Synthesis, biological evaluation and mechanism study of a class of cyclic combretastatin A-4 analogues as novel antitumour agents

Article abstract of DOI:10.1039/c5ra19270f

In the course of our search for novel antitumor agents, a series of cyclic combretastatin A-4 (CA-4) analogues bearing an amide group, A-B or B-C ring condensation, and CC or CN bond in the B ring were designed, synthesized and identified as new microtubule inhibitors. The structure-activity relationship (SAR) studies showed that the hexa-cyclic compounds bearing B-C ring condensation, containing a CC bond in the B ring (4a) provided excellent antiproliferative activities at nanomolar concentrations against various cancer cell lines (IC₅₀ = 46-80 nM). 4a inhibited tubulin assembly at a micromolar range (IC₅₀ = 2.56 ± 0.15 μM) as evidenced by a molecular docking study, which revealed that 4a exerted tubulin polymerisation inhibitory activity by binding to the colchicine binding site of tubulin. Further molecular biology studies showed that 4a disrupted intracellular microtubule polymerisation and thus induced G2/M phase arrest and apoptotsis in A549 cells. Altogether, these results we obtained can guide the design of novel potent molecules for future development.

Full text of DOI:10.1039/c5ra19270f

RSC Advances
View Article Online
View Journal | View Issue
PAPER
Synthesis, biological evaluation and mechanism
study of a class of cyclic combretastatin A-4
analogues as novel antitumour agents†
Cite this: RSC Adv., 2015, 5, 98527
*
Jun Yan, Yanqing Pang, Jie Chen, Jianfei Sheng, Jinhui Hu, Ling Huang
*
and Xingshu Li
In the course of our search for novel antitumor agents, a series of cyclic combretastatin A-4 (CA-4)
analogues bearing an amide group, A–B or B–C ring condensation, and C]C or C]N bond in the B
ring were designed, synthesized and identified as new microtubule inhibitors. The structure–activity
relationship (SAR) studies showed that the hexa-cyclic compounds bearing B–C ring condensation,
containing a C]C bond in the B ring (4a) provided excellent antiproliferative activities at nanomolar
concentrations against various cancer cell lines (IC50 ¼ 46–80 nM). 4a inhibited tubulin assembly at
a micromolar range (IC50 ¼ 2.56 ꢀ 0.15 mM) as evidenced by a molecular docking study, which revealed
that 4a exerted tubulin polymerisation inhibitory activity by binding to the colchicine binding site of
tubulin. Further molecular biology studies showed that 4a disrupted intracellular microtubule
polymerisation and thus induced G2/M phase arrest and apoptotsis in A549 cells. Altogether, these
results we obtained can guide the design of novel potent molecules for future development.
Received 18th September 2015
Accepted 3rd November 2015
DOI: 10.1039/c5ra19270f
www.rsc.org/advances
disodium phosphate (CA-4P) has been approved for clinical
trials.5–7 However, the cis-stilbene structure in CA-4 is prone to
Introduction
isomerize to the more thermodynamically stable, but less
bioactive trans isomers.8,9 To overcome this shortcoming, lots of
modications to stabilize the conformation by replacement of
the olenic bridge with a ring or carbonyl have been developed
in recent decades.10–14
The tubulin–microtubule system, which is crucially involved in
cell mitosis, is an attractive target for the development of highly
efficient anticancer drugs.1–3 Among them, combretastatin A-4
(CA-4, Fig. 1), rst isolated from the bark of the African willow
tree in 1982,4 dramatically disrupts tubulin assembly and its
In our previous work,15,16 via a simple cross-coupling strategy
in aqueous solution with polyethylene glycol as additive under
very mild conditions, we have synthesized and evaluated a new
series of ortho-(3,4,5-trimethoxybenzoyl)-acetanilides as tubulin
polymerisation inhibitors. Among them, compound 1 (Fig. 1)
exhibited excellent anti-proliferative activity against various
human cancer cell lines.16 Subsequently, on reaction of the
ortho-(3,4,5-trimethoxybenzoyl)-bromoacetanilides with NH3 in
water, we obtained a series of cyclic compounds including
compound 2 (Fig. 1) which displayed excellent anti-tumour
activity in vitro and in vivo.17 To investigate the detailed struc-
ture and activity relationships (SARs), we designed and
synthesized another seventeen cyclic CA-4 analogues, and
mainly focused on three aspects: the number of atoms in the B
ring (hexa-cyclic or hepta-cyclic compounds), the pattern of ring
condensation (A–B ring condensation or B–C ring condensa-
tion), and the effect of a C]C or C]N bond in the B ring. In
addition to the synthesis work, we evaluated the anti-
proliferative activities of all the synthesized compounds
toward different types of cancer cells, and also investigated the
mechanism of their anti-proliferative activity.
Fig. 1 Design strategy for combretastatin A-4 analogues.
School of Pharmaceutical Sciences, Sun Yat-sen University, Guangzhou, 510006,
China. E-mail: huangl72@mail.sysu.edu.cn; lixsh@mail.sysu.edu.cn
† Electronic supplementary information (ESI) available. See DOI:
10.1039/c5ra19270f
This journal is © The Royal Society of Chemistry 2015
RSC Adv., 2015, 5, 98527–98537 | 98527

View Article Online
RSC Advances
Paper
Results and discussion
Chemistry
Compounds 4a–d, 5, 9, and 12 were synthesized using the route
in Scheme 1. First, according to the previous method, key
intermediates 3a–d, 8, and 11 were conveniently obtained by the
reaction of acetanilides with 3,4,5-trimethoxybenzaldehyde in
the presence of 5% Pd(OAc)2 and polyethylene glycol as the
additive. The intermolecular aldol condensation was performed
in the presence of CH3ONa to afford cyclic compounds 4a–d, 9,
and 12. Then compound 4a was reacted with methyl iodine to
give 5.
Scheme 2 Synthesis of compound 16. Reagents and conditions: (a)
benzyl chloroformate, Et3N, CH2Cl2, rt, 1 h; (b) 5 mol% Pd(OAc)2, 4 eq.
TBHP, 26 mol% TFA, PEG-2000, water, rt, 12 h; (c) Pd/C, H2, THF, rt, 4
h; (d) NH2CONH2, AcOH, 110 ꢁC, 20 h.
The reaction of 3-methoxaniline with benzyl chloroformate
provided intermediate 13. By the carbon–carbon cross-coupling
reaction described above, compound 14 was obtained, which
was de-protected and then reacted with urea at a temperature of
ꢁ
carbon cross coupling and intermolecular aldol condensation
as above (Scheme 3).
110 C for 20 h to give compound 16 (Scheme 2).
Compounds 20a–e were obtained with 2,3,4-trimethoxyani-
line as starting material, by the same procedure of carbon–
Finally, the cross coupling reaction of compound 21, which
came from 2,3,4-trimethoxyaniline, with different aldehydes
gave ketone derivatives 22a–d. Compounds 22a–d via ammo-
nolysis and cyclization provided target compounds 23a–
d uently (Scheme 4).
Biological evaluation
In vitro anti-proliferative activity. To evaluate the anti-
proliferative activities of the synthesized cyclic compounds,
six human cancer cell lines, A549 (non-small-cell-lung cancer
cell line), Bel-7402 (human liver carcinoma cell line), Hela
(human epithelial cervical cancer cell line), MCF-7 (human
breast carcinoma cell line), SW480 (human colon carcinoma
cell line), and MGC-803 (human gastric cancer cell line) were
tested
using
MTT
(3-(4,5-dimethylthiazol-2-yl)-2,5-
diphenyltetrazolium bromide) assay with CA-4 as reference.
The results shown in Tables 1–3 demonstrate that most of this
series of compounds displayed good anti-proliferative activities
Scheme 1 Synthesis of compounds 4a–d, 5, 9, and 12. Reagents and
conditions: (a1) CH3COCl, CH2Cl2, rt, 1 h; (a2) CH3CH2COCl, CH2Cl2, rt, Scheme 3 Synthesis of compounds 20a–e. Reagents and conditions:
1 h; (b) 5 mol% Pd(OAc)2, 4 eq. TBHP, 26 mol% TFA, PEG-2000, water, (a) CH3COCl, CH2Cl2, rt, 1 h; (b) 5 mol% Pd(OAc)2, 4 eq. TBHP, 26 mol%
rt, 12 h; (c) 10 eq. CH3ONa, DMF, 90 ꢁC, 12 h; (d) CH3I, THF, rt, 12 h.
TFA, PEG-2000, water, rt, 12 h; (c) 10 eq. CH3ONa, DMF, 90 ꢁC, 12 h.
98528 | RSC Adv., 2015, 5, 98527–98537
This journal is © The Royal Society of Chemistry 2015

View Article Online
Paper
RSC Advances
seems slightly unfavourable for the activity, compound 5 with
CH3 instead of H at R3 position of 4a, provided 0.52–1.14 mM
IC50 values towards the six kinds of cancer cell lines. Compound
9, where the benzene ring of 4 was replaced with a naphthalene
ring gave almost 10-fold lower anti-proliferative activity
compared with compound 4a. When introducing a methyl
group on the double bond of the B ring, we obtained compound
12, which caused a nearly 20.4–110.9 fold decrease in anti-
proliferative activity.
Besides the C]N double bond in the B ring, compound 16
has a very similar structure to compound 4a. Unfortunately,
surprisingly 16 exhibited poor activity toward all the six cancer
cell lines (IC50 ¼ 1.24 mM for Hela, and 7.98–12.1 mM for other
cancer cell lines). These results showed that the C]C bond in B
ring was very crucial for the anti-proliferative activity in this
series compounds.
Scheme 4 Synthesis of compounds 23a–d. Reagents and conditions:
(a) BrCH2COCH2Br, CH2Cl2, rt, 1 h; (b) 5 mol% Pd(OAc)2, 4 eq. TBHP,
26 mol% TFA, PEG-2000, water, rt, 12 h; (c) NH3–H2O, EtOH, rt, 6 h,
reflux, 2 h.
Compounds 20a–e, with the characteristic of A–B ring
condensation, were also evaluated for their anti-proliferative
activities (Table 2). Compound 20a, which has a similar struc-
ture to 4a (besides B–C ring condensation in 4a), provided
activities that range from 1.05 mM to 5.82 mM, which were much
lower than for its isomer 4a. Compound 20b, which has
a hydroxyl in the R1 position, gave the best result with IC50
values in a range from 0.204–0.968 mM. Other compounds,
without a methoxyl group in the C ring (20d), or bearing
a methoxyl group in the R3 position (20e), or two methoxyl
groups in the R2 and R3 position (20c), exhibited poor anti-
proliferative activities. When compared to compound 4a,
compound 20a showed a 14.2–87.4 fold decrease towards the six
cancer cell lines, which demonstrated that A–B ring condensa-
tion is not favorable for the hexa-cyclic compounds.
with IC50 values in the micromolar to sub-micromolar range. It
can be seen from Table 1, that compound 4a, with three
methoxyl groups on the A ring and one methoxyl on the C ring at
the R1 position, exhibited the best anti-proliferative activity
towards all six cancer cell lines (IC50 ¼ 0.046 mM for Hela, 0.074
mM for Bel-7402, 0.072 mM for A549, 0.08 mM for MCF-7, 0.074
mM for MGC-803 and 0.069 mM for SW480). Compounds 4b and
4c, with a methoxyl group at the R2 position, or two methoxyl
groups at both the R1 and R2 position, gave relatively poor
activities (4b provided IC50 values in the range 0.33 to 1.20 mM,
4c gave IC50 values in the range 0.80 to 4.45 mM). The replace-
ment of the methoxyl group with a hydroxyl group at the R2
position led to an almost complete loss of activity (4d, gave an
IC50 value of more than 10 mM). An N-substituent in the B ring
Table 1 Anti-proliferative activities of compound 4a–d, 5a, 5b, 9, 12, and 16 against different human cancer cell lines
IC50,a mean ꢀ SE (mM)
Compd.
Hela
Bel-7402
A549
MCF-7
MGC-803
SW480
4a
4b
4c
4d
5
0.046 ꢀ 0.01
0.47 ꢀ 0.11
0.99 ꢀ 0.17
>10
0.67 ꢀ 0.08
0.34 ꢀ 0.006
0.94 ꢀ 0.20
1.24 ꢀ 0.05
0.003 ꢀ 0.002
0.074 ꢀ 0.06
0.79 ꢀ 0.02
1.48 ꢀ 0.67
>10
0.59 ꢀ 0.05
1.95 ꢀ 0.52
1.30 ꢀ 0.30
8.22 ꢀ 0.03
0.012 ꢀ 0.001
0.072 ꢀ 0.009
0.817 ꢀ 0.04
0.80 ꢀ 0.14
>10
0.52 ꢀ 0.04
2.56 ꢀ 0.08
1.72 ꢀ 0.44
8.74 ꢀ 0.31
0.005 ꢀ 0.001
0.08 ꢀ 0.009
0.637 ꢀ 0.287
1.67 ꢀ 0.83
>10
0.74 ꢀ 0.02
3.395 ꢀ 0.658
6.16 ꢀ 2.05
7.98 ꢀ 0.06
0.013 ꢀ 0.001
0.074 ꢀ 0.19
0.33 ꢀ 0.06
4.45 ꢀ 0.02
>10
0.72 ꢀ 0.25
5.20 ꢀ 0.91
1.92 ꢀ 0.32
11.0 ꢀ 0.14
0.009 ꢀ 0.002
0.069 ꢀ 0.03
1.20 ꢀ 0.25
3.69 ꢀ 0.25
>10
1.14 ꢀ 0.01
7.60 ꢀ 0.85
7.65 ꢀ 0.19
12.1 ꢀ 0.15
0.0017 ꢀ 0.002
9
12
16
CA-4
a
Data are presented as the mean ꢀ SE from the dose–response curves of at least three independent experiments.
This journal is © The Royal Society of Chemistry 2015
RSC Adv., 2015, 5, 98527–98537 | 98529

View Article Online
RSC Advances
Paper
Table 2 Anti-proliferative activities of compound 20a–e against different human cancer cell lines
IC50,a mean ꢀ SE (mM)
Compd.
Hela
Bel-7402
A549
MCF-7
MGC-803
SW480
20a
20b
20c
20d
20e
CA-4
4.02 ꢀ 0.46
1.05 ꢀ 0.08
0.789 ꢀ 0.13
20.6 ꢀ 0.26
28.6 ꢀ 0.17
10.6 ꢀ 0.31
0.012 ꢀ 0.001
4.82 ꢀ 0.09
0.307 ꢀ 0.06
18.6 ꢀ 0.69
37.6 ꢀ 0.84
15.6 ꢀ 0.59
0.003 ꢀ 0.002
3.69 ꢀ 0.19
0.487 ꢀ 0.35
19.6 ꢀ 0.28
31.2 ꢀ 0.14
11.2 ꢀ 0.19
0.013 ꢀ 0.001
5.82 ꢀ 0.13
0.968 ꢀ 0.29
18.4 ꢀ 0.17
30.6 ꢀ 0.15
14.6 ꢀ 0.21
0.009 ꢀ 0.002
1.16 ꢀ 0.03
0.892 ꢀ 0.06
18.6 ꢀ 0.24
0.204 ꢀ 0.03
15.8 ꢀ 0.23
25.1 ꢀ 0.56
10.4 ꢀ 0.25
0.005 ꢀ 0.001
24.9 ꢀ 0.07
9.84 ꢀ 0.28
0.0017 ꢀ 0.002
a
Data are presented as the mean ꢀ SE from the dose–response curves of at least three independent experiments.
To further examine the relationship between the structure modication,17,18 with CA-4 as the reference compound. The
and the anti-proliferative activity of cyclic-CA-4 analogues, 23a– results summarized in Table 4 show that compound 4a, which
e (A–B ring condensation) were synthesized and tested against gave the best anti-proliferative activity in vitro, also displayed
the different cancer cell lines. As shown in Table 3, compounds the most potent inhibitory activity on tubulin polymerisation,
23a–e exhibited good activities against the six cancer cell lines. with an IC50 of 2.56 mM. Compounds 20b, 23a, and 23b, with
When compared to 20a, compound 23a showed an almost 16.8- lower anti-proliferative activity, provided slightly reduced
fold increase of antiproliferative activity toward Hela cell lines, inhibitory activity toward tubulin polymerisation accordingly.
which indicated that the hepta-ring was more favourable for the The correlation between the anti-proliferative activity and the
activity than the hexa-ring.
tubulin polymerisation inhibitory activity indicated that the
In vitro inhibition of tubulin polymerisation. To study synthesized compounds indeed target the tubulin–microtubule
whether these compounds target the tubulin–microtubule system, which was consistent with most antimitotic agents.19
system, compounds 4a, 20b, 23a, and 23b, which showed rela-
It is well known that tubulin-binding agents can be broadly
tively better anti-proliferative activity in vitro, were chosen to classied into two types: stabilizing agents (such as taxanes)
investigate their ability to block microtubule assembly by the and destabilizing agents (such as colchicine and CA-4).14,20 To
method originally described by Bonne, D. et al. with moderate conrm which type compound 4a belongs to, we carried out
Table 3 Anti-proliferative activities of compound 23a–e against different human cancer cell lines
IC50,a mean ꢀ SE (mM)
Hela
Compd.
Bel-7402
A549
MCF-7
MGC-803
SW480
23a
23b
23c
23d
CA-4
0.24 ꢀ 0.17
0.282 ꢀ 0.06
0.184 ꢀ 0.01
5.69 ꢀ 0.05
4.16 ꢀ 0.31
0.012 ꢀ 0.001
0.306 ꢀ 0.03
0.175 ꢀ 0.32
4.304 ꢀ 0.71
7.62 ꢀ 0.59
0.261 ꢀ 0.21
0.31 ꢀ 0.15
4.05 ꢀ 0.07
3.27 ꢀ 0.19
0.013 ꢀ 0.001
0.381 ꢀ 0.28
0.58 ꢀ 0.18
5.94 ꢀ 0.13
4.62 ꢀ 0.21
0.009 ꢀ 0.002
0.493 ꢀ 0.29
0.384 ꢀ 0.01
3.87 ꢀ 0.21
0.21 ꢀ 0.27
3.95 ꢀ 0.22
5.64 ꢀ 0.25
0.005 ꢀ 0.001
3.84 ꢀ 0.28
0.003 ꢀ 0.002
0.0017 ꢀ 0.002
a
Data are presented as the mean ꢀ SE from the dose–response curves of at least three independent experiments.
98530 | RSC Adv., 2015, 5, 98527–98537
This journal is © The Royal Society of Chemistry 2015

View Article Online
Paper
RSC Advances
Table 4 The IC50 values of selected compounds on inhibition of tubulin polymerisation
Compd.
4a
20b
23a
23b
CA-4
IC50,ameanꢀ SE (mM)
2.56 ꢀ 0.15
4.15 ꢀ 0.21
3.34 ꢀ 0.35
3.01 ꢀ 0.03
1.20 ꢀ 0.04
a
Data are presented as the mean ꢀ SE from the dose–response curves of at least three independent experiments.
microtubule dynamics assays with 0.938, 1.875, 3.75, 7.5, and 15 colchicine binding site of a,b-tubulin. The MOE package was
mM of compound 4a, respectively. The representative raw data employed to investigate the docking between compound 4a and
for the polymerisation assay of compound 4a are shown in a,b-tubulin. As shown in Fig. 4a, when we re-docked colchicine
Fig. 2. As shown in Fig. 2, compound 4a inhibited tubulin back to the colchicine binding site, the similar conformation of
polymerisation in a concentration dependent manner, and thus the X-ray structure of the colchicine (PDB code: 1SA0) veries
is a microtubule destabilizing agent.
the reliability of our docking method. Then we employed this
Disruption of microtubule dynamics. Given the important docking model for the docking of compound 4a. The docking
function of the tubulin–microtubule system in maintaining the studies in Fig. 4B showed that 4a occupied the colchicine
cellular morphology,1,12 immunouorescence assays were per- binding site of tubulin, as evidenced by two hydrogen bonds
formed to reveal whether compound 4a could affect microtu- and two s–p conjugate actions were formed between 4a and the
bule dynamics in living cells. Confocal microscopy studies residues of tubulin. The trimethoxyphenyl moiety of compound
clearly showed heavy disruption of the microtubule system of 4a was positioned in the binding cavity buried in the b-subunit
Hela cells aer treatment with 4a. As shown in Fig. 3, in the and formed a hydrogen bond with the thiol group of Cysb241,
vehicle-treated group, microtubules were observed under the and two key amino acids of b-tubulin (Leub248 and Leub255)
normal state, as evidenced by the regular assembly, slim, formed hydrophobic interactions (s–p conjugate) with the tri-
brous microtubules (green) wrapped around the cell nucleus methoxyphenyl moiety. Overall, the in silico results correlate
(blue). While as the concentration of compound 4a increased, well with the tubulin polymerisation inhibition activity.
the network of microtubules became disorganized and nuclear
chromatin condensation appeared at a concentration of 100
nM. Furthermore, when the concentration increased to 200 nM,
the spindle microtubules strikingly shrink around the centre of
the cells, and the tubulin–microtubule system was heavily dis-
rupted. These morphology changes of the microtubules implied
that compound 4a could easily interfere with the microtubule
dynamics eventually leading to cell death.
Molecular docking study of 4a. As can be seen from the
above results 4a effectively inhibited the tubulin polymerisation
in vitro and dramatically disrupted intracellular tubulin–
microtubule dynamics. Then we performed a molecular dock-
ing study to elucidate the potential pose of compound 4a to the
Fig. 3 Effects of 4a on the cellular microtubule dynamics by immu-
nofluorescence. Hela cells were incubated with 0.1% DMSO or 50,
Fig. 2 The effect of compound 4a on microtubule dynamics. Tubulin 100, and 200 nM compound 4a at 37 ꢁC for 24 h, and then direct
was mixed with the compound at different concentrations in general microscopy detection of the fixed and stained cells was performed.
tubulin buffer containing 1 mM GTP and 20% glycerol. Microtubule Images were taken using an LSM 570 laser confocal microscope (Carl
polymerisation was monitored using the kinetic model of FlexStation 3 Zeiss, Germany). The experiments were performed three times, and
multimode reader (excitation, 360 nm; emission, 450 nm).
the results of representative experiments are shown.
This journal is © The Royal Society of Chemistry 2015
RSC Adv., 2015, 5, 98527–98537 | 98531

View Article Online
RSC Advances
Paper
Fig. 4 Proposed binding mode for compounds colchicine (A) and 4a
(B), obtained by molecular docking simulation into the colchicine
binding pocket of b-tubulin.
Fig. 5 Cell cycle distribution of Hela cells with compound 4a. Cells
were treated with compound 4a (50, 100, and 200 nM) or DMSO
(0.01%) for 48 h. The DNA contents of each cell phase were analyzed
by flow cytometry (A) quantitative analysis of the percentage of cells in
each cell cycle phase was analysed by EXPO32 ADC analysis software
(B). The experiments were performed three times, and the results of
representative experiments are shown.
Cell cycle arrest at the G2/M phase. It is well known that most
tubulin polymerisation inhibition agents induce cytoskeleton
disruption and then block the cell cycle in the G2/M phase.21 The
excellent potency of compound 4a on tubulin polymerisation
inhibition prompted us to evaluate its effects on the cell cycle.
Aer Hela cells were treated with compound 4a at different
concentrations (50, 100, and 200 nM) and DMSO (0.01%) as the
reference for 48 h, an analysis of the results of harvested cells by
ow cytometry indicated that 4a resulted in a signicant cell-
cycle arrest at the G2/M phase, and showed a dose-dependent
manner (Fig. 5). As shown in Fig. 5, the control group showed
18.3% of the cells at the G2/M phase. Whereas, when the cells
were treated with 4a at 50, 100, and 200 nM, the cells at the G2/M
phase correspondingly increased to 23.5%, 40.7% and 75.9%,
respectively. Furthermore, as the concentration of compound
4a was increased, a typical apoptosis sub-peak was also obvi-
ously observed in the DNA histogram (Fig. 5A), which revealed
that compound 4a might have the ability to induce cell
apoptosis. Therefore, we concluded that compound 4a could
arrest cell-cycle progression at the mitotic phase, which is in
agreement with most of the antimitotic agents.
ow cytometry. The data summarized in Fig. 6 show that 4a
induced cell apoptosis in a dose-dependent manner. The total
percentage of early and late apoptotic cells were 2.21%, 23.40%,
and 40.90% in the cases of 50, 100, and 200 nM 4a treated Hela
cells, respectively. In contrast, the apoptotic cells were almost
negligible in the control group, which revealed that compound
4a could exclusively trigger apoptotic cell death of the Hela cells.
These results further demonstrated that the anti-proliferative
activity of compound 4a was due to its interaction with
tubulin, resulting in a prolonged cell cycle arrest that eventually
leads to cell apoptosis.
Metabolic stability. As well discussed in previous literature,
the cis double bond in CA-4 is prone to isomerize to the more
thermodynamically stable, but less bioactive trans isomers.8,9 To
evaluate the stability of the synthesized cyclic CA-4 analogues,
the optimal compound 4a and the reference compound CA-4
were investigated for their metabolic stability in rat liver
microsomes. When incubated in the presence of rat liver
microsomes (Table 5), CA-4 displayed a relatively fast metabolic
prole, with substrate remaining ranging from 68.1% to 20.7%
aer 30 to 120 min. Under the same conditions, 4a exhibited
Compound 4a induces apoptosis of Hela cells. As an obvious
typical apoptosis sub-peak appeared in the cell cycle analysis
(Fig. 5A), we hypothesized that compound 4a might induce cell
apoptosis. To conrm this property, compound 4a treated Hela
cells were stained with Annexin V-FITC and PI, and analyzed by
98532 | RSC Adv., 2015, 5, 98527–98537
This journal is © The Royal Society of Chemistry 2015

View Article Online
Paper
RSC Advances
BioSpin GmbH spectrometer at 400 and 101 MHz, respectively.
High-resolution mass spectra (HR-MS) were recorded using
a Shimadzu LCMS-ITTOF mass spectrometer. The melting
points were determined using an SRS-OptiMelt automated
melting point instrument. The reactions were monitored by
thin layer chromatography (TLC) on glass-packed percolated
silica gel plates and visualized in an iodine chamber or with
a UV lamp. Flash column chromatography was performed using
silica gel (200–300 mesh) purchased from Qingdao Haiyang
Chemical Co. Ltd. The purity ($95%) of the synthesised
compounds was determined by high-performance liquid chro-
matography (HPLC) with a TC-C18 column (4.6 ꢂ 250 mm, 5
mm), an acetonitrile–water or methanol–water mobile phase,
and a ow rate of 1.00 mL minꢃ1
.
General procedure for the preparation of substituted acet-
anilide substrates (2a–d, 7, 10, 21).22,23 All the substituted ani-
lides were prepared from the corresponding anilines and acyl
chlorides in CH2Cl2 according to the literature method without
modications.
General procedure for the coupling reactions.15,16,24,25
Substituted anilide, 3,4,5-trimethoxybenzaldehyde (2.0 equiv.),
Pd(OAc)2 (0.05 equiv.) and PEG-2000 (0.2 equiv.) were loaded
into a Schlenk tube, and then water was added to the mixture.
Aer stirring for 2 min, TBHP (4.0 equiv.) and triuoroacetic
acid (0.26 equiv.) were added to the mixture at room tempera-
ture. The mixture was stirred at 40 ꢁC for 12 hours and then
cooled to room temperature. A saturated solution of K2CO3 was
added and the mixture was extracted with ethyl acetate. The
combined organic phase was washed (brine) and concentrated
under reduced pressure to give the crude products which were
puried by ash column chromatography on silica gel (200–300
mesh) to afford the desired product.
General procedure for the synthesis of 4a–d, 9, 12, 20a–e. To
a stirred solution of intermediates 3a–d, 8, 11 and 19a–e in
anhydrous DMF, CH3ONa (10.0 equiv.) was added. Aer
reuxing for 12 h, the solvents were evaporated under vacuum,
and the residue was extracted with CH2Cl2, washed (brine),
dried (Na2SO4), ltered and concentrated under vacuum to
provide the crude product, which was puried by column
chromatography to afford a white solid.
Fig. 6 Compound 4a induced apoptosis of Hela cells. Cells were
treated with compound 4a (50, 100, and 200 nM) or DMSO (0.01%) for
48 h. Then the cells were collected and analyzed by flow cytometry
after being stained by Annexin V-FITC and PI (A). The percentages of
cells in each stage of cell apoptosis were quantitated by flow cytom-
etry. (A1: upper left quadrant) necrotic cells; (A2: upper right quadrant)
late apoptotic cells; (A3: bottom left quadrant) live cells; and (A4:
bottom right quadrant) early apoptotic cells (B). The experiments were
performed three times, and representative experiments are shown.
Table 5 The metabolic stability of 4a and CA-4
Substrate remaining
Time/min
7-Methoxy-4-(3,4,5-trimethoxyphenyl)quinolin-2(1H)-one (4a).
Yellow solid; yield 52.2%; mp: 261.9–262.6 ꢁC; 1H NMR (400
MHz, CDCl3) d 11.99 (s, 1H), 7.54 (d, J ¼ 9.0 Hz, 1H), 6.89 (d, J ¼
2.4 Hz, 1H), 6.79 (dd, J ¼ 9.0, 2.4 Hz, 1H), 6.67 (s, 2H), 6.54 (s,
1H), 3.94 (s, 3H), 3.93 (s, 3H), 3.88 (d, J ¼ 7.0 Hz, 6H).13C NMR
(101 MHz, CDCl3) d 164.61, 161.93, 153.40, 153.31, 140.87,
138.44, 132.93, 128.08, 117.38, 113.73, 112.35, 106.13, 98.69,
60.97, 56.28, 55.73. HRMS (ESI) (m/z) [M + H]+ calcd for
30
60
90
120
4a
CA-4
69.6%
68.1%
49.2%
28.6%
42.9%
23.1%
36.5%
20.7%
69.6–36.5%, which indicated that 4a has relatively better
metabolic stability than CA-4.
C
19H19NO5, 342.1336; found, 342.1340. Purity: 99.7% (by
HPLC).
6-Methoxy-4-(3,4,5-trimethoxyphenyl)quinolin-2(1H)-one (4b).
Experimental section
Chemistry
White solid; yield 61%; mp: 283.0–283.5 ꢁC; 1H NMR (400 MHz,
CDCl3) d 12.64 (s, 1H), 7.48 (d, J ¼ 8.9 Hz, 1H), 7.20 (dd, J ¼ 8.9,
General methods. All reagents used in the synthesis were 2.6 Hz, 1H), 7.11 (d, J ¼ 2.6 Hz, 1H), 6.72 (d, J ¼ 6.8 Hz, 3H), 3.95
obtained commercially and used without further purication, (d, J ¼ 8.3 Hz, 3H), 3.90 (s, 6H), 3.76 (s, 3H).13C NMR (101 MHz,
unless otherwise specied. The H NMR and 13C NMR spectra CDCl3) d 163.83, 155.17, 153.40, 152.78, 138.44, 133.63, 132.67,
1
were recorded using TMS as the internal standard on a Bruker 120.92, 120.11, 120.02, 118.07, 108.26, 106.03, 61.00, 56.28,
This journal is © The Royal Society of Chemistry 2015
RSC Adv., 2015, 5, 98527–98537 | 98533

View Article Online
RSC Advances
Paper
55.67. HRMS (ESI) (m/z) [M + H]+ calcd for C19H19NO5, 342.1336; calcd for C19H19NO6, 358.1285; found, 358.1277. Purity: 100%
found, 342.1327. Purity: 99.9% (by HPLC).
(by HPLC).
6,7-Dimethoxy-4-(3,4,5-trimethoxyphenyl)quinolin-2(1H)-one (4c).
4-(3,4-Dimethoxyphenyl)-6,7,8-trimethoxyquinolin-2(1H)-one (20c).
White solid; yield 58%; mp: 226.3–227.0 ꢁC; 1H NMR (400 MHz, Yellow solid; yield 61%; mp: 221.0–222.4 ꢁC; 1H NMR (400 MHz,
CDCl3) d 12.77 (s, 1H), 7.06 (s, 1H), 6.98 (s, 1H), 6.71 (s, 2H), 6.60 CDCl3) d 9.14 (s, 1H), 7.05–7.00 (m, 2H), 6.97 (d, J ¼ 1.7 Hz, 1H),
(s, 1H), 4.04 (s, 3H), 3.96 (s, 3H), 3.90 (s, 6H), 3.79 (s, 3H).13
C
6.82 (s, 1H), 6.56 (s, 1H), 4.05 (s, 3H), 3.98 (d, J ¼ 2.1 Hz, 6H),
NMR (101 MHz, CDCl3) d 164.32, 153.39, 152.86, 152.79, 145.80, 3.91 (s, 3H), 3.75 (s, 3H).13C NMR (101 MHz, CDCl3) d 161.87,
138.40, 135.11, 133.08, 117.62, 112.76, 107.00, 105.99, 98.67, 152.06, 149.55, 148.98, 143.91, 139.28, 129.77, 127.87, 121.21,
60.99, 56.40, 56.28, 56.15. HRMS (ESI) (m/z) [M + H]+ calcd for 120.63, 115.06, 111.87, 111.24, 103.23, 61.43, 61.20, 56.17,
C20H21NO6, 372.1442; found, 372.1439. Purity: 99.8% (by 56.03, 55.98. HRMS (ESI) (m/z) [M + H]+ calcd for C20H21NO6,
HPLC).
6-Hydroxy-7-methoxy-4-(3,4,5-trimethoxyphenyl)quinolin-2(1H)-
372.1442; found, 372.1434. Purity: 100% (by HPLC).
6,7,8-Trimethoxy-4-phenylquinolin-2(1H)-one (20d). Yellow
1
1
one (4d). White solid; yield 51%; mp: 184.5–187.6 C; H NMR solid; yield 68%; mp: 214.0–215.3 ꢁC; H NMR (400 MHz, CDCl3)
(400 MHz, DMSO) d 11.72 (s, 1H), 6.98 (s, 1H), 6.94 (s, 1H), 6.73 d 9.15 (s, 1H), 7.51 (q, J ¼ 5.2 Hz, 3H), 7.46–7.42 (m, 2H), 6.72 (s,
(s, 2H), 6.29 (s, 1H), 3.83 (s, 9H), 3.74 (s, 3H).13C NMR (101 MHz, 1H), 6.55 (s, 1H), 4.05 (s, 3H), 3.97 (s, 3H), 3.73 (s, 3H).13C NMR
DMSO) d 161.41, 153.38, 152.18, 152.06, 143.43, 138.14, 134.11, (101 MHz, CDCl3) d 161.95, 152.27, 149.01, 144.02, 139.35,
133.10, 117.07, 112.93, 110.42, 106.56, 98.80, 60.55, 56.52, 137.25, 128.83, 128.67, 128.58, 127.95, 120.92, 114.98, 103.25,
56.14. HRMS (ESI) (m/z) [M + H]+ calcd for C19H19NO6, 358.1285; 61.45, 61.20, 56.16. HRMS (ESI) (m/z) [M + H]+ calcd for
ꢁ
found, 358.1279. Purity: 97.3% (by HPLC).
4-(3,4,5-Trimethoxyphenyl)benzo[h]quinolin-2(1H)-one
White solid; yield 56%; mp: 262.1–263.5 ꢁC; 1H NMR (400 MHz,
C
18H17NO4, 312.1230; found, 312.1238. Purity: 99.3% (by
(9). HPLC).
6,7,8-Trimethoxy-4-(3-methoxyphenyl)quinolin-2(1H)-one (20e).
CDCl3) d 12.01 (s, 1H), 8.81 (d, J ¼ 8.5 Hz, 1H), 7.92 (d, J ¼ 7.5 Yellow solid; yield 60%; mp: 273.0–273.5 ꢁC; 1H NMR (400 MHz,
Hz, 1H), 7.79–7.74 (m, 1H), 7.69 (t, J ¼ 7.2 Hz, 1H), 7.64 (d, J ¼ CDCl3) d 9.10 (s, 1H), 7.39 (d, J ¼ 8.6 Hz, 2H), 7.04 (d, J ¼ 8.6 Hz,
8.9 Hz, 1H), 7.58 (d, J ¼ 8.9 Hz, 1H), 6.87 (s, 1H), 6.73 (s, 2H), 2H), 6.78 (s, 1H), 6.53 (s, 1H), 4.05 (s, 3H), 3.97 (s, 3H), 3.90 (s,
3.98 (s, 3H), 3.92 (s, 6H).13C NMR (101 MHz, CDCl3) d 163.68, 3H), 3.75 (s, 3H).13C NMR (101 MHz, CDCl3) d 161.85, 160.10,
154.17, 153.37, 138.38, 136.08, 134.10, 133.11, 128.58, 128.30, 152.04, 148.97, 143.89, 139.21, 129.91, 129.51, 127.82, 120.60,
127.19, 123.35, 122.91, 122.29, 121.81, 120.53, 115.67, 106.24, 115.15, 114.13, 103.33, 61.41, 61.20, 56.19, 55.37. HRMS (ESI)
61.04, 56.33. HRMS (ESI) (m/z) [M + H]+ calcd for C22H19NO4, (m/z) [M + H]+ calcd for C19H19NO5, 342.1336; found, 342.1342.
362.1387; found, 362.1371. Purity: 99.5% (by HPLC).
Purity: 100% (by HPLC).
7-Methoxy-3-methyl-4-(3,4,5-trimethoxyphenyl) quinolin-2(1H)-
7-Methoxy-1-methyl-4-(3,4,5-trimethoxyphenyl)quinolin-2(1H)-
1
ꢁ
one (12). White solid; yield 48%; mp: 294.2–294.8 C; H NMR one (5). Compound 5 was synthesised following a previously
(400 MHz, CDCl3) d 11.78 (s, 1H), 7.07 (d, J ¼ 9.0 Hz, 1H), 6.87 (d, reported procedure.13 Briey, to a solution of 4a in anhydrous
J ¼ 2.4 Hz, 1H), 6.71 (dd, J ¼ 9.0, 2.4 Hz, 1H), 6.44 (s, 2H), 3.95 (s, tetrahydrofuran, sodium hydride (2.0 equiv.) was added. Aer
3H), 3.90 (s, 3H), 3.86 (s, 6H), 2.08 (s, 3H).13C NMR (101 MHz, stirring for 30 min, CH3I (1.5 equiv.) was added and then the
CDCl3) d 164.78, 160.80, 153.49, 148.92, 138.60, 137.42, 132.70, mixture was stirred at room temperature for 12 h. The solvents
128.14, 124.16, 115.28, 111.58, 105.75, 98.21, 61.00, 56.22, were evaporated under vacuum, and the residue was extracted
55.57, 14.01. HRMS (ESI) (m/z) [M + H]+ calcd for C20H21NO5, with CH2Cl2, washed with brine, dried over Na2SO4, ltered and
356.1492; found, 356.1487. Purity: 99.1% (by HPLC).
concentrated under vacuum to give the crude product, which
6,7,8-Trimethoxy-4-(4-methoxyphenyl)quinolin-2(1H)-one (20a). was puried by column chromatography to give a yellow solid.
Yellow solid; yield 61%; mp: 228.4.0–229.7 ꢁC; 1H NMR (400 Yield 48%; mp: 179.1–180.2 ꢁC; 1H NMR (400 MHz, CDCl3)
MHz, CDCl3) d 9.16 (s, 1H), 7.04 (d, J ¼ 1.6 Hz, 1H), 6.98 (d, J ¼ d 7.57 (d, J ¼ 8.9 Hz, 1H), 6.87 (d, J ¼ 2.3 Hz, 1H), 6.79 (dd, J ¼
8.2 Hz, 1H), 6.94 (dd, J ¼ 8.2, 1.7 Hz, 1H), 6.82 (s, 1H), 6.53 (s, 8.9, 2.4 Hz, 1H), 6.63 (s, 2H), 6.55 (s, 1H), 3.93 (d, J ¼ 1.8 Hz, 6H),
1H), 5.96 (s, 1H), 4.04 (s, 3H), 3.98 (d, J ¼ 4.1 Hz, 6H), 3.76 (s, 3.87 (s, 6H), 3.74 (s, 3H).13C NMR (101 MHz, CDCl3) d 162.39,
3H).13C NMR (101 MHz, CDCl3) d 161.77, 159.77, 152.19, 149.06, 161.77, 153.26, 150.81, 142.04, 138.30, 132.86, 129.14, 117.97,
143.99, 139.22, 138.54, 129.74, 127.81, 120.94, 120.80, 114.90, 114.52, 109.29, 106.18, 99.00, 60.95, 56.25, 55.60, 29.53. HRMS
114.52, 114.15, 103.28, 61.44, 61.21, 56.22, 55.38. HRMS (ESI) (ESI) (m/z) [M + H]+ calcd for C20H21NO5, 356.1492; found,
(m/z) [M + H]+ calcd for C19H19NO5, 342.1336; found, 342.1321. 356.1497. Purity: 99.1% (by HPLC).
Purity: 96.4% (by HPLC).
(2-Amino-4-methoxyphenyl)(3,4,5-trimethoxyphenyl)methanone
4-(3-Hydroxy-4-methoxyphenyl)-6,7,8-trimethoxyquinolin-2(1H)- (15). To a stirred suspension of benzyl 5-methoxy-2-(3,4,5-
one (20b). White solid; yield 51%; mp: 235.3.0–236.1 ꢁC; 1H trimethoxybenzoyl)phenylcarbamate (14) in THF, 10% Pd/C
NMR (400 MHz, CDCl3) d 9.16 (s, 1H), 7.42 (t, J ¼ 7.9 Hz, 1H), was added and the mixture was hydrogenated in 30 Mpa of
7.05–7.00 (m, 2H), 6.99–6.96 (m, 1H), 6.76 (s, 1H), 6.56 (s, 1H), hydrogen pressure at room temperature for 4 h. The mixture
4.05 (s, 3H), 3.97 (s, 3H), 3.86 (s, 3H), 3.74 (s, 3H).13C NMR (101 was ltered and dried under vacuum to give the crude product,
MHz, CDCl3) d 161.72, 152.14, 149.04, 147.22, 145.79, 143.86, which was puried by column chromatography (petroleum
139.06, 130.37, 127.62, 120.55, 120.51, 115.09, 115.04, 110.81, ether–ethyl acetate) to give a yellow solid.
103.44, 61.39, 61.20, 56.25, 56.04. HRMS (ESI) (m/z) [M + H]+
7-Methoxy-4-(3,4,5-trimethoxyphenyl)quinazolin-2(1H)-one (16).
To a stirred suspension of compound 15 in acetic acid,
98534 | RSC Adv., 2015, 5, 98527–98537
This journal is © The Royal Society of Chemistry 2015

View Article Online
Paper
RSC Advances
carbamide (2.0 equiv.) was added and then reuxed at 110 ꢁC for 56.96, 56.25, 56.13. HRMS (ESI) (m/z) [M + H]+ calcd for
20 h. Aer the solvents were evaporated under vacuum, the 20H19N2O5F3, 425.1319; found, 425.1301. Purity: 100% (by
C
mixture was extracted with ethyl acetate, washed with brine, HPLC).
dried over Na2SO4, ltered and concentrated under vacuum to
give the crude product, which was puried by column chro-
Biology
matography to give a white solid. Yield 56%; mp: 254.9–255.8
ꢁC; 1H NMR (400 MHz, CDCl3) d 13.39 (s, 1H), 7.82 (d, J ¼ 9.1 Hz,
Cell lines and culture. The human cancer cell lines (Hela,
1H), 7.00 (s, 2H), 6.96 (d, J ¼ 2.3 Hz, 1H), 6.82 (dd, J ¼ 9.1, 2.4 A549, Bel-7402, MCF-7, SW-480, MGC-803) used in this study
Hz, 1H), 3.96 (s, 3H), 3.95 (s, 3H), 3.93 (s, 6H).13C NMR (101 were cultivated in DMEM containing 10% (v/v) heat-inactivated
MHz, CDCl3) d 175.12, 165.54, 158.89, 153.14, 146.31, 140.27, FBS, 100 units per mL penicillin, and 100 mg mLꢃ1 strepto-
ꢁ
132.16, 130.36, 114.32, 110.05, 107.20, 97.29, 61.00, 56.39, mycin. The cells were incubated at 37 C under a 5% CO2 and
56.23. HRMS (ESI) (m/z) [M + H]+ calcd for C18H18N2O5, 90% relative humidity (RH) atmosphere.
343.1288; found, 343.1294. Purity: 98.5% (by HPLC).
MTT assay. The anti-proliferative activity towards the human
General procedure for the synthesis of 23a–e. The interme- cancer cell lines and cell cytotoxicity towards human normal
diates 22a–e were dissolved in ethanol, 25% ammonia solution cells of the target compounds were examined by MTT assay.
was then added. Aer stirring at room temperature for 6 h, the Briey, when the cells were growing in the logarithmic phase,
reaction mixture was then reuxed for 2 h. The solvents were 5 ꢂ 103 cells per well cells were harvested and plated into the 96-
evaporated under vacuum to give the crude product, which was well plates for 24 h, and then the cells were exposed to different
puried by column chromatography (petroleum ether–ethyl concentrations of the test compounds for 48 h in three repli-
acetate, 1 : 1) to give the product.
cates. Aerward, 20 mL of 5 mg mLꢃ1 MTT (Sigma) was added
7,8,9-Trimethoxy-5-(4-methoxyphenyl)-1H-benzo[e][1,4]diazepin- and incubated for another 4 h. Then, the suspension was dis-
2(3H)-one (23a). Yellow solid; yield: 78%; mp: 217.8–218.9 ꢁC; 1H carded and 150 mL of DMSO was added to each well. The plates
NMR (400 MHz, CDCl3) d 8.39 (s, 1H), 6.97 (d, J ¼ 8.6 Hz, 2H), were shaken to dissolve the dark blue crystals (formazan) for 10
6.74 (d, J ¼ 8.6 Hz, 2H), 6.39 (s, 1H), 3.79 (d, J ¼ 1.4 Hz, 6H), 3.71 min, then the absorbance at 570 nm was measured using
(s, 6H), 3.41 (s, 2H).13C NMR (101 MHz, CDCl3) d 171.97, 157.98, a multifunction microplate reader (Moleculardevices, Flex
152.33, 149.39, 140.81, 134.32, 132.33, 129.66, 121.57, 113.85, Station 3). All experiments were repeated at least three times.
108.86, 61.17, 60.92, 56.03, 55.26, 44.93, 37.44. HRMS (ESI) (m/z) The IC50 values were calculated using Graph Pad Prism version
[M + H]+ calcd for C19H20N2O5, 357.1445; found, 357.1455. 5.0.
Purity: 99.0% (by HPLC).
Tubulin polymerisation assay in vitro. Tubulin polymerisa-
5-(3-Fluoro-4-methoxyphenyl)-7,8,9-trimethoxy-1H-benzo
[e] tion is followed by uorescence enhancement due to the
[1,4]diazepin-2(3H)-one (23b). White solid; yield: 75%; mp: incorporation of a uorescent reporter into the microtubules as
206.8–207.9 ꢁC; 1H NMR (400 MHz, CDCl3) d 7.83 (s, 1H), 7.40– polymerisation occurs. The tubulin polymerisation assay was
7.35 (m, 1H), 7.32 (d, J ¼ 9.1 Hz, 1H), 6.95 (t, J ¼ 8.5 Hz, 1H), 6.54 monitored by the increase in uorescence emission at 410 nm
(s, 1H), 4.28 (s, 2H), 3.99 (s, 6H), 3.94 (s, 3H), 3.74 (s, 3H).13
C
over a 60 min period at 37 ꢁC (excitation wavelength is 340 nm)
NMR (101 MHz, CDCl3) d 170.79, 168.36, 153.06, 150.61, 149.55, using a modication of the methods described by Bonne, D.
148.70, 144.15, 142.91, 132.28, 127.24, 126.14, 121.55, 117.35, et al.18 Puried brain tubulin polymerisation kit was purchased
112.41, 108.02, 61.32, 61.08, 56.91, 56.23. HRMS (ESI) (m/z) [M + from Cytoskeleton (BK110P, Denver, CO). The nal buffer
H]+ calcd for C19H19FN2O5, 375.1351; found, 375.1365. Purity: concentration for tubulin polymerisation contained 80.0 mM
99.4% (by HPLC).
piperazine-N,N0-bis(2-ethanesulfonic acid) sequisodium salt
7,8,9-Trimethoxy-5-(3-methoxyphenyl)-1H-benzo[e][1,4]diazepin- (pH 6.9), 2.0 mM MgCl2, 0.5 mM EGTA, 1 mM GTP, and 10.2%
2(3H)-one (23c). Yellow solid; yield: 78%; mp: 234.8–235.7 ꢁC; 1H glycerol. Firstly, 5 mL of tested compounds were added in
ꢁ
NMR (400 MHz, CDCl3) d 7.81 (s, 1H), 7.29 (t, J ¼ 5.1 Hz, 1H), different concentrations, and then were warmed to 37 C for 1
7.22–7.18 (m, 1H), 7.08–7.03 (m, 1H), 7.02–6.98 (m, 1H), 6.54 (s, min. The reaction was initiated by the addition of 55 mL of the
1H), 4.31 (s, 2H), 4.02–3.93 (m, 6H), 3.84 (d, J ¼ 6.4 Hz, 3H), 3.71 tubulin reaction mix.
(d, J ¼ 6.4 Hz, 3H).13C NMR (101 MHz, CDCl3) d 170.66, 169.90,
Immunouorescence microscopy. Hela cells were cultured
159.52, 148.62, 144.08, 142.70, 140.64, 129.05, 127.10, 122.48, in a confocal culture-dish at 3 ꢂ 104 cells per dish and cultured
121.92, 116.49, 114.51, 108.33, 61.31, 61.09, 56.99, 56.23, 55.42. for 24 h. The tested compound 4a was added at the indicated
HRMS (ESI) (m/z) [M + H]+ calcd for C19H20N2O5, 357.1445; concentrations for another 24 h. Then the cells were briey
found, 357.1435. Purity: 100% (by HPLC).
ꢁ
washed with PBS, and xed with 4% pre-warmed (37 C) para-
7,8,9-Trimethoxy-5-(4-methoxy-3-(triuoromethyl)phenyl)-1H- formaldehyde for 15 min. Then cells were permeabilized with
benzo[e][1,4]diazepin-2(3H)-one (23d). Yellow solid; yield: 69%; 0.5% Triton X-100 for 15 min and blocked for 30 min in 10%
1H NMR (400 MHz, CDCl3) d 8.09 (s, 1H), 7.87 (d, J ¼ 2.0 Hz, 1H), goat serum. FITC-conjugated mouse anti-tubulin antibody
7.73 (dd, J ¼ 8.7, 2.1 Hz, 1H), 7.01 (d, J ¼ 8.7 Hz, 1H), 6.50 (s, (Sigma) was added and incubated at 4 ꢁC overnight. Cells were
1H), 4.30 (s, 2H), 4.01 (t, J ¼ 5.1 Hz, 6H), 3.95 (d, J ¼ 6.7 Hz, 3H), washed with PBS three times and incubated with goat anti-
3.73 (d, J ¼ 6.7 Hz, 3H).13C NMR (101 MHz, CDCl3) d 170.66, mouse IgG/Alexa-Fluor 488 (Invitrogen, USA) for 1 h. The
168.25, 159.06, 148.83, 144.25, 142.94, 134.76, 131.26, 128.71, nuclei of cells were labelled with Hochest 33342 (Sigma) in
128.66, 127.33, 124.60, 121.34, 111.46, 107.82, 61.33, 61.11, darkness at room temperature for 30 min. Aer washing, 1 mL
This journal is © The Royal Society of Chemistry 2015
RSC Adv., 2015, 5, 98527–98537 | 98535

View Article Online
RSC Advances
Paper
PBS was added and the samples were immediately visualized on glucose 6-phosphate, 0.4 units per mL glucose 6-phosphate
a Zeiss LSM 570 laser scanning confocal microscope.
dehydrogenase, and 3.3 mM MgCl2. Then, 1 mg mLꢃ1 of rat liver
Molecular docking study. In our study, the X-ray structure of microsomes were added. The samples were shaken for 60 min
tubulin in complex with DAMA-colchicine (PDB code 1SA0)24 at 37 ꢁC in a water bath and the incubation was stopped by
was applied (from the RCSB Protein Data Bank (http:// adding 0.5 mL of the ice-cold acetonitrile. Aer centrifugation
www.rcsb.org/pdb)) as receptor model for the docking proce- at 12 500g for 10 min, the supernatant was then directly
dure using the MOE2012 soware. The colchicine binding site analyzed by HPLC. The control group was lacking the NADPH
on the a,b-tubulin was dened as the site of ligand binding over regenerative system or the microsomes.
a,b-tubulin.26,27 Structure les of the ligand were prepared for
molecular docking by dening the number of torsion angles
and addition of hydrogen atoms. To validate the use of the MOE
Conclusions
program, the docking studies were performed on the reference
compound colchicine. MOE successfully reproduced the
binding conformations reported in the X-ray structure with
In summary, on the basis of our previous work reporting ortho-
(3,4,5-trimethoxybenzoyl)-acetanilides
and
cyclic
CA-4
analogues as anti-tumour agents, in order to investigate the
detailed SARs, we designed and synthesized a series of cyclic
compounds as CA-4 analogues and evaluated their anti-
proliferative activity. Three aspects of the SARs which
included the effect of the B-ring (hexa-cyclic or hepta-cyclic), the
pattern of ring condensation (A–B ring condensation or B–C
ring condensation), and C]C or C]N bond in the B ring were
studied. In detail, for the hexa-cyclic compounds, A–B ring
condensation is not favourable over B–C ring condensation,
which was the same for the hepta-cyclic compounds. From the
in vitro cytotoxicity results, we summarized the following three
items that the hepta-cyclic structure in the B-ring, B–C ring
condensation and the C]C bond in the B ring were crucial for
the anti-proliferative activity.
The mechanism studies including cell cycle analysis, tubulin
polymerisation inhibition assay, immunouorescence assay,
molecular docking study, and apoptosis analysis further vali-
dated that compound 4a could effectively inhibit tubulin poly-
merisation, target the tubulin–microtubule system by binding
to the colchicine binding site, interfere with cell mitosis,
resulting in a prolonged G2/M cell cycle arrest and ultimately
lead to cell apoptosis of cancer cells. In conclusion, our work
revealed a new attractive scaffold for mitosis-targeting drug
˚
acceptable root-mean-square deviation (rmsd < 1 A) of the atom
coordinates. All structural images were prepared using PyMOL.
Cell cycle analysis. For ow cytometric analysis of DNA
content, Hela cells were plated in 6-well plates (3 ꢂ 105 cells per
well) and incubated in the presence or absence of compound 4a
at the indicated concentrations for 48 h. Then, the cells were
harvested and xed with ice-cold 70% ethanol overnight.
Ethanol was removed by centrifugation, and the cells were
washed with cold 10% phosphate buffer solution (PBS), then,
treated with RNAse A (100 mg mLꢃ1, Beyotime) at 37 C for 30
ꢁ
ꢁ
min and DNA staining solution PI (Sigma) at 4 C for 15 min.
The DNA contents of 10 000 events were measured by a ow
cytometer (Beckman Coulter, Epics XL) at 488 nm. The data
regarding the number of cells in different phases of the cell
cycle were analyzed by EXPO32 ADC analysis soware.
Apoptosis analysis. Hela cells were plated in 6-well plates (3
ꢂ 105 cells per well) and incubated in the presence or absence of
compound 4a at the indicated concentrations for 48 h to induce
cell apoptosis; 0.1% DMSO was used as a vehicle control. The
percentages of apoptotic cells were estimated by staining with
Annexin-V-FITC and PI (Annexin-V-FITC Apoptosis Detection
Kit, Beyotime) according to the manufacturer’s instructions
with moderate modication. Aer the cells were induced for 48
h, both treated and untreated cells were harvested and incu-
bated with 5 mL of Annexin-V-FITC in binding buffer (10 mM
HEPES, 140 mM NaCl, and 2.5 mM CaCl2 at pH 7.4) at room
temperature for 15 min. PI was then added dropwise to the
medium at cold temperature, and the mixture was incubated for
10 min. Almost 10 000 events were collected for each sample
and were analyzed by ow cytometry (Beckman Coulter, Epics
discovery. As
a promising new tubulin binding agent,
compound 4a is worthy of further in vivo antitumor evaluation
as a potential chemotherapeutic agent. Further in vitro and in
vivo study of compounds with B–C ring condensation, bearing
C]C and hepta-cyclic structure in B ring as anti-tumour agents
is also in progress.
XL). The percentage of cells undergoing apoptosis was calcu- Acknowledgements
lated using EXPO32 ADC analysis soware.
We thank the National Natural Science Foundation of China
Metabolic stability study. Rat liver microsomes were
prepared as previously described with slight modications.28,29
Fresh rat liver was homogenized in freshly prepared sucrose
solution (pH 7.4) consisting of 0.25 M sucrose, 10 mM Tris–HCl,
1 mmol Lꢃ1 EDTA, and centrifuged at 16 000g for 30 min at 4 ꢁC.
The supernatant was then centrifuged at 10 000g for 60 min at 4
ꢁC. The supernatant, which contained the soluble microsomes,
was carefully decanted and stored at ꢃ80 ꢁC until use.
Compounds 4a and CA-4 at a nal concentration of 50 mM were
incubated in the Tris–HCl buffer containing NADPH regenera-
tive system which consists of 1.3 mM b-NADPNa2, 3.3 mM
(No. 21302235, 20972198), Guangdong Natural Science Foun-
dation (2014A030313124) and the Fundamental Research
Funds for the Central Universities (15ykpy04) for nancial
support of this study.
Notes and references
1 F. Pellegrini and D. R. Budman, Cancer Invest., 2005, 23, 264–
273.
2 M. Kavallaris, Nat. Rev. Cancer, 2010, 10, 194–204.
98536 | RSC Adv., 2015, 5, 98527–98537
This journal is © The Royal Society of Chemistry 2015

View Article Online
Paper
RSC Advances
3 M. A. Jordan and L. Wilson, Nat. Rev. Cancer, 2004, 4, 253– 14 A. Brancale and R. Silvestri, Med. Res. Rev., 2007, 27, 209–
265. 238.
4 G. R. Pettit, G. M. Cragg, D. L. Herald, J. M. Schmidt and 15 Y. Wu, B. Li, F. Mao, X. Li and F. Y. Kwong, Org. Lett., 2011,
P. Lohavanijaya, Can. J. Chem., 1982, 60, 1374–1376. 13, 3258–3261.
5 G. J. S. Rustin, S. M. Galbraith, H. Anderson, M. Stratford, 16 J. Sheng, F. Mao, J. Yan, L. Huang and X. Li, RSC Adv., 2014,
L. K. Folkes, L. Sena, L. Gumbrell and P. M. Price, J. Clin.
Oncol., 2003, 21, 2815–2822.
6 G. J. Rustin, G. Shreeves, P. D. Nathan, A. Gaya,
4, 41510–41520.
17 J. Yan, Y. Pang, J. Sheng, Y. Wang, J. Chen, J. Hu, L. Huang
and X. Li, Biochem. Pharmacol., 2015, 97, 51–61.
´
T. S. Ganesan, D. Wang, J. Boxall, L. Poupard, 18 D. Bonne, C. Heusele, C. Simon and D. Pantaloni, J. Biol.
D. J. Chaplin, M. R. L. Stratford, J. Balkissoon and
M. Zweifel, Br. J. Cancer, 2010, 102, 1355–1360.
7 D. W. C. Siemann, D. J. Chaplin and P. A. Walicke, Expert
Opin. Invest. Drugs, 2009, 18, 189–197.
Chem., 1985, 260, 2819–2825.
19 J. F. A. Diaz and J. Manuel, Biochemistry, 1993, 32, 2747–
2755.
20 F. G. Mollinedo and C. Gajate, Apoptosis, 2003, 8, 413–450.
8 M. Cushman, D. Nagarathnam, D. Gopal, A. K. Chakraborti, 21 C. E. Walczak, Curr. Opin. Cell Biol., 2000, 12, 52–56.
C. M. Lin and E. Hamel, J. Med. Chem., 1991, 34, 2579–2588. 22 P. R. A. Clarke and L. A. Allan, Trends Cell Biol., 2009, 19, 89–
9 G. R. Pettit, B. E. Toki, D. L. Herald, M. R. Boyd, E. Hamel,
R. K. Pettit and J. C. Chapuis, J. Med. Chem., 1999, 42, 23 F. Zhou, X. L. Han and X. Y. Lu, Tetrahedron Lett., 2011, 52,
1459–1465. 4681–4685.
10 H. Chen, Y. Li, C. Sheng, Z. Lv, G. Dong, T. Wang, J. Liu, 24 V. R. Choudhary and D. K. Dumbre, Catal. Commun., 2011,
98.
M. Zhang, L. Li, T. Zhang, D. Geng, C. Niu and K. Li, J.
Med. Chem., 2013, 56, 685–699.
12, 1351–1356.
25 F. Szabo, D. Simko and Z. Novak, RSC Adv., 2014, 4, 3883–
11 Y. Z. J. Shan, Z. Liu, M. Wang and Y. Dong, Curr. Med. Chem.,
2011, 18, 523–538.
3886.
´
´
26 F. Szabo, J. Daru, D. Simko, T. Z. Nagy, A. Stirling and
´
12 R. Kaur, G. Kaur, R. K. Gill, R. Soni and J. Bariwal, Eur. J. Med.
Chem., 2014, 87, 89–124.
13 B. L. Flynn, G. S. Gill, D. W. Grobelny, J. H. Chaplin, D. Paul,
A. F. Leske, T. C. Lavranos, D. K. Chalmers, S. A. Charman,
Z. Novak, Adv. Synth. Catal., 2013, 355, 685–691.
27 R. B. G. Ravelli, B. Gigant, P. A. Curmi, I. Jourdain,
S. Lachkar, A. Sobel and M. Knossow, Nature, 2004, 428,
198–202.
E. Kostewicz, D. M. Shackleford, J. Morizzi, E. Hamel, 28 A. Brunschweiger, J. Iqbal, F. Umbach, A. B. Scheiff,
´
M. N. Munkonda, J. Sevigny, A. F. Knowles and
M. K. Jung and G. Kremmidiotis, J. Med. Chem., 2011, 54,
6014–6027.
¨
C. E. Muller, J. Med. Chem., 2008, 51, 4518–4528.
¨
29 L. Yan and C. E. Muller, J. Med. Chem., 2004, 47, 1031–1043.
This journal is © The Royal Society of Chemistry 2015
RSC Adv., 2015, 5, 98527–98537 | 98537