Novel cinnamaldehyde-based aspirin derivatives for the treatment of colorectal cancer

Article abstract of DOI:10.1016/j.bmcl.2018.07.032

Colorectal cancer (CRC) is a leading cause of mortality worldwide. Current treatments of CRC involve anti-cancer agents with relatively good efficacy but unselectively target both cancer and non-cancer cells. Thus, there is a need to discover and develop

Full text of DOI:10.1016/j.bmcl.2018.07.032

Accepted Manuscript
Novel cinnamaldehyde-based aspirin derivatives for the treatment of colorectal
cancer
Shan Lu, Obinna N. Obianom, Yong Ai
PII:
S0960-894X(18)30612-7
https://doi.org/10.1016/j.bmcl.2018.07.032
BMCL 25963
DOI:
Reference:
Bioorganic & Medicinal Chemistry Letters
To appear in:
Received Date:
Revised Date:
Accepted Date:
14 May 2018
1 July 2018
18 July 2018
Please cite this article as: Lu, S., Obianom, O.N., Ai, Y., Novel cinnamaldehyde-based aspirin derivatives for the
treatment of colorectal cancer, Bioorganic & Medicinal Chemistry Letters (2018), doi: https://doi.org/10.1016/
j.bmcl.2018.07.032
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.

Novel cinnamaldehyde-based aspirin derivatives for the treatment of
colorectal cancer
Shan Lu, a, * Obinna N Obianom, b Yong Ai b, **
aCollege of Pharmacy, Hubei University of Chinese Medicine, Hubei 430065, PR
China
bDepartment of Pharmaceutical Sciences, University of Maryland School of
Pharmacy, Baltimore, MD 21201, United States
* Corresponding author. College of Pharmacy, Hubei University of Chinese Medicine,
Hubei 430065, PR China. E-mail address: lushan9805@163.com (S. Lu).
** Corresponding author. Department of Pharmaceutical Sciences, University of
Maryland School of Pharmacy, Baltimore, MD 21201, United States. E-mail address:
aiyong0508@126.com (Y. Ai).
1

Abstract
Colorectal cancer (CRC) is a leading cause of mortality worldwide. Current
treatments of CRC involve anti-cancer agents with relatively good efficacy but
unselectively target both cancer and non-cancer cells. Thus, there is a need to discover
and develop novel CRC therapeutics that have potent anti-cancer effects, but show
reduced off-target cell effects. Here, a novel series of cinnamaldehyde-based aspirin
derivatives were designed and synthesized. Biological evaluation indicated that the
most active compound 1f exhibited more than 10-fold increase in the
anti-proliferation efficacy in HCT-8 cells compared to the parent compounds. Its
effects were similarly reproduced in another CRC cell line, DLD-1, but with 7- to
11-fold less inhibitory activity in non-tumorigenic colon cells. Flow cytometry
analysis showed that 1f induced cell cycle arrest and apoptosis, which was further
validated with immunoblot analysis of the relative protein levels of cleaved caspase 3
and PARP as well as the ROS production in CRC cells. More so, 1f significantly
inhibited the growth of implanted CRC in vivo in mouse xenograft model. Taken
together, our results show that cinnamaldehyde-based aspirin derivatives such as 1f
show promise as novel anti-CRC agent for further pharmaceutical development.
Keywords: aspirin, cinnamaldehyde, colorectal cancer, anti-proliferative activity,
apoptosis
2

Colorectal cancer (CRC) is the third most commonly diagnosed malignancy and
the fourth leading cause of cancer-related death worldwide [1, 2]. Chemotherapeutic
drugs like 5-Fluorouracil (5-FU), Capecitabine (Xeloda), Irinotecan (Camptosar),
Oxaliplatin (Eloxatin), and Trifluridine and tipiracil (Lonsurf, a combination drug in
pill form) are widely used for the treatment of CRC. However, in many cases, they
produce mild to grave side effects such as hair loss, mouth sores, nausea, diarrhea,
immunosuppression, bleeding, and fatigue [3-7]. In recent years, with various novel
targets and therapeutic strategies discovered in targeting CRC, a number of groups
have developed newer drugs to specifically alter these new targets. For example, the
rapid increase in development of vascular endothelial growth factor (VEGF) and
epidermal growth factor receptor (EGFR) inhibitors [8]. Nonetheless, these new
treatments still pose serious side effects that include blood clotting, severe bleeding,
colonic perforations, slow wound healing, cardiomyopathy and nephropathy.
Therefore, novel therapeutics for CRC with reduced side effects is still urgently
needed.
Aspirin (ASA, 2, Fig. 1), a widely used nonsteroidal anti-inflammatory drug
(NSAID), has been shown to be effective for the prevention and remission of cancers,
especially in CRC [9-13]. In fact, the Colorectal Adenoma/Carcinoma Prevention
Program (CAPP2) randomized clinical trial in carriers of hereditary forms of CRC,
lynch syndrome, demonstrated substantial protection by ASA in these patients [14].
They found that long-term treatment with 600 mg of ASA per day substantially
reduced cancer incidence in lynch syndrome patients in a span of 55.7 months [14].
As a result, many synthetic ASA derivatives have now been prepared (Fig. 1),
including NO- and HNO-releasing ASA (3 and 4) [15-19], HS-releasing ASA (5)
[20-22], dual nitric oxide- and hydrogen sulfide-releasing ASA (6) [23, 24],
phospho-ASA (7) [25], 6-gingerol-based ASA prodrug (8) [26], resveratrol-based
ASA prodrug (9) [27], and Se-ASA (10) [28]. These synthetic ASA derivatives
3

possess stronger cytotoxicity and chemo-preventive effects especially against CRC
cells than ASA. Another ASA derivative extensively reported in the literature is the
Asplatin (11), a Pt (IV) prodrug of cisplatin containing the ASA moiety, which
exhibits significantly higher cytotoxicity than cisplatin towards tumor cells and almost
fully overcomes drug resistance in cisplatin resistant cells [29, 30]. In the same way,
further discovery and development of novel ASA derivatives with improved
anti-cancer selectivity, but lower off-target cellular effects is warranted. One means to
achieve this objective is by hybrid drug approach as demonstrated with Asplatin.
Cinnamaldehyde (CA, 12, Fig. 1) is an important bioactive ingredient found in the
cinnamon [31]. It is widely used in food and beverages and has been known to exert
anticancer activity through reactive oxygen species (ROS) generation [32, 33].
Although CA contains active Michael acceptor pharmacophore, it is not genotoxic or
carcinogenic in vivo [34]. Furthermore, it is considered generally recognized as safe
(GRAS) and widely used in foodstuffs in the United States and European countries
[35]. More importantly, CA has been reported to induce ROS generation mainly in the
mitochondria and inhibit growth of human cancer cells, with minimal cytotoxicity to
normal cells [36, 37].
4

Fig. 1. Chemical structures of ASA, representative synthetic ASA derivatives and CA.
Considering the aforementioned chemo-preventive effects of ASA and its
derivatives and the biological effects of CA, along with the previous reports that
support the modification of ASA to obtain more potent and selective antitumor agents,
we designed novel hybrid CA-ASA derivatives. We hypothesized that these hybrids
will exert potent cytotoxicity against CRC cells with good selectively compared to
normal cells. Accordingly, we synthesized this novel class of CA-based ASA
derivatives (1a-h, Scheme 1), and evaluated their bioactivity in vitro and in vivo.
The target compounds 1a-h were synthesized as depicted in Scheme 1. Initially,
aspirin 2 was converted to the acid chloride 13 upon treatment with thionyl chloride.
Then the corresponding cinnamaldehyde 15a-h were prepared by the Wittig reaction
[38-41] starting from aldehydes 14a-h in higher yields (55-79%). Next, reaction of
acid chloride 13 with 15a-h in dry CH2Cl2, in the presence of the non-nucleophilic
base triethylamine (TEA), furnished the respective target compounds 1a-h in good
yields (75-89%). The structures of all hybrids were fully characterized by their 1H and
13C NMR spectroscopic data as well as by mass spectral analysis (data in
Supplementary Materials).
Scheme 1. Synthesis of cinnamaldehyde-based aspirin derivatives 1a-h. Reagents and
conditions: a) SOCl2, anhydrous CH2Cl2, reflux, 2h; b) Ph3P=CHCHO, PhMe, 60°C,
8h, N2; c) 3, TEA, anhydrous CH2Cl2, rt, 12h.
5

Compounds 1a-h were screened for their inhibitory activity against HCT-8 by
measuring cell proliferation using MTT assay (Supplementary Materials) with
5-Fluorouracil (5-FU) as a positive control. As shown in Table 1, compounds 1b and
1e-h exhibited more potent cell growth inhibitory activity (IC50 = 2.6-9.3 μM) against
HCT-8 cells compared to 5-FU (IC50 = 28.9 μM), ASA (IC50 > 100 μM) and CA (IC50
= 29.4 μM), and even the combination of ASA and CA (IC50 = 15.6 μM, physical
mixture, mole ratio, 1:1). Among them, compounds 1f exhibited the most potent
inhibitory activity (IC50 = 2.6 μM), which was 11, 38, 4, and 2.5 fold higher than that
of the positive control 5-FU, two individual moieties 16 (IC50 > 100 μM, methyl ester
of aspirin) and 17 (IC50 = 10.5 μM, O-acetylated cinnamaldehyde derivative,
Supplementary Materials), and the physical combination of 16 and 17 (IC50 = 6.6 μM),
respectively. These results suggest that the antitumor activity of 1f may be most likely
attributed to the synergic effects derived from both the cinnamaldehyde and aspirin
moieties. It was obvious that R substitution on the benzene ring of the CA as in 1b-h
was crucial for their anticancer activity. Generally, the compounds with substituents
like methoxy, or methyl, or halogen (1b-h) displayed better inhibitory activity than
that of 1a without any substitution at the R position. Substitutions on the carbon 4 (1b)
led to preferable anticancer activity than others (1c and 1d). More distinctly, the
compounds with an electron-withdrawing group such as halogens (1f-h) on the
benzene ring showed more potent inhibitory activity than those bearing an
electron-donating group such as methoxy, or methyl (1b and 1e).
Table 1. The anti-proliferative activity of compounds 1a-h.
6

IC50 (μM)a
HCT-8
Compounds
R
H
20.3 ± 1.7
9.3 ± 0.7
15.8 ± 1.1
17.6 ± 1.0
8.5 ± 0.4
2.6 ± 0.6
4.3 ± 0.9
5.6 ± 0.7
>100
1a
1b
4-OMe
5-OMe
6-OMe
4-Me
4-F
1c
1d
1e
1f
4-Cl
1g
4-Br
1h
ASA
CA
29.4 ± 1.3
15.6 ± 1.6
>100
ASA+CA
16
10.5 ± 1.3
6.6 ± 0.9
28.9 ± 2.7
17
16+17
5-FU
aData are expressed as mean IC50 (µM) ± SD of each compound from three
independent experiments. IC50 is the drug concentration inhibiting 50% of the cell
proliferation.
Subsequently, the most potent compound 1f was further investigated for its
antitumor efficacy and selectivity in other cell lines by MTT assay (Table 2). We
found that 1f displayed more potent activity (IC50 = 4.3 μM) than 5-FU (IC50 = 33.0
μM) against a human colorectal adenocarcinoma cell line, DLD-1. While 5-FU had
similar inhibitory activity in both CRC and nontumor CCD841 cells, 1f exhibited a 7-
to 11-fold less anti-proliferative activity against CCD841 cells (IC50 = 35.7 μM). This
suggests that 1f may have anti-proliferative activity selectively against CRC cells
relative to normal cells.
7

Table 2. The anti-proliferative activity of compound 1f.
IC50 (μM) a
Compounds
DLD-1
4.3 ± 0.5
33.0 ± 2.1
CCD841
30.7 ± 2.9
35.7 ± 3.2
1f
5-FU
aData are expressed as mean IC50 (µM) ± SD of each compound from three
independent experiments. IC50 is the drug concentration inhibiting 50% of the cell
proliferation.
To evaluate the in vivo anticancer activity of 1f, BALB/c nude mice were
inoculated subcutaneously with HCT-8 cells. After the establishment of solid tumor,
the mice were randomly treated intraperitoneally with 1f (60 mg/kg) or the vehicle
consisting of Tween 80/PBS (5:95) daily for 32 consecutive days, respectively.
Treatment with 60 mg/kg of 1f significantly reduced the volume of implanted colon
tumors (P < 0.001 vs. the vehicle-treated control, Figs. 3A and 3B). More so, the
tumor weights in the mice treated with 1f at 60 mg/kg were 47% (w/w) less than that
in the vehicle-treated controls (1.95 ± 0.42 g versus 1.03 ± 0.18 g, P < 0.001, Fig. 3C)
while the body weights of the mice were not significantly changed by treatment with
1f (Fig. 3D). Together, our data clearly demonstrated that 1f inhibited the growth of
implanted tumors in vivo with little off-target cell toxicity.
8

A
B
1000
800
600
400
200
0
Vehicle
1f 60 mg/kg
Vehicle
1f 60 mg/kg
0
10
20
30
40
Days
C
D
2.5
2.0
1.5
1.0
0.5
0.0
26
24
22
20
18
Vehicle
1f 60 mg/kg
***
0
10
20
30
40
Vehicle
1f 60 mg/kg
Days
Fig. 3. Anticancer effects of 1f in mice inoculated with HCT-8 cells. (A) Images of the
tumors. The effect of intraperitoneal 1f (60 mg/kg) on tumor volumes (B), tumor
weight (C) and body weight (D) in a HCT-8 xenograft model. Data are expressed as
the mean ± SD from each group of mice (n = 6 per group). ( ∗∗∗) P < 0.001 vs the
vehicle-treated.
To understand the mechanisms underlying the action of 1f, the effects of 1f on
CRC cell cycle were examined. HCT-8 cells were treated with varying concentrations
(0, 3, 6, and 12 μM) of 1f for 24 h and then stained with propidium iodide (PI),
followed by flow cytometry analysis. Figure 4 shows that treatment with 1f induced
HCT-8 cell cycle arrest at G1 in a concentration dependent manner – 40.05% (0 μM),
43.17% (3 μM), 48.66% (6 μM), and 51.62% (12 μM) respectively, which were
significantly higher than that of vehicle-treated control. As a result, the fraction of
CRC cells at the G2 phase was significantly lower in compound treated than
vehicle-treated control. This suggests that the anti-cancer mechanisms of 1f may
include its ability to induce cycle arrest in the cells.
9

Fig. 4. Effects of 1f on cell cycle distribution in HCT-8 cells. Cells were treated with
the indicated concentrations of 1f for 24 h and stained with PI, followed by flow
cytometry analysis.
To determine whether the inhibitory effects of 1f on CRC cell proliferation are
accompanied by enhanced apoptosis, we next carried out Annexin V-FITC and PI
staining on 1f treated cells. HCT-8 cells were incubated with vehicle, or 1f (3, 6, and
12 μM) for 24 h. Then flow cytometry analysis was used to determine the apoptotic
status of the cells. We observed that treatment with 1f induced apoptosis in HCT-8
cells in a concentration dependent manner (Fig. 5). Increasing concentration of 1f
specifically increased the fraction of cells undergoing early apoptosis as indicated by
the lower right quadrant. There was not no significant difference in the early apoptosis
of the cells between 6 and 12 µM. However, in the upper right quadrant indicating
cells in the late apoptotic state, increase from 0 to 3 to 6 to 12 µM 1f led to an
10

increase from 2.40 to 5.42 to 17.03 to 27.71 % of cells in this quadrant. This suggests
that 1f treatment significantly increases apoptosis in cancer cells as a result of its
anti-cancer effects. Furthermore, apoptosis rate was significantly decreased in
N-acetyl-L-cysteine (NAC, ROS scavenger) co-treated cells (12.98%) compared to 1f
(6 μM)-treated cells (32.59%). These data demonstrate that apoptosis resulting from
1f treatment is due to its induction of intracellular ROS in HCT-8 cells.
Fig. 5. Effects of 1f on the induction of apoptosis. HCT-8 cells were treated with the
indicated concentrations of 1f or vehicle (control) for 24 h. The cells were stained
with Annexin V-FITC and propidium iodide (PI), followed by flow cytometry
analysis.
To understand the mechanism of 1f-induced apoptosis, we further examined the
changes of the intracellular proteins related to apoptosis such as caspase-3, and PARP
in HCT-8 cells treated with 1f (0, 3, 6, and 12 μM). As shown in Figure 6, compound
1f significantly induced the relative levels of cleaved caspase 3 and
poly(ADP-ribose)polymerase (PARP) in a dose-dependent manner in HCT-8 cells.
Overall, our results confirm that compound 1f inhibits the growth of CRC cells by
inducing apoptosis in addition to its effects on cell cycle.
11

Fig. 6. Effect of 1f on the expression of apoptotic markers. The relative protein
expression levels of PARP and Caspase 3 were measured following 1f treatment in
HCT-8 cells. Cells were treated with increasing concentrations of 1f or control (0.1%
DMSO) for 24 h and immunoblotting of the whole cell lysate was carried out to
determine the protein levels.
Studies have shown that the anticancer effect of CA is associated with increase in
ROS levels [32, 37, 42]. To investigate whether induced cytotoxic effect of 1f is
related to ROS production, we measured ROS levels in 1f-treated CRC cells by flow
cytometry analysis. Again, we treated HCT-8 with different concentrations (0, 3, 6,
and 12 μM) of 1f, and measured the ROS levels. As shown in Figure 7, a dramatic
increase (from 3.67 to 47.83) in the levels of ROS generation in HCT-8 cells was
observed after treatment with 1f. These results suggest that the induction of cell death
by 1f may be due to ROS production.
12

Fig. 7. Effect of compound 1f on ROS levels in CRC cells. Flow cytometry analysis
of HCT-8 cells treated with the indicated concentrations of 1f for 24 h and stained
with the fluorogenic dye 2', 7'-dichlorofluorescin diacetate (DCFH-DA).
In summary, we have designed and synthesized a novel group of CA-based ASA
derivatives 1a-h. We found that compound 1f had potent and selective
anti-proliferative activity against CRC cells relative to normal cells in vitro.
Furthermore, treatment with 1f significantly inhibited the growth of implanted CRC
cancer in mice. In addition, treatment with 1f induced CRC cell cycle arrest at G1
phase and apoptosis in vitro, increased the relative levels of cleaved caspase 3 and
PARP as well as the ROS in HCT-8 cells. These findings may provide a proof of
principle that 1f may be potential chemotherapeutic agents for the intervention of
CRC.
Acknowledgments
This work was supported by grants from Qing Miao Plan of Hubei University of
Chinese Medicine (2016ZZX013).
References
[1] R.L. Siegel, K.D. Miller, A. Jemal, CA Cancer J. Clin. 66 (2016) 7-30.
[2] M. Arnold, M.S. Sierra, M. Laversanne, et al. Gut 66 (2016) 683-691.
13

[3] D.B. Longley, D.P. Harkin, P.G. Johnston, Nat. Rev. Cancer. 3 (2003) 330-338.
[4] B.R. Hirsch, S.Y. Zafar, Cancer Manag. Res. 3 (2011) 79-89.
[5] K. Fujita, Y. Kubota, H. Ishida, et al. World J. Gastroenterol. 21 (2015)
12234-12248.
[6] P. Comella, R. Casaretti, C. Sandomenico, et al. Ther. Clin. Risk Manag. 5 (2009)
229-238.
[7] T. Kish, P. Uppal, P T. 41 (2016) 314-317, 325.
[8] K. Jin, H. Lan, F. Cao, et al Int. J. Oncol. 41 (2012) 583-588.
[9] C. Sostres, C.J. Gargallo, A. Lanas, World J. Gastrointest. Pharmacol. Ther. 5
(2014) 40-49.
[10] G.J. Tsioulias, M.F. Go, B. Rigas, Curr. Pharmacol. Rep. 1 (2015) 295-301.
[11] P.C. Elwood, A.M. Gallagher, G.G. Duthie, et al. Lancet 373 (2009) 1301-1309.
[12] M.A. Thorat, J. Cuzick, Curr. Oncol. Rep. 15 (2013) 533-540.
[13] M.J. Thun, E.J. Jacobs, C. Patrono, Nat. Rev. Clin. Oncol. 9 (2012), 259-267.
[14] J. Burn, A.M. Gerdes, F. Macrae, et al. Lancet 378 (2011) 2081-2087.
[15] D. Basudhar, G. Bharadwaj, R.Y. Cheng, et al. J. Med. Chem. 56 (2013)
7804-7820.
[16] V. Kozoni, T. Rosenberg, B. Rigas, Acta Pharmacol. Sin. 28 (2007) 1429-1433.
[17] C.V. Rao, B.S. Reddy, V.E. Steele, et al. Mol. Cancer Ther. 5 (2006) 1530-1538.
[18] J.L. Williams, S. Borgo, I. Hasan, et al. Cancer Res. 61(2001) 3285-3289.
[19] J.L. Williams, N. Nath, J. Chen, et al. Cancer Res. 63 (2003) 7613-7618.
[20] M. Chattopadhyay, R. Kodela, N. Nath, et al. Biochem. Pharmacol. 83 (2012)
723-732.
[21] M. Chattopadhyay, R. Kodela, N. Nath, et al. Biochem. Pharmacol. 83 (2012)
715-722.
[22] M. Chattopadhyay, R. Kodela, N. Nath, et al. Biochem. Pharmacol. 83(2012) 733
-740.
[23] M. Chattopadhyay, R. Kodela, K.R. Olson, et al. Biochem. Biophys. Res.
Commun. 419 (2012) 523-528.
[24] R. Kodela, M. Chattopadhyay, K. Kashfi, ACS Med. Chem. Lett. 3 (2012)
14

257-262.
[25] L. Huang, G.G. Mackenzie, Y. Sun, et al. Cancer Res. 71 (2011) 7617 -7627.
[26] Y. Zhu, F. Wang, Y. Zhao, et al. Sci. Rep. 7 (2017) 40119.
[27] Y. Zhu, J. Fu, K.L. et al. J. Med. Chem. 58 (2015) 6494-6506.
[28] D. Plano, D.N. Karelia, M.K. Pandey, et al. J. Med. Chem. 59(2016) 1946-1959.
[29] Q. Cheng, H. Shi, H. Wang, et al. Chem. Commun. 50 (2014) 7427-7430.
[30] Q. Cheng, H. Shi, H. Wang, et al. Metallomics 8 (2016) 672-678.
[31] M. Friedman, J. Agric. Food Chem. 65 (2017) 10406-10423.
[32] H. Ka, H.J. Park, H.J. Jung, et al. Cancer Lett. 196 (2003) 143-152.
[33] C.M. Cabello, W.B. 3rd Bair, S.D. Lamore, et al. Free Radic. Biol. Med. 46(2009)
220-231.
[34] R. Kiwamoto, D. Ploeg, I.M. Rietjens, et al. Toxicol. In Vitro 31(2016) 114-125.
[35] J. Cocchiara, C.S. Letizia, J. Lalko, et al. Food Chem. Toxicol. 43 (2005)
867-923.
[36] B. Kim, E. Lee, Y. Kim, et al. Adv. Funct. Mater. 23 (2013) 5091-5097.
[37] D.C. Han, M.Y. Lee, K.D. Shin, et al. J. Biol. Chem. 279 (2004) 6911-6920.
[38] L. Lyu, M.Y. Jin, Q. He, et al. Org. Biomol. Chem. 14 (2016) 8088-8091.
[39] R. Maity, S.C. Pan, Org. Biomol. Chem. 16 (2018) 1598-1608.
[40] X.L. Sun, Y.H. Chen, D.Y. Zhu, et al. Org. Lett. 18 (2016) 864-867.
[41] H. Liu, Y. Wang, X. Guo, et al. Org. Lett. 20 (2018) 546-549.
[42] J. Noh, B. Kwon, E. Han, et al. Nat. Commun. 6 (2015) 6907.
15

Graphical Abstract
16

Highlights
A novel series of cinnamaldehyde-based aspirin derivatives were designed and
synthesized.
The most active compound 1f has anti-proliferative activity selectively against CRC
cells in vitro.
Compound 1f significantly inhibited the growth of implanted CRC in vivo.
Compound 1f induced cell cycle arrest and apoptosis, increased the relative levels of
cleaved caspase 3 and PARP as well as the ROS in CRC cells.
17