Synthesis and bio-evaluation of novel quinolino-stilbene derivatives as potential anticancer agents

Article abstract of DOI:10.1016/j.bmc.2015.11.007

A series of 25 novel quinolino-stilbene derivatives were designed, synthesized and evaluated for their potential as anticancer agents. Three of them not only displayed quite potent antiproliferative activity with IC₅₀ values <4 μM but also showed approximately twofold selectivity against cancer cells, compared to non-cancerous cells. Three other compounds exhibited comparatively good activity with IC₅₀ values in the range of 4-10 μM, and the rest was moderately active or inactive. One of these viz. 3-[E-(4-fluorostyryl)]-2-chloroquinoline (compound 7B) caused substantial DNA damage and arrested cell cycle in S phase. Interestingly, 7B was very active against MDA-MB468 (IC₅₀ = 0.12 μM), but not against other cell lines examined. Compound 3-[Z-(3-(trifluoromethyl)styryl)]-2-chloroquinoline (12A), the most effective against all cancer cell lines examined, caused prolonged cell cycle arrest at mitosis and eventually apoptosis. Data from an in vitro study showed that compound 12A inhibited microtubule polymerization in a similar fashion to nocodazole. Further study using in silico molecular modeling revealed that 12A causes the impediment of microtubule polymerization by binding to tubulin at the same cavity where podophyllotoxin binds.

Full text of DOI:10.1016/j.bmc.2015.11.007

Accepted Manuscript
Synthesis and bio-evaluation of novel quinolino-stilbene derivatives as potential
anticancer agents
Vandana Srivastava, Hoyun Lee
PII:
S0968-0896(15)30132-2
DOI:
http://dx.doi.org/10.1016/j.bmc.2015.11.007
BMC 12655
Reference:
To appear in:
Bioorganic & Medicinal Chemistry
Received Date:
Revised Date:
Accepted Date:
31 August 2015
29 October 2015
6 November 2015
Please cite this article as: Srivastava, V., Lee, H., Synthesis and bio-evaluation of novel quinolino-stilbene
derivatives as potential anticancer agents, Bioorganic & Medicinal Chemistry (2015), doi: http://dx.doi.org/10.1016/
j.bmc.2015.11.007
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Synthesis and bio-evaluation of novel quinolino-stilbene derivatives as potential
anticancer agents
1
,2
1,2,3*
Vandana Srivastava , Hoyun Lee
1
Advanced Medical Research Institute of Canada, Health Sciences North, 41 Ramsey Lake
Road, Sudbury, Ontario, Canada P3E 5J1
2
Division of Medical Sciences, Northern Ontario School of Medicine, 935 Ramsey Lake Road,
Sudbury, Ontario, Canada P3E 2C6
3
Departments of Medicine, the University of Ottawa, Ottawa, Ontario K1H 5M8, Canada
*
Corresponding author
Hoyun Lee, Advanced Medical research Institute of Canada, 41 Ramsey Lake Road, Sudbury,
Ontario, P3E 5J1, Canada. Phone: +1 705 522-6237, ext. 2703. E-mail: hlee@amric.ca
Running title: Anticancer activity of quinolino-stilbene derivatives
1

Abstract
A series of 25 novel quinolino-stilbene derivatives were designed, synthesized and evaluated for
their potential as anticancer agents. Three of them not only displayed quite potent
antiproliferative activity with IC values <4 µM but also showed approximately two-fold
5
0
selectivity against cancer cells, compared to non-cancerous cells. Three other compounds
exhibited comparatively good activity with IC values in the range of 4-10 µM, and the rest was
5
0
moderately active or inactive. One of these viz. 3-[E-(4-fluorostyryl)]-2-chloroquinoline
compound 7B) caused substantial DNA damage and arrested cell cycle in S phase. Interestingly,
B was very active against MDA-MB468 (IC 0.12 µM), but not against other cell lines
(
7
5
0=
examined. Compound 3-[Z-(3-(trifluoromethyl)styryl)]-2-chloroquinoline (12A), the most
effective against all cancer cell lines examined, caused prolonged cell cycle arrest at mitosis and
eventually apoptosis. Data from an in vitro study showed that compound 12A inhibited
microtubule polymerization in a similar fashion to nocodazole. Further study using in silico
molecular modeling revealed that 12A causes the impediment of microtubule polymerization by
binding to tubulin at the same cavity where podophyllotoxin binds.
Key words:
Quinoline; stilbene; anticancer agent; tubulin; cell cycle; spindle checkpoint; apoptosis
2

1
. Introduction
Cancer is the second most common cause of death in the United States, and the American
1
Cancer Society estimates 1,658,370 new cases and 583,430 cancer-related deaths in 2015.
Although the development of early detection methods and timely intervention have resulted in
the substantial improvement of survival and quality of life in certain cancer populations,
effective cancer treatments are still difficult to achieve. One crucial limitation of current cancer
therapies is that a curable dose often cannot be given due to severe side effects. Moreover,
cancer curability is further hampered by the emergence of tumor cells that are resistant to
2
therapeutic agents. Therefore, there is a great need of developing new anticancer therapeutics
with better pharmacological properties including increased selectivity for tumor cells. As a part
of achieving this goal, we previously synthesized and characterized quinoline-based anticancer
3
–6
agents, some of which showed substantial promises. Quinoline, a heterocyclic aromatic
compound present in a wide range of natural and synthetic pharmaceuticals, is a privileged
skeleton in drug discovery. Quinoline derivatives possess many different biological and
pharmacological activities including antimicrobial, antimalarial, antifungal, anti-inflammatory,
7
,8
analgesic, antiviral, antiprotozoal, cardiovascular, CNS effective and antineoplastic. Several
well-known anticancer agents including camptothecin, topotecan and irinotecan contain a
9
quinoline moiety. It is also well-known that quinoline analogues often inhibit tubulin
polymerization, DNA repair, tyrosine kinase activity, proteasome, histone acetyl-transferases
(HATs) and histone deacetylase (HDACs), all of which are considered to be effective cancer
1
0
therapeutic targets.
Stilbenes (1,2 diarylethylene, Fig.1) attract considerable interest because of their wide
1
1–15
range of biological activities and potential therapeutic values, especially against cancer.
Resveratrol (Fig. 1) is a phytoalexin stilbene found in berries, grapes, peanuts and red wine.
Resveratrol and its analogues exhibit various biological and pharmacological activities including
E-
1
6–18
anticancer.
Combretastatin A4 (Fig. 1), isolated from Combretum caffrum, is another natural
9
,19
stilbene analogue possessing potent tubulin inhibition activity. CA4P (combretastatin-4, 3-O
phosphate), a prodrug of CA4, significantly reduces blood flow to the tumor cells, leading to
extensive tumor necrosis. Several other stilbene derivatives have been synthesized previously in
2
0–23
an attempt to develop effective anticancer drugs.
Tamoxifen (Fig. 1), one of the stilbene
3

derivatives, shows selective inhibition activity against estrogen receptor and is currently used as
24,25
chemotherapeutic agent for breast cancer treatment.
Figure 1. Prototype of quinoline-based stilbene derivatives. Shown are various stilbene
derivatives being used as anticancer agents (tamoxifen, resveratrol and combretastatin A4) and
the quinoline-based anticancer agent camptothesin.
The introduction of fluorine to bioactive molecules may alter their physiological
properties and bioactivities. For instance, a fluorinated taxane is more active than the parental
2
6
taxane against several cancer cell lines. High electronegativity, chemical reactivity and the
small size of fluorine all contribute to the enhancement of binding affinity, metabolic stability
2
7
and selective target reactivity. As the number of drug candidates with one or more fluorine
atoms continues to increase in recent years, several groups have reviewed the role of fluorine in
2
7–30
medicinal chemistry.
4

To extend our ongoing efforts to develop effective and safe anticancer agents, we
designed and synthesized a series of novel stilbene derivatives that are composed of a substituted
quinoline moiety as ring A and a fluorine- or trifluoromethyl-substituted phenyl group as ring B.
Subsequently, we examined their anticancer activities and cancer cell selectivity using four
different cancer cell lines and a matching non-cancer cell line.
2
2
. Result and discussion
.1. Chemistry
All the quinolino-stilbene derivatives were synthesized by performing Wittig reactions on
substituted quinoline-3-carbaldehyles (5a-5d), Wittig salts (2) using dimethyl sulfoxide as polar
3
1
aprotic solvent, and sodium hydroxide as the base (Scheme1). Quinoline-3-carbaldehydes (5a-
d; Scheme 1) were either synthesized starting with substituted acetanilide (4) via Vilsmeier-
o
Haack reaction using dimethylformamide (DMF) and phosphorus oxychloride (POCl ) at 85 C-
3
3
2
9
0°C as described previously or purchased from sigma. Benzyl triphenylphosphene bromide
Wittig salts, 2) were synthesized by the reaction of corresponding benzyl bromides with
triphenylphosphine and toluene under refluxing conditions. All of the final products were
(
1
13
characterised by H, C NMR and mass spectroscopy, which were found to be consistent with
the assigned structures (Table1).
5

Br
CH PPh Br
2
3
R1
R1
R2
i
R4
R2
R4
R3
2
R3
1
R5
R6
NHCOCH3
R5
CHO
Cl
R5
NH2
ii
iii
R6
N
R6
5
4
3
5
a. R = R = H
5 6
5
b. R = H, R = Cl
5 6
5
c. R = OCH , R6 = H
5 3
5
d. R +R = -OCH O-
5
6
2
R2
R3
R4
R1
R5
CH PPh Br
2
3
R5
R6
CHO
Cl
R1
iv
R6
N
Cl
(21.6-75.6%)
6-21A
N
R4
R2
R2
R3
5
2
R1
R3
R4
R5
R6
N
Cl
6
-20B (2-10%)
Scheme 1. Reagents and conditions: (i) PPh , toluene reflux; (ii) dil HCl, acetic anhydride, rt;
3
◦
(
iii) POCl : DMF 3:1, ice, then 90 C, reflux; and (iv) NaOH, DMSO, rt.
3
6

Table 1: Chemical structures of novel stilbene derivatives
C H NClR R R R R R
6
1
7
6
1
2
3
4
5
S. No
Compounds
R1
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
CF3
CF3
CF3
CF3
H
H
H
H
CF3
R2
H
H
H
H
H
H
H
H
R3
R4
R5
H
H
H
H
R6
H
H
H
Cl
Cl
H
H
H
H
H
H
H
H
H
H
Cl
H
H
H
H
1
2
3
4
5
6
7
8
9
6A
7A
7B
8A
9A
CF3
F
H
H
F
H
CF3
F
H
H
H
10A
10B
11A
11B
12A
12B
13A
14A
14B
15A
16A
17A
17B
18A
18B
19A
19B
20A
20B
21A
CF3
CF3
F
H
H
H
OCH3
OCH3
OCH3
OCH3
H
H
F
H
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
CF3
CF3
CF3
CF3
CF3
CF3
CF3
H
H
H
H
CF3
CF3
CF3
CF3
H
H
H
H
H
H
H
H
OCH3
OCH3
OCH3
H
H
H
H
H
CF3
CF3
CF3
CF3
H
CF3
CF3
CF3
CF3
H
H
H
H
H
H
H
OCH3
OCH3
CF3
CF3
H
-OCH O-
2
H
H
-OCH O-
2
-OCH O-
2
H
H
-OCH O-
2
H
H
OCH3
H
Cis (6A-21A) and trans (6B-20B) isomers, the major and minor products of the Wittig
reaction respectively, were purified by column chromatography. We sometimes encountered
difficulties in separating these two stereoisomers due to their close retention factors (RFs),
resulting in medium to low yield of pure isomers along with the mixture of isomers. In some
cases only cis products were isolated in the pure chemical form. The geometry of stilbenes (cis
1
and trans) was identify by HNMR on the basis of H-H coupling constant (J). For some cis
derivatives (for instance 12A), we unexpectedly got only one signal for two ethylene protons
when spectra were run in CDCl . On the other hand, two signals (one doublet, J = 15 Hz, & one
3
merged doublet) for two ethylene protons were found for trans isomer (12B). The resultant mass
spectra of both the isomers were in agreement with expected structures. To confirm geometry of
7

cis isomer, we ran the sample 12A in acetone d6, by which we could clearly resolve doublets (J =
1
2
2
2.15 Hz) for two ethylene protons.
.2. Biological activity
.2.1. In vitro cytotoxicity and structural activity relationship (SAR)
We examined the novel compounds for their antiproliferative activities against four
cancer cell lines: HeLa (cervical carcinoma), MDA-MB231 (ER-negative undifferentiated
metastatic breast cancer), MCF7 (ER-positive well-differentiated breast cancer), and MDA-
MB468 (PTEN mutated, intermediately differentiated breast cancer). To determine differential
effects between cancer and non-cancer cells, we also examined the effect of these compounds on
1
84B5, a non-cancer breast epithelial cell line. Data from sulforhodamine B (SRB)-based
3
3
assays showed that the stilbene analogues displayed cell line-dependent IC values (Table 2).
50
Three of the compounds examined (12A, 13A and 21A) showed IC values in the range of 2.6-
5
0
4
.0 µM against all four cancer cell lines. Furthermore, they were approximately two-fold more
effective on cancer cells than non-cancerous cells (Table 2). Three of the derivatives (6A, 14A
and 15A) were also relatively active with IC values in the range of 4-10 µM. The rest was
5
0
generally not very active with IC of 10 µM or higher.
5
0
When considered in the context of SAR, the results suggest that stereochemistry and the
positions of different functional groups are likely to play important roles in their activities. We
found that the cis configuration of stilbene derivatives is generally more favorable for higher
activity and better cancer selectivity than trans, except for 7B. This data is in line with previous
report indicating that s-trans conformation of chalcones possess better anti-invasive activities
than s-cis counterpart. Indeed, the most anti-invasive compound shown in that study, which was
geometrically similar to corresponding s-trans chalcone, was designed as cis (Z) stilbene
34
mimic. Compound 7B (trans) was very active against MDA-MB468 (IC , 0.12 µM) and
5
0
moderately active against MCF7 (IC , 15.13 µM) breast cancer cells. However, it was not active
5
0
against HeLa, MDA-MB231 and 184B5 cells. This result suggests that SAR is substantially
affected by the genetic background of the (tumor) cells. In this context, it may be worth to note
that PTEN is mutated in the MDA-MB468 cell line.
8

Table 2: Antiproliferative activity of stilbene derivatives against cancer and non-cancer cells.
IC Values in µM ±S.E.M. against different cell lines
5
0
S. No Compounds
MDA-
MB231
MDA-
MB468
HeLa
MCF7
184B5
1
2
3
4
5
6
7
8
9
6A
4.22±0.32
4.26±0.58
5.36±0.33
23.98±0.14
>50
7.38±0.12
7.94±0.63
7A
24.91±1.89 20.29±2.30
5.29±0.83 35.73±1.00
0.12±0.02 38.45±9.5
16.13±0.46 26.40±0.61
10.91±0.10 26.89±10.98
22.65±1.94 40.04
7B
>50
15.13±5.79
6.45±1.24
8A
12.64±0.22
13.55±4.12
42.56±2.03
22.83
9A
38.62±5.46 16.89±1.92
10A
10B
11A
11B
12A
12B
13A
14A
15A
16A
17A
18A
19A
20A
21A
Paclitaxel*
16.21
12.41±1.99
46.97±2.93
8.97±3.03
14.05±3.61
3.53±0.74
>50
48.83±1.56
11.40±0.02
11.03±1.28
3.75±0.29
45.01±2.08
3.28±0.24
4.47±0.11
10.13±0.35
>50
>50
33.48±1.06
9.67±0.84
>50
15.17±1.06 19.18±1.4
14.84±3.66
3.70±0.37
23.99
9.74±1.8
6.15±0.40
24.73±3.8
5.58±0.13
7.97±1.95
1
0
1
2
3
4
5
6
7
8
9
0
1
2
3
2.85±0.02
1
1
1
1
1
1
1
1
1
2
2
2
2
20.27±0.27 24.43±3.64
3.03±0.12
4.14±0.30
6.78±0.66
2.78±0.68
4.70±1.21
7.46±1.16
3.91±0.75
5.79±0.63
10.52±0.71 13.67±0.98
41.35±6.50 41.71±4.41
35.89±1.96 10.45
>50
42.95±11.43
ND
41.76±11.61
>50
>50
>50
>50
>50
>50
>50
>50
>50
>50
>50
20.09±0.20 18.90±0.23
26.0±6.32
2.77±0.08
2.56±0.80
5.77
16.97±0.13
5.31±0.71
2.32±0.25
4.96±2.18
2.60± 0.08
2.29±0.03
4.47±0.08
3.99±0.14
4.98±0.80
3.87±0.50
Camptothecin* 6.13±0.63 ND**
Chloroquine 29.96±3.39 ND
14.76±2.02
36.53±3.08
19.86±0.84 63.08
*
IC in nM; ** Not determined
50
The presence of a CF group on the phenyl ring generally renders good activity as shown
3
by compounds 6A, 12A, 13A and 21A. The replacement of a CF group with an F atom reduced
3
the activity by 2-5 fold (6A vs 7A and 8A vs 9A). An addition of a chlorine group at the position
7
2
(8A) or a methoxy group at the position 6 (10A) on the quinoline ring reduced the activity by
-4 fold when combined with a 4 CF substituted phenyl ring. The shift of the CF group from
3
3
9

the position 4 (6A) on the benzene ring to the position 3 (12A) improved activity by ~2 fold.
However, an addition of a methoxy group at the position 6 on the quinoline ring did not show
any effect on the antiproliferative activity when combined with a CF group on the phenyl ring
3
(
12A vs 13A). The shift of a CF group from the position 3 to the position 2 along with a
3
methoxy group at the position 6 (13A vs 21A) did not show any significant effect on the activity.
The addition of one more CF group on the phenyl ring affected the activity profile. The
3
presence of CF groups at the 3 and 5 positions rendered 5-10 fold better activity than presence
3
at the 2 and 4 positions (15A vs17A & 14A vs 18A). An addition of a chlorine group at the
position 7 on the quinoline ring along with CF groups at the 3 and 5 positions of the phenyl ring
3
reduced the activity by ~5 fold (15A vs 16A), whereas the addition of a methoxy group at the
position 6 improved the activity by ~2 fold (15A vs 14A). An addition of a methylenedioxy
group on the quinoline ring reduced activity by ~10 fold (19A vs 20A).
According to SAR studies, a CF group at the position 2 or 3 of the phenyl ring is best
3
suited for better antiproliferative activity. A methoxy group at the positon 6 of the quinoline ring
is tolerable; however, it does not contribute to the enhancement of activity. Compound 12A, the
most active compound in this series was also examined for its activity against MCF10A, another
non-cancer breast cell line. The IC value against MCF10A was 6.01±0.28 µM (Fig. 2), which
5
0
is similar to that of 184B5 non-cancer cell line (Table 2). This result confirms that compound
2A is selective for cancer cells in inhibiting cell proliferation. We carried out further biological
1
studies with compound 12A, the lead compound, and compound 7B that showed high activity
against MDA-MB468 cells.
1
0

1
50
MDA-MB231
HeLa
100
MDA-MB468
MCF7
5
0
184B5
MCF10A
0
0.0
0.5
1.0
1.5
2.0
Log concentration (M)
-50
Figure 2. Antiproliferative effects of compound 12A on cancer and matching non-cancer cell
lines. An SRB assay was carried out with exponentially growing cells as described in Materials
and methods.
2
.2.2. Compounds 12A and 7B arrested cell cycle at mitosis and S-phase, respectively,
eventually leading to cell death by apoptosis
When HeLa cells were treated with 6 µM of compound 12A, the number of cells in G2-
M was 78% by 12 h, a substantial increase from 19% shown in time 0 h (Fig. 3A, 12 h). By 24 h
post-treatment, cells arrested at G2/M were substantially decreased, with a concomitant increase
in cell population with sub-G1 DNA content (Fig. 3A, 24 h). Together, this data suggests that
compound 12A impedes normal G2-M progression, eventually leading to cell death as a cell tries
to complete its cell division cycle without properly segregating its chromosomes. Compound
1
2A also caused a similar pattern of G2/M arrest and DNA fragmentation in the MDA-MB231
metastatic breast cancer cells (Fig. 3B and Supplementary Fig. S1). In contrast, MCF10A, a non-
cancer breast cell line, showed neither substantial G2/M arrest nor DNA fragmentation in the
presence of compound 12A (Fig. 3B and Supplementary Fig. S2).
The treatment of MDA-MB 468 cells with compound 7B produced substantially
increased S-phase population by 24 h (41.7%), compared to sham control (16.9%). The cell
population in S phase was decreased by 48 h post-treatment with a concomitant increase in sub-
G1 population, suggesting that a death pathway was activated in those cells arrested in S-phase
without going through proper cell division process (Supplementary Fig. S3).
1
1

Figure 3. Compound 12A caused cell cycle arrest at G2/M by 12 h and extensive cell death by
8 h post-treatment. (A) Flow cytometry profiles of asynchronous HeLa cells treated with 12A
6 µM), which showed dramatic increases in sub-G1 population by 48 h post-treatment. Post-
4
(
treatment time (h) shown is in hour(s). (B) % of sub-G1 population of cells at different time
points. Data is from an average of two replicates of flow cytometry (as shown in panel A). Bars
represent mean ±S.E.M. of two independent experiments.*p<0.05, **p<0.005.
1
2

Figure 4. Compound 12A did not cause DNA damage. HeLa cells were treated with 6 µM of
2A or 50 µM etoposide for 24 or 48 h, followed by immunostaining with an anti-γH2A.X
1
(phospho-Ser139) antibody. DNA was counterstained with DRAQ5. Scale bar is 20 µm
To determine if compounds 12A and 7B cause DNA damage, we carried out
immunostaining of 12A-treated HeLa cells and 7B-treated MDA-MB468 cells with an anti-
γH2A.X (phospho-Ser139) antibody. Data shown in Fig. 4 indicates that 12A does not cause
DNA damage at least up to 48 h post-treatment. Therefore, the arrest of cell cycle at G2/M
observed by flow cytometry (Fig. 3A) is not caused by DNA damage. In contrast, MDA-MB468
cells treated with 1 µM 7B showed substantial DNA damage as most of the treated cells showed
γH2A.X positive by 24 h post-treatment (Supplementary Fig. S4). This data thus suggests that
7
B causes DNA damage which in turn causes S-phase arrest, eventually leading to cell death.
Cell cycle may be arrested at metaphase by the activation of spindle assembly checkpoint
3
5
(
SAC) for a variety of reasons. To gain insight into whether 12A caused cell-cycle arrest by
SAC activation, we investigated the levels of (phosphorylated) proteins involved in the G2-M
transition. We found that the protein level of Wee 1 kinase was barely detectable in the presence
of 12A, while the level of histone 3 phosphorylation on serine 10 was much higher than that of
1
3

control (Fig. 5A and B). This data is consistent with the notion that 12A-treated cells have
3
6
already passed G2 and chromosome condensation stage by 12 h-post treatment. The level of
cyclin B was very high in cells treated with 12A at all of the time points examined (Fig. 5A and
B). To our surprise, the level of cyclin B was still very high even at 48 h post-treatment when a
large number of cells contained sub-G1 DNA content (Fig. 3A). These data indicate that cells
were arrested prolonged duration at the spindle checkpoint step (Fig. 5A and B).
A large number of 12A-treated cells contained sub-G1 DNA content by 48 h post-
treatment, suggesting that they were dying by apoptosis (Fig. 5A). This conclusion was
confirmed by the data that PARP1 protein was cleaved by 48 h post-12A treatment (Fig. 5A).
We then documented 12A-treated cells by a time-lapse camera. As expected, HeLa cells treated
with 12A showed severe membrane blebbing and popcorn-like morphology, a typical phenotype
3
7
of cells undergoing apoptosis (Fig. 5C).
Figure 5. Cells arrested at mitosis by 12A underwent massive apoptosis by 48 h post-treatment.
A) Western blot analysis indicates that cells are arrested at mitosis in the presence of 12A. HeLa
(
1
4

cells were sham-treated or treated with 6 µM 12A, and collected at indicated time points. Cell
extracts were then subjected to polyacrylamide gel electrophoresis, followed by Western blotting
with antibodies specific for proteins listed at the left of the panel. GAPDH is a loading control.
(B) The relative abundance of proteins was normalized with that of GAPDH, and plotted in a
graph form. (C) 12A caused apoptosis. HeLa cells treated with 12A were subjected to
documentation by time-lapse microscopy. Slides shown are pictures taken at the indicated time
points. Scale bar is 50 µm.
2
.2.3. 12A-treated cells are defective in chromosome alignment and segregation
Interestingly, securin band was up-shifted in cells treated with 12A (Fig. 5A), of which
pattern is often observed in cells treated with microtubule-interacting drugs such as paclitaxel,
3
8
colchicine and vinblastine. Our data thus raised the possibility that 12A might impede
microtubule functions. To gain insights into this possibility, we examined cell morphology by
immunofluorescence microscopy. By 12 h post-treatment, sham-treated cells were mostly in
normal interphase morphology (Fig. 6A, Sham). In contrast, the majority of cell treated with 12A
were arrested at mitosis with disoriented tubulin structure (Fig. 6A, 12A, α-tubulin stained cell).
The chromosomes of cells exposed to 12A were condensed, but not properly aligned at center
plate (Fig. 6A, arrows). As their nuclear membrane was already broken (as evident by the lack of
membrane staining by an anti-lamin A antibody), they have already entered mitosis. However,
these mitotic cells were “permanently” arrested at mitosis as the entire chromosomes were not
aligned at the centre plate (arrows in Fig. 6B and C).
1
5

Figure 6. Compound 12A caused defects in tubulin function and chromosome alignment at
mitosis. (A) Cell morphology was abnormal in the presence of 12A, likely due to defects in
microtubule function. Pictures were taken 12 h post-treatment. DNA was stained with DAPI.
Arrows show abnormal alignment of chromosomes. Scale bar is 5 µm. (B) Chromosome
alignment at the center plate was not properly completed in the presence of 12A. HeLa cells
were analyzed 12 h post-treatment. DNA was counterstained with DRAQ5. Arrows show
misaligned chromosomes. (C) Numbers of cells with defective chromosome alignment were
quantified at 12 h and 24 h post-treatment. 100 cells were counted for each experiment, and the
entire experiment was repeated once more.
2
.2.4. 12A inhibits tubulin polymerization
Since securin band-shift (Fig. 5A) and data from immunofluorescence microscopy (Fig.
) raised the possibility of 12A being a microtubule interacting agent, we evaluated the effect of
2A on tubulin polymerization with an in vitro tubulin assay kit. The kit was based on the
6
1
1
6

principle that light scatter by the microtubule is proportional to the concentration of polymerized
microtubules, which can be detected at 340 nm wave-length. Therefore, tubulin stabilizing and
destabilising agents would produce higher and lower absorbance, respectively, than the tubulin
buffer control. As expected, paclitaxel and nocodazole (Noco) produced high and low
absorbance at 340 nm, respectively, as the former stabilizes microtubules and the latter inhibits
microtubule polymerization (Fig. 7). The absorbance pattern of 12A at 6 and 12 µM was similar
to that of nocodazole, indicating that it is an inhibitor of microtubule polymerization (Fig. 7).
Taken together, our data indicated that the inhibition of microtubule polymerization by 12A
leads to prolonged cell cycle arrest at the mitotic spindle checkpoint stage, eventually resulting in
apoptosis.
Figure 7. Compound 12A inhibited tubulin polymerization. Microtubule polymerization kinetics
are shown in the presence of different ligands: paclitaxel 10 µM, nocodazole (Noco) 5 µM,
compound 12A 6 µM or 12 µM. Tubulin buffer was used as a control.
1
7

2
.2.5. The molecular target of 12A coincides with the podophyllotoxin (PDT)-binding site
on the tubulin
Since our data is consistent with the notion that 12A impedes tubulin polymerization, we
searched for the potential binding site of 12A on the tubulin using the MOE Sitefinder program
(See Material and methods) and the co-crystalized structure of the tubulin α-β heterodimers
3
9
(
PDBID: 1SA1). We found that the most likely 12A-binding site determined by the lowest
docking energy (S) and root mean square deviation (RMSD) coincides with the PDT binding
pocket on the tubulin (Fig. 8): it shows the lowest S-score (-6.1615) and RMSD (0.0126). This
modeling data predicts that compound 12A forms a pi-bond interaction with amino acid
asparagine B258 (Asn B258) with a bond length of 3.77Å and energy -0.6 kcal/mol (Fig. 8C).
Potential Van der Waals interaction sites of amino acids lysine B352, asparagine B258, threonine
A179, alanine B250, valine B315, alanine B316, leucines B255, B248 and B255, valine A181
and alanine B354 are common in 12A and PDT binding interactions. PDT, a natural product
belongs to lignan class of chemicals, exhibits strong anticancer activity against various cancer
9
cell lines. PDT, similarly to 12A, interacts with tubulin, inhibits its assembly and arrests cell
cycle in metaphase.
1
8

Figure 8. Data from in silico molecular modeling suggests that compound 12A binds to tubulin
at the same cavity bound by PDT. (A) An image of docking both 12A (magenta) and PDT (blue)
at the PDT-binding site on the tubulin. (B) An image of 12A alone docking at the PDT-binding
site on the tubulin. (C) Proximity contour of 12A showing Van der Waals and pi-bond
interactions with adjacent amino acid residues of tubulin in the PDT pocket. (D) The relative
bindings of 12A and podophyllotoxin at the PDT-binding cavity on the tubulin, which was
derived from the figure in panel A.
1
9

3
. Conclusion
A total of 25 novel quinolino-stilbene derivatives were designed and synthesized, among
which 20 were studied for their antiproliferative activity. Although trans derivatives were
generally less active, one of them, 7B, showed surprisingly strong activity against MDA-MB468
breast cancer cells (IC = 0.12 µM). Data from flow cytometry and immunofluorescence
5
0
microscopy suggested that 7B causes considerable DNA damage, resulting cell cycle arrest in S-
phase and eventually apoptosis. However, 7B was not effective on other cell lines examined,
indicating that genetic background is an important factor in its cell killing ability.
Some of the cis derivatives exhibited good anticancer activity against all four cancer cell
lines examined. Three compounds in this series (12A, 13A, 21A) exhibited quite potent
antiproliferative activities with IC less than 4 µM; three other compounds (6A, 14A and 15A)
5
0
displayed relatively good antiproliferative activities with IC in the range of 4-10 µM.
5
0
Compound 12A, the most active one in this series, caused prolonged cell cycle arrest at the
spindle checkpoint step, eventually leading to cell death by apoptosis. Data from Western
blotting, an in vitro tubulin polymerization assay, and an immunofluorescence-based study
strongly indicated that 12A causes cell cycle arrest by impeding tubulin polymerization. The
result of in silico molecular modeling is consistent with this conclusion, as the model predicts
that 12A binds to the pocket known to be bound by PDT. As 12A preferentially kills cancer over
non-cancer cells, it is a promising anticancer lead compound that can be further optimized to
enhance anticancer activity and cancer selectivity.
4
4
. Experimental
.1. Chemistry
All the chemicals and dry organic solvents were purchased from Sigma-Aldrich or
Fisher, and used without further purification. Thin layer chromatography (TLC, aluminum plate
coated with silica gel 60 F254, 0.25 mm thickness, Sigma) was used to monitor the progress of
reactions. Iodine vapour and/or UV light was used to detect products on TLC. Melting points
were measured with Sturat SMP-20 instrument in open capillaries and were uncorrected. All the
2
0

1
H and 13 Carbon NMR (Nuclear Magnetic Resonance) spectra were recorded either with a
Bruker Avance 400 or 500 MHz spectrometer (Bruker Scientific Corporation Ltd., Switzerland).
Tetramethylsilane or solvent signal was used as internal standard, and chemical shift is reported
in part per million (ppm). Splitting patterns are defined as follows: s, for singlet; bs for broad
singlet; d for doublet; t, for triplet, m for multiplet, and coupling constant (J) is reported in hertz
(Hz). Mass spectra were carried out with a Kratos Concept 1-S high resolution magnetic sector
mass spectrometer with electron impact to identify mass of synthesized compounds.
4
.1.1. General procedure of Wittig salt (2) synthesis
1
.0 mmol of corresponding benzyl bromide was dissolved in toluene with 1.2 mmole of
triphenylphosphene. Reaction mixture was refluxed at 110ºC for 1-3 h; generated precipitate was
filtered to get Wittig salt (2) in 60-80% yield.
4
.1.2 .General synthetic procedure of compounds 6A and 6B
In a round-bottom flask, 1.56 mmol of Wittig salt (2) and 2.5 mmol sodium hydroxide
were dissolved in 3 ml of dimethylsulfoxide (DMSO) and stirred for 30 min. To this reaction
mixture, 1.3 mmol of 2-chloroquinoline-3-carbaldehyde was added, followed by stirring it for 5
h to overnight (~15 h) at room temperature. After completing the reaction, mixture was
neutralized with diluted HCl and extracted with chloroform. Organic layer was washed with
water and dried over anhydrous sodium sulfate; solvent was evaporated; and crude mixture was
purified by silica gel column chromatography to purify cis and trans stilbene derivatives. 1H
spectra of all the stilbenes are shown in supplementary information.
4
8
8
1
1
.1.2.1. 3[Z-(4-(trifluoromethyl)styryl)]-2-chloroquinoline (6A). Solid; yield = 57.4%; mp.
1
9-91°C; H NMR (400 MHz, CDCl ): δ 6.86 (s, 2H), 7.26 (d, 2H, J = 8.0 Hz), 7.42 (d, 2H, J =
3
Hz), 7.45 (t, 1H, J = 8Hz), 7.54 (d, 1H, J = 8Hz), 7.67 (1, 1H, J = 8Hz), 7.86, (s, 1H), 7.89 (d,
13
H, J = 12 Hz); C NMR (100 MHz, CDCl ):δ 122.60, 125.42 (q, 2C, J = 4 Hz), 126.86, 127.27,
3
27.54, 127.59, 129.12, 129.12, 129.33, 129.41, 130.68, 131.89, 133.46, 138.40, 139.44, 147.03,
2
1

+
+
+
1
1
3
50.28; MS (EI) 333.1 (M ); HRMS calculated for C H ClF N [M] 333.05321, found [M]
18 11 3
33.05436.
4
.1.2.2. 3[Z-(4-fluorostyryl)]-2-chloroquinoline (7A). Solid; yield = 21.6%; mp. 86-88°C; H
NMR (400 MHz, CDCl ): δ 6.69 (d, 1H, J=12Hz), 6.78 (d, 1H, J=12Hz), 6.86 (m, 2H), 7.13 (m,
3
2
1
1
1
H), 7.44 (t, 1H, J=8.1Hz), 7.54 (d, 1H, J=8.1Hz), 7.65 (t, 1H, J=8.5Hz), 7.90 (s, 1H), 7.98 (d,
13
H, J=8.36Hz); C NMR (100 MHz, CDCl ):δ 115.43, 115.65, 125.37, 126.95, 127.13, 127.58,
3
28.27, 129.79, 130.45, 130.56, 130.64, 131.82 (d, J =3.75 Hz), 132.22, 138.37, 146.89, 150.54,
+
+
60.85 (d, J=246Hz); MS (EI) 283.1 (M ); HRMS Calculated for C H ClFNO [M] 283.05641,
1
7
11
+
found [M] 283.06003.
4
.1.2.3. 3-[E-(4-fluorostyryl)]-2-chloroquinoline (7B). Solid; yield = 5.4%; mp. 144-146°C;
1
H NMR (400 MHz, CDCl ): δ 7.05 (t, 2H, J=8.0Hz), 7.13 (d, 1H, J=16.0Hz), 7.41 (d, 1H,
3
J=16.0Hz), 7.55 (m, 3H), 7.67 (t, 1H, J=8.0Hz), 7.82 (d, 1H, J=8.0Hz), 7.98 (d, 1H, J=8.0Hz),
1
3
8
1
.65 (s,1H); C NMR (100 MHz, CDCl ):δ 115.79, 116.00, 123.32, 123.34, 127.24, 127.53,
3
28.36, 128.58, 128.66, 130.20, 130.29, 132.03, 132.79, 133.69, 146.87, 150.17, 161.66 (d,
+
+
J=248Hz); ); MS (EI) 283.1 (M ); HRMS calculated for C H ClFNO [M] 283.05641, found
1
7
11
+
[
M] 283.05303.
4
.1.2.4. 3-[Z-(4-(trifluoromethyl)styryl)]-2,7-dichloroquinoline (8A). Solid; yield = 26.9%;
1
mp. 121-124°C; H NMR (500 MHz, CDCl ): δ 6.85 (d, 1H, J = 12.1Hz), 6.91 (d, 1H, J =
3
1
3
1
2.1Hz), 7.29 (d, 2H, J = 8.5Hz), 7.46 (m, 4H), 7.86 (s, 1H), 8.02 (d, 1H, J = 1.7Hz); C NMR
125 MHz, CDCl ): δ 122.82, 125.19, 125.51 ( q, 2C, J = 3.75 Hz), 127.25, 127.38, 128.39,
(
3
1
28.66, 128.97, 129.11, 129.11, 129.60, 132.24, 136.71, 138.08, 139.30, 147.23, 151.50; MS
+
+
+
(
EI) 367.0 (M ); HRMS calculated for C H Cl F N [M] 367.01424, found [M] 367.02624.
18 10 2 3
4
.1.2.5. 3-[Z-(4-fluorostyryl)]-2,7-dichloroquinoline (9A). Solid; yield = 22.6%; mp. 103-
1
1
05°C; H NMR (500 MHz, CDCl ): δ 6.71 (d, 1H, J = 12.1 Hz), 6.84 (d, 1H, J = 12.1 Hz), 6.89
3
(t, 2H, J = 8.7Hz), 7.15 (m, 2H), 7.45 (d, 1H, J = 8.7 Hz), 7.51 (d, 1H, J = 8.7Hz), 7.91 (s, 1H),
1
3
8
1
.01 (s, 1H); C NMR (125 MHz, CDCl ): δ 99.99, 115.52, 115.69, 125.05, 125.28, 127.35,
3
28.24, 128.61, 130.07, 130.54, 130.61, 132.58, 136.45, 138.05, 147.09, 151.75, 161.15 (d, J =
2
2

+
+
2
46.84 Hz); MS (EI) 317.0 (M ); HRMS calculated for C H ClFN [M] 317.01743, found
17 10
+
[
M] 317.0174.
4
.1.2.6. 3-[Z-(4-(trifluoromethyl)styryl)]-2-chloro-6-methoxyquinoline (10A). Solid; yield =
1
7
5.6%; mp. 127-129°C; H NMR (400 MHz, CDCl ): δ 3.87 (s, 3H), 6.82 (d, 1H, J=2.5Hz), 6.85
3
(s, 2H), 7.31 (d, 2H, J=8.0Hz), 7.36 (dd, 1H, J=9 &2.5Hz), 7.46 (d, 2H, J=8.0Hz), 7.79 (s, 1H),
1
3
7
.91 (d, 1H, J = 9.0Hz); C NMR (125 MHz, CDCl ):δ 55.58, 104.92, 122.88, 123.39, 125.43
3
(q, 2C J=3.75Hz), 127.83, 127.99, 129.16, 129.16, 129.48, 129.48, 129.66, 136.67, 137.11,
+
1
39.52, 143.09, 147.55, 153.29;); MS (EI) 363.1 (M ); HRMS calculated for C H ClF NO
1
9
13
3
+
+
[
M] 363.06378, found [M] 363.06483.
4
.1.2.7. 3-[E-(4-(trifluoromethyl)styryl)]-2-chloro-6-methoxyquinoline(10B). Solid; yield =
1
8
.6%; mp. 129-131°C; H NMR (500 MHz, CDCl ): δ 3.96 (s, 3H), 7.11 (d, 1H, J=2.5Hz), 7.18
3
(d, 1H, J=16.0Hz), 7.36 (dd, 1H, J= 9.0 &2.5Hz), 7.58(d, 1H, J=16.0Hz), 7.68 (s, 4H), 7.89 (d,
1
3
1
1
1
H, 9.0Hz), 8.29 (s, 1H); C NMR (125 MHz, CDCl ):δ 55.61, 105.05, 123.00, 123.31,
3
25.79(q, 2C, J=3.75Hz), 126.22, 127.07, 127.07, 128.50, 129.66, 129.71, 130.01, 131.37,
+
32.95, 139.99, 143.17, 147.46, 158.42;MS (EI) 363.1 (M ); HRMS calculated for
+
+
C H ClF NO [M] 363.06378, found [M] 363.06473.
19
13
3
4
.1.2.8. 3-[Z-(4-fluorostyryl)]-2-chloro-6-methoxyquinoline (11A). Solid; yield = 33.9%; mp.
1
1
34-136°C; H NMR (500 MHz, CDCl ): δ 3.86 (s, 3H), 6.71 (d, 1H, J=10Hz), 6.80 (d, 2H,
3
J=10Hz), 6.89 (t, 1H, J=8.7Hz)7.16 (m, 2H), 7.34 (dd, 1H, J=10Hz), 7.84 (s, 1H), 7.90 (d, 1H,
13
J=10Hz); C NMR (125 MHz, CDCl ):δ55.57, 104.88, 115.41, 115.58, 123.15, 125.55, 128.04,
3
1
29.63, 129.94, 130.58, 130.65, 131.90, 131.98, 137.08, 142.95, 147.82, 158.91, 161.07 (d,
+
+
J=246.25 Hz); MS (EI) 313.1 (M ); HRMS calculated for C H ClFNO [M] 313.06697, found
1
8
13
+
[
M] 313.06504.
4
.1.2.9. 3-[E-(4-fluorostyryl)]-2-chloro-6-methoxyquinoline (11B). Solid; yield = 6.5%; mp.
1
1
85-189°C; H NMR (500 MHz, CD OD): δ 3.84 (s, 3H), 6.71 (d, 1H, J=15Hz), 6.90 (d, 1H,
3
J=10Hz), 7.06 (t, 2H, J=10Hz), 7.18 (t, 2H, J = 10Hz), 7.30 (d, 1H, J = 5Hz), 7.42 (dd, 1H,J =
1
3
1
0Hz), 7.86 (d, 1H, J = 10Hz), 8.09 (s, 1H); C NMR (125 MHz, CD OD):δ 56.11, 106.26,
3
2
3

1
1
15.89, 116.06, 123.63, 125.79, 128.58, 129.51, 130.36, 131.13, 131.19, 132.40, 132.44, 137.73,
+
42.68, 147.13, 158.33, 160.89 (d, J=243.75); MS (EI) 313.1 (M ); HRMS calculated for
+
+
C H ClFNO [M] 313.06697, found [M] 313.06575.
18
13
4
.1.2.10. 3-[Z-(3-(trifluoromethyl)styryl)]-2-chloroquinoline (12A). Solid; yield = 57.5%;
1
mp. 86-89°C; H NMR (500 MHz, Acetone D ): δ 6.94 (d, 1H J = 12.15 Hz), 7.05 (d, 1H, J =
6
1
1
1
1
2.15 Hz), 7.43 (t, 1H, J = 7.6 Hz), 7.49 (d, 1H, J = 7.7 Hz), 7.54 (m, 3H), 7.78 (m, 2H), 7.96 (d,
1
3
H, J = 8.45), 816 (s, 1H); C NMR (125 MHz, CDCl ):δ 122.75, 124.34 (q, J = 3.75 Hz),
3
25.82 (q, J = 3.75 Hz), 126.82, 127.19, 127.23, 127.44, 128.28, 128.92, 129.19, 130.63, 130.95,
+
31.77, 131.83, 136.61, 138.30, 147.01, 150.30;MS (EI) 333.1 (M ); HRMS calculated for
+
+
C H ClF N [M] 333.05321, found [M] 333.05651.
18
11
3
4
.1.2.11. 3-[E-(3-(trifluoromethyl)styryl)]-2-chloroquinoline (12B). Solid; yield = 6.9%; mp.
1
1
08-111°C; H NMR (500 MHz, CDCl ): δ 7.20 (d, 1H, J = 15.0Hz), 7.53 (m, 4H), 7.21 (t, 1H, J
3
=
10Hz), 7.78 (d, 1H, J = 10Hz), 7.82, (s, 1H), 7.86 (d, 1H, J = 10Hz), 8.02 (d, 1H, J = 10Hz),
1
3
8
1
1
3
.40 (s, 1H); C NMR (125 MHz, CDCl ):δ 122.91, 123.77 (q, J = 3.75), 124.99 (q, J = 3.75Hz),
3
25.41, 127.43, 127.43, 127.62, 128.37, 129.35, 129.71, 129.82, 130.55, 131.23 131.65, 134.12,
+
+
37.33, 147.05, 150.07;MS (EI) 333.1 (M ); HRMS calculated for C H ClF N [M]
1
8
11
3
+
33.05321, found [M] 333.05670.
4
7
6
1
1
.1.2.12. 3-[Z-(3-(trifluoromethyl)styryl)]-2-chloro-6-methoxyquinoline (13A). Solid yield =
1
3.1%; mp. 130-135°C: H NMR (500 MHz, CDCl ): δ3.86 (s, 3H), 6.81 (d, 1H,J = 2.5Hz),
3
.86 (s, 2H), 7.28 (d, 1H, J = 7.5Hz), 7.32 (m, 2H), 7.46 (d, 1H, J = 7.5Hz 7.53 (s, 1H), 7.82 (s,
13
H), 7.90 (d, 1H, J = 10Hz); C NMR (125 MHz, CDCl ): δ 55.57, 104.86, 122.78, 123.35,
3
24.29 (q, J=3.75 Hz), 125.85 (q, J=3.75 Hz), 127.40, 127.95, 128.89, 129.34, 129.65, 130.90 (q,
+
J = 31.25Hz), 131.55, 131.88, 136.66, 137.00, 143.09, 147.59, 158.28;MS (EI) 363.1 (M );
+
+
HRMS calculated for C H ClF NO [M] 363.06378, found [M] 363.06246.
1
9
13
3
4
.1.2.13. 3-[Z-(3,5-bis(trifluoromethyl)styryl)]-2-chloro-6-methoxyquinoline (14A). Solid;
1
yield = 73.9%; mp. 139-141°C; H NMR (500 MHz, CDCl ): δ 3.86 (s, 3H), 6.81 (d, 1H,
3
J=2.6Hz), 6.86 (d, 1H, J=12.2 Hz), 6.98 (d, 1H, J=12.2Hz), 7.37 (dd, 1H, J=9.2 &2.1Hz), 7.66 (s,
2
4

1
3
2
1
1
H), 7.71 (s, 1H), 7.81 (s, 1H), 7.91 (d, 1H, J=9.2Hz); C NMR (125 MHz, CDCl ):δ 55.58,
3
04.69, 121.19 (m), 121.89,121.89, 123.74, 127.77, 128.35, 128.86 (2C, bS), 129.26, 129.70,
+
29.84, 131.47, 136.71, 137.95, 143.27, 147.18, 158.45; MS (EI) 431.1 (M ); HRMS calculated
+
+
for C H ClF NO [M] 431.05116, found [M] 431.05265.
2
0
12
6
4
.1.2.14. 3-[E-(3,5-bis(trifluoromethyl)styryl)]-2-chloro-6-methoxyquinoline (14B). Solid
1
yield = 2.1%; mp. 170-172°C; H NMR (500 MHz, CDCl ): δ 3.93 (s, 3H, -OCH ), 7.14 (d, 1H,
3
3
J = 2.5Hz), 7.24 (d, 1H, J = 15 Hz), 7.41 (dd, 1H, J = 2.5 & 9.2Hz) 7.67 (d, 1H, J = 15Hz), 7.85
+
(
s, 1H), 9.95 (d, 1H, J = 9.0Hz), 8.01 (s, 2H), 8.34 (s, 1H); MS (EI) 431.1 (M ); HRMS
+
+
calculated for C H ClF NO [M] 431.05116, found [M] 431.05083.
2
0
12
6
4
.1.2.15. 3-[Z-(3,5-bis(trifluoromethyl)styryl)]-2-chloroquinoline (15A). Solid; yield =
1
3
3.5%; mp. 103-105°C; H NMR (500 MHz, CDCl ): δ6.89 (d, 1H, J=12.2 Hz), 7.00 (d, 1H,
3
J=12.2 Hz), 7.50 (t, 1H, J=7.5Hz), 7.57 (d, 1H, J=7.5Hz), 7.65 (s, 2H), 7.71 (s, 1H), 7.73 (t, 1H,
13
J=8.25Hz), 7.90 (s, 1H), 8.03 (d, 1H J=8.45Hz) C NMR (125 MHz, CDCl ): δ 121.23 (m),
3
1
1
4
21.86, 121.86, 126.60, 127.29, 127.48, 128.23, 128.81(bs, 2C), 129.06, 130.07, 131.80 (2C),
+
+
37.90, 138.08, 147.17, 149.90; MS (EI) 401.0 (M ); HRMS Calculated for C H ClF N [M]
1
9
10
6
+
01.04060, found [M] 401.04327.
4
.1.2.16. 3-[Z-(3,5-bis(trifluoromethyl)styryl)]-2,7-dichloroquinoline (16A). Solid, yield
1
=
24.7%; mp. 138-139°C: H NMR (500 MHz, CDCl ): δ 6.91 (d, 1H, J = 12Hz), 6.97 (d, 1H, J =
3
1
8
2Hz), 7.46 (d, 1H, J = 10Hz), 7.50 (d, 1H, J = 12 Hz), 7.63 (s, 2H), 7.72 (s, 1H), 7.87 (s, 1H),
+
+
.02 (s, 1H); MS (EI) 435.0 (M ); HRMS calculated for C H Cl F N [M] 435.00162, found
1
9
9
2 6
+
[
M] 435.00456.
4
2
1
=
.1.2.17. 3-[Z-(2,4-bis(trifluoromethyl)styryl)]-2-chloroquinoline (17A). Solid; yield =
1
6.7%; mp. 147-150°C; H NMR (500 MHz, CDCl ): δ 7.07 (d, 1H, J=12.0Hz), 7.15 (d, 1H, J =
3
2.0 Hz), 7.23 (d, 1H, J=10Hz), 7.47 (m, 3H), 7.60 (s, 1H), 7.70 (t, 1H, J = 10Hz), 9.99 (d, 2H, J
13
12Hz); C NMR (125 MHz, CDCl ): δ 122.36, 123.44 (m), 124.54, 126.71, 127.34, 127.56,
3
1
28.26, 128.56 (bs), 128.93, 129.64, 129.82, 129.88, 130.15, 130.84, 131.86, 133.72, 133.98,
2
5

+
+
1
47.00, 150.12; MS (EI) 401.0 (M ); HRMS calculated for C H ClF N [M] 401.0406, found
19 10 6
+
[
M] 401.03806.
4
.1.2.18. 3-[E-(2,4-bis(trifluoromethyl)styryl)]-2-chloroquinoline (17B). Solid; yield = 5.7%;
1
mp. 165-167°C; H NMR (500 MHz, CDCl ): δ 7.54 (d, 1H, J=15Hz), 7.61 (m, 2H), 7.76 (t, 1H
3
1
3
J=5Hz), 7.88 (d, 1H, J = 5Hz), 7.91 (d, 1H, J = 10Hz), 7.99 (m, 3H), 8.41 (s, 1H); C NMR (125
MHz, CDCl ): δ 122.27 , 123.41 (m), 124.44 , 127.19, 127.30, 127.58, 127.87, 128.21, 128.39,
3
1
4
28.51, 128.96 (bs), 129.10, 130.03, 130.35, 130.99, 134.90, 139.14, 147.37, 149.89;MS (EI)
+
+
+
01.0 (M ); HRMS calculated for C H ClF N [M] 401.0406, found [M] 401.04026.
1
9
10
6
4
.1.2.19. 3-[Z-(2,4-bis(trifluoromethyl)styryl)]-2-chloro-6-methoxyquinoline (18A). Solid;
1
yield = 57.4%; mp. 140-143°C; H NMR (500 MHz, CDCl ): δ 3.84 (s, 3H), 6.74 (d, 1H,
3
J=2.5Hz), 7.04 (d, 1H, J=12Hz), 7.12 (d, 1H, J=12Hz), 7.23 (d, 1H, 8.1Hz), 7.33 (dd, 1H, J =
1
3
9
.2&2.6Hz), 7.48 (md, 2H, J=7Hz), 7.87 (d, 1H, J=9.2Hz), 7.98 (s, 1H); C NMR (125 MHz,
CDCl ):δ 55.58, 104.92, 122.39, 123.39 (m), 124.57, 127.86, 128.51(bs), 128.70, 129.60, 129.60,
3
1
29.77, 129.84, 130.07, 131.94, 131.94, 137.45, 139.04, 143.04, 147.39, 158.35; MS (EI) 431.1
+
+
+
(
M ); HRMS Calculated for C H ClF NO [M] 431.05116, found [M] 431.05484.
20 12 6
4
.1.2.20. 3-[E-(2,4-bis(trifluoromethyl)styryl)]-2-chloro-6-methoxyquinoline (18B). Solid;
1
yield = 4.5%; mp. 175-177°C; H NMR (500 MHz, CDCl ): δ 3.98 (s, 3H), 7.16 (d, 1H,
3
J=2.5Hz), 7.39 (dd, 1H, J=9.1 & 2.6Hz), 7.52 (d, 1H, J = 16Hz), 7.59 (d, 1H, J = 16Hz), 7.87 (d,
1
3
1
H, J = 8.3Hz), 7.91 (d, 1H, J = 9.1Hz), 7.98 (s, 1H), 8.01 (d, 1H, J = 8.1Hz), 8.30 (s, IH); C
NMR (125 MHz, CDCl ):δ55.67, 105.13, 122.27, 123.41 (m), 123.86, 124.57, 126.95, 128.19,
3
1
1
4
28.44, 128.98 (bs), 129.12, 129.70, 130.03, 130.17, 130.30, 133.62, 139.18, 143.54, 147.27,
+
+
+
58.54; MS (EI) 431.1 (M ); HRMS calculated for C H ClF NO [M] 431.05116, found [M]
2
0
12
6
31.05378.
4
.1.2.21. 3-[Z-(3,5-bis(trifluoromethyl)styryl)]-2-chloro-6,7-methelenedioxyquinoline (19A).
1
Solid; yield = 64%; mp. 164-165°C; H NMR (500 MHz, CDCl ): δ 6.21 (s, 2H), 6.77 (s, 1H),
3
6
1
.83 (d, 1H, J = 12.1 Hz), 6.95 (d, 1H, J = 12.1Hz), 7.30 (s, 1H), 7.65 (s, 2H), 7.69 (d, 2H, J =
13
2.8 Hz); C NMR (125 MHz, CDCl ): δ 102.16. 102.19, 105.06, 121.08 (m), 121.91, 121.91,
3
2
6

1
1
4
23.72, 126.05, 128.82, 128.82 (bs), 129.19, 129.41, 131.71 (2C), 136.64, 138.08, 145.73,
+
+
47.67, 148.59, 152.00; MS (EI) 445.0 (M ); HRMS calculated for C H ClF NO [M]
2
0
10
6
2
+
45.03043, found [M] 445.02965.
4
.1.2.22. 3-[E-(3,5-bis(trifluoromethyl)styryl)]-2-chloro-6,7-methelenedioxyquinoline (19B).
1
Solid; yield = 5.2%; mp. 201-204°C; H NMR (500 MHz, CDCl ): δ 6.17 (s, 2H), 7.11 (s, 1H),
3
7
.19 (d, 1H, J = 16.2 Hz), 7.33 (s, 1H), 7.63 (d, 1H, J = 16.2 Hz), 7.83 (s, 1H), 7.99 (s, 2H), 8.23
+
+
+
(
s, 1H); MS (EI) 445.0 (M ); HRMS calculated for C H ClF NO [M] 445.03043, found [M]
20 10 6 2
4
45.03337.
4
.1.2.23. 3-[Z-(3-(trifluoromethyl)styryl)]-2-chloro-6,7-methelenedioxyquinoline (20A).
1
Solid; yield = 64%; mp. 123-125°C; H NMR (500 MHz, CDCl ): δ 6.10 (s, 2H), 6.77 (s, 1H),
3
6
7
1
1
.83 (s, 2H), 7.26 (d, 1H, J=10Hz), 7.29 (s, 1H), 7.32 (d, 1H, J = 7.85 Hz), 7.45 (d, 1H, J=10Hz),
13
.50 (s, 1H), 7.69 (s, 1H); C NMR (125 MHz, CD OD):δ 102.06, 102.30, 105.00, 122.77,
3
23.88, 124.22 (q, J=3.75), 125.76 (q, J=3.75), 127.01, 127.30, 128.85, 130.86, 131.13, 131.89,
+
36.76, 137.00, 145.45, 148.00, 148.37, 151.71;MS (EI) 377.0 (M ).
4
.1.2.24. 3-[E-(3-(trifluoromethyl)styryl)]-2-chloro-6,7-methelenedioxyquinoline (20B).
1
Solid; yield = 10%; mp. 162-164°C; H NMR (500 MHz, CDCl ): δ 6.12 (s, 2H), 7.05 (s, 1H),
3
7
7
1
1
.10 (d, 1H, J=15Hz), 7.26 (s, 1H), 7.49 (m, 2H), 7.56 (d, 1H, J = 10 Hz), 7.74 (d, 1H, J=5Hz),
13
.78 (s, 1H), 8.15 (s, 1H); C NMR (125 MHz, CD OD):δ 102.11, 102.49, 105.03, 122.96,
3
23.61 (q, J=5Hz), 124.47, 124.76 (q, J=3.75), 125.37, 127.42, 129.29, 129.69, 130.53, 131.15,
+
32.82, 137.46, 145.42, 147.80, 148.53, 151.67; MS (EI) 377.0 (M ); HRMS calculated for
+
+
C H ClF NO [M] 377.04304, found [M] 377.04464.
19
11
3
2
4
4
6
.1.2.25. 3-[Z-(2-(trifluoromethyl)styryl)]-2-chloro-6-methoxyquinoline (21A). Solid; yield =
1
6.3%; mp. 143-145°C; H NMR (500 MHz, CDCl ): δ 3.83 (s, 3H), 6.70 (d, 1H, J = 2.5Hz),
3
.95 (d, 1H, J = 11.95), 7.10 (d, 1H, J = 7.70 Hz), 7.15 (d, 1H, J = 11.95 Hz), 7.22 (t, 1H, J = 7.5
1
3
Hz), 7.28 (m, 2H), 7.52 (s, 1H), 7.24 (d, 1H, J = 7.85), 7.86 (d, 1H, J = 9.25); C NMR (125
MHz, CDCl ):δ 55.52, 105.07, 123.07, 123.17, 125.99 (q, J = 5.3Hz), 127.53, 127.92, 128.19,
3
2
7

1
28.74, 129.35, 129.54, 130.29, 131.13, 131.73, 135.19, 137.51, 142.86, 147.80, 158.18; MS
+
+
+
(
EI) 377.0 (M ); HRMS calculated for C H ClF NO [M] 363.06378, found [M] 363.06815.
19 13 3
4
.2. Biological experiments
All of the cell lines used were purchased from ATCC and authenticated using a short
tandem repeat (STR) profiling method by Genetica DNA Laboratories (Burlington, NC) (March
015; July 2015). HeLa, MDA-MB231 and MDA-MB468 were grown in RPMI 1640 and MCF7
2
in DMEM. The culture media were supplemented with 10% fetal bovine serum (FBS) and a
combination of penicillin (100 units/ml) and streptomycin (100 µg/ml). Non-cancer cells
(MCF10A and 184B5) were grown in DMEM/F12 supplemented with 10% FBS, 20 ng/ml
epidermal growth factor, 10 μg/ml insulin, 0.5 g/ml hydrocortisone and a combination of
penicillin (100 units/ml) and streptomycin (100 µg/ml). Cells were grown under humidified
condition at 37ºC with 5% CO and 95% air.
2
4
.2.1. Sulforhodamine B (SRB) cell proliferation assay
Antiproliferative activities of compounds were examined by an SRB assay as described
3
3
previously. Briefly, 4,000-5,000 cells/well were incubated for 16 h in a 96-well clustered plate.
After 16 h of incubation, culture medium was replaced with fresh medium containing different
dilutions of test compounds dissolved in DMSO. Some wells were treated with 100 µl of 10%
trichloroacetic acid (TCA) as a negative control, and a sham (medium with DMSO) treated
sample was used as a positive control. Culture medium was removed after 72 h of incubation,
and cells were fixed with 10% TCA at 4ºC for 1 h. TCA was removed and cells were washed
with cold tap water, and plate was air-dried, followed by addition of 50 µl of 0.4% SRB staining
solution to each well. After 30 min incubation, SRB staining solution was removed. Cells were
washed with 1% acetic acid solution, and then washed with tap water to remove unbound SRB,
followed by air-drying. 200 µl of 10 mmol (pH 10.5) trizma base buffer was added to each well
to solubilise cells, and plate was then read at 540 nm wave-length using an automated plate
reader (Synergy H4 Hybrid Multi-Mode Microplate Reader, BioTek). Cell growth was
calculated by the following formula: % cells proliferation = [(ATCT)/ (STCT)] x100
2
8

Where AT is absorbance of treated cells; CT is absorbance of negative control cells; and ST is
absorbance of sham treated cells. IC values were calculated from sigmoidal dose-response
5
0
curves generated by two independent biological replicates, with four technical replicates in each
set by using Graph Pad Prism v.5.04 software.
4
.2.2. Cell cycle analysis
6
Approximately1×10 cells per plate were seeded overnight. Cells were then treated the
next morning with test compounds, and harvested at scheduled time points post-treatment. Cell
TM
pellet was collected by centrifugation at 1,100 rpm (Allegra X-12 centrifuge, Beckman
Coulter) for 5min followed by washing twice with PBS, and fixing cells with 75% ethanol for
TM
1
2-24 h at -20ºC. Ethanol was removed by centrifugation at 1,100 rpm (Allegra X-12
centrifuge) for 5 min, and cells were suspended in PBS and centrifuged again 1,100 rpm
TM
(
Allegra X-12 centrifuge) for 5 min. PBS was removed and cell pellet was resuspended in
propidium iodide (PI) staining solution containing 0.3% nonidet P-40,100 µg/ml RNase A and
00 µg/ml PI in PBS for 1 h. DNA content in the different phase of the cell cycle was analysed
by flow cytometry using a Beckmann Coulter Cytomics FC500.
1
4
.2.3. Western blot analysis
Samples treated with compound 12A and sham control were collected at scheduled time
TM
points post-treatment and centrifuged for 5 min. at 1,100 rpm (Allegra X-12 centrifuge,
Backman Coulter). Cell pellets were washed with PBS by centrifugation for 5 min. at 1,100 rpm
TM
(
Allegra X-12 centrifuge), and were then lysed with 100 µl Lysis buffer (150 mmol NaCl, 5
mmol EDTA, 1% triton X-100, 10 mmol tris pH 7.4, 1 mmol PMSF, 5 mmol EDTA and 5 mmol
protease inhibitors cocktail) by maintaining on ice for 10-15 min, followed by centrifugation at
TM
1
,100 rpm (Allegra X-12 centrifuge) for 10 min at 4 ºC. Supernatant was collected and protein
concentration was determined using a BCA protein assay kit (Thermo Scientific, USA). Cell
lysates were diluted with 2× Laemmli sample buffer, and then boiled at 95-100ºC for 5 min. 30-
4
0 µg protein was resolved by polyacrylamide gel (8 or 10%) electrophoresis. The resolved
proteins were transferred to a PVDF membrane using a semi-dry gel transfer apparatus (75 min
2
9