroxy-1′-propy-
loxy) xanthone analogues as novel topoisomerase II alpha catalytic
inhibitor. Eur. J. Med. Chem. 2011, 46, 1964−1971.
(56) Graves, D. E. Drug−DNA Interactions. Methods Mol. Biol. 2001,
95, 161−169.
(57) Peixoto, P.; Bailly, C.; Helene, M.; David-Cordonnier, M.-H.
Topoisomerase I-mediated DNA relaxation as a tool to study
intercalation of small molecules into supercoiled DNA. In Drug−DNA
Interaction Protocols, 2nd ed.; Fox, K. R., Ed.; Humana Press; Totowa,
NJ, 2010; Vol. 613, pp 235−256.
(58) Suh, D.; Chaires, J. B. Criteria for the mode of binding of DNA
binding agents. Bioorg. Med. Chem. 1995, 3, 723−728.
(59) Palchaudhuri, R.; Hergenrother, P. J. DNA as a target for
anticancer compounds: methods to determine the mode of binding and
the mechanism of action. Curr. Opin. Biotechnol. 2007, 18, 497−503.
(60) Perrin, D.; van Hille, B.; Barret, J. M.; Kruczynski, A.; Etievant, C.;
Imbert, T.; Hill, B. T. F 11782, a novel epipodophylloid non-
intercalating dual catalytic inhibitor of topoisomerases I and II with an
original mechanism of action. Biochem. Pharmacol. 2000, 59, 807−819.
(61) Huang, H.; Chen, Q.; Ku, X.; Meng, L.; Lin, L.; Wang, X.; Zhu, C.;
Wang, Y.; Chen, Z.; Li, M.; Jiang, H.; Chen, K.; Ding, J.; Liu, H. A series
of alpha-heterocyclic carboxaldehyde thiosemicarbazones inhibit
topoisomerase IIalpha catalytic activity. J. Med. Chem. 2010, 53,
3048−3064.
(79) Discovery Studio 3.0; Accelrys Software Inc.: San Diego, CA, 2010;
http://www.accelrys.com.
T
dx.doi.org/10.1021/jm300605m | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
(80) Shaw, D. E. A fast, scalable method for the parallel evaluation of
distance-limited pairwise particle interactions. J. Comput. Chem. 2005,
26, 1318−1328.
(81) Bowers, K. J.; Chow, E.; Xu, H.; Dror, R. O.; Eastwood, M. P.;
́
Gregersen, B. A.; Klepeis, J. L.; Kolossvary, I.; Moraes, M. A.; Sacerdoti,
J. K.; Shan, Y.; Shaw, D. E. Scalable Algorithms for Molecular Dynamics
Simulations on Commodity Clusters. Proceedings of the ACM/IEEE
Conference on Supercomputing (SC06), Tampa, Florida, November
11−17, 2006.
(82) Bowers, K. J.; Dror, R. O.; Shaw, D. E. The midpoint method for
parallelization of particle simulations. J. Chem. Phys. 2006, 124, 184109.
(83) Bowers, K. J.; Dror, R. O.; Shaw, D. E. Zonal Methods for the
Parallel Execution of Range-Limited N-Body Simulations. J. Comput.
Phys. 2007, 221, 303−329.
U
dx.doi.org/10.1021/jm300605m | J. Med. Chem. XXXX, XXX, XXX−XXX