Synthesis and in?vitro and in?vivo antitumour activity study of 11-hydroxyl esterified bergenin/cinnamic acid hybrids

Article abstract of DOI:10.1016/j.ejmech.2017.03.053

Fourteen bergenin/cinnamic acid hybrids were synthesized, characterized and evaluated for their antitumour activity both in?vitro and in?vivo. The most potent compound, 5c, arrested HepG2 cells (IC₅₀?=?4.23?±?0.79?μM) in the G2/M phase and induced cellular apoptosis. Moreover, compound 5c was also found to suppress the tumour growth in Heps xenograft-bearing mice with low toxicity. In the mechanistic study, 5c administration ignited a mitochondria-mediated apoptosis pathway of HepG2 cell death. Furthermore, 5c activated Akt-dependent pathways and further decreased the expression of the Bcl-2 family of proteins. The downstream mitochondrial p53 translocation was also significantly activated, accompanied by an increase of the caspase-9, caspase-3 activation. These data imply that bergenin/cinnamic acid hybrids could serve as novel Akt/Bcl-2 inhibitors for further preclinical studies.

Full text of DOI:10.1016/j.ejmech.2017.03.053

Accepted Manuscript
Synthesis and in vitro and in vivo antitumour activity study of 11-hydroxyl esterified
bergenin/cinnamic acid hybrids
Chengyuan Liang, Shaomeng Pei, Weihui Ju, Minyi Jia, Danni Tian, Yonghong Tang,
Gennian Mao
PII:
S0223-5234(17)30220-9
10.1016/j.ejmech.2017.03.053
EJMECH 9313
DOI:
Reference:
To appear in: European Journal of Medicinal Chemistry
Received Date: 3 January 2017
Revised Date: 23 March 2017
Accepted Date: 24 March 2017
Please cite this article as: C. Liang, S. Pei, W. Ju, M. Jia, D. Tian, Y. Tang, G. Mao, Synthesis and
in vitro and in vivo antitumour activity study of 11-hydroxyl esterified bergenin/cinnamic acid hybrids,
European Journal of Medicinal Chemistry (2017), doi: 10.1016/j.ejmech.2017.03.053.
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.

ACCEPTED MANUSCRIPT
Synthesis and in vitro and in vivo antitumour activity
study of 11-hydroxyl esterified bergenin/cinnamic
acid hybrids
Chengyuan Liang1,2,* , Shaomeng Pei1, Weihui Ju1, Minyi Jia1, Danni Tian1, Yonghong Tang2, Gennian Mao1,*
1. Faculty of Pharmacy, Shaanxi University of Science and Technology, Xi’an 710021, People’s
Republic of China;
2. Xi’an Taikomed Pharmaceutical Technology Co., LTD, Xi’an 710077, People’s Republic of China;
*Correspondence: C.Y. Liang: E-mail: chengyuanliang@gmail.com Phone: +86-18292490697;
G..N. Mao: E-mail: maogn@sust.edu.cn
Novel bergenin/cinnamic acid hybrids were synthesized and evaluated for their antitumour
activity both in vitro and in vivo. The most potent compound, 5c, arrested HepG2 cells
(IC50=4.23±0.79μM) in the G2/M phase and induced mitochondria-mediated apoptosis, which
could serve as a novel Akt/Bcl-2 inhibitor for further preclinical studies
Abstract
1

ACCEPTED MANUSCRIPT
Fourteen bergenin/cinnamic acid hybrids were synthesized, characterized and evaluated for their
antitumour activity both in vitro and in vivo. The most potent compound, 5c, arrested HepG2 cells
(IC50=4.23±0.79μM) in the G2/M phase and induced cellular apoptosis. Moreover, compound 5c
was also found to suppress the tumour growth in Heps xenograft-bearing mice with low toxicity.
In the mechanistic study, 5c administration ignited a mitochondria-mediated apoptosis pathway of
HepG2 cell death. Furthermore, 5c activated Akt-dependent pathways and further decreased the
expression of the Bcl-2 family of proteins. The downstream mitochondrial p53 translocation was
also significantly activated, accompanied by an increase of the caspase-9, caspase-3 activation.
These data imply that bergenin/cinnamic acid hybrids could serve as novel Akt/Bcl-2 inhibitors
for further preclinical studies.
Keywords: bergenin, cinnamic acid, antitumour, apoptosis, Akt/Bcl-2
1 Introduction
Polyphenols of plant origin are important resources in natural medicament, which are often
used for pharmacological applications [1-3]. Bergenin (Fig. 1), a dihydroisocoumarin derivative,
is the major bioactive component of Saxifragaceae [4]. Bergenin is a C-glucoside of 4-O-methyl
gallic acid isolated from several medicinal plants, such as Bergenial crassifolia [5], Corylopsis
spicata [6], Mallotus japonicus [7], Mallotus philippinensis [8], Caesalpinia digyna [9],
Peltophorum africanum [10], etc. Bergenin is found to possess versatile biological activities,
including anticancer [11], antihepatotoxic [7], antioxidant [12], antiarrhythmic [12], anti-HIV [13],
anti-inflammatory [14] and neuroprotective properties [15]. Because of the few side effects, low
toxicity and lack of drug resistance of bergenin, structural modifications and biological activity
research have been conducted by medicinal chemists [16-21]. Several studies have demonstrated
that bergenin can induce cell cycle arrest and enhance apoptosis in human prostate cancer cell
lines and murine breast cancer cell lines. Additionally, bergenin has an anti-angiogenesis effect,
which could suppress vessel endothelial cell growth [22-24].
Fig. 1
Cinnamic acid (Fig. 2) and its derivatives, which consist of a phenyl ring substituted with
an acrylic acid group, are widely present in propolis, coffee, fruits and wine [25]. Cinnamic acid
exhibits extinguishing pharmacological properties, such as antioxidant [26] and antimicrobial [27].
2

ACCEPTED MANUSCRIPT
Recent studies have also revealed that cinnamic acid possesses potential antitumour activity in
melanoma, glioblastoma, gastric adenocarcinoma, prostate and lung carcinoma cells [28, 29] by
inducing apoptosis and cytoskeleton damage. However, medicinal applications of cinnamic acid
have been hindered by its poor solubility in water and low bioavailability caused by the propenoic
chain.
Fig. 2
In contrast, highly water-soluble bergenin was endowed with two phenolic hydroxyl
groups and three alcoholic hydroxyl groups (one of which is a primary hydroxyl group), which
can be easily modified under molecular hybridization drug R&D practices. Molecular
hybridization is a valuable structural modification concept, which involves the design of new
hybrid compounds with ameliorated efficacy by the combination of pharmacophoric units of
different bioactive moieties targeted at increasing the biological affinity and efficacy and reducing
the undesirable defects compared to the parent compounds [30-33]. Herein, bergenin and cinnamic
acid are both perceived as promising lead compounds for antitumour agents [34, 35]. The rational
design is based on the molecular hybridization between bergenin and a variety of cinnamic acids
aimed at discovering new hybrid compounds with improved aqueous solubility, bioavailability and
anticancer activity. Additionally, with drug-like criteria in mind, the M.W. (reflection of lattice
energy) and logP (key parameter to solubility and permeability) of bergenin/cinnamic acid hybrids
do not violate Lipinski’s Rules of Five.
Based on this hypothesis, a series of bergenin/cinnamic acid hybrids were synthesized and
evaluated for antitumour activity against a panel of human tumour cell lines using 5-FU as a
potent antitumour reference. The most promising molecule was selected for further in vivo
antitumour activity and pharmacological mechanistic studies.
2 Chemistry
The synthetic route of 11-hydroxyl-modified bergenin derivatives (5a-n) is outlined in
Scheme 1. Cinnamic acid (1) and bergenin (3) were bought from J&K Scientific and used as
starting materials. Acid chlorides (2a-n) were synthesized through the reaction between cinnamic
acid analogues (1a-n) and thionyl chloride with CH2Cl2 as a catalyst. For the synthesis of ester
3

ACCEPTED MANUSCRIPT
analogues (5a–n) of bergenin at the C-11 position, the phenolic hydroxyl groups of bergenin were
protected by converting bergenin into dibenzyl bergenin (4) with benzyl bromide at room
temperature using DMF as a catalyst. Finally, compound 4 was treated with acid chlorides (2a-n)
in the presence of dimethylaminopyridine (DMAP）in pyridine and CH2Cl2, followed by
hydrogenation with Pd/C to obtain the bergenin ester derivatives (5a–n), as shown in Table 1.
More significantly, the esterification reaction occurred only at the primary alcohol group in
all cases. Compound 5a in this study was previously synthesized; however, there were no
immediate reports of its antitumour potential. Thus, both compound 5a and the novel compounds
5b-n synthesized herein were evaluated for their antitumour activity. The structures of the
resulting esters were characterized by 1H and 13C NMR.
Scheme 1
Table 1
3 Biological Activity
3.1 Cytotoxicity evaluation
The synthesized compounds, 5a-n, were preliminarily evaluated in vitro for their cytotoxicity
to a panel of five human cancer cell lines consisting of PC-3, SGC-7901, A549, MCF-7 and
HepG2 by applying the standard MTT method with 5-FU as the positive control. The
half-inhibitory concentration (IC50) values are listed in Table 2. The results show that most of the
evaluated compounds displayed potent antiproliferative activities against the tested cell lines. For
PC-3 cells, compounds 5a, 5b, 5c, 5d, 5f and 5j displayed more potential cytotoxic activity with
IC50 values ranging from 12.06 μM to 30.15 μM compared to 5-FU (IC50 = 38.61 μM). For A549
cells, compounds 5b, 5c, 5d and 5f exhibited significant cytotoxicities compared to that of 5-FU
(IC50 = 10.65 μM) with IC50 values of 5.46 μM, 10.66 μM, 12.35 μM and 10.32 μM, respectively.
For SGC-7901 cells, compounds 5a, 5b, 5c and 5f showed equal to or stronger cytotoxicity effects
than that of 5-FU (IC50 =13.25 μM) with IC50 values of 9.67 μM, 7.53 μM, 8.32 μM, 11.38 μM,
respectively. For MCF-7 cells, 5c and 5d displayed stronger cytotoxic activity with IC50 values of
7.34 μM and 6.25 μM, respectively, compared to that of 5-FU (IC50 = 7.67 μM). For HepG2 cells,
4

ACCEPTED MANUSCRIPT
the cytotoxic activity of compounds 5b, 5c, 5d and 5j were more potent than that of 5-FU (IC50
=
17.44 μM) with IC50 values of 6.62 μM, 5.23 μM, 16.83 μM, 17.15 μM, respectively. Compound
5c exhibited the strongest cytotoxicity activity against all five cancer cells compared to 5-FU with
IC50 values ranging from 5.23 μM to 26.96 μM. Fig. 3 indicates that 5c exhibits both dose- and
time-dependent effects on HepG 2 proliferation. Thus, compound 5c was regarded as the most
promising compound and was selected for further investigation.
Table 2
Fig. 3
3.2 Structure-activity relationship (SAR) of the bergenin/cinnamic acid hybrids
A structure activity relationship (SAR) analysis for the test compounds 5a-n showed that the
LogP values of all the newly synthesized hybrids were generally between that of bergenin and the
corresponding cinnamic acids (Bergenin’s LogP: -1.4, Table 1), which indicates equilibrium
between lipid and water solubilities. The substitution of cyan or methoxy groups (5b-e) led to a
significant increase in activity, which is in accordance with the literature results [36-38] that
revealed that the introduction of methoxy groups could apparently enhance the antitumour activity.
The cytotoxic activities were found to be more potent when two methoxyl groups were located
adjacent to one another (5c compared to 5b, 5d and 5e). Inversely, the substitution of halogen
(chloro- and fluoro-) and methyl groups (5f-k) led to a slight loss of activity. The replacement of
substituents with electron-donating groups, such as methoxy (5b-e) and methyl (5f), would
enhance the antiproliferative activity. However, the introduction of a nitro group (5l-n) could lead
to a significant loss of activity. In addition, the cytotoxic activity of bergenin/cinnamic hybrids (5a)
was visibly superior to that of bergenin, which may be related to the α,β-unsaturated ketone
pharmacophore in the cinnamic acid moiety.
3.3 The effect of 5c on cell cycle
To further understand the mechanism of the antiproliferative effect of compound 5c, we
examined the effect of 5c on cell cycle distribution by flow cytometry. As shown in Fig. 4, 5c can
induce a cell arrest at the G2/M phase. Cell populations in the control group were approximately
64.31% in the G0/G1 phase, 25.28% in the S phase, and 10.41% in the G2/M phase. After
incubation with 2 μM/L 5c for 24 h, the cells in the S phase and the G2/M phase increased to
5

ACCEPTED MANUSCRIPT
19.92% and 28.93%, respectively.
Fig. 4
3.4 The effect of compound 5c on apoptosis
To examine whether or not compound 5c induces cell apoptosis, HepG 2 cells were treated
with compound 5c at concentrations of 0.1 μM, 2 μM and 4 μM for 24 h, stained with Annexin
V-FITC and propidium iodide, and then analysed by flow cytometry. As shown in Fig. 5, after
incubation for 24 h, the apoptotic cell rates at 1 μM, 2 μM and 4 μM were 5.58%, 10.68% and
20.68%, respectively, which were higher than that of the control group. The results of the cell
apoptosis assay suggested that compound 5c can induce apoptosis in a dose-dependent manner in
HepG 2 cells.
Fig. 5
3.5 In vivo antitumour activity of compound 5c
Compound 5c showed the highest cytotoxic activity in vitro among all of the
bergenin/cinnamic acid ester derivatives. Consequently, 5c was selected for in vivo antitumour
evaluation on the tumour-bearing mice. Compound 5c was injected intraperitoneally every day for
14 days at doses of 15 and 30 mg/kg. Saline of blank control and 5-FU of positive control were
injected in the same manner. As indicated in Fig. 6A, the average body weights of the
tumour-bearing mice treated with 5c were higher than those of the control group at a dose of 15
mg/kg. In contrast, the average body weights of the tumour-bearing mice treated with
5-fluorouracil (5-FU, 20 mg/kg) significantly decreased compared to the control group, which
indicate that compound 5c is less toxic than 5-FU.
However, compound 5c significantly attenuated the tumour-bearing mice in both tumour
weight (Fig. 6B) and tumour size (Fig. 6C) in a dose-dependent manner. Treatment with
compound 5c (30 mg/kg) caused a reduction in the tumour weight by 61.02%, whereas 5-FU (20
mg/kg) treatment resulted in a 52.18% tumour growth regression (Table 3), suggesting that
compound 5c can significantly inhibit tumour growth in vivo, which was consistent with its in
vitro cytotoxicity. Furthermore, treatment with compound 5c resulted in improved thymus and
spleen indexes, which were superior to those treated with 5-FU (Fig. 6D, Fig. 6E). These results
6

ACCEPTED MANUSCRIPT
proved that compound 5c possesses stronger antitumour activity with fewer side effects than 5-FU
and could be a potential antitumour compound.
Table 3
Fig. 6
3.6 Western blot assay
To clarify the mechanism of apoptosis caused by compound 5c, activation of procaspase-3,
procaspase-9, Akt, p53, Bcl-2, Bcl-XL and Bax were evaluated by western blot analysis. HepG2
cells were treated with different concentrations (1 μM, 2 μM, 4 μM, and 8 μM) of compound 5c,
to which a vehicle control was added (0.5% DMSO). As shown in Fig. 7, upon treatment with
compound 5c, the down regulation of expression of apoptotic procaspase-3 and procaspase-3 is
associated with decreased expression of Bcl-2, Bcl-XL protein. Moreover, decreased expression
levels of Akt and p53 were also observed, suggesting that down-regulation of Akt protein
expression would affect the expression of downstream mediator p53 protein. In addition,
compound 5c down-regulated the expression of Bcl-2, Bcl-XL and up-regulated the expression of
proapoptotic proteins, Bax, in both a time- and dose-dependent manner in HepG2 cells. All of the
results indicate that the mitochondria-dependent apoptosis pathway was fully activated.
Compound 5c could activate Akt and regulate the expression of the Bcl-2 family of proteins as
well as p53 and Bax, which play key roles in the regulation of apoptosis by controlling the
mitochondrial membrane [39,40].
Fig. 7
4 Conclusion
A series of novel 11-hydroxyl-modified bergenin/cinnamic acid hybrids were synthesized
and evaluated in an effort to seek safe and effective anti-tumour lead compounds. The present
study demonstrated that most compounds significantly suppressed the viability of five human
tumour cells (PC-3, SGC-7901, A549, MCF-7, and HepG2). After screening, compound 5c was
found to display the most potent antitumour activity both in vitro and in vivo, significantly
attenuating tumour growth in a dose-dependent manner without decreasing tumour-bearing mice
body weight. Further mechanistic studies showed that compound 5c arrested the cell cycle at the
G2/M phase and induced typical apoptotic features in a dose-dependent manner. Moreover, the
western blot assay revealed that compound 5c can activate Akt and regulate the expression of the
7

ACCEPTED MANUSCRIPT
Bcl-2 family of proteins, promote Bax translocation to mitochondria, then trigger caspase-9 and
caspase-3, which are essential prerequisites for induction of apoptosis. A preliminary
structure-activity relationship (SAR) analysis suggested that the cyan or methoxy substitution of
the cinnamic acid phenyl group was essential for antiproliferative activity. Further chemical
modifications and preclinical studies, including PK (pharmacokinetic)/PD(pharmacodynamic)
properties are currently underway in our laboratories.
5 Experimental protocols
5.1. Chemistry
Unless noted, all solvents and reagents in this study were commercially available. All the
reactions were monitored by thin-layer chromatography (TLC) on pre-coated silica GF 254 plates
and visualized with UV lamp (254 nm and 365 nm). Melting points (mp) were determined on X-4
1
micro melting point apparatus (Beijing) and without correction. H NMR and 13C NMR spectra
were measured in DMSO solution on a 400 Bruker NMR spectrometer at ambient temperature
using tetramethylsilane (TMS) as the internal standard. Chemical shifts (δ values) and coupling
constant (J values) were reported in parts per million (ppm) and hertz (Hz), respectively.
Multiplicities were given as s (singlet), d (doublet), t (triplet), m (multiplet), dd (double doublet).
Flash column chromatography was carried out on silica gel 200–300 mesh. Mass spectra
experiments were performed on electrospray ionization (ESI) techniques. The purities of all
compounds were over 96% identified by HPLC.
5.2. General procedure for the synthesis of 5a-n
Cinnamic acid derivatives 1a-n (1 equiv.) was dissolved in thionyl chloride (SOCl2, 5 equiv.),
the reaction mixture was refluxed in the presence of sufficient amount of DMF as the catalyst for
3-5 h. The solvent was removed by reduced pressure distillation to obtain the product 2a-n in the
form of solid residue. Meanwhile, bergenin (1 equiv.) and K2CO3 (3 equiv.) were dissolved in dry
DMF at 0 °C and benzyl bromide (3 equiv.) was added. After the mixture was stirred for 12 h at
room temperature, the compound 4 was obtained through extraction with ethyl acetate, drying
with anhydrous sodium sulfate and desolventizing by evaporated in vacuo. The compound 4 was
directly added to an ice-cold stirred solution of above solid residue in dry pyridine (200mg/mL).
8

ACCEPTED MANUSCRIPT
Then warm the mixture to room temperature and stir for 20 h with CH2Cl2 added and DMAP as a
catalytic. After completion of the reaction (monitored by TLC: ethyl acetate /petroleum ether
=2:1), the mixture was hydrogenated by Pd/C under H2 atmosphere for 15 h. The mixture was
filtered, evaporated under vacuum and purified using column chromatography to afford the
compound 5a-n (Scheme 1). Characteristic data for all bergenin ester derivatives are as follows:
5.2.1. Compound 5a:11-O-cinnamaldehyde bergenin
1
White crystals. Yield: 48%, mp: 167-169°C. H-NMR(400MHz, DMSO-d6) δ: 9.78 (1H, s),
8.46 (1H, s), 7.64-7.73 (2H, m), 7.49-7.58 (2H, m), 7.46 (1H, d, J=11.6 Hz), 7.35 (1H, m), 6.98
(1H, s), 6.32 (1H, d, J=9.8 Hz), 5.57 (1H, d, J=10.3 Hz), 5.05 (1H, d, J=15.7 Hz), 4.12-4.24 (1H,
m), 3.96-4.08 (2H, m), 3.72-3.84 (1H, m), 3.58-3.61 (1H, m), 3.45 (3H, s), 3.40 (2H, s), 3.27 (1H,
dd, J=11.2, 11.2 Hz); 13C-NMR (100MHz, DMSO-d6) δ: 167.6, 167.4, 150.7,148.0, 147.3, 140.5,
134.2, 130.2, 128.9, 128.8, 128.2, 119.2, 118.0, 109.4, 81.9, 79.7, 73.9, 72.1, 70.6, 61.2, 59.8;
HRMS (ESI) calcd. for C23H22O10 [M + H]+, 458.1213; found, 458.1219.
5.2.2. Compound 5b: 11-O-(E)-3-(3,4,5-trimethoxyphenyl)acrylaldehyde bergenin
1
White crystals. Yield: 26%, mp: 199-201°C. H-NMR(400MHz, DMSO-d6) δ: 9.76 (1H, s),
8.42 (1H, s), 7.50 (1H, d, J=13.4 Hz), 6.95 (1H, s), 6.80 (2H, s), 6.32 (1H, d, J=11.8 Hz), 5.47 (1H,
d, J=11.8), 4.52-4.61 (1H, m), 4.26-4.35 (2H, m), 4.25 (1H, m), 3.84 (12H, s), 3.88-3.87 (1H, m),
3.56 (2H, s), 3.40 (1H, dd, J=10.5, 10.5 Hz); 13C-NMR (100MHz, DMSO-d6) δ: 167.5, 167.3,
153.5, 149.2, 146.7, 145.9, 145.2, 138.6, 126.5, 125.9, 118.8, 117.2, 109.2, 104.0, 81.4, 79.7, 75.3,
74.6, 63.4, 61.1, 60.0, 56.2; HRMS (ESI) calcd. for C26H28O13 [M + H]+, 548.1530; found,
548.1534.
5.2.3. Compound 5c: 11-O-(E)-4-(3-oxoprop-1-en-1-yl)benzonitrile bergenin
1
White crystals. Yield: 37%. mp: 194-196°C. H-NMR(400MHz, DMSO-d6) δ: 9.82 (1H, s),
8.46 (1H, s), 7.88 (2H, d, J=10.6 Hz), 7.58 (1H, d, J=9.6 Hz), 7.48 (2H, d, J=12.9 Hz), 6.98 (1H,
s), 6.32 (1H, d, J=12.0 Hz), 5.48 (1H, d, J=10.5), 4.53-4.62 (1H, m), 4.26-4.35 (2H, m), 4.27 (1H,
m), 3.84 (3H, s), 3.93-4.01 (1H, m), 3.58 (2H, s), 3.42 (1H, dd, J=10.8, 10.8 Hz); 13C-NMR
(100MHz, DMSO-d6) δ: 166.6, 166.4, 149.2, 146.7, 145.9, 145.2, 139.7, 132.3, 128.9, 125.9,
118.9, 117.8, 116.4, 109.2, 81.4, 79.6, 75.2, 74.6, 72.0, 63.4, 61.0; HRMS (ESI) calcd. for
9

ACCEPTED MANUSCRIPT
C24H21NO10 [M + H]+, 483.4242; found, 483.4246.
5.2.4. Compound 5d: 11-O-(E)-3-(3,5-dimethoxyphenyl)acrylaldehyde bergenin
White crystals. Yield: 37%, mp: 172-174°C. 1H-NMR(400MHz, DMSO-d6) δ: 9.74 (1H, s),
8.26 (1H, s), 7.48 (1H, d, J=12.8 Hz), 6.99 (1H, s), 6.74 (2H, s), 6.32 (1H, d, J=10.6 Hz), 6.27 (1H,
s), 5.49 (1H, d, J=11.2 Hz), 4.54-4.63 (1H, m), 4.26-4.32 (2H, m), 4.24 (1H, m), 3.92-4.01 (1H,
m), 3.84 (9H, s), 3.58 (2H, s), 3.40 (1H, dd, J=9.8, 9.8 Hz); 13C-NMR (100MHz, DMSO-d6) δ:
166,7 166.5, 161.7, 149.2, 146.7, 145.9, 145.2, 134.2, 126.0, 117.9, 116.4, 109.2, 105.6, 99.8, 81.5,
79.7, 75.3, 74.6, 72.1, 63.4, 61.1, 56.0; HRMS (ESI) calcd. for C25H26O12 [M + H]+, 518.4667;
found, 518.4670.
5.2.5. Compound 5e: 11-O-(E)-3-(4-methoxyphenyl)acrylaldehyde bergenin
1
White crystals. Yield: 31%, mp: 175-177°C. H-NMR(400MHz, DMSO-d6) δ: 9.86 (1H, s),
8.54 (1H, s), 7.64 (2H, d, J=8.8 Hz), 7.49 (1H, d, J=12.8 Hz), 7.02 (1H, s), 6.96 (2H, d, J=8.5 Hz),
6.32 (1H, d, J=11.0 Hz), 5.48 (1H, d, J=11.8), 4.52-4.61 (1H, m), 4.27-4.36 (2H, m), 4.25 (1H, m),
3.84 (6H, s), 3.91-4.00 (1H, m), 3.56 (2H, s), 3.40 (1H, dd, J=10.9, 10.9 Hz); 13C-NMR (100MHz,
DMSO-d6) δ: 166.6, 166.4, 159.9, 149.2, 146.7, 145.9, 145.2, 130.4, 127.6, 125.9, 117.9, 116.4,
114.4, 109.2, 81.4, 79.6, 75.3, 74.6, 72.0, 63.4, 61.0, 55.9; HRMS (ESI) calcd. for C24H24O11 [M +
H]+, 488.4408; found, 488.4412.
5.2.6. Compound 5f: 11-O-(E)-3-(p-tolyl)acrylaldehyde bergenin
1
White crystals. Yield: 22%, mp: 220-222°C. H-NMR(400MHz, DMSO-d6) δ: 9.80 (1H, s),
8.48 (1H, s), 7.60 (2H, d, J=7.9 Hz), 7.49 (1H, d, J=12.5 Hz), 7.20 (2H, d, J=8.2 Hz), 7.00 (1H, s),
6.32 (1H, d, J=11.5 Hz), 5.48 (1H, d, J=10.8), 4.52-4.60 (1H, m), 4.28-4.37 (2H, m), 4.26 (1H, m),
3.84 (3H, s), 3.91-4.00 (1H, m), 3.58 (2H, s), 3.42 (1H, dd, J=12.1, 11.1 Hz), 2.36 (3H, s);
13C-NMR (100MHz, DMSO-d6) δ: 166.6, 166.4, 149.2, 146.7, 145.9, 145.2, 137.8, 132.4, 129.1,
128.7, 125.9, 117.9, 116.4, 109.2, 81.4, 79.6, 75.3, 74.6, 72.0, 63.4, 61.0, 21.5; HRMS (ESI)
calcd. for C24H24O10 [M + H]+, 472.4414; found, 472.4416.
5.2.7. Compound 5g: 11-O-(E)-3-(3,4-dimethylphenyl)acrylaldehyde bergenin
1
White crystals. Yield: 28%, mp: 211-213°C. H-NMR(400MHz, DMSO-d6) δ: 9.77 (1H, s),
8.12 (1H, s), 7.48 (1H, d, J=13.7 Hz), 7.42 (1H, d, J=7.8 Hz), 7.18 (1H, d, 8.6 Hz), 7.00 (1H, s),
10

ACCEPTED MANUSCRIPT
6.98 (1H, s), 6.33(1H, d, J=12.4 Hz), 5.48 (1H, d, J=11.8 Hz), 4.54 (1H, m), 4.28 (1H, m),
4.16-4.25 (2H, m),3.84 (3H, m), 3.58 (2H, s), 3.44-3.51 (1H, m), 3.40 (1H, dd, J=11.2, 11.2 Hz),
2.36 (6H, s); 13C-NMR (100MHz, DMSO-d6) δ: 166.8, 166.5, 149.2, 146.7, 145.9, 145.3, 136.9,
136.0, 132.3, 132.2, 131.5, 125.9, 125.7, 117.9, 116.4, 109.1, 81.6, 79.8, 75.3, 74.6, 72.1, 63.4,
61.0, 19.4, 18.9; HRMS (ESI) calcd. for C25H26O10 [M + H]+, 486.4679; found, 486.4684.
5.2.8. Compound 5h: 11-O-(E)-3-(3,4-dichlorophenyl)acrylaldehyde bergenin
White crystals. Yield: 43%, mp: 187-189°C. 1H-NMR(400MHz, DMSO-d6) δ: 9.76 (1H, s),
8.23 (1H, s), 7.58 (1H, d, J=8.5 Hz), 7.48 (1H, d, J=12.6 Hz), 7.40 (1H, d, 9.5 Hz), 7.26 (1H, s),
6.98 (1H, s), 6.33(1H, d, J=12.9 Hz), 5.49 (1H, d, J=10.5 Hz), 4.51-4.60 (1H, m), 4.28-4.35 (2H,
m), 4.27 (1H, m), 3.85 (3H, m), 3.65-3.74 (1H, m), 3.58 (2H, s), 3.41 (1H, dd, J=10.3, 10.3 Hz);
13C-NMR (100MHz, DMSO-d6) δ: 166.8, 166.5, 149.2, 146.7, 145.9, 145.3, 134.9, 133.5, 132.8,
130.3, 127.9, 127.3, 125.9, 117.9, 116.4, 109.1, 81.6, 79.8, 75.3, 74.6, 72.1, 63.4, 60.9; HRMS
(ESI) calcd. for C23H20Cl2O10 [M + H]+, 527.3049; found, 527.3054.
5.2.9. Compound 5i: 11-O-(E)-3-(4-chlorophenyl)acrylaldehyde bergenin
1
White crystals. Yield: 25%, mp: 179-181°C. H-NMR(400MHz, DMSO-d6) δ: 9.82 (1H, s),
8.57 (1H, s), 7.70 (2H, d, J=7.4 Hz), 7.48 (1H, d, J=11.5 Hz), 7.45 (2H, d, J=8.5 Hz), 6.98 (1H, s),
6.32 (1H, d, J=12.5 Hz), 5.48 (1H, d, J=11.3), 4.52-4.60 (1H, m), 4.28-4.37 (2H, m), 4.26 (1H, m),
3.84 (3H, s), 3.91-4.00 (1H, m), 3.58 (2H, s), 3.42 (1H, dd, J=11.7, 11.7 Hz); 13C-NMR (100MHz,
DMSO-d6) δ: 166.6, 166.4, 149.2, 146.7, 145.9, 145.2, 133.8, 133.4, 129.1, 128.7, 125.9, 117.9,
116.4, 109.2, 81.4, 79.6, 75.3, 74.6, 72.0, 63.4, 61.0; HRMS (ESI) calcd. for C23H21ClO10 [M +
H]+, 492.8598; found, 492.8602.
5.2.10. Compound 5j: 11-O-(E)-3-(4-fluorophenyl)acrylaldehyde bergenin
1
White crystals. Yield: 29%, mp:163-165°C. H-NMR(400MHz, DMSO-d6) δ: 9.87 (1H, s),
8.50 (1H, s), 7.74 (2H, d, J=8.4 Hz), 7.48 (1H, d, J=12.6 Hz), 7.21 (2H, d, J=8.9 Hz), 6.98 (1H, s),
6.32 (1H, d, J=12.8 Hz), 5.48 (1H, d, J=10.8), 4.53-4.62 (1H, m), 4.26-4.37 (2H, m), 4.27 (1H, m),
3.84 (3H, s), 3.93-4.01 (1H, m), 3.58 (2H, s), 3.44 (1H, dd, J=12.6, 12.6 Hz); 13C-NMR (100MHz,
DMSO-d6) δ: 166.6, 166.4, 162.2, 149.2, 146.7, 145.9, 145.2, 130.9, 130.6, 125.9, 117.9, 115.5,
116.4, 109.2, 81.4, 75.3, 74.6, 72.0, 63.4, 61.0; HRMS (ESI) calcd. for C23H21FO10 [M + H]+,
11

ACCEPTED MANUSCRIPT
476.4052; found, 476.4056.
5.2.11. Compound 5k: 11-O-(E)-3-(3,5-difluorophenyl)acrylaldehyde bergenin
1
White crystals. Yield: 35%, mp: 156-158°C. H-NMR(400MHz, DMSO-d6) δ: 9.83 (1H, s),
8.52 (1H, s), 7.48 (1H, d, J=13.1 Hz), 6.99 (1H, s), 6.74 (2H, s), 6.58 (1H, s), 6.32 (1H, d, J=11.6
Hz), 5.49 (1H, d, J=9.5 Hz), 4.54-4.62 (1H, m), 4.26-4.32 (2H, m), 4.24 (1H, m), 3.94-4.02 (1H,
m), 3.84 (3H, s), 3.58 (2H, s), 3.46 (1H, dd, J=9.8, 9.8 Hz); 13C-NMR (100MHz, DMSO-d6) δ:
166,7 166.5, 161.7, 160.9, 149.2, 146.7, 145.9, 145.2, 138.6, 126.0, 117.9, 116.4, 109.6, 104.0,
81.5, 79.7, 75.3, 74.6, 72.1, 63.4, 61.0; HRMS (ESI) calcd. for C23H20F2O10 [M + H]+, 494.1025;
found, 494.1029.
5.2.12. Compound 5l: 11-O-(E)-3-(3,4-dimethoxyphenyl)acrylaldehyde bergenin
1
White crystals. Yield: 29%, mp: 188-190°C. H-NMR(400MHz, DMSO-d6) δ: 9.80 (1H, s),
8.38 (1H, s), 7.48 (1H, d, J=12.6 Hz), 7.23 (1H, s), 7.19 (1H, d, J=7.7 Hz), 7.00 (1H, s), 6.98(1H,
d, 9.8 Hz), 6.33(1H, d, J=11.8 Hz), 5.48 (1H, d, J=10.8 Hz), 4.54 (1H, m), 4.28 (1H, m), 4.16-4.25
(2H, m),3.84 (9H, m), 3.69-3.76 (1H, m), 3.58 (2H, s), 3.40 (1H, dd, J=10.4, 10.4 Hz); 13C-NMR
(100MHz, DMSO-d6) δ: 166.8, 166.5, 150.0,149.2, 146.7, 145.9, 145.3, 127.5, 126.0, 122.7, 117.9,
116.4, 112.0, 111.7, 109.2, 81.6, 79.8, 75.3, 74.6, 72.1, 63.4, 61.0, 56.3; HRMS (ESI) calcd. for
C25H26O12 [M + H]+, 518.4667; found, 518.4669.
5.2.13. Compound 5m: 11-O-(E)-3-(4-nitrophenyl)acrylaldehyde bergenin
1
White crystals. Yield: 46%, mp: 167-169°C. H-NMR(400MHz, DMSO-d6) δ: 9.78 (1H, s),
8.54 (1H, s), 8.23 (2H, d, J=12.1 Hz), 8.05 (2H, d, J=10.7 Hz), 7.64 (1H, d, J=11.4 Hz), 6.99 (1H,
s), 6.62 (1H, d, J=12.6 Hz), 5.49 (1H, d, J=9.6), 4.52-4.60 (1H, m), 4.28-4.36 (2H, m), 4.26 (1H,
m), 3.85 (3H, s), 3.96-4.05 (1H, m), 3.60 (2H, s), 3.45 (1H, dd, J=13.2, 13.2 Hz); 13C-NMR
(100MHz, DMSO-d6) δ: 166.6, 166.4, 149.2, 147.3, 146.7, 145.9, 145.2, 141.5, 129.2, 125.9,
124.0, 117.8, 116.4, 109.2, 81.4, 79.6, 75.2, 74.6, 72.0, 63.4, 60.9; HRMS (ESI) calcd. for
C23H21NO12 [M + H]+, 503.4123; found, 503.4126.
5.2.14. Compound 5n: 11-O-(E)-3-(3,5-dinitrophenyl)acrylaldehyde bergenin
1
White crystals. Yield: 22%, mp: 202-204°C. H-NMR(400MHz, DMSO-d6) δ: 9.85 (1H, s),
8.72 (2H, s), 8.66 (1H, s), 8.27 (1H, s), 7.60 (1H, d, J=12.8 Hz), 6.98 (1H, s), 6.58 (1H, d, J=13.6
12

ACCEPTED MANUSCRIPT
Hz), 5.48 (1H, d, J=8.7 Hz), 4.54-4.62 (1H, m), 4.28-4.35 (2H, m), 4.25 (1H, m), 3.94-4.02 (1H,
m), 3.85 (3H, s), 3.59 (2H, s), 3.41 (1H, dd, J=10.4, 10.4 Hz); 13C-NMR (100MHz, DMSO-d6) δ:
166,7 166.5, 149.2, 148.8, 146.7, 145.9, 145.2, 137.2, 128.9, 126.0, 117.9, 117.4, 116.4, 109.2,
81.5, 79.7, 75.3, 74.6, 72.0, 63.4, 60.8; HRMS (ESI) calcd. for C23H20N2O14 [M + H]+, 548.4099;
found, 548.4103.
5.3. Cytotoxicity assay
5.3.1. Cell lines and culture
Five human cancer cell lines, including PC-3 (human prostate cancer), SGC-7901 (human
gastric cancer), A549 (human non-small cell lung cancer), MCF-7 (human breast cancer), and
HepG2 (human hepatoma cell) were obtained from Cancer Cell Repository (Shanghai cell bank).
Cell lines were maintained in DMEM supplemented with 10% (v/v) heat inactivated foetal bovine
serum (FBS), 100 IU/mL penicillin, and 100 mg/mL streptomycin. All cancer cells were incubated
in a humidified atmosphere of 5% CO2 at 37°C.
5.3.2. In vitro anti-proliferative activity assays
All of the synthesized compounds (5a-n) were assessed for their cytotoxic activity via the
MTT assay using the PC-3, SGC-7901, A549, MCF-7 and HepG2 cancer cell lines. 5000 cells per
well were planted into 96-well microtiter plates in 180 µL growth medium (5% FBS in DMEM)
and incubated for 24 h. Then, the cells were exposed to each compound in DMSO at four different
concentrations (0.1, 1, 10, and 100 µmol/L) for 72 h. The mixture was incubated at 37°C under a
humidified 5% CO2 atmosphere. After incubation, 20 µL of MTT (5 mg/mL in PBS) was added to
each well and the cells were further incubated for 4 h at 37°C. The supernatant of each well was
removed and the formazan contained in the cells was dissolved in 150 µL of DMSO. The plate
was shaken to ensure the crystals were dissolved completely. Then, the absorbance (OD values)
was recorded by spectrophotometric quantification at 570 nm as a marker for cell viability. The
IC50 (50% cell viability inhibition) values were calculated by means of PRISM 6, Graph Pad
software. The independent experiment was repeated at least three times.
5.4 Flow cytometry cell cycle analysis
13

ACCEPTED MANUSCRIPT
HepG2 cells were seeded 3×105 per well in a 6-well plate and incubated for 24 h at 37°C.
Then, the media was replaced and the cells were treated with compound 5c at different
concentrations for 24 h. The cells were harvested and washed two times with cold PBS, and fixed
with cold 70% alcohol at -20°C for at least 24 h. After centrifugation of the suspensions at 1500
rpm for 5 min, the cell pellets were treated with RNase A (50 mg/ml in PBS) and stained with
propidium iodide (50 mg/mL) at 37°C for 30 min. Flow cytometry analysis was performed by a
flow cytometer (Becton, Dickinson and Company).
5.5. Apoptosis analysis
HepG2 cells were planted at 1.5×105 per well in a 6-well plate and incubated at 37°C. After
24 h, the media was replaced and the cells were exposed to different concentrations of compound
5c for 48 h. The cells were harvested and washed two times with cold PBS. The percentages of
apoptotic cells were evaluated by staining with Annexin V-FITC/PI Apoptosis Detection Kit
(APOAF; Sigma, St. Louis, MO, USA) according to the manufacturer's instructions. The cells
were analysed with a flow cytometer (Becton, Dickinson and Company).
5.6. In vivo antitumour activity assay with compound 5c
Female Swiss mice with body weights of 20-22 g, obtained from the Experimental Animal
Center of Xi’an Jiaotong University Health Science Center, were maintained in a conventional
animal colony and freely available for food and water. HepG2 cells (3×106) were injected
subcutaneously into the right flank to generate the xenograft-bearing mice model. After the tumour
cells were inoculated for 7-21 days, 24 tumour-bearing mice were randomly divided into four
groups, with 6 mice in each group: saline control group, 5c 15 mg/kg group, 5c 30 mg/kg group
and 5-FU (positive control) 20 mg/kg group. All groups were given a daily intraperitoneal
injection. The body weights were recorded every day and the tumour wet weights were measured
on the last day of the experiment (all mice were killed on Day 14). In addition, the toxicity of 5c
on the thymus and spleen were also observed. The thymus and spleen indexes were calculated by
using the following formulas: thymus index = (thymus weight/final body weight); spleen index =
(spleen weight/final body weight).
14

ACCEPTED MANUSCRIPT
5.7. Western blot assay
HepG2 cells were seeded into 12-well plates at 5×105 cells per well with different
concentrations of compound 5c and incubated at 37°C for 24 h in the presence of 10% FBS. The
controls were treated with vehicle (0.5% DMSO). The cells were lysed into ice-cold cell lysis
buffer for 30 min and centrifuged at 13000 g for 15 min at 4°C. The protein concentrations of the
supernatant were determined by a BCA protein Assay Kit. Equal amounts of protein lysates were
separated on 10% SDS-PAGE (sodium dodecyl sulphate–polyacrylamide gel electrophoresis) and
transferred onto PVDF membranes. The membranes were blocked with 5% non-fat milk in TBST
(0.1% Tween 20 in TBS) buffer for 2 h, incubated with primary antibodies at 1:2000 dilutions at
4°C, and washed by TBST four times. Then, the membranes were incubated with horseradish
peroxidase-conjugated secondary antibodies at 1:2000 dilutions at room temperature for 2 h and
washed four times with TBST. The western blotting was detected by enhanced chemiluminescence
reagent (Beyotime Institute of Biotechnology, China). GAPDH was used as a loading control.
5.8. Statistical analysis
The experimental results are expressed as the mean ± standard deviation (SD) and mean ±
standard error of the mean (S.E.M). The statistical differences were analysed according to analysis
of variance (ANOVA) test, one-way and two-way. P values less than 0.05 (P<0.05) were
considered statistically significant and represented by: *P<0.05, **P<0.01. All statistics were
calculated using the statistical progam PRISM 6, Graph Pad software.
Acknowledgment
This work was supported by the National Natural Science Foundation of China (Grant
No.81602967) and the China Postdoctoral Science Foundation (Grant No.2016M592898XB), the
Shaanxi Postdoctoral Science Foundation (2015), the Basic Research Plan of the Education
Department of Shaanxi Province National (Grant No.15JK1076), the College Students’ Innovative
Entrepreneurial Training Program (Grant No.201510708172, 201610708019). We appreciate the
support from Center for Instrumental Analysis, China Pharmaceutical University, for their
contribution to the structural confirmation.
15

ACCEPTED MANUSCRIPT
References
[1] Asensi M, Ortega A, Mena S, Feddi F, Estrela JM. Natural polyphenols in cancer therapy[J]. Critical Reviews
in Clinical Laboratory Sciences. 2011,48(48):197-216.
[2] Ebrahimi A, Schluesener H. Natural polyphenols against neurodegenerative disorders: potentials and
pitfalls[J]. Ageing Research Reviews. 2012,11(2):329-45.
[3] Lecour S, Lamont KT. Natural polyphenols and cardioprotection[J]. Mini Reviews in Medicinal Chemistry.
2011,11(14):1191-9(9).
[4] Tanaka N, Kraus FJ, Fujioka T, Fujii H, Mihashi K, Ishimaru K. Bergenin and fragrance molecules in
Saxifraga stolonifera root cultures[J]. Current Topics in Phytochemistry, Volume 5: 85-89. 2002.
[5] Chen XM, Yoshida T, Hatano T, Fukushima M, Okuda T. Galloylarbutin and other polyphenols from
Bergenia purpurascens[J]. Phytochemistry. 1987,26(2):515–7.
[6] Patel DK, Patel K, Kumar R, Gadewar M, Tahilyani V. Pharmacological and analytical aspects of bergenin: a
concise report[J]. Asian Pacific Journal of Tropical Disease. 2012,2(2):163-7.
[7] Lim HK, Kim HS, Choi HS, Oh S, Choi J. Hepatoprotective effects of bergenin, a major constituent of
Mallotus japonicus, on carbon tetrachloride-intoxicated rats[J]. Journal of Ethnopharmacology.
2000,72(3):469–74.
[8] Haribabu K, Ajitha M, Ramesh B, Babu K, Rao J. Quantification of bergenin from Mallotus philippinensis by
HPTLC-MS and study on different extraction methods[J]. JPC - Journal of Planar Chromatography - Modern
TLC. 2012,25:445-9.
[9] Kumar R, Patel DK, Prasad SK, Laloo D, Krishnamurthy S, Hemalatha S. Type 2 antidiabetic activity of
bergenin from the roots of Caesalpinia digyna Rottler[J]. Fitoterapia. 2012,83(2):395–401.
[10] Mebe PP, Makuhunga P. 11-(E)-p-Coumaric acid ester of bergenin from Peltophorum africanum[J].
Phytochemistry. 1992,31(9):3286-7.
[11] Chen Z, Liu YM, Yang S, Song BA, Xu GF, Bhadury PS, et al. Studies on the chemical constituents and
anticancer activity of Saxifraga stolonifera (L) Meeb[J]. Bioorganic
2008,16(3):1337-44.
& Medicinal Chemistry.
[12] Arfan M, Amin H, Khan I, Shah MR, Shah H, Khan AZ, et al. Molecular simulations of bergenin as a new
urease inhibitor[J]. Medicinal Chemistry Research. 2012,21(9):2454-7.
[13] Piacente S, ., Pizza C, ., Tommasi N, De, Mahmood N, . Constituents of Ardisia japonica and their in vitro
anti-HIV activity[J]. Journal of Natural Products. 1996,59(6):565-9.
[14] Gao XJ, Guo MY, Zhang ZC, Wang TC, Cao YG, Zhang NS. Bergenin Plays an Anti-Inflammatory Role via
the Modulation of MAPK and NF-κB Signaling Pathways in a Mouse Model of LPS-Induced Mastitis[J].
Inflammation. 2014:1-9.
[15] Takahashi H, Kosaka M, Watanabe Y, Nakade K, Fukuyama Y. Synthesis and neuroprotective activity of
bergenin derivatives with antioxidant activity[J]. Bioorganic & Medicinal Chemistry. 2003,11(8):1781-8.
[16] D., H., T., Y., J., C., et al. A carbon-carbon-coupled dimeric bergenin derivative biotransformed by Pleurotus
ostreatus[J]. Bioorganic & Medicinal Chemistry Letters. 2005,15(18):4073–5.
[17] De-Biao, ZHANG, Dong-Ping, Ming, Wen-Yuan, FENG, et al. Synthesis and cytotoxic activity of 3, 4,
11-trihydroxyl modified derivatives of bergenin[J]. Chinese Journal of Natural Medicines.
2014,12(12):929-36.
[18] Kumar TV, Tiwari AK, Robinson A, Babu KS, Kumar RSC, Kumar DA, et al. Synthesis and antiglycation
potentials of bergenin derivatives[J]. Bioorganic & Medicinal Chemistry Letters. 2011,21(16):4928-31.
16

ACCEPTED MANUSCRIPT
[19] Nighat N, Surrinder K, Mushtaq Ahmad Q, Muzaffar Hussain N, Mohammed Iqbal Z. Evaluation of
antioxidant and antimicrobial activities of Bergenin and its derivatives obtained by chemoenzymatic
synthesis[J]. European Journal of Medicinal Chemistry. 2011,46(6):2415-20.
[20] Shah MR, Arfan M, Amin H, Hussain Z, Qadir MI, Choudhary MI, et al. Synthesis of new bergenin
derivatives as potent inhibitors of inflammatory mediators NO and TNF-α[J]. Bioorganic & Medicinal
Chemistry Letters. 2012,22(8):2744–7.
[21] Zamarrud, Ali I, Hussain H, Ahmad VU, Qaiser M, Amyn A, et al. Two new antioxidant bergenin derivatives
from the stem of Rivea hypocrateriformis[J]. Fitoterapia. 2011,82(4):722–5.
[22] Manh Heun K, Sung Yi H, Myeong Hwan O, Hyuk KH, So Ra K, Won LM. Anti-oxidative and
anti-proliferative activity on human prostate cancer cells lines of the phenolic compounds from Corylopsis
coreana Uyeki[J]. Molecules. 2013,18(5):4876-86.
[23] Megumi S, Toshikazu S, Nijsiri R, Kazuei I, Fumio I. Ardisiphenols and other antioxidant principles from the
fruits of Ardisia colorata[J]. Chemical & Pharmaceutical Bulletin. 2002,50(11):1484-7.
[24] Newell AMB, Yousef GG, Lila MA, Ramírez-Mares MV, Mejia EGD. Comparative in vitro bioactivities of
tea extracts from six species of Ardisia and their effect on growth inhibition of HepG2 cells[J]. Journal of
Ethnopharmacology. 2010,130(3):536–44.
[25] Hoskins JA. The occurrence, metabolism and toxicity of cinnamic acid and related compounds[J]. Journal of
Applied Toxicology. 1984,4(6):283-92.
[26] Gülçin İ. Antioxidant activity of caffeic acid (3,4-dihydroxycinnamic acid)[J]. Toxicology. 2006,217(s
2–3):213-20.
[27] Sova M. Antioxidant and antimicrobial activities of cinnamic acid derivatives[J]. Mini Reviews in Medicinal
Chemistry. 2012,12(8):749-67(19).
[28] De P, ., Baltas M, ., Bedos-Belval F, . Cinnamic acid derivatives as anticancer agents-a review[J]. Current
Medicinal Chemistry. 2011,18(11):1672-703(32).
[29] Qi-Fu L, Song-Lin S, Qing-Rong L, Jian T, Jianye S, Ying L. Anticancer effects of ginsenoside Rg1, cinnamic
acid, and tanshinone IIA in osteosarcoma MG-63 cells: Nuclear matrix downregulation and cytoplasmic
trafficking of nucleophosmin[J]. International Journal of Biochemistry & Cell Biology. 2008,40(9):1918–29.
[30] Bosquesi PL, Trf M, Vizioli EO, dos Santos JL, Chung MC. Anti-Inflammatory Drug Design Using a
Molecular Hybridization Approach[J]. Pharmaceuticals. 2011,4(12):1450-74.
[31] Cloete TT, De KC, Smith PJ, N'Da DD. Synthesis, in vitro antiplasmodial activity and cytotoxicity of a series
of artemisinin–triazine hybrids and hybrid-dimers[J]. European Journal of Medicinal Chemistry.
2014,76(7):470–81.
[32] Meunier B. Hybrid molecules with a dual mode of action: dream or reality?[J]. Accounts of Chemical
Research. 2008,41(1):69-77.
[33] Singh K, Kaur H, Chibale K, Balzarini J. Synthesis of 4-aminoquinoline-pyrimidine hybrids as potent
antimalarials and their mode of action studies[J]. European Journal of Medicinal Chemistry.
2013,66C(15):314-23.
[34] Ting Z, Qian S, Bastow KF, Kuo-Hsiung L. Antitumor agents 286. Design, synthesis, and structure-activity
relationships of 3'R,4'R-disubstituted-2',2'-dimethyldihydropyrano[2,3-f]chromone (DSP) analogues as potent
chemosensitizers to overcome multidrug resistance[J]. Journal of Medicinal Chemistry. 2010,53(24):8700-8.
[35] Wang-Fun F, Xiao-Ling S, Christoph G, Michael W, Guang-Ying C, Guo-Yuan Z, et al. Methoxylation of
3',4'-aromatic side chains improves P-glycoprotein inhibitory and multidrug resistance reversal activities of
7,8-pyranocoumarin against cancer cells[J]. Bioorganic & Medicinal Chemistry. 2008,16(7):3694–703.
17

ACCEPTED MANUSCRIPT
[36] Xu CC, Deng T, Fan ML, Lv WB, Liu JH, Yu BY. Synthesis and in vitro antitumor evaluation of
dihydroartemisinin-cinnamic acid ester derivatives[J]. European Journal of Medicinal Chemistry. 2015,107.
[37] Yan J, Guo Y, Wang Y, Fei M, Ling H, Li X. Design, synthesis, and biological evaluation of
benzoselenazole-stilbene hybrids as multi-target-directed anti-cancer agents[J]. European Journal of
Medicinal Chemistry. 2015,95:220-9.
[38] Yang X, Wei W, Tan J, Song D, Ming L, Dan L, et al. Synthesis of a series of novel dihydroartemisinin
derivatives containing a substituted chalcone with greater cytotoxic effects in leukemia cells. Bioorg Med
Chem Lett[J]. Bioorganic & Medicinal Chemistry Letters. 2009,19(15):4385-8.
[39] Cory S, Huang DC, Adams JM. The Bcl-2 family: roles in cell survival and oncogenesis[J]. Oncogene.
2003,22(53):8590-607.
[40] Nazir N, Koul S, Qurishi M A, et al. Evaluation of antioxidant and antimicrobial activities of Bergenin and its
derivatives obtained by chemoenzymatic synthesis.[J]. European Journal of Medicinal Chemistry, 2011,
46(6):2415-20.
18

ACCEPTED MANUSCRIPT
Table 1. Structure of 11-hydroxyl-modified bergenin/cinnamic acid hybrids
Substituents
Yield
（%）
48
MP (°C)
LogPa
(Compound)
1.07
LogP
(Cinnamic acid)
1.93
Compound
R1
H
R2
H
R3
H
167-169
199-201
194-196
172-174
175-177
220-222
211-213
187-189
179-181
163-165
156-158
188-190
167-169
202-204
5a
5b
5c
5d
5e
5f
OMe
H
OMe
CN
H
OMe
H
26
0.69
1.55
37
1.11
1.97
OMe
H
OMe
H
37
0.81
1.68
OMe
Me
Me
Cl
31
0.94
1.81
H
H
22
1.55
2.42
Me
Cl
H
28
2.04
2.91
5g
5h
5i
H
43
2.18
3.05
H
Cl
H
25
1.62
2.49
H
F
H
29
1.22
2.09
5j
F
H
F
35
1.38
2.25
5k
5l
OMe
H
OMe
NO2
H
H
29
0.81
1.68
H
46
1.21
1.92
5m
5n
NO2
NO2
22
1.47
1.88
aTheoretical values of logP were calculated using commercially available ACD LAB/logP release
10, product version 10.08. Bergenin’s LogP: -1.4
Table 2. In vitro antiproliferative activity (IC50 in μM) of the synthesized compounds 5a-n against
PC-3, SGC-7901, A549, MCF-7 and HepG2 cell lines.
IC50 (μM)a
Compound
PC-3
A549
SGC-7901
23.56±1.97
9.67±1.68
7.53±0.16
8.32±1.62
27.82±2.18
15.86±1.25
11.38±1.46
>100
MCF-7
>100
HepG2
34.56±3.12
18.46±1.55
6.62±0.53
5.23±0.79
16.83±1.90
29.71±3.67
>100
>100
77.28±3.47
56.89±2.65
5.46±0.26
10.66±0.10
12.35±0.95
26.38±1.07
10.32±1.86
>100
bergenin
5a
>100
13.46±0.33
12.06±0.24
26.96±0.15
26.32±1.76
>100
11.69±1.01
7.34±0.88
6.25±1.74
18.65±3.23
20.22±1.49
59.54±2.68
>100
5b
5c
5d
5e
30.15±3.12
>100
5f
>100
5g
>100
>100
>100
63.24±3.36
>100
5h
86.52±2.21
>100
>100
>100
5i

ACCEPTED MANUSCRIPT
>100
>100
22.87±1.98
46.24±1.25
>100
35.75±3.18
72.13±2.88
>100
17.15±1.55
>100
5j
>100
28.36±2.21
>100
5k
58.94±2.50
>100
86.13±1.71
>100
5l
>100
>100
>100
5m
5n
5-FUb
>100
>100
>100
>100
>100
38.61±2.16
10.65±0.43
13.25±0.32
7.67±0.18
17.44±0.85
All data are presented as the mean±SD (n = 3).
aConcentration that promotes 50% of cell growth inhibition after 48 h of treatment. IC50 was
determined by the MTT assay. The independent experiment was run at least three times.
bUsed as a positive control.
cWhen a 50% inhibition could not be reached at the highest compound concentration, >100 μM was
given.
Table 3. Inhibitory effect of compound 5c on HepS tumour xenograft.
Compounds
Rate of inhibition (%)
5c (15 mg/kg)
5c (30 mg/kg)
5-FU (20 mg/kg)
45.36
61.02
52.18

ACCEPTED MANUSCRIPT
Figure Legends
Fig. 1 Structure of bergenin
Fig. 2 Structures of natural and synthetic cinnamic acids
Fig.3 Dose- and time-dependent effect of compound 5c on HepG 2 proliferation

ACCEPTED MANUSCRIPT
Fig.4 Compound 5c (2μmol/L) induced G2/M arrest in HepG 2 cells for 24 h, harvested and
stained with propidium iodide (PI) for 20 min. Cells were then subjected to flow cytometric
analysis for cell distributions at each phase of cell cycle.
Fig.5 Compound 5c induced apoptosis in HepG2 cells was assessed by flow cytometric analysis
after staining with Annexin V-PI.
A
3 0
c o n tr o l
2 8
2 6
2 4
2 2
2 0
1 8
5 c (1 5 m g /k g )
5 c (3 0 m g /k g )
5 -F U (2 0 m g /k g )
0
1
2
3
4
5
6
7
8
9
1 0 1 1 1 2 1 3
T im e (d )
Fig.6 In vivo antitumor efficacy assay of compound 5c. Swiss HepS xenografts-bearing mice were

ACCEPTED MANUSCRIPT
given a daily intraperitoneal injection for 14 days with normal saline, 5c (15, 30 mg/kg), 5-FU
(20mg/kg). The body weight changes as a function of time (A); the tumor weight (B); the tumor
mode (C); effects of 5c against thymus (D) and spleen index (E). Data represent mean ± S.E.M (n
= 6) from three independent experiments, *P < 0.05, vs control.
Fig. 7 Western blot analysis of Akt, p53, Bcl-2, Bcl-XL and Bax in HepG2 cells treated with
different concentrations (for 24 h) and different time (for 4 μmol/L) of compound 5c. GAPDH was
used as a loading control.

ACCEPTED MANUSCRIPT
Scheme 1 The synthesis of 5a-n.
Reaction conditions and reagents: (i) thionyl chloride, CH2Cl2; (ii) K2CO3, BnBr, DMF/acetone;
(iii) pyridine, DMAP; (iv) H2/Pd/C, MeOH/CH2Cl2, rt.

ACCEPTED MANUSCRIPT
Research Highlights
1. Novel bergenin/cinnamic acid hybrids with better aqueous and lipid solubility balance.
2. Superior in vitro and moderate in vivo antitumor activity as well as low toxicity.
3. 5c arrested HepG2 cells in G2/M phase and induced mitochondria-mediated apoptosis.
4. 5c inhibited the expression of Akt and Bcl-2 family proteins in a dose- and time- dependent
manner.
5. 5c could be novel Akt/Bcl‑ 2 signaling pathway inhibitor for further preclinical studies.