Synthesis and biological evaluation of aryloxazole derivatives as antimitotic and vascular-disrupting agents for cancer therapy

Article abstract of DOI:10.1021/jm400840p

A series of aryloxazole, thiazole, and isoxazole derivatives was synthesized as vascular-targeting anticancer agents. Antiproliferative activity and tumor vascular-disrupting activity of all of the synthesized compounds were tested in vitro using various human cancer cell lines and HUVECs (human umbilical vein endothelial cells). Several compounds with an arylpiperazinyl oxazole core showed excellent cytotoxicity and metabolic stability in vitro. Among this series, two representative compounds (6-48 and 6-51) were selected and tested for the evaluation of anticancer effects in vivo using tumor-bearing mice. Compound 6-48 effectively reduced tumor growth (42.3% reduction in size) at the dose of 100 mg/kg. We believe that compound 6-48 will serve as a good lead compound for antimitotic and vascular-disrupting agents; further investigation to improve the in vivo efficacy of this series is underway.

Full text of DOI:10.1021/jm400840p

Article
pubs.acs.org/jmc
Synthesis and Biological Evaluation of Aryloxazole Derivatives as
Antimitotic and Vascular-Disrupting Agents for Cancer Therapy
Min Jeong Choi,†,‡ Eun Sun No,† Dhanaji Achyutrao Thorat,†,§ Jae Wan Jang,†,§ Hakkyun Yang,†,§
Jaeick Lee,∥ Hyunah Choo,† Soo Jin Kim,⊥ Chang Sik Lee,⊥ Soo Young Ko,‡ Jiyoun Lee,#
GhilSoo Nam,*,†,§ and Ae Nim Pae*,†,§
†Center for Neuro-Medicine, Brain Science Institute, Korea Institute of Science and Technology, P.O. Box 131, Cheongryang, Seoul
130-650, Republic of Korea
‡Chemistry & Nano Science, Ewha Womans University, 11-1 Daehyun-dong, Seodaemun-Gu, Seoul 120-750, Republic of Korea
§Department of Medicinal & Pharmaceutical Chemistry, School of Science, University of Science and Technology, 217 Gajeong-ro,
Yuseong-gu, Daejeon 305-350, Republic of Korea
∥Doping Control Center, Korea Institute of Science and Technology, P.O. Box 131, Cheongryang, Seoul 130-650, Republic of Korea
⊥Chong Kun Dang Research Institute, CKD Pharmaceuticals, Jung-dong, Giheung-gu, Yongin-si, Gyeonggi-do 464-3, Republic of
Korea
#Department of Global Medical Science, Sungshin University, Seoul 142-732, Republic of Korea
S
* Supporting Information
ABSTRACT: A series of aryloxazole, thiazole, and isoxazole derivatives was synthesized as vascular-targeting anticancer agents.
Antiproliferative activity and tumor vascular-disrupting activity of all of the synthesized compounds were tested in vitro using
various human cancer cell lines and HUVECs (human umbilical vein endothelial cells). Several compounds with an
arylpiperazinyl oxazole core showed excellent cytotoxicity and metabolic stability in vitro. Among this series, two representative
compounds (6-48 and 6-51) were selected and tested for the evaluation of anticancer effects in vivo using tumor-bearing mice.
Compound 6-48 effectively reduced tumor growth (42.3% reduction in size) at the dose of 100 mg/kg. We believe that
compound 6-48 will serve as a good lead compound for antimitotic and vascular-disrupting agents; further investigation to
improve the in vivo efficacy of this series is underway.
and colchicine are known to bind β-tubulins and to block the
formation of functional microtubules. Taxol and their
derivatives, however, bind to microtubules and disrupt their
functions, halting cells in the mitotic state.2 By the early 1990’s,
hundreds of effective mitosis inhibitors had been developed;
unfortunately, the majority of spindle poisons failed in clinical
trials because of their poor therapeutic index. Many compounds
demonstrated inadequate efficacy or high toxicity for clinical
use, which possibly derived from chemical instability and
unrecognized interactions with multiple protein targets.3,4
INTRODUCTION
■
Mitosis is a series of mechanical events that generate two new
nuclei containing identical copies of DNA during cell division.
Because of the crucial role of the mitotic spindle in mitosis, the
spindle fiber has been a valuable drug target in cancer therapy
for decades.1 The discovery of natural products, such as taxol
and vinca alkaloids, which interact with the spindle fibers during
mitosis in malignant cell growth, has led scientists to explore
diverse sets of natural compounds and synthetic small
molecules as anticancer drugs. These mitotic spindle poisons
halt cell cycle progression by two distinct mechanisms. One is
preventing tubulins from polymerizing into microtubules,
thereby inhibiting cellular division. For example, vinca alkaloids
Received: June 6, 2013
Published: October 27, 2013
© 2013 American Chemical Society
9008
dx.doi.org/10.1021/jm400840p | J. Med. Chem. 2013, 56, 9008−9018

Journal of Medicinal Chemistry
Article
Figure 1. (A) Structures of 1 (colchicine), 2 (ABT751), and 3. (B) Docked pose of compounds 3 (green) and 1 (purple) in the colchicine binding
pocket (PDB code: 3UT5). (C) Docked pose of compounds 3 (green) and 2 (orange) in the colchicine binding pocket (PDB code: 3HKC).
Despite the early clinical failures of the first-generation
spindle poisons, paclitaxel, a microtubule depolymerization
inhibitor, has achieved huge success in the clinic and is known
to be the most broadly effective and commercially successful
anticancer agent. This success led to extensive efforts to
develop other tubulin poisons with improved pharmacokinetic
profiles and selectivity. Interestingly, these tubulin binders were
also found to interact with tumor endothelial cells, leading to a
rapid occlusion of tumor vasculature and resultant necrotic cell
death.5,6 These vascular-disrupting agents (VDAs) selectively
block established tumor vasculature and prevent tumor
progression rapidly, whereas conventional antiangiogenesis
agents inhibit the formation of new blood vessels, thereby
preventing tumor growth chronically. Currently, several notable
VDAs including combretastatin A-4 phosphate (CA4P),7
DMXAA,8 ZD6126,9 and BNC10510 are undergoing clinical
trials.
Considering the unique selectivity and effectiveness of
tubulin binding VDAs, we aimed to develop a new class of
anticancer drugs having dual effects: tumor vasculature
disruption and mitotic arrest. In this article, we report the
design, synthesis, and biological evaluation of aryl oxazole
derivatives as dual-action anticancer agents and present a
complete SAR study of the new core to identify essential
structural properties required for dual effects. We designed and
synthesized a series of heterocyclic small molecules containing
oxazoles and isoxazoles based on the previous structure−
activity relationship (SAR) study on the colchicine binding sites
of tubulin.11
leukemia cells (HL-60) with our compounds to evaluate their
inhibitory effect on tumor cell growth in vitro. Second, we
measured the vascular-disrupting effect of our compounds using
human umbilical vein endothelial cells (HUVEC). Finally, we
measured the inhibitory effect of our compounds against
tubulin polymerization. Combined with microsomal stability
test, we were able to identify two final compounds that are not
only stable against metabolism but also inhibit tubulin
polymerization and HUVEC growth at low concentrations.
Anticancer activities of two selected compounds were evaluated
in vivo using tumor-bearing mice.
COMPUTATIONAL DESIGN
■
Colchicine (1), a natural product from plants, is known to bind
β-tubulins and to inhibit tubulin polymerization. Despite the
high toxicity of colchicine, which is comparable to arsenic
poisoning, its excellent binding affinity toward spindle fibers
makes it an attractive lead compound as an anticancer agent.
To date, most of the related studies have focused on developing
small molecules targeting the colchicine binding site with low
toxicity. One of the most notable examples is ABT-751 (2)
(Figure 1A).12 Compound 2 is a synthetic small molecule
selected from random screening of antibacterial sulfonamides,
and it recently demonstrated promising anticancer effects in
phase II trials.13 By taking a similar approach, we have tested an
in-house library of compounds with N-phenylpyrazole, 4,5-
hydropyrazole, and oxazole core for cytotoxicity against various
tumor cell lines (details not shown).
Our preliminary results showed that some of these
compounds demonstrated potent cytotoxicity against tumor
cell lines as well as inhibited tubulin polymerization, which are
We evaluated biological activity of the newly synthesized
compounds in vitro and in vivo. First, we incubated human
9009
dx.doi.org/10.1021/jm400840p | J. Med. Chem. 2013, 56, 9008−9018

Journal of Medicinal Chemistry
Article
a
Scheme 1
a
Reagents and conditions: (a) (i) HDNIB, CH3CN, reflux, 1 h, (ii) acetamide (or thioacetamide), microwave 80 W, 1 min; (b) (i) 1N NaOH, (ii) 1
N HCl, (c) arylpiperazine, EDCI, HOBt, NMM, CH2Cl2.
a
Scheme 2
a
Reagents and conditions: (a) (i) NEt3, NCS, CH2Cl2, (ii) propargyl alcohol, NEt3, THF, (iii) Jones reagent, acetone, 0 °C; (b) (i) n-BuLi, THF,
−78 °C, then I2, −30 °C; (c) 1-(3,5-dimethoxy- phenyl)piperazine, EDCI, HOBt, NMM, CH2Cl2; (d) arylboronic acid, Pd(PPh3)4, Na2CO3,
dioxane.
required for dual effects (Table S1, Supporting Information).
On the basis of these preliminary data, we hypothesized that
these tested compounds might bind to tubulin in the same way
as colchicine. Therefore, we further carried out molecular-
modeling studies to design a series of analogs that occupy the
colchicine binding domain of tubulin. We selected compound
3, one of the most potent cytotoxic agents in the screening
library, docked in the colchicine binding domain, and we
superimposed this with colchicine and 2 in the crystal
structures, as shown in Figure 1B,C (PDB codes 3UT5 and
3HKC, respectively). Compound 3 appeared to fit in the
colchicine binding site and partially overlapped with a docked
colchicine in the hydrophobic binding pocket, as shown in
Figure 1B. The estimated binding energy of compound 3
(GScore = −4.88 kcal/mol) was much higher than that of
colchicine 1 (GScore = −6.789 kcal/mol). It has been reported
that the A ring with three methoxy groups serves as an anchor
to place the C ring of colchicine in the proper orientation
within the binding site.14,15 Therefore, the apparent lack of this
anchoring effect in 3 may lead to a slight perturbation inside the
binding pocket, resulting in the higher binding energy.
However, it should be noted that compound 3 has an extra
carbonyl group, which provides an additional hydrogen bond
with the amide nitrogen of Ala317 in the backbone of β-strand
9 compared to colchicine. Interestingly, the methyl group of the
oxazole ring in 3 is pointed toward α-tubulin T5 loop, which
also appears to partially overlap with the N-acetyl substituent of
colchicine. Conformational changes of the T5 loop are known
to be involved in the assembly of tubulin αβ heterodimers,
indicating that the methyl group of the oxazole moiety may be
critical for its biological activity.16,17
is that the hydroxyl group on the aniline moiety of 2 forms
hydrogen bonds with Tyr202 and Val238, which is absent in 3.
However, the 3-chlorophenyl ring of 3 extended deeper to
strand S6 of the N-terminal nucleotide binding domain; thus, it
provided additional interactions with hydrophobic residues,
such as Leu252, Leu242, Val238, Leu255, and Tyr202. On the
basis of the binding mode analyses of compound 3, the oxazole
analogs fit in the colchicine binding domain and can be further
optimized by additional structural modifications. Therefore, a
series of analogs containing the various substituted aryl oxazole
and thiazole group was designed and synthesized.
CHEMISTRY
■
Aryl oxazole and thiazole derivatives were prepared according
to Scheme 1. Oxazole and thiazole cores were synthesized from
commercially available ethyl 3-oxo-3-phenyl-propanoate in the
presence of HDNIB ([hydroxy(2,4-dinitrobenzene-
sulfonyloxy)iodo]benzene) in one pot. The reaction mixture
first formed an α-sulfonyloxyl ketone as an intermediate, and
upon the treatment of acetamide (or thioacetamide) in
combination with a microwave irradiation at 80 W for 1 min,
heterocyclic core 4 was obtained in over 20% yield. Hydrolysis
of 4 afforded the acid fragment 5 in 80% yield. A peptide
coupling reaction was carried out with various aryl piperazines
using EDCI (1-ethyl-3-(3-dimethylaminopropyl)carbodiimide)
to obtain the desired derivatives, series 6 and 7, in over 60%
yield. Most of the substituted arylpiperazines were commer-
cially available, but 1-(3-ethoxyphenyl)piperazine, 1-(2,3-
dichlorobenzyl)piperazine, 1-(benzo[d][1,3]dioxol-5-yl)-
piperazine, 3-(piperazin-1-yl)benzonitrile, 2-(3-(piperazin-1-
yl)phenyl)propan-2-ol, 1-(3,5-dimethoxyphenyl)piperazine, 1-
(3,5-dichlorophenyl) piperazine, 1-(5-chloro-2-methylphenyl)-
piperazine, and 1-(3,5-difluoro-phenyl)piperazine were synthe-
sized by employing an SN2 reaction with substituted anilines
and bis(2-chloroethyl)amine (Scheme S1, Supporting Informa-
tion).
When compound 3 was docked in the colchicine binding
domain, 3 appeared to overlap very well with the bound 2, as
shown in Figure 1C. However, the estimated binding energy of
compound 3 (GScore = −4.88 kcal/mol) was also higher than
that of 2 (GScore = −6.89 kcal/mol). One possible explanation
9010
dx.doi.org/10.1021/jm400840p | J. Med. Chem. 2013, 56, 9008−9018

Journal of Medicinal Chemistry
Article
Table 1. Cytotoxic Effects of (2-Methyl-4-phenyloxazol-5-yl)(4-phenylpiperazin-1-yl)methanone Derivatives against Human
Leukemia Cells (HL-60)
a
a
R
IC50 (nM)
R
IC50 (nM)
3
3-chloro
39.2
6-15
4-pro-1-en-2-yl
4-isopropanoyl
3-carbamoyl
NC
6-1
6-2
6-3
6-4
6-5
6-6
6-7
6-8
6-9
6-10
6-11
6-12
6-13
6-14
H
39.1
6-16
NC
3-methoxy
3-ethoxy
34.8
6-17
71
151
6-18
benzo[1,3]dioxol-5-yl
2,6-dimethyl
3,5-dimethoxy
2,4-dimethyl
3,4-dimethyl
2,3-dimethyl
2,5-dimethyl
3,4-dichloro
251
146.1
19.2
100.5
43
3-cyano
39.6
6-19
4-fluoro
>1000
45
6-20
3-fluoro
6-21
2-fluoro
146.8
>1000
152.9
119.5
124.1
>1000
329
6-22
4-trifluoromethyl
3-trifluoromethyl
2-trifluoromethyl
2-chloro
6-23
35
6-24
31.3
124.3
34.5
41.8
133
112.3
6-25
6-26
3,5-dimethyl
3,5-dichloro
4-methyl
6-27
3-aminoacetyl
4-acetyl
6-28
2,3-dichloro
NC
12 (CYT997)
a
IC50 = compound concentration required to inhibit cell growth by 50% after cells were treated with compounds for 72 h. Data are expressed as the
mean from at least three independent experiments.
Isoxazole derivatives with a methyl substituent were prepared
according to Scheme 2. (E)-Acetaldehyde oxime was treated
with triethylamine and N-chlorosuccinimide successively
followed by the addition of propargyl alcohol to afford the
isoxazole core. The primary alcohol group was then converted
to a carboxylic acid via Jones oxidation to obtain the isoxazole
fragment 8. Iodination of intermediate 8 using n-BuLi and
iodine produced compound 9 in 30% yield. It should be noted
that this iodination step requires a constant temperature at −78
°C and absolute anhydrous conditions to obtain the desired
product. After a peptide coupling reaction with 1-(3,5-
dimethoxyphenyl)piperazine, intermediate 10 was obtained
and further subjected to Suzuki cross-coupling with arylboronic
acids to give a series of the isoxazole analogues, 11, in over 60%
yield.
Additionally, when these bulkier groups were placed on the 4-
position, these compounds completely lost cytotoxicity,
indicating that the size and the position of the substituents
seem to be more important factors rather than electro-
negativity. Some of the derivatives with multiple substituents
also showed excellent cytotoxicity, for example, 3,5-dimethoxy-
substituted compounds (6-20) exhibited the lowest IC50 value
(19.2 nM) among this series. Both 3,5-dimethyl- and -dichloro-
substituted derivatives also demonstrated good cytotoxicity;
however, when these substituents were moved to different
positions, such as 2,6-, 2,4-, or 2,3-, these compounds became
much less potent, again indicating that a steric factor is critical
for the cytotoxicity of these derivatives.
To investigate the effect of the substituents on the phenyl
ring directly linked to the oxazole core, we modified the phenyl
ring with methoxy groups (6-29−6-31), hydrophilic groups (6-
32−6-37), and halogen groups (6-38−6-63) and tested
cytotoxicity as shown in Table 2. All of the derivatives with
hydrophilic substituents and methoxy groups (6-29−6-37)
became much less potent compared to their nonsubstituted
counterparts, suggesting that van der Waals interaction may
play an important role in this particular binding region. All
fluoro-substituted derivatives with 3,5-dimethoxy on R2 (6-51−
6-53) exhibited excellent cytotoxic effect, with IC50 values
ranging from 10.3 to 38.5 nM. When the 3,5-dimethoxy groups
were replaced with a 3-chloro substituent (6-57−6-59), little or
no effects were observed, confirming that electronegativity on
this position does not affect overall binding. Interestingly,
among all of the oxazole compounds in this series, only the
derivatives with the 3,5-dimethoxy groups on the R2 position
showed slightly improved activity compared to the parent
compound without any substituent (6-1, R1 = R2 = H),
suggesting that hydrogen-bond interactions may exist in this
region.
RESULTS AND DISCUSSION
■
Structure−Activity Relationship and in Vitro Cytotox-
icity. Cytotoxic effects of all of the synthesized derivatives were
first evaluated by measuring cell viability of human leukemia
cells (HL-60) upon being treated with each compound at
different concentrations. CYT99718 (12), a known cytotoxic
vascular-disrupting agent, was also tested as a reference. Among
the oxazole derivatives with a phenyl group (Table 1), the 3-
substituted aryl piperazine derivatives (3, 6-2, 6-4, and 6-6)
showed good cytotoxic effects, having IC50 values in the low
nanomolar range. Despite having substituents with different
electronegativity values, all four compounds have similar IC50
values (35−45 nM), suggesting that electronegativity on the 3-
position does not affect cytotoxicity significantly; however,
when a fluoride was placed on either 2- or 4-position (6-5 and
6-7), the IC50 values increased markedly. It should be also
noted that if the 3-position is substituted with slightly bulkier
groups, such as trifluoromethyl (6-9) or aminoacetyl group (6-
13), then the IC50 values increased up to 2- to 3-fold.
Next, we replaced the methyl group on the oxazole core with
various alkyl groups to see if the bulkier substituent affects
9011
dx.doi.org/10.1021/jm400840p | J. Med. Chem. 2013, 56, 9008−9018

Journal of Medicinal Chemistry
Article
Table 2. Cytotoxic Effects of (2-Methyl-4-phenyloxazol-5-yl)(4-phenylpiperazin-1-yl)methanone Derivatives against Human
Leukemia Cells (HL-60)
a
R1
R2
IC50 (nM)
6-29
6-30
6-31
6-32
6-33
6-34
6-35
6-36
6-37
6-38
6-39
6-40
6-41
6-42
6-43
6-44
6-45
6-46
6-47
6-48
6-49
6-50
6-51
6-52
6-53
6-54
6-55
6-56
6-57
6-58
6-59
6-60
6-61
6-62
6-63
4-methoxy
3,4-dimethoxy
3-methoxy
3-hydroxy
3-chloro
3-chloro
3-chloro
3-chloro
3-chloro
3-chloro
3-chloro
3-chloro
3-chloro
NC
NC
NC
104.9
>1000
85.9
303
3-sodium phosphate
4-hydroxy-3-methoxy
3-methoxy-4-sodium phosphate
3-hydroxy-4-methoxy
3-aminoacetyl
2-fluoro
NC
208
4-pro-1-en-2-yl
4-chloro
NC
4-fluoro
NC
4-chloro
3-isopropanoyl
2,5-dimethyl
2,5-dimethyl
4-isopropanoyl
2,5-dimethyl
2,5-dimethyl
2,5-dimethyl
2,5-dimethyl
3,5-dichloro
3,5-dichloro
3,5-dichloro
3,5-dimethoxy
3,5-dimethoxy
3,5-dimethoxy
3,5-difluoro
5-chloro-2-methyl
5-chloro-2-methoxy
3-chloro
NC
4-fluoro
358
2-fluoro
318.9
NC
4-fluoro
3-fluoro
204.2
>1000
345.2
NC
2-chloro
3-chloro
4-chloro
2-fluoro
60.2
41.7
52.1
10.3
28.2
38.5
186.2
228.7
152.9
39.4
33.7
40.1
132.4
56.1
NC
3-fluoro
4-fluoro
2-fluoro
3-fluoro
4-fluoro
2-fluoro
2-fluoro
2-fluoro
2-fluoro
3-fluoro
3-chloro
4-fluoro
3-chloro
2-chloro
3-chloro
2-fluoro
3-methoxy
3-fluoro
3-methoxy
4-fluoro
3-methoxy
112.3
a
IC50 = compound concentration required to inhibit cell growth by 50% after cells were treated with compounds for 72 h. Data are expressed as the
mean from at least three independent experiments.
cytotoxicity. All of the modified analogues (6-64−6-84)
exhibited a marked decrease in cytotoxicity, as shown in
Table 3. As the size of the substituent becomes bigger, these
derivatives completely lost cytotoxicity, suggesting that the
oxazole binding region can accommodate only a small
substituent for favorable interaction. We also modified the
oxazole core to thiazole (7-1−7-2) and isoxazole (11-1−11-5)
to investigate the structural effects of the main core on
cytotoxicity, as shown in Table 4. We chose to modify the
oxazole core of compound 6-51, which showed the lowest IC50
value (10.3 nM). Unfortunately, all of the derivatives with a
thiazole or an isoxazole core became significantly less potent,
showing 10−20-fold increased IC50 values, which indicates that
having an oxazole core is critical for cytotoxicity. We also
replaced the arylpiperazine group on the oxazole with various
heterocycles including arylpiperidines and homopiperazines
(Supporting Information, Tables S2 and S3); however, all of
the modified compounds in this series completely lost
cytotoxicity, which indicates that both of the nitrogen atoms
in the piperazine contribute to possible hydrogen-bond
interactions in this region. In addition, the spatial arrangement
of the substituted aryl group also appears to be crucial for
proper binding.
Additionally, we tested several selected compounds against
various cancer cell lines including human lung cancer cells and
colon cancer cells, as described in Table 5. Not surprisingly, all
of the tested compounds showed moderate to potent
cytotoxicity similar to what these compounds already
demonstrated against HL-60 cells. It was also observed that
compounds with low IC50 values against HL-60 (6-51−6-53)
showed low IC50 values against other cancer cell lines,
9012
dx.doi.org/10.1021/jm400840p | J. Med. Chem. 2013, 56, 9008−9018

Journal of Medicinal Chemistry
Article
Table 3. Cytotoxic Effects of (2-Substituted-4-phenyloxazol-
5-yl)(phenylpiperazin-1-yl)methanone Derivatives against
Human Leukemia Cells (HL-60)
Table 5. Cytotoxic Effects of Selected Compounds against
Various Cancer Cell Lines
a
b
IC50 (nM)
H-460
HCT-116
A-549
6-34
6-48
6-49
6-50
6-51
6-52
6-53
12
211
165
221
243
42
375
319
366
484
26
302
204
354
296
27
a
R1
R2
R3
IC50 (nM)
67
216
404
356
98
6-64
6-65
6-66
6-67
6-68
6-69
6-70
6-71
6-72
6-73
6-74
6-75
6-76
6-77
6-78
6-79
6-80
6-81
6-82
6-83
6-84
2-fluoro
ethyl
3-chloro
3-chloro
3-chloro
3-chloro
3-chloro
3-chloro
3-chloro
3-chloro
3-chloro
3-chloro
3-chloro
3-chloro
3-chloro
3-chloro
3-methoxy
3-methoxy
3-methoxy
3-methoxy
3-methoxy
3-methoxy
3-methoxy
138.8
>1000
NC
68
158
338
2-fluoro
t-butyl
272
2-fluoro
cyclohexyl
trifluoromethyl
ethyl
a
H-460, human lung cancer cells; HCT-116, human colorectal
carcinoma cell line; A-549, human lung adenocarcinoma cell line.
IC50 = compound concentration required to inhibit cell growth by
50% after cells were treated with compounds for 72 h. Data are
expressed as the mean from at least three independent experiments.
2-fluoro
NC
3-fluoro
317.3
NC
b
3-fluoro
trifluoromethyl
ethyl
4-fluoro
168.5
NC
4-fluoro
trifluoromethyl
ethyl
2-chloro
3-chloro
4-methoxy
4-methoxy
3,4-dimethoxy
3,4-dimethoxy
2-fluoro
98.9
NC
effective concentration to inhibit 50% of HUVEC growth
(GI50). In addition, the GI50 value for each compound was
measured at the different densities of HUVECs (HUVEC
distribution, ng/mL) where inhibition of cell growth was
observed. The relative ratio of GI50 versus HUVEC distribution
(normalized GI50) was also calculated and is shown in Table 6.
trifluoromethyl
ethyl
>1000
NC
propyl
ethyl
NC
propyl
NC
ethyl
606.7
200.5
NC
3-fluoro
ethyl
Table 6. Effects of Selected Compounds on the Capillary
Formation by Human Umbilical Vein Endothelial Cells
(HUVECs)
3-fluoro
trifluoromethyl
ethyl
t-butyl
NC
isopropyl
4-fluoro
ethyl
886.5
410.6
NC
a
b
ethyl
HUVEC distribution (ng/mL)
GI50 (ng/mL)
b/a
4.6
4-fluoro
trifluoromethyl
3
100
300
100
100
30
463
a
IC50 = compound concentration required to inhibit cell growth by
50% after cells were treated with compounds for 72 h. Data are
expressed as the mean from at least three independent experiments.
6-1
9,631
2,520
1052
319
32.1
25.2
10.5
10.6
11.7
3.8
6-4
6-6
6-20
6-22
6-23
6-24
6-26
6-27
6-32
6-34
6-48
6-49
6-50
6-51
6-52
6-53
12
Table 4. Cytotoxic Effects of (2-Methyl-4-phenylthiazol)-5-
yl)methanone Derivatives and (3-Methyl-4-phenylisoxazol-
5-yl)methanone Derivatives against Human Leukemia Cells
(HL-60)
300
100
100
100
100
1000
300
100
100
300
30
3522
380
1732
356
17.3
3.6
191
1.9
177
134
1.77
1.34
a
a
R
IC50 (nM)
R
IC50 (nM)
118
126
3.93
1.26
7-1
7-2
3,5-dimethoxy
3,5-dichloro
96.8
11-1
11-2
11-3
11-4
11-5
H
152.3
165.3
151.9
475.3
191.4
100
300
100
156.8
2-fluoro
3-fluoro
4-fluoro
2-furyl
2098
20.98
a
Data represents the starting concentrations of each compound for the
b
disruption of capillary-like structure formation. GI50 = effective
concentration of the tested compounds to inhibit 50% of HUVEC
growth.
a
IC50 = compound concentration required to inhibit cell growth by
50% after cells were treated with compounds for 72 h. Data are
expressed as the mean from at least three independent experiments.
confirming the potent antiproliferative activities of these
compounds.
Representative images of the HUVEC growth assay are shown
in Figure 2 in comparison with 12 (CYT997, IC50 = 112.3 nM)
as reference. The two most cytotoxic compounds, 6-20 (R1 =
H, R2 = 3,5-dimethoxy, IC50 = 19.2 nM) and 6-51 (R1 = 2-F, R2
= 3,5-dimethoxy, IC50 = 10.3 nM) inhibited the growth of
HUVECs at the lowest concentration, 30 ng/mL, whereas 6-32
(R1 = 3-hydroxy, R2 = 3-chloro, IC50 = 104.9 nM), which
showed the highest IC50 among the selected compounds,
Vascular-Disrupting Activity (HUVEC Assay). To
determine the vascular-disrupting effect of the synthesized
compounds, we selected 18 compounds with low IC50 values
ranging from 10 to 100 nM for further evaluation. We
incubated human umbilical vein endothelial cells (HUVECs)
with each compound at different concentrations to calculate the
9013
dx.doi.org/10.1021/jm400840p | J. Med. Chem. 2013, 56, 9008−9018

Journal of Medicinal Chemistry
Article
Figure 2. Representative images of HUVECs (human umbilical vein endothelial cells) in the vascular-disrupting activity assay. HUVECs were placed
onto a thick layer of Matrigel in the presence of vehicle (control) or tested compounds at the indicated concentrations.
inhibited the growth of HUVECs at only 1000 ng/mL. All
other compounds inhibited growth of HUVEC at 100−300 ng/
mL. This result indicates that the in vitro cytotoxicity of these
compounds correlates with their vascular-disrupting activity.
Tubulin Polymerization Inhibition (TPI Test). To
examine the ability to block tubulin polymerization of the
synthesized compounds, we tested several compounds with
high cytotoxicity, as shown in Table 7. Interestingly, all tested
Table 8. Percentages of Remaining Drug after Incubation of
Synthesized Compounds in the Presence of NADPH at 37
°C
microsomal stability
human (%)
mouse (%)
3
13.5
8
2
6-4
1.6
6-20
6-23
6-24
6-26
6-27
6-34
6-48
6-40
6-50
6-51
6-52
6-53
12
15.8
29.8
33.8
25.6
66.3
70.3
82.3
75.9
71.8
15
0.7
83.6
1.4
Table 7. Inhibition of Tubulin Polymerization of Selected
Compounds
12
a
IC50 (μg/mL)
43.4
72.5
49
6-1
0.62
0.53
0.67
0.73
0.77
0.71
1.01
6-23
6-27
40.7
31.1
16.6
11.6
5.6
6-48
6-49
6-51
11.3
12.9
54.6
13 (AC7739)
a
87.5
The tubulin concentration was 4 mg/mL. Data are expressed as the
mean from at least three independent experiments.
3,5-dichloro) seemed more stable in human liver microsomes
than in mouse liver microsomes, and these compounds also
showed much higher stability than the compounds with
methoxy groups (6-51−6-53, R2 = 3,5-dimethoxy). It was
disappointing that compounds 6-51−6-53 demonstrated low
metabolic stability because these compounds showed excellent
cytotoxic effects and vascular-disrupting activity in vitro.
Overall, we decided to move forward with compounds 6-48
(R1 = 2-F, R2 = 3,5-dimethyl) and 6-51 (R1 = 2-F, R2 = 3,5-
dimethoxy) for further evaluation of their antitumor activity in
vivo.
Tumor Growth Inhibition in Vivo. Considering in vitro
biological activity and metabolic stability, we selected 6-48 and
6-51 and evaluated antitumor activity in vivo using a simple
tumor xenograft model (HCT-116, human colorectal carcino-
ma) (Figure 3). Tumor-bearing mice were injected intra-
peritoneally with each compound once a day for 14 days, and
tumor volumes were determined by caliper measurements.
After 5 days of the treatment with 6-48, significant inhibition of
tumor growth was observed at 50 and 100 mg/kg. When
compared to untreated mice, the treatment with compound 6-
compounds inhibited tubulin polymerization at lower concen-
trations compared to known polymerization inhibitor 13
(AC7739),19 indicating that cytotoxicity and vascular-disrupting
activity of these compounds are due to the inhibition of tubulin
polymerization during mitosis.
Microsomal Stability. To evaluate metabolic stability, we
selected 14 compounds that showed good vascular-disrupting
effects at the concentrations less than 100 ng/mL and
measured microsomal stability using liver microsomes from
human and mouse (Table 8). We incubated each compound in
buffers containing liver microsomes and measured the quantity
of remaining drugs after 1 h. Compounds 6-23 and 6-48 were
the most stable, having 83.6 (in mouse liver microsomes) and
82.3% (in human liver microsomes) remaining, respectively. It
was remarkable that over 70% of 6-34 was remaining in both
microsomes (70.3% for human, 72.5% for mouse), and we
believe that hydrophilic groups on the R1 position may
contribute to metabolic stability in liver microsomes despite
having low cytotoxicity and weak vascular-disrupting ability.20
The compounds with halide groups on R2 (6-48−6-50, R2 =
9014
dx.doi.org/10.1021/jm400840p | J. Med. Chem. 2013, 56, 9008−9018

Journal of Medicinal Chemistry
Article
Figure 3. Antitumor effects of daily administration of compounds 6-48 and 6-51: (a) tumor growth inhibition by 6-48 and 6-51 in HCT-116 colon
cancer xenograft model (calculated on the basis of the vehicle control) and (b) changes in body weight over the course of the treatment.
48 reduced the size of the tumors to 42% at 100 mg/kg and
16.1% at 50 mg/kg, whereas 19% of body weight was lost at
100 mg/kg and 10.8% at 50 mg/kg. When we injected 200 mg/
kg of 6-48, tumor growth was inhibited markedly; however, two
out of six mice died during the experiment, indicating that 6-48
is toxic at high concentrations. Unfortunately, despite having
potent activity in vitro, the treatment of 6-51 did not affect
tumor growth in vivo and did not seem to change body weight.
On the basis of the microsomal stability test, we believe that
because of poor metabolic stability 6-51 may have degraded
during systemic circulation before the compound reached its
therapeutic target.
because of its metabolic stability. It is interesting to note that 6-
48 showed greater effects on tumor growth in vivo despite
having a much lower in vitro potency than 6-51. Considering
that many tubulin binders are too toxic and have short half-
lives, which prevents them from shutting down tumor
vasculature effectively and thus possibly leaving an outer rim
of viable cancer cells,21 we believe that this compound will serve
as an excellent lead to discover potent dual-acting agents.
EXPERIMENTAL SECTION
■
Docking Studies. The crystal structures of complexes (PDB codes
3UT5 and 3HKC) were initially prepared by adding hydrogen atoms
at pH 7.4 and charges using the Protein Preparation Wizard in
Maestro9.2. All waters were deleted during preparation. The ligand
structures were drawn and prepared by ionizing at pH 7.4 using the
LigPrep module in Maestro9.2. All ligand structures were minimized
using the OPLS_2001 force field. The ligands were docked into the
protein structures using Glide XP module in Maestro 9.2. A receptor
grid was generated before docking the compounds. An outer box was
used to generate the grid and accommodating ligands, and an inner
box was used for ligand center site point search. The outer and inner
boxes were set to lengths of less than 20 and 10 Å, respectively. For
grid generation, the scaling factor for receptor atom van der Waals
radius was set to 1.0.
CONCLUSIONS
■
We have developed a series of aryloxazole, thiazole, and
isoxazole for antimitotic and vascular-disrupting agents. In vitro
cytotoxicity was evaluated using various tumor cell lines, and
tumor vascular-disrupting activity was also tested using a
HUVEC assay. Several compounds with excellent cytotoxicity
and vascular-disrupting activity were selected, and metabolic
stability and inhibition of tubulin polymerization were
examined. As a result, the two most potent compounds, 6-48
and 6-51, were chosen for further in vivo evaluation of
antitumor activity. Compound 6-51 (R1 = 2-F, R2 = 3,5-OMe)
exhibited the highest cytotoxicity, with an IC50 value of 10.3
nM against HL-60, and excellent vascular-disrupting activity.
Although compound 6-51 seemed only moderately stable in
liver microsomes (15% for human, 16.6% for mice), compound
6-48 (R1 = 2-F, R2 = 3,5-Cl) demonstrated an outstanding
microsomal stability (82.3% for human, 49% for mice),
although it showed relatively lower cytotoxicity (IC50 = 60.5
nM) and vascular-disrupting activity compared to 6-51. Both of
these compounds inhibited tubulin polymerization at low
concentrations, suggesting that their biological activity comes
from tubulin binding. When injected into tumor-bearing mice,
compound 6-48 effectively reduced tumor growth (42.3% in
size) at 100 mg/kg; however, compound 6-51 did not affect
tumor growth, apparently because of its poor metabolic
stability.
Synthesis. General Methods. All chemicals were reagent grade
and used as purchased. Moisture-sensitive reactions were performed
under an inert atmosphere of dry nitrogen with dried solvents.
Reactions were monitored by TLC analysis using Merck silica gel 60
F-254 thin-layer plates. Flash column chromatography was carried out
1
on Merck silica gel 60 (230−400 mesh). H NMR and 13C NMR
spectra were recorded on a Bruker (AVANCE II) at 400 and 300 MHz
1
for H NMR and at 100 and 75 MHz for 13C NMR. The coupling
constants (J) are reported in Hz. High-resolution mass spectra
(HRMS) were recorded on a LTQ Orbitrap (Thermo Electron
Corporation). Microwave-mediated reactions were carried out in a
Personal Chemistry (Biotage) microwave synthesizer. Purity was
measured using a Waters e2695/2489 HPLC instrument (1 mL/min
flow rate; H2O/CH3CN 90:10 → 0:100 in 15 min; 10 μL of injection
volume; λ = 254, 280 nm). All test compounds exhibited >95% purity
except fo compounds 6-30 and 6-34, which were 94 and 90% pure,
respectively.
Detailed synthetic procedures and characterization data for
intermediates 4, 5, and aryl piperazine fragments can be found in
the Supporting Information.
Ethyl 2-Methyl-4-phenyloxazole-5-carboxylate (4-1). Ethyl 3-oxo-
3-phenylpropanoate (8 g, 41 mmol) was mixed with HDNIB (25 g, 54
mmol), prepared in advance, and 7.2 g of acetamide (0.12 mol) was
We have developed a dual-action anticancer agent containing
an aryloxazole moiety. Compound 6-48 showed an excellent
antimitotic effect and vascular-disrupting activity in vitro and
demonstrated promising antitumor activity in vivo, possibly
9015
dx.doi.org/10.1021/jm400840p | J. Med. Chem. 2013, 56, 9008−9018

Journal of Medicinal Chemistry
Article
added to the mixture. The reaction mixture was heated at 120 °C for
30 s under microwave irradiation (80 W). After being cooled to rt, the
suspension was diluted with saturated NaHCO3 solution (10 mL),
extracted with EA, dried over MgSO4, filtered, and concentrated in
vacuo. Purification of the residue by flash column chromatography on
silica gel (EA/Hex 1:7) afforded desired product 1 (6.01 g, 26 mmol,
63%). 1H NMR (300 MHz, CDCl3) δ 8.07−8.02 (m, 2H), 6.98−6.94
(m, 3H), 4.40 (q, J = 7.12 Hz, 2H), 2.57 (s, 3H), 1.43 (t, J = 7.14 Hz,
3H).
3-Methylisoxazole-5-carboxylic Acid (8). To a solution of
acetaldehyde oxime (1 g, 16.9 mmol) in dry THF (15 mL) was
added NCS (2.7 g, 20 mmol) at rt. After 2 h of stirring, a diluted
solution of propargyl alcohol (1.42 g, 25 mmol) in dry THF (4 mL)
was added slowly, and after 30 min, a diluted solution of TEA (2.78
mL, 20 mmol) in dry THF was added dropwise successively. After 1 h
with vigorous stirring, water (10 mL) was added to the reaction
mixture. The suspension was extracted with EA (20 mL × 3), and the
organic layer was dried over MgSO4, filtered, and concentrated in
vacuo. Purification of the residue by flash column chromatography on
silica gel (EA/Hex 1:1) afforded (3-methylisoxazol-5-yl)methanol
2-Methyl-4-phenyloxazole-5-carboxylic Acid (5-1). To a solution
of 1 (1.74 g, 7.55 mmol) in ethanol (10 mL) was added 1 N NaOH
solution (22.6 mL) at rt. The mixture was heated at 40 °C for 1 h.
After cooling, the mixture was adjusted to pH 2 with 1 N HCl solution
at 0 °C, and white solid was precipitated and filtered. The residue was
extracted with EA (50 mL × 3), and the combined organic layers were
dried over MgSO4, filtered, and concentrated in vacuo. A white solid
was obtained without purification as desired product 2 (1.22 g, 6.04
1
(0.93 g, 8.28 mmol, 49%). H NMR (300 MHz, CDCl3) δ 6.09 (s,
1H), 4.73 (s, 2H), 2.52 (br, 1H), 2.34 (s, 3H). 13C NMR (75 MHz,
CDCl3) δ 171.09, 159.85, 102.53, 56.46, 11.37.
To a solution of (3-methylisoxazol-5-yl)methanol (0.5 g, 4.42
mmol) in acetone (19 mL) was added Jones reagent (2.21 mL) at 0
°C, and the mixture was stirred for 3 h. The reaction was monitored by
TLC, and after termination of the reaction, the mixture was adjusted to
pH 2 with 1 N HCl solution. The residue was extracted with EA (20
mL × 3), and the combined organic layer was dried over MgSO4,
filtered, and concentrated in vacuo. Purification of the residue by flash
column chromatography on silica gel (EA/Hex 1:1) afforded desired
product 8 (0.63 g, 4.97 mmol, 100%). 1H NMR (300 MHz, CH3OD)
δ 6.90 (s, 1H), 2.35 (s, 3H). 13C NMR (75 MHz, CH3OD) δ 161.05,
160.82, 158.16, 109.46, 9.81.
4-Iodo-3-methylisoxazole-5-carboxylic Acid (9). To a solution of 8
(85 mg, 0.66 mmol) in dry THF (5 mL) was added n-BuLi (2.64 mL,
2 M in hexane) at −78 °C. After the mixture was stirred for 1 h, iodine
(0.2 g, 0.8 mmol) in dry THF was added dropwise at −30 °C. After 5
h with vigorous stirring at −78 °C, 1 N HCl solution (2 mL) was
added to the reaction mixture. This suspension was extracted with EA
(5 mL × 3), and the combined organic layer was washed with Na2SO3
satd solution and brine water (5 mL), dried over MgSO4, filtered, and
concentrated in vacuo. Purification of the residue by flash column
chromatography on silica gel (CH2Cl2/CH3OH 5:2) afforded desired
compound 6 (48 mg, 0.19 mmol, 29%). 1H NMR (300 MHz,
CH3OD) δ 2.34 (s, 3H). 13C NMR (75 MHz, CH3OD) δ 163.99,
159.88, 157.70, 67.76, 11.06.
1
mmol, 80%). H NMR (300 MHz, CH3OD) δ 7.99 (q, J = 1.96 Hz,
2H), 7.42 (q, J = 1.80 Hz, 3H), 2.56 (s, 3H).
General Procedure for Compounds 3 and 6-1 through 6-91. To a
mixture of 5 (45.5 mg, 0.224 mmol), EDCI (90 mg, 0.448 mmol),
HOBt (50 mg, 0.39 mmol), and NMM (40 μL, 0.39 mmol) in dry
CH2Cl2 (10 mL) was added phenylpiperazine (40 μL, 0.27 mmol).
After 1 h of stirring, water (10 mL) was added and extracted with
CH2Cl2 (10 mL × 3), and the organic layer was dried over MgSO4,
filtered, and concentrated in vacuo. The residue was purified by flash
column chromatography on silica gel (CH2Cl2/CH3OH 40:1) to
obtain the desired product.
(4-(3-Chlorophenyl)piperazin-1-yl)(2-methyl-4-phenyloxazol-5-
1
yl)methanone (3). H NMR (400 MHz, CDCl3) δ 7.75−7.73 (m,
2H), 7.41−7.36 (m, 3H), 7.15−7.13(m, 1H), 6.85−6.83 (m, 1H),
6.81(s, 1H), 6.80 (d, J = 2.04 Hz, 1H), 3.89 (br, 2H), 3.46 (br, 2H),
3.22 (br, 2H), 2.91 (br, 2H), 2.59 (s, 3H). 13C NMR (100 MHz,
CDCl3) δ 161.74, 159.86, 151.74, 140.58, 137.69, 135.05, 130.38,
130.20, 129.16, 128.80, 127.65, 120.33, 116.48, 114.52, 50.22, 48.80,
46.62, 42.13, 14.07. mp 191−193 °C. HPLC purity 99%. HR-MS m/z
[M + H]+ (ESI+) calcd for C21H21ClN3O2, 382.1317; found, 382.1328.
(2-Methyl-4-phenyloxazol-5-yl)(4-phenylpiperazin-1-yl)-
methanone (6-1). 6-1 was obtained in 48% yield (40 mg, 0.18 mmol).
1H NMR (300 MHz, CDCl3) δ 7.76 (d, J = 7.50 Hz, 2H), 7.44−7.37
(4-(3,5-Dimethoxyphenyl)piperazin-1-yl)(4-iodo-3-methylisoxa-
zol-5-yl)methanone (10). In a flask, 4-iodo-3-methylisoxazole-5-
carboxylic acid 6 (30 mg, 0.12 mmol), EDCI (33 mg, 0.18 mg),
HOBt (23 mg, 0.18 mg), and NMM (0.02 mL, 0.18 mmol) were
dissolved in 5 mL of dry CH2Cl2. The reaction mixture was stirred for
1 h. 1-(3,5-Dimethoxyphenyl)piperazine (32 mg, 0.14 mmol) was
added in the reaction mixture and stirred for 1 h. Water (5 mL) was
added to the mixture, the mixture was extracted with CH2Cl2, and the
organic layer was washed with Na2SO3 satd solution (5 mL) and brine
water (5 mL), dried over MgSO4, filtered, and concentrated in vacuo.
Purification of the residue by flash column chromatography on silica
gel (EA/Hex 3:5) afforded the desired compound (21 mg, 0.046
(m, 3H), 7.27 (t, J = 7.91 Hz, 2H), 6.96−6.86 (m, 3H), 3.93 (br, 2H),
3.45 (br, 2H), 3.24 (br, 2H), 2.93 (br, 2H), 2.57 (s, 3H). 13C NMR
(75 MHz, CDCl3) δ 161.62, 159.87, 150.75, 140.38, 137.82, 130.43,
129.25, 129.06, 128.76, 127.61, 127.12, 120.72, 116.74, 50.22, 47.88,
14.03. mp 180−182 °C. HPLC purity 98%. HR-MS m/z [M + H]+
(ESI+) calcd for C21H22N3O2, 348.1707; found, 348.1723.
(4-(3,5-Dichlorophenyl)piperazin-1-yl)(4-(2-fluorophenyl)-2-
1
methyloxazol-5-yl)methanone (6-48). H NMR (300 MHz, CDCl3)
δ 7.75 (td, J = 7.51, 1.58 Hz, 1H), 7.41−7.34 (m, 1H), 7.27−7.22 (m,
1H), 7.15−7.09 (m, 1H), 6.84 (s, 1H), 6.71 (s, 2H), 3.86 (br, 2H),
3.53 (br, 2H), 3.22 (br, 2H), 3.03 (br, 2H), 2.65 (s, 3H). 13C NMR
(75 MHz, CDCl3) δ 161.64, 159.67 (d, JC−F = 248.5 Hz), 159.43,
152.14, 140.02, 135.59, 135.10, 130.84 (d, JC−F = 8.3 Hz), 130.30,
130.27, 124.60 (d, JC−F = 3.4 Hz), 119.79, 118.85, 118.67, 115.92 (d,
JC−F = 21.7 Hz), 114.43, 48.39, 47.56, 14.02. mp 248−249 °C. HPLC
purity 95%. FABMS (m/z): [M+ + H] calcd for C21H19Cl2FN3O2.
434.0760; found, 434.0843.
1
mmol, 40%). H NMR (300 MHz, CDCl3) δ 6.10 (q, J = 1.46 Hz,
3H), 3.92 (t, J = 4.64 Hz, 2H), 3.82 (s, 6H), 3.65 (t, J = 4.50 Hz, 2H),
3.28 (t, J = 5.16 Hz, 2H), 3.23 (t, J = 4.95 Hz, 2H), 2.18 (s, 3H).
General Procedure for Compound 11. To a solution of 10 (21 mg,
0.5 mmol), phenyl-boronic acid (7 mg, 0.55 mmol), and Pd(PPh3)4
(44 mg, 3.5 × 10−7 mol) in dry dioxane (5 mL) was added sodium
carbonate (8 mg, 0.75 mmol) in water (1.0 mL) at rt. The reaction
mixture was heated at 100 °C for 10 h. After being cooled to rt, the
mixture was filtered over a Celite pad, and the filtrate was evaporated
to dryness. The residue was purified by flash column chromatography
on silica gel (EA/Hex 1:2), affording the desired compound.
Biological Assays. Cytotoxicity Assay. The HL-60 (leukemia) cell
line was obtained from American Type Culture Collection (ATCC,
USA). HL-60 cells were grown in RPMI-1640 medium containing
10% heat-inactivated fetal bovine serum (FBS) at 37 °C under a
humidified 5% CO2 atmosphere. All cells were seeded into 96-well
plates, and the compound diluents were added. After a 72 h incubation
at 37 °C in a humidified 5% CO2 atmosphere, cell viability was
determined by addition of MTT (final concentration 0.25 mg/mL).
(4-(3,5-Dimethoxyphenyl)piperazin-1-yl)(4-(2-fluorophenyl)-2-
1
methyloxazol-5-yl)methanone (6-51). H NMR (300 MHz, CDCl3)
δ 7.75 (td, J = 7.51, 1.27 Hz, 1H), 7.39−7.33 (m, 1H), 7.27−7.21 (m,
1H), 7.15−7.09 (m, 1H), 6.05 (s, 3H), 3.79 (br, 2H), 3.77 (s, 6H),
3.53 (br, 2H), 3.21−2.96 (br, 4H), 2.58 (s, 3H). 13C NMR (100 MHz,
CDCl3) δ 161.59, 161.54, 159.69 (d, JC−F = 248.0 Hz), 159.40, 152.77,
140.19, 134.83, 130.73 (d, JC−F = 8.5 Hz), 130.30, 130.26, 124.52 (d,
JC−F = 3.3 Hz), 118.94, 118.75, 115.92 (d, JC−F = 21.7 Hz), 95.71,
92.34, 55.26, 49.30, 46.56, 42.14, 14.00. mp 221−222 °C. HPLC
purity 95%. FABMS (m/z): [M+ + H] calcd for C23H25FN3O4,
426.1751; found, 426.1829. HR-MS m/z [M + H]+ (ESI+) calcd for
C23H25FN3O4, 426.1824; found, 426.1841.
9016
dx.doi.org/10.1021/jm400840p | J. Med. Chem. 2013, 56, 9008−9018

Journal of Medicinal Chemistry
Article
HUVEC Tube-Disruption Assay. HUVECs were obtained from
Modern Cell & Tissue Technologies (MCTT, Korea). HUVECs were
cultured in endothelial cell growth medium (EGM-2) and used
between passages 3 and 7. HUVECs were seeded on a Matrigel-coated
96-well plate. After capillary-like tube formation was observed,
HUVECs were treated with DMSO or serial-diluted test compounds
for 1 h.
Tubulin Polymerization Assay. Tubulin polymerization was carried
out using a kit supplied by Cytoskeleton. A DMSO solution of the test
compounds was prepared at varying concentrations (10, 3, 1, and 0.3
μg/mL). Compounds that interact with tubulin alter the polymer-
ization of tubulin can be detected using a spectrophotometer
(FlexStation, Molecular Devices, LLC). The absorbance at 340 nm
over a period of 1 h at 37 °C was monitored. The experimental
procedure of the assay was performed as described in version 8.2 of the
tubulin polymerization assay kit manual.
Microsomal Stability. The reaction mixture consisted of potassium
phosphate buffer (50 mM, pH 7.4), 2.5 mM magnesium chloride,
NADPH-generating system (1.0 NADP, +1.0 NADPH, 2.5 glucose 6-
phosphate, 1.0 U glucose 6-phosphate dehydrogenase), and testing
compounds (final concentrations at 5 μM) in a total volume of 200
μL. The reaction mixture was incubated at 37 °C for 60 min in a
shaking water bath, and the reaction was stopped by adding 200 μL of
acetonitrile. The samples were then spun at 13 000 rpm for 5 min, and
aliquots (10 μL) were analyzed by LC/MS. The metabolic stability of
test compounds was evaluated by measuring the disappearance of each
compound compared to the control (no NADPH and NADP+)
sample.
Inhibition of Tumor Growth in Vivo. Male BALB/C nu/nu mice
were obtained from the Central Lab Animal Inc. (Seoul, Korea).
Procedures involving animals and their care were conducted in
conformity with institutional guidelines, which are in compliance with
the Korean Animal Welfare Act. Mice were used between 5 to 6 weeks
of age. The antitumor activity of selected compounds was measured in
a human colorectal carcinoma (HCT-116) xenograft model in male
BALB/C nu/nu mice. HCT-116 cells were implanted subcutaneously
in the flanks of nude mice. Mice bearing tumors of approximately the
same size (100−200 mm3) were dosed intraperitoneally with vehicle
or testing compounds once a day for 14 days. Tumor volumes were
determined twice per week by caliper measurements and were
recorded along with body weights and mortality as an indicator of
gross toxicity.
ACKNOWLEDGMENTS
■
This study was supported by the Seoul R&D program
(2G08020) and Korea Institute of Science and Technology
(KIST, 2V03110 and 2E23870).
ABBREVIATIONS USED
■
VDA, vascular-disrupting agents; HUVECs, human umbilical
vein endothelial cells; SAR, structure−activity relationship;
IC50, inhibitory concentration 50%; HR-MS, high-resolution
mass spectrometry; HPLC, high-performance liquid chroma-
tography
REFERENCES
■
(1) Wood, K. W.; Cornwell, W. D.; Jackson, J. R. Past and future of
the mitotic spindle as an oncology target. Curr. Opin. Pharmacol. 2001,
1, 370−377.
(2) Dumontet, C.; Jordan, M. A. Microtubule-binding agents: A
dynamic field of cancer therapeutics. Nat. Rev. Drug. Discovery 2010, 9,
790−803.
(3) Goldspiel, B. R. Clinical overview of the taxanes. Pharmacotherapy
1997, 17, 110−125.
(4) Budman, D. R. Vinorelbine (Navelbine): A third-generation vinca
alkaloid. Cancer Invest. 1997, 15, 475−490.
(5) Lippert, J. W., 3rd. Vascular disrupting agents. Bioorg. Med. Chem.
2007, 15, 605−615.
(6) Patterson, D. M.; Rustin, G. J. Vascular damaging agents. Clin.
Oncol. (R. Coll. Radiol.) 2007, 19, 443−456.
(7) Tron, G. C.; Pirali, T.; Sorba, G.; Pagliai, F.; Busacca, S.;
Genazzani, A. A. Medicinal chemistry of combretastatin A4: Present
and future directions. J. Med. Chem. 2006, 49, 3033−3044.
(8) Baguley, B. C. Antivascular therapy of cancer: DMXAA. Lancet
Oncol. 2003, 4, 141−148.
(9) Davis, P. D.; Dougherty, G. J.; Blakey, D. C.; Galbraith, S. M.;
Tozer, G. M.; Holder, A. L.; Naylor, M. A.; Nolan, J.; Stratford, M. R.;
Chaplin, D. J.; Hill, S. A. ZD6126: A novel vascular-targeting agent
that causes selective destruction of tumor vasculature. Cancer Res.
2002, 62, 7247−7253.
(10) Flynn, B. L.; Gill, G. S.; Grobelny, D. W.; Chaplin, J. H.; Paul,
D.; Leske, A. F.; Lavranos, T. C.; Chalmers, D. K.; Charman, S. A.;
Kostewicz, E.; Shackleford, D. M.; Morizzi, J.; Hamel, E.; Jung, M. K.;
Kremmidiotis, G. Discovery of 7-hydroxy-6-methoxy-2-methyl-3-
(3,4,5-trimethoxybenzoyl)benzo[b]furan (BNC105), a tubulin poly-
merization inhibitor with potent antiproliferative and tumor vascular
disrupting properties. J. Med. Chem. 2011, 54, 6014−6027.
(11) Nguyen, T. L.; McGrath, C.; Hermone, A. R.; Burnett, J. C.;
Zaharevitz, D. W.; Day, B. W.; Wipf, P.; Hamel, E.; Gussio, R. A
common pharmacophore for a diverse set of colchicine site inhibitors
using a structure-based approach. J. Med. Chem. 2005, 48, 6107−6116.
(12) Yoshino, H.; Ueda, N.; Niijima, J.; Sugumi, H.; Kotake, Y.;
Koyanagi, N.; Yoshimatsu, K.; Asada, M.; Watanabe, T.; Nagasu, T.;
et al. Novel sulfonamides as potential, systemically active antitumor
agents. J. Med. Chem. 1992, 35, 2496−2497.
(13) Rudin, C. M.; Mauer, A.; Smakal, M.; Juergens, R.; Spelda, S.;
Wertheim, M.; Coates, A.; McKeegan, E.; Ansell, P.; Zhou, X.; Qian, J.;
Pradhan, R.; Dowell, B.; Krivoshik, A.; Gordon, G. Phase I/II study of
pemetrexed with or without ABT-751 in advanced or metastatic non-
small-cell lung cancer. J. Clin. Oncol. 2011, 29, 1075−1082.
(14) Andreu, J. M.; Perez-Ramirez, B.; Gorbunoff, M. J.; Ayala, D.;
Timasheff, S. N. Role of the colchicine ring A and its methoxy groups
in the binding to tubulin and microtubule inhibition. Biochemistry
1998, 37, 8356−8368.
ASSOCIATED CONTENT
■
S
* Supporting Information
Initial screening hits and anti-proliferative effects against various
tumor cell lines, cytotoxic effects of (2-methyl-4-phenyloxazol
(or phenylthiazol)-5-yl) methanone derivatives and (3-methyl-
4-phenylisoxazol-5-yl)methanone derivatives against human
leukemia cells (HL-60), cytotoxic effects of (2-methyl-4-
phenyloxazol-5-yl) (phenylhomopiperazin-1-yl)methanone de-
rivatives against human leukemia cells (HL-60), changes in
tumor volume of compounds 6-48 and 6-51 treated mice, and
detailed synthetic procedures and characterization data of all
test compounds. This material is available free of charge via the
Internet at http://pubs.acs.org.
AUTHOR INFORMATION
■
Corresponding Authors
*Tel.: +82-2-958-5165; Fax: +82 2 958 5189 ; E-mail: gsnam@
kist.re.kr (G.N.).
(15) Sharma, S.; Poliks, B.; Chiauzzi, C.; Ravindra, R.; Blanden, A. R.;
Bane, S. Characterization of the colchicine binding site on avian
tubulin isotype betaVI. Biochemistry 2010, 49, 2932−2942.
(16) Dorleans, A.; Gigant, B.; Ravelli, R. B.; Mailliet, P.; Mikol, V.;
Knossow, M. Variations in the colchicine-binding domain provide
*Tel.: +82 2 958 5185; Fax: +82 2 958 5189; E-mail: anpae@
kist.re.kr (A.N.P.).
Notes
The authors declare no competing financial interest.
9017
dx.doi.org/10.1021/jm400840p | J. Med. Chem. 2013, 56, 9008−9018

Journal of Medicinal Chemistry
Article
insight into the structural switch of tubulin. Proc. Natl. Acad. Sci. U.S.A.
2009, 106, 13775−13779.
(17) Ravelli, R. B.; Gigant, B.; Curmi, P. A.; Jourdain, I.; Lachkar, S.;
Sobel, A.; Knossow, M. Insight into tubulin regulation from a complex
with colchicine and a stathmin-like domain. Nature 2004, 428, 198−
202.
(18) Burns, C. J.; Harte, M. F.; Bu, X.; Fantino, E.; Joffe, M.;
Sikanyika, H.; Su, S.; Tranberg, C. E.; Wilson, N.; Charman, S. A.;
Shackleford, D. M.; Wilks, A. F. Discovery of CYT997: A structurally
novel orally active microtubule targeting agent. Bioorg. Med. Chem.
Lett. 2009, 19, 4639−4642.
(19) Ohsumi, K.; Nakagawa, R.; Fukuda, Y.; Hatanaka, T.; Morinaga,
Y.; Nihei, Y.; Ohishi, K.; Suga, Y.; Akiyama, Y.; Tsuji, T. Novel
combretastatin analogues effective against murine solid tumors: Design
and structure-activity relationships. J. Med. Chem. 1998, 41, 3022−
3032.
(20) Walle, T. Methylation of dietary flavones increases their
metabolic stability and chemopreventive effects. Int. J. Mol. Sci. 2009,
10, 5002−5019.
(21) Gaya, A. M.; Rustin, G. J. Vascular disrupting agents: A new
class of drug in cancer therapy. Clin. Oncol. (R. Coll. Radiol.) 2005, 17,
277−290.
9018
dx.doi.org/10.1021/jm400840p | J. Med. Chem. 2013, 56, 9008−9018