Protective effect of caffeic acid derivatives on tert-butyl hydroperoxide-induced oxidative hepato-toxicity and mitochondrial dysfunction in HepG2 cells

Article abstract of DOI:10.3390/molecules22050702

Oxidative stress results in structural and functional abnormalities in the liver and is thought to be a crucial factor in liver diseases. The aim of this study was to investigate the cytoprotective and antioxidant effects of caffeic acid (CA) derivatives

Full text of DOI:10.3390/molecules22050702

molecules
Article
Protective Effect of Caffeic Acid Derivatives on
tert-Butyl Hydroperoxide-Induced Oxidative
Hepato-Toxicity and Mitochondrial Dysfunction in
HepG2 Cells
Tzung-Hsun Tsai 1,†, Chun-Hsien Yu 2,3,†, Yu-Ping Chang 4, Yu-Ting Lin 4, Ching-Jang Huang 5,
Yueh-Hsiung Kuo 6,7,* and Po-Jung Tsai 4,
*
1
Department of Dentistry, Keelung Chang-Gung Memorial Hospital, Keelung 204, Taiwan;
tts1725@gmail.com
Department of Pediatrics, Taipei Tzu-Chi Hospital, Buddhist Tzu-Chi Medical Foundation,
New Taipei City 231, Taiwan; chryu@mail.tcu.edu.tw
2
3
4
Department of Pediatrics, College of Medicine, Buddhist Tzu-Chi University, Hualien 970, Taiwan
Department of Human Development and Family Studies, National Taiwan Normal University, Taipei 106,
Taiwan; elf_elisa@yahoo.com.tw (Y.-P.C.); linyuting830812@gmail.com (Y.-T.L.)
Institute of Microbiology and Biochemistry, and Department of Biochemical Science and Technology,
National Taiwan University, Taipei 106, Taiwan; cjjhuang@ntu.edu.tw
Department of Chinese Pharmaceutical Sciences and Chinese Medicine Resources, China Medical University,
Taichung 404, Taiwan
Department of Biotechnology, Asia University, Taichung 413, Taiwan
Correspondence: kuoyh@mail.cmu.edu.tw (Y.-H.K.); pjtsai@ntnu.edu.tw (P.-J.T.);
Tel.: +886-4-2205-3366 (ext. 5709) (Y.-H.K.); +886-2-7734-1455 (P.-J.T.)
These authors contributed equally to this work.
5
6
7
*
†
Academic Editor: Diego Muñoz-Torrero
Received: 20 February 2017; Accepted: 25 April 2017; Published: 28 April 2017
Abstract: Oxidative stress results in structural and functional abnormalities in the liver and is thought
to be a crucial factor in liver diseases. The aim of this study was to investigate the cytoprotective
and antioxidant effects of caffeic acid (CA) derivatives on tert-butyl hydroperoxide (t-BHP)-induced
oxidative stress in HepG2 cells. Nine CA derivatives were synthesized, including N-phenylethyl
caffeamide (PECA), N-(3-florophen)methyl caffeamide (FMCA), N-(4-methoxy-phen)methyl
caffeamide (MPMCA), N-heptyl caffeamide (HCA), N-octyl caffeamide (OCA), octyl caffeate (CAOE),
phenpropyl caffeate (CAPPE), phenethyl caffeate (CAPE), and phenmethyl caffeate (CAPME).
The results showed that CA and its derivatives significantly inhibited t-BHP-induced cell death
of HepG2 cells. The rank order of potency of the CA derivatives for cytoprotection was CAOE > HCA
> OCA > FMCA > CAPPE > CAPME > CAPE > PECA > MPMCA > CA. Their cytoprotective activity
was associated with lipophilicity. The antioxidant effect of these compounds was supported by
the reduction in the levels of thiobarbituric acid reactive substrates, a biomarker of lipid peroxidation,
in HepG2 cells. Pre-treatment of CA derivatives significantly prevented the depletion of glutathione,
the most important water-soluble antioxidant in hepatocytes. Pre-treatment of CA derivatives
before t-BHP exposure maintained mitochondrial oxygen consumption rate and ATP content in the
injured HepG2 cells. CA derivatives except OCA and HCA significantly suppressed t-BHP-induced
hypoxia-inducible factor-1α (HIF-1α) protein level. In addition, all of these CA derivatives markedly
increased the nuclear factor erythroid 2-related factor 2 (Nrf2) accumulation in the nucleus, indicating
that their cytoprotection may be mediated by the activation of Nrf2. Our results suggest that CA
derivatives might be a hepatoprotective agent against oxidative stress.
Keywords: caffeic acid derivatives; tert-butyl hydroperoxide; antioxidant; liver protection
Molecules 2017, 22, 702; doi:10.3390/molecules22050702
www.mdpi.com/journal/molecules

Molecules 2017, 22, 702
2 of 16
1. Introduction
Increased oxidative stress and associated high levels of free radical generation have been described
to occur during the pathogeneses of various diseases. Excessive reactive oxygen species (ROS)
production plays a crucial role in liver diseases, including paracetamol-induced liver damage, alcoholic
liver disease, nonalcoholic steatohepatitis, liver cirrhosis and fibrosis [1,2]. Chronic liver inflammation
leads to oncogenic mutations in numerous cellular signaling cascades. Elevated levels of ROS, induced
by hepatitis viruses, are considered as a key factor underlying the oncogenic effects of hepatitis
B and C viruses [
mitochondria DNA, stimulate apoptotic pathways, and increase the tendency for hepatocellular
necrosis [ ]. Therefore, antioxidant therapy alone or in combination with other pharmacological
3,4]. Oxidative damage is the most likely causative process to alter and deplete
5
strategies has been proposed as a reasonable treatment of a variety of liver diseases.
Caffeic acid (CA) occurs widely in food plants and is thus widely consumed in the human diet.
CA is absorbed in humans after oral intake and specific metabolites can be detected in the urine [
CA exerts various biological activities, such as anti-oxidation, anti-inflammatory, and immune
regulation effects [ ]. In addition, its derivatives also show unique pharmaceutical functions.
6].
7
For example, the phenethyl ester of caffeic acid (CAPE), an active component of honeybee propolis
extract, possesses antimicrobial, antioxidant, anti-inflammatory, antitumoral, and neuroprotective
properties [8]. Due to their potential health benefits, CA and its derivatives have attracted considerable
interest in the past few years. As a result, CA has been widely used as a template for the development
of new chemical entities with a potential therapeutic interest in human diseases associated with
oxidative stress [9]. As the synthesis of esters, amides and hybrids base on the CA scaffold could
yield new or effective agents against oxidative stress, we prepared nine CA derivatives using CA as
a substrate, including N-phenylethyl caffeamide (PECA), N-(3-florophen)methyl caffeamide (FMCA),
N-(4-methoxyphen)methyl caffeamide (MPMCA), N-heptyl caffeamide (HCA), N-octyl caffeamide
(OCA), octyl caffeate (CAOE), phenpropyl caffeate (CAPPE), phenethyl caffeate (CAPE; phenethyl
ester of caffeic acid), and phenmethyl caffeate (CAPME) in this study (Figure 1).
Figure 1. Chemical structures of caffeic acid derivatives.

Molecules 2017, 22, 702
3 of 16
tert-Butyl hydroperoxide (t-BHP), a pro-oxidant, induces oxidative stress and cell injury that result
from the intracellular production of ROS. t-BHP has been used as a model compound to investigate the
mechanisms of cell damage initiated by oxidative stress. In hepatocyte cultures and livers, t-BHP can be
metabolized to free radical intermediates, and these intermediates result in initiating lipid peroxidation,
decreasing mitochondrial membrane potential, changes in mitochondrial membrane integrity, depletion
of cellular stores of glutathione (GSH), damage to cell integrity and hepatic inflammation [10–12].
HepG2 cells have been considered to be a good tool to study the in vitro xenobiotic metabolisms,
and cytoprotective, genotoxic, and antigenotoxic effects of compounds, as they retain the activity of
many phase I, phase II, and antioxidant enzymes [13–16]. Here, we evaluated the protective effects
of the nine CA derivatives on t-BHP-induced oxidative injury and mitochondrial dysfunction in
HepG2 cells.
2. Results
2.1. Protective Effect of CA Derivatives against t-BHP Induced Cytotoxicity
Treatment of the cells with CA and its derivatives (5, 10 and 20
no apparent cytotoxicity. When the HepG2 cells treated for 72 h with increasing concentrations
of CA and its derivatives, OCA, HCA, CAOE, and CAPPE (100 M) significantly decreased the
cell viability (Figure 2c,d). Then, we chose non-toxic concentrations (1–20 M) for all subsequent
µM) for 24 h (Figure 2a,b) had
µ
µ
experiments. The cytoprotection effect of CA derivatives against t-BHP induced cell death was
evaluated by pre-incubating the cells with CA derivatives for 24 h, followed by treatment with 0.7 mM
of t-BHP for 24 h. Silymarin (20 µM) was used as a control.
120
100
80
60
40
20
0
120
100
80
60
40
20
0
5
10 20
5
10 20
PECA
5
10 20
5
10 20
OCA
Control DMSO
5
10 20
HCA
(μM)
20
5
10 20
5
10 20
CAPE
5
10 20
5
10 20 (μM)
5
10 20
Control DMSO
Silymarin
FMCA
MPMCA
CAPPE
CAPME
CA
CAOE
(a)
(b)
120
100
80
60
40
20
0
120
*
100
80
60
40
20
0
*
*
*
(μΜ)
(μΜ)
Control DMSO 20 50 100
CAOE
20 50 100 20 50 100 20 50 100
20 50 100
Control DMSO 20 50 100 20 50 100
20 50 100 20 50 100
20 50 100 20 50 100
HCA Silymarin
PECA FMCA MPMCA OCA
CAPE
CAPPE CAPME CA
(c)
(d)
Figure 2. Effect of caffeic acid derivatives on HepG2 cell viability. HepG2 cells were treated with
different concentration of samples for 24 h ( ) and 72 h ( ). Cell viability is expressed as the
a,
b
c,d
percentage of vehicle control (0.1% DMSO). Control, medium alone without DMSO. Values were
shown as mean ± SD (n = 3). *, p < 0.05 vs. DMSO-vehicle.

Molecules 2017, 22, 702
4 of 16
As shown in Figure 3, all CA derivatives (10 and 20
µM) showed significant cytoprotection effects
against t-BHP-induced cell death. At the concentration of 20
inhibiting t-BHP-mediated toxicity and was more potent than silymarin. CA was less efficient and
did not achieve the IC50 in the experiments with t-BHP up to the highest concentration tested (20 M).
All CA derivatives were more potent than CA (Figure 3). The potency of the CA and its derivatives
was expressed as IC50 values. The rank order of potency of them for cytoprotection was CAOE (5.9 M)
M) >
µM, FECA was the least effective in
µ
µ
> HCA (11.4
µM) > OCA (11.7
µM) > FMCA (11.9
µM) > CAPPE (12.1
µM) > CAPME (12.4
µ
CAPE (12.7 µM) > PECA (14.1 µM) > MPMCA (18.8 µM) > CA (not applicable).
120
120
a
a
a
a
100
80
60
40
20
0
100
80
60
40
20
0
bc
bc
c
b
b
b
cd
d
c
c
e
d
e
fg
ef
f
e
fg
gh
gh
ghi
g
hij
hi
hi
ijk
ghi
gh
ij
hij
hij
hij
ijk
jk
jk
jk
hij
jk
ij
jk
j
ij
k
hi
i
1
5
10 20
1
5
10 20
1
5
10 20
1
5
10 20
1
5
10 20
1
5
10 20
1
5
10 20
1
5
10 20
1
5
10 20
1
5
10 20
DMSO
Control
Control Vehicle
20
Vehicle DMSO
CAOE (μM) CAPE (μM) CAPPE (μM) CAPME (μM)
CA (μM)
HCA (μM)
PECA (μM)
FMCA (μM) MPMCA (μM)
OCA (μM)
Silymarin
(μM)
t-BHP (0.7 mM)
t-BHP(0.7 mM)
(a)
(b)
Figure 3. Protective effects of caffeic acid derivatives of t-BHP-induced cytotoxicity. HepG2 cells
were treated with different concentrations of caffeamide derivatives ( ) and caffeate derivatives ( ) for
24 h before being exposed to t-BHP (0.7 mM) for 20 h. The vehicle and positive controls respectively
received DMSO (0.1%) and silymarin (20 M). Cell viability is expressed as the percentage of control.
Values were shown as mean SD (n = 3). Data were analyzed by one-way ANOVA and Duncan
a
b
µ
±
multiple comparison test. Values not sharing the same letter are significantly different at p < 0.05.
2.2. Effects of CA Derivatives on t-BHP-Induced Lipid Peroxidation
The thiobarbituric acid reactive substrates (TBARS) assay is still the most commonly used assay
for the measurement of lipid peroxidation in biomaterials. It is relatively simple, fast, and cost-effective.
1.2
1.0
0.8
0.6
0.4
0.2
0.0
1.2
1.0
0.8
0.6
0.4
0.2
0.0
l
i
k
h
jk
h
j
h
h
gh
i
g
f
d
h
h
ef
h
h
d
fg
de
g
g
g
ef
cd
c
d
cd
c
bc
c
cd
c
cd
bc
bcd
b
ab
a
a
a
a
i
a
Control DMSO
1
5 10 20
1
5 10 20
20
1
5 10 20
1
5 10 20
1
5 10 20
1
5
10 20
1
5 10 20
Control DMSO
1
5 10 20
1
5 10 20
1
5 10 20
PECA (μM)
OCA (μM)
FMCA (μM) MPMCA (μM)
HCA (μM) Silymarin
(μM)
CAPME (μM)
CA (μM)
CAPE (μM)
CAOE (μM)
CAPPE (μM)
t-BHP (0.5 mM)
t-BHP (0.5 mM)
(a)
(b)
Figure 4. Inhibitory effects of caffeic acid derivatives on t-BHP-induced lipid peroxidation in HepG2
cells. Pre-treatment with caffeamide derivatives ( ) and caffeate derivatives ( ) for 24 h, HepG2 cells
were stimulated with t-BHP (0.5 mM) for 24 h. The vehicle and positive controls respectively received
DMSO (0.1%) and silymarin (20 M). The amount of TBARS formation is expressed as nmol/mg protein.
Values were shown as mean SD (n = 3). Data were analyzed by one-way ANOVA and Duncan
multiple comparison test. Values not sharing the same letter are significantly different at p < 0.05.
a
b
µ
±

Molecules 2017, 22, 702
5 of 16
As shown in Figure 4, the effect of CA derivatives on TBARS formation in t-BHP-treated HepG2
cells was examined. Treatment of HepG2 cells with 0.5 mM t-BHP for 24 h resulted in a significant
increase in TBARS concentration. However, pretreatment with the CA derivatives (at concentrations
of 5, 10, and 20
Pre-treatments with 20
CAPME reduced t-BHP-induced TBARS formation by 44.4%, 43.0%, 42.2%, 47.1%, 45.8%, 44.7%,
47.9%, 49.2%, and 49.6%, respectively. Five CA derivatives (20 M), including OCA, CAOE, CAPE,
M) in the inhibition of t-BHP-induced
µM) effectively inhibited the increase in TBARS production induced by t-BHP.
µM of PECA, FMCA, MPMCA, OCA, HCA, CAOE, CAPE, CAPPE, and
µ
CAPPE, and CAPME, are more potent than silymarin (20
lipid peroxidation.
µ
2.3. Effects of CA Derivatives on Glutathione Levels
Treatment with 0.5 mM t-BHP induced a remarkable decrease in GSH levels compared to
the control and DMSO vehicle (60.5 and 24.7 nmol/mg protein, respectively). Pre-treatments with
20 µM of PECA, FMCA, MPMCA, OCA, HCA, CAOE, CAPE, CAPPE, and CAPME prevented GSH
depletion by 76.0%, 76.0%, 79.8%, 74.5%, 75.7%, 74.5%, 89.2%, 82.8%, and 74.5%, respectively (Figure 5).
Among the tested compounds, CAPE presented the highest efficiency in a concentration-dependent
manner (Figure 5).
80
80
a
a
60
40
20
0
60
40
20
0
cd
b
bc
cdef
b
cde
bc
defg
bc
bc
bcd
cde
fgh
de
de
cde
efgh
e
cde
bcd
f
e
f
f
hi
gh
f
hi
ij
fg
fgh
ij
j
hi
ghi
i
j
i
hi
hi
i
j
j
20
1
5
10 20
1
5
10 20
1
5
10 20
1
5
10 20
1
5
10 20
Control DMSO
Control DMSO
1
5
10 20
1
5
10 20
1
5
10 20
1
5
10 20
1
5
10 20
PECA (μM)
FMCA (μM) MPMCA (μM) OCA (μM)
HCA (μM)
Silymarin
(μM)
CA (μM)
CAOE (μM)
CAPE (μM)
CAPPE (μM) CAPME (μM)
t-BHP (0.5mM)
t-BHP (0.5mM)
(a)
(b)
Figure 5. Effects of caffeic acid derivatives on the glutathione levels in HepG2 cells after the oxidative
damage induced by t-BHP. Pre-treatment with caffeamide derivatives ( ) and caffeate derivatives (
for 24 h, HepG2 cells were stimulated with t-BHP (0.5 mM) for 24 h. The vehicle and positive controls
respectively received DMSO (0.1%) and silymarin (20 M). The amount of cellular glutathione level
is expressed as nmol/mg protein. Values were shown as mean SD (n = 3). Data were analyzed
a
b)
µ
±
by one-way ANOVA and Duncan multiple comparison test. Values not sharing the same letter are
significantly different at p < 0.05.
2.4. Effects of CA Derivatives on t-BHP-Induced Mitochondrial Dysfunction
Further investigation on mitochondrial oxygen consumption was conducted with CA derivatives
at the concentration of 20 µM. As shown in Figure 6, t-BHP abolished mitochondrial respiration
capacity including basal, maximal, and spare respirations as well as ATP production, all of which were
significantly improved by CA derivatives pre-treatment. In contrast, at the same concentrations, CA is
ineffective in maintaining basal, maximal, and spare respirations. The CA derivatives are superior to
CA in protecting against oxidant-induced mitochondrial dysfunction.

Molecules 2017, 22, 702
6 of 16
1000
Oligomycin
FCCP
Antimycin A & Rotenone
control
t-BHP
800
600
400
200
0
t-BHP + CAOE
t-BHP + CAPPE
t-BHP + CAPME
t-BHP + CAPE
t-BHP + PECA
t-BHP + FMCA
t-BHP + MPMCA
t-BHP + HCA
t-BHP + OCA
t-BHP + CA
0
20
40
60
80
100
120
Time (minutes)
(a)
Basal respiration
ATP production
300
250
200
150
100
500
400
300
200
100
0
a
a
b
b
b
bc
bc
bc
bc
cd
c
bc
bc
cd
bc
bcd
bc
cd
bc
bc
cd
de
d
50
0
e
t-BHP (0.5mM)
t-BHP (0.5mM)
Maximal respiration
Spare respiratory capacity
1000
400
300
200
100
0
a
a
800
600
400
200
0
b
bc
b
bc
bcd
cd
cd
cd
cde
cde
cde
bcd
cd
cde
de
cde
cde
f
de
f
e
e
t-BHP (0.5mM)
t-BHP (0.5mM)
(b)
Figure 6. Effect of CA derivatives on t-BHP-induced mitochondria dysfunction. HepG2 cells were
treated with CA and CA derivatives (20 M) or control (vehicle, 0.1% DMSO) for 6 h, followed by
0.5 mM t-BHP for an additional 2 h. Respiration curve ( ) and statistical analyses of basal, maximal and
spare mitochondrial respiration capacity, and ATP production ( ) were analyzed. Values were shown
as mean SD (n = 3). Data were analyzed by one-way ANOVA and Duncan multiple comparison test.
µ
a
b
±
Values not sharing the same letter are significantly different at p < 0.05.
2.5. Effects of CA Derivatives on HIF-1α and Nrf2 Levels
Next, we determined whether t-BHP affects hypoxia-inducible factor-1α (HIF-1α) protein level.
As shown in Figure 7a-1, HIF-1α protein expression was markedly induced by t-BHP at 12 h and then

Molecules 2017, 22, 702
7 of 16
declined. All of these CA derivatives except OCA and HCA significantly inhibited the t-BHP-induced
HIF-1 protein levels. While CA did not significantly attenuate HIF-1 expression induced by t-BHP
α
α
(Figure 7a-2). The translocation of nuclear factor erythroid-2-related factor (Nrf2) from the cytosol to
nucleus is a key step for Nrf2 activation [17]. Thus, we explore whether CA derivatives could increase
Nrf2 accumulation in the nucleus. As shown in Figure 7b, when HepG2 cells were treated with t-BHP
for 12 h, t-BHP significantly decreased nuclear Nrf2 level as compared to the untreated control cells.
Pretreatment with CA and silymarin (20
µM) markedly increased Nrf2 accumulation in the nucleus.
All of these CA derivatives (20 M) also significantly enhanced nuclear Nrf2 level when cells were
µ
incubated with t-BHP for 12 h (Figure 7b). Our data demonstrated that nuclear Nrf2 expression was
enhanced by CA derivatives in t-BHP-treated HepG2 cells.
(a)
(b)
Figure 7. Effect of CA derivatives on HIF-1
α
and Nrf2 protein levels in t-BHP-treated HepG2 cells.
Cells were treated with t-BHP (0.5 mM) alone for the indicated times (
compounds (20 M) for 24 h, cells were then stimulated with t-BHP (0.5 mM) for 12 h (
amounts of whole cell lysate (WCL) were subjected to immunoblotting to determine the HIF-1
Quantification of protein levels was normalized to -actin ( ). Pre-treatment with test compounds
(20 M) for 24 h, cells were treated with t-BHP (0.5 mM) for 12 h. Then, equal amounts of nuclear
protein were subjected to western blot analysis to determine the Nrf2 level. Quantification of protein
levels was normalized to lamin B ( ). The vehicle and positive controls respectively received DMSO
(0.1%) and silymarin (20 M). Results were expressed as mean SD of three experiments. Data were
a-
1
). Pre-treatment with test
). The equal
level.
µ
a-2
α
β
a
µ
b
µ
±
analyzed by one-way ANOVA and Duncan multiple comparison test. Values not sharing the same
letter are significantly different at p < 0.05.
3. Discussion
Numerous studies have demonstrated the biological and pharmacological actions of CA
derivatives, especially CAPE. CAPE is a potent antioxidant, which effectively scavenges ROS and
protects the cell membrane against lipid peroxidation [
8]. Moreover, CAPE possesses hepato-protective
activity against tetrachloride or cholestasis-induced liver injury [18
,
19] and t-BHP-induced cytotoxicity
in HepG2 cells and rat liver [20]. In addition, CAOE exerts anti-virus activity against hepatitis C
virus, which is a major causative agent of chronic liver disease including cirrhosis and hepatocellular
carcinoma [21]. Caffeamide derivatives are more stable compound than CAPE and could potentially
extend the beneficial effects of CAPE [22]. In particular, PECA prevents photo-damage and protects

Molecules 2017, 22, 702
8 of 16
against the metabolic consequences of diabetes mellitus [23
,24]. Several catechol ring-fluorinated
derivatives of CA phenethylamide show cytoprotective activity against H2O2 induced oxidative stress
in human umbilical vein endothelial cells [25]. To the best of our knowledge, there is no study relating
the biological and pharmacological activities of FMCA, MPMCA, HCA, OCA, and CAPME. Moreover,
the information concerning the hepato-protective action of CA derivatives is still limited, with the
exception of CAPE. Therefore, the possible cytoprotective effect of the nine CA derivatives against
t-BHP-induced cell death of hepatocytes was investigated in this study.
Silymarin displays antioxidant, anti-inflammatory, and immunomodulatory actions [26] and
has been widely used as a therapeutic agent for a variety of acute and chronic liver diseases.
Silymarin reverses the GSH content and reduces lipid peroxides (malondialdehyde, MDA) in
isolated rat hepatocytes and human erythrocytes that were altered by t-BHP [27–29]. Due to its
proven hepatoprotective and antioxidant properties, we used silymarin as a positive control in this
study. Death of liver cells is a characteristic feature of many liver diseases, such as cholestasis,
hepatitis, or ischemia/reperfusion [30]. In this study, all nine CA derivatives exerted similar effective
cytoprotection as compared to silymarin (Figure 3). In addition, cells treated with the nine CA
derivatives showed higher protective effect compared to cells treated with CA against oxidative cell
injury (Figure 3). Interestingly, CA-derived ester analogues are decomposed by esterases that lead to
their low bioavailability in vivo [31]. However, CAOE exhibited the most potent cytoprotective activity
in this study. Concerning structure activity relationship, Wu et al. [32] reported that the antioxidative
activity of CAPE and its related compounds (CA, ferulic acid, and ethyl ferulate) depends on the
hydrogen groups or catechol rings, the polarity, hydrophobicity (or partitioning properties between
lipid and aqueous phases) and stability of the antioxidants. All CA derivatives used in this work have
the same ortho-dihydroxyl functionality in the catechol ring. In our present experimental condition,
the crucial structural feature responsible for the better cytoprotection activities of CAOE, OCA and
HCA is their hydrophobic side chains. For CAOE, OCA and HCA, the conversion of the acid group
to the ester or amide group decrease the molecular polarity. The higher lipophilicity might facilitate
the transport of these compounds through cell membrane and reach cell interior, which then these
compounds could scavenge the ROS where they were produced.
Next, investigation of whether cytoprotective action of CA derivatives is related to the antioxidant
capacity was conducted. Oxidative stress is an imbalance between pro-oxidants and anti-oxidants.
Overproduction of ROS or inadequate antioxidant defenses (e.g., low levels of vitamins, GSH), or both,
can lead to liver injury. Intracellular MDA concentration is a direct result of membrane unsaturated
fatty acid peroxidation and was used as a biomarker for t-BHP-induced oxidation in HepG2 cells.
All CA derivatives efficiently inhibited the TBARS formation (Figure 4). Moreover, GSH depletion is
considered a potential biomarker of drug-induced hepatotoxicity. Because hepatic mitochondria lack
catalase, GSH plays a critical role in protecting mitochondria against oxidative stress [33]. Our data
demonstrated that all CA derivatives prevented GSH depletion in t-BHP-treated HepG2 cells (Figure 5).
Hepatocytes are normally rich in mitochondria and each hepatocyte contains about 800
mitochondria occupying about 18% of the entire liver cell volume. Mitochondria play an important role
in hepatocyte metabolism and ATP production. The mitochondrial dysfunction contributes to oxidative
stress. Oxidative stress is believed to play an important role in the development of steatosis and liver
lesion [34]. We further examined whether CA derivatives preserve mitochondrial respiration and
energy production capacity in the presence of t-BHP. Mitochondrial metabolic function was determined
by measuring mitochondrial oxygen consumption rate (OCR) using an XF24 analyzer. After t-BHP
treatment, OCR of HepG2 cells rapidly declined, confirming mitochondrial dysfunction (Figure 6)
and consistent with previous studies in myotubes [35]. Pre-treatment with CA derivatives preserved
mitochondrial OCR in t-BHP-treated cells (Figure 6). All CA derivatives herein prevent mitochondrial
dysfunction and energy deficits caused by oxidant exposure. Mitochondrial dysfunction is a common
mechanism in the etiology of organ injuries and diseases caused by metabolic insufficiency. Our results
highlight the potential of CA derivatives to treat disorders characterized by mitochondrial impairment.

Molecules 2017, 22, 702
9 of 16
HIF-1
transmitting information regarding cellular oxygen levels. There is evidence to suggest that ROS
regulate HIF-1 stabilization under normoxia. Addition of exogenous H2O2 stabilizes HIF-1
protein under normoxia [36]. After 12 h treatment of t-BHP, an exogenous inducer of oxidative
stress, increased HIF-1 protein level (Figure 7a-1). While, CA derivatives except OCA and HCA
significantly decreased t-BHP-induced HIF-1 protein level under normoxia (Figure 7a-2). However,
the mechanisms explaining how HIF-1 could mediate t-BHP-induced damage remains unclear.
α is a heterodimeric transcription factor that plays a key role in the signaling pathways
α
α
α
α
α
Nrf2 signal plays a crucial role against oxidative stress and cytotoxicity. The enhanced activation of
Nrf2 exerts hepatoprotective activity in different oxidative stress models, such as acetaminophen or
CCl4-induced liver injury [5,37]. Activation of Nrf2 protects mitochondria from oxidative stress [5].
CAPE has been shown to protect liver cells against CCl4-induced oxidative stress and inhibit activation
of hepatic stellate cells through up-regulated Nfr2 expression [38]. In this study, we demonstrated that
all CA derivatives increased nuclear Nrf2 accumulation (Figure 7b). Taken together, these findings
suggested that Nrf2 signaling pathways could play a key role in mediating the hepatoprotective effects
of CA derivatives.
In conclusion, our data suggested that CA derivatives have protective effects against
oxidant-induced hepatotoxicity, at least in part, by attenuation of lipoperoxidation, preservation
of glutathione, inhibition of mitochondrial dysfunction, and regulation of the master regulator of
antioxidant response Nrf2. However, further studies are still required to validate the detailed molecular
mechanisms involved in the protective effect of CA derivatives.
4. Materials and Methods
4.1. General Information
All chemicals, including caffeic acid, N,N-dimethylformamide, trimethylamine, BOP, CH2Cl2,
EtOAc, HCl, thionyl chloride, SOCl2, NaHCO3 (aq.), and MgSO4, were of analytical-grade purity.
They were purchased from the branch of Merck in Taipei, Taiwan, and the branch of Sigma Aldrich in
Taichung, Taiwan. UV spectra were recorded on a Shimadzu UV-1601PC spectrophotometer. Infrared
1
(IR) spectra were measured on a Perkin-Elmer-983G FT-IR spectrophotometer. H- and 13C-NMR and
2D NMR spectra were obtained on a Bruker DRX-500 FT-NMR spectrometer with tetramethylsilane
(TMS) as the internal standard. EI-MS were measured with a Jeol-JMSHX300 mass spectrometer. Silica
gel (230–400 mesh; Merck & Co., Inc.) was used for column chromatography (CC). Pre-coated silica gel
(60 F-254; Merck & Co., Inc.) plates were used for TLC. The spots on TLC were detected by spraying
with 5% H2SO4 and then heating at 100 ◦C.
4.2. Synthesis of Caffeamide Derivatives
Compounds were obtained by the following method: caffeic acid (100 mg, 0.56 mmol) was
dissolved in the mixture of N,N-dimethylformamide (DMF, 1 mL) and trimethylamine (80
µL), and then
the appropriate amine derivative (80 L, assumes that they all have the same density and molecular
µ
◦
weight, 1.2 eq.) was added to the reaction solution cooled in an ice bath (0 C), and then BOP (295 mg,
1.2 eq.) in CH2Cl2 (5 mL) was added to the reaction mixture, followed by stirring for 30 min (Scheme 1).
The reaction mixture was kept at room temperature for 2 h. The solvent was then evaporated under
vacuum, and the crude mixture was partitioned between EtOAc and H2O. The organic layer was
washed with 3 N aqueous HCl and then with 10% aqueous NaHCO3 solution. The product was
purified by chromatography on silica gel. The yields of PECA, FMCA, MPMCA, OCA, and HCA
were 70%~80%.

Molecules 2017, 22, 702
10 of 16
Scheme 1. Synthesis of caffeic acid amide derivatives.
N-Phenylethyl caffeamide (PECA): Pale white needle-like crystals; m.p.: 148–149 ◦C; IR
3288, 1642, 1591, 1523, 1361, 1279, 1036, 975, 849. 1H-NMR (CD3COCD3, 500 MHz):
ν
δ
(cm−1):
2.84 (2H, t,
max
J = 6.8 Hz), 3.53 (2H, q, J = 6.8 Hz), 6.43 (1H, d, J = 15.2 Hz), 6.83 (1H, d, J = 8.1 Hz), 6.92 (1H, dd, J = 8.1
,
1.8 Hz), 7.07 (1H, d, J = 1.8 Hz), 7.15–7.30 (5H, m), 7.35 (-NH, br. s), 7.43 (1H, d, J = 15.2 Hz), 8.20 (-OH,
s), 8.42 (-OH, s). EI-MS m/z (%): 283 (M+, 17), 178 (22), 163 (100); UV (MeOH)
λmax (logε): 322 (4.42),
296 (4.36), 245 (4.30), 216 (4.61)nm.
N-(3-Florophen)methyl caffeamide (FMCA): White solid; m.p.: 186–188 ◦C; IR
1619, 1520, 1440, 1358, 1115, 1015, 976, 852, 818. H-NMR (CD3OD, 400 MHz):
ν
max (cm−1): 3436, 1652,
4.47 (2H, s), 6.40 (1H,
1
δ
d, J = 15.6 Hz), 6.75 (1H, d, J = 8.0Hz), 6.90 (1H, dd, J = 8.0, 2.0 Hz), 6.98 (1H, m), 7.01 (1H, d, J = 2.0 Hz),
7.03 (1H, m), 7.12 (1H, m), 7.33 (1H, m), 7.43 (1H, d, J = 15.6 Hz). EI-MS m/z (%): 287 (M+, 100), 247 (35),
163 (95), 124 (90), 109 (50); UV(MeOH) λmax (logε): 324 (4.37), 296 (4.30), 245 (4.28), 251(4.54) nm.
N-(4-Methoxyphen)methyl caffeamide (MPMCA): Solid; m.p.: 170–171 ◦C; IR
ν
(cm−1): 3283, 1653,
max
1
1613, 1520, 1447, 1374, 1116, 1009, 850. H-NMR (CD3COCD3, 500 MHz): 3.75 (3H, s), 4.44 (2H, d,
J = 6.2 Hz), 6.49 (1H, d, J = 15.8 Hz), 6.81–6.94 (4H, m), 7.07 (1H, d, J = 1.6 Hz), 7.25 (2H, d, J = 8.8 Hz),
7.45 (1H, d, J = 15.8 Hz), 7.59 (1H, br. s, -NH), 8.17 (1H, s, -OH), 8.38 (1H, s, -OH). EI-MS m/z (%):
299 (M+, 7), 163 (100); UV (MeOH) λmax (logε): 321(4.16), 295 (4.13), 284 (4.13), 245 (4.16) nm.
N-Octyl caffeamide (OCA): White solid; m.p.: 111–112 ◦C; IR
ν
(cm−1): 3286, 1642, 1588, 1520, 1363,
max
1
1277, 1112, 975, 811. H-NMR (CD3COCD3, 400 MHz):
δ 0.84 (3H, t, J = 6.6 Hz), 1.24 (10H, m), 1.52 (2H,
quin, J = 6.6 Hz), 3.30 (2H, q, J = 6.6 Hz), 6.47, 7.42 (each 1H, d, J = 15.6 Hz), 6.82 (1H, d, J = 8.2 Hz),
6.90 (1H, dd, J = 8.2, 1.8 Hz), 7.09 (1H, d, J = 1.8 Hz). EI-MS m/z (%): 291 (M+, 18), 220 (8), 193 (11),
178 (31), 163 (100), 145 (8), 135 (13), 128 (22), 117 (11), 98 (8), 89 (19), 84 (12); UV (MeOH)
322 (4.32), 294 (4.26), 238 (3.92), 219 (4.37) nm.
λmax (logε):
◦
N-Heptyl caffeamide (HCA): White solid; m.p.: 126–127 C; IR
ν
(cm−1): 3347, 1642, 1588, 1545, 1510,
0.82 (3H, t, J = 6.6 Hz), 1.20 (8H, m),
max
1
1363, 1266, 1112, 975, 809. H-NMR (CD3COCD3, 400 MHz):
δ
1.52 (2H, quin, J = 6.6 Hz), 3.32 (2H, q, J = 6.6 Hz), 6.51, 7.48 (each 1H, d, J = 15.6 Hz), 6.83 (1H, d,
J = 8.1 Hz), 6.92 (1H, dd, J = 8.1, 1.5 Hz), 7.11 (1H, d, J = 1.5 Hz). EI-MS m/z (%): 277 (M+, 42), 192 (10),
178 (32), 163 (100), 145 (12), 135 (20), 114 (14), 98 (8); UV (MeOH)
2.34 (4.16), 217 (4.30) nm.
λ
max
(logε): 322 (4.23), 295 (4.17),
4.3. Synthesis of Caffeate Derivatives
Caffeate derivatives were obtained as follows: Caffeic acid (200 mg) and thionyl chloride (4 mL)
dissolved in dry dichloromethane (10 mL), were heated under reflux for 4 h. The solvent and excess
SOCl2 was removed under vacuum, and then ROH (1.2 equiv.) in triethylamine (0.08 mL) was added
dropwise under dry conditions. The reaction mixture was stirred for 24 h at ambient temperature,
and then evaporated under vacuum. The residue was partitioned successively between EtOAc and

Molecules 2017, 22, 702
11 of 16
H2O, and then the EtOAc layer was washed with 3N aqueous HCl then 10% NaHCO3 (aq.), dried over
MgSO4 and concentrated under vacuum. The product was purified by column chromatography on
silica gel. The final products (65%~75% yield) were recrystallized from acetone to obtain pure crystals.
(Scheme 2).
Scheme 2. Synthesis of caffeic acid ester derivatives.
Octyl caffeate (CAOE): White solid; m.p.: 98–100 ◦C; IR
ν
(cm−1): 3488, 3340, 1675, 1630, 1275, 1184,
max
972, 812; 1H-NMR (CD3COCD3):
δ 0.85 (3H, t, J = 6.7 Hz), 1.26 (10H, m), 1.67 (2H, quin, J = 6.7 Hz),
4.16 (2H, t, J = 6.7 Hz), 6.23, 7.54 (each 1H, d, J = 15.9 Hz), 6.84 (1H, d, J = 8.2 Hz), 6.96 (1H, dd, J = 8.2,
2.0 Hz), 7.06 (1H, d, J = 2.0 Hz), 8.26 (2H, brs, -OH). EI-MS m/z (%): 292 (M+, 27), 180 (100), 163 (47),
145 (8), 136 (18), 134 (12), 89 (13).
◦
Phenethyl caffeate (CAPE): Pale white needle-like crystals; m.p.: 126–128 C, IR (KBr)
ν
3400,
max
1
1709, 1605, 1519, 1352, 1282, 1192 cm−1; H-NMR (400 MHz, DMSO-d6)
δ
2.95 (2H, t, J = 6.9 Hz,
-OCH2C
H
2Ph), 4.32 (2H, t, J = 6.9 Hz, -OC
H
2CH2Ph), 6.23, 7.58 (each 1H, d, J = 15.9 Hz), 6.76 (1H, d,
J = 8.1 Hz), 6.99 (1H, d, J = 8.2, 2.0 Hz), 7.03 (1H, d, J = 2.0 Hz), 7.22 (1H, m, H-4’), 7.18–7.30 (5H, m);
EI-MS m/z (%): 284 (M+), 179, 161, 135 (100%), 117.
Phenpropyl caffeate (CAPPE): Yellow solid; m.p.: 116–118 ◦C; IR (KBr)
ν
(cm−1): 3482, 3327, 1671,
max
1
1629, 1597, 1179, 973, 809, 696; H-NMR (CDCl3):
δ 2.01 (2H, quin, J = 6.8 Hz, -OCH2CH2CH2Ph),
2.72, 4.20 (each 2H, t, J = 6.8 Hz), 6.25, 7.55 (each 1H, d, J = 15.9 Hz), 6.85 (1H, d, J = 8.2 Hz), 6.98 (1H,
dd, J = 8.2, 1.8 Hz), 7.08 (1H, d, J = 1.8 Hz), 7.10–7.30 (5H, m); EI-MS m/z (%): 298 (M+, 18), 180 (100),
163 (19), 135 (8), 118 (30), 117 (30), 91 (24).
◦
Phenmethyl caffeate (CAPME): Light yellowish powder; m.p.: 150–152 C, IR
ν
(cm−1): 3464,
max
1
3327, 1689, 1600, 1277, 1185. H-NMR (DMSO-d6, 400 MHz): δH 5.19 (1H, s), 6.32, 7.52 (each 1H, d,
J = 15.9 Hz), 6.75 (1H, d, J = 8.1 Hz), 7.02 (1H, dd, J = 2.0, 8.1 Hz), 7.05 (1H, d, J = 2.0 Hz), 7.34–7.41
(5H, m), 9.20, 9.59 (each 1H, s, -OH). EI-MS m/z (%): 270 (9%, M+), 208 (19), 163 (50), 136 (33), 91 (100),
89 (32), 77 (19), 65 (20), 51 (18).
4.4. HPLC Determination of the CA Derivatives
The purity of CA derivatives was measured by HPLC on a system (Ecom, Prague, Czech Republic)
equipped with gradient pumps (Ecom LCP 4100), a UV detector (Ecom LCD 2084) and a LiChrospher®
100 RP-18E (5
µm) HPLC column (125 mm
×
4 mm i.d., Merck Millipore (Darmstadt, Germany).
The mobile phase consisting of a mixture of solvent A (water/methanol, 98:2), and solvent B
(methanol/acetic acid, 98:2) was run in the following gradient mode: 0–7 min, from 50% A to 40% A
with a flow rate of 0.5 mL/min; 7–12 min, from 40% A to 30% A with a flow rate of 0.3 mL/min; and
12–28 min, from 30% A to 20% A with a flow rate of 0.2 mL/min. The UV detector was at 280 nm.
Chromatographic processing was done using the Peak-ABC Chromatography Data Handling System.
The purities of PECA, FMCA, MPMCA, OCA, HCA, CAOE, CAPE, CAPPE, and CAPME were 96.4%,
97.1%, 99.7%, 95.2%, 95.6%, 98.6%, 96.7%, 96.4%, and 98.3%, respectively (Figure 8).

Molecules 2017, 22, 702
12 of 16
(a)
(b)
Figure 8. Retention time and purity analysis of caffeamide derivatives (a) and caffeate derivatives (b)
by HPLC.
4.5. Determination of the Viability of HepG2 Cells
The human HepG2 cell line (BCRC 60025) was obtained from the Bioresource Collection and
Research Center (Hsinchu, Taiwan). Cells were maintained in DMEM (with 5.5 mM glucose)
supplemented with 10% heat-inactivated fetal bovine serum (FBS, Gibco, Grand Island, NY, USA),
◦
penicillin (100 U/mL), and streptomycin (100
5% CO2. HepG2 cells (5
µ
g/mL) at 37 C in a humidified atmosphere with
×
104 cells/well) were cultured in 96-well culture plates and 24 h later
treated with various concentrations of tested samples for the indicated times. Cell viability was
determined using the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide (MTT) assay.
MTT (Sigma-Aldrich, St. Louis, MO, USA; 100
µL, 0.5 mg/mL) was added to each well and incubated
at 37 ◦C for 3 h. The reaction was terminated by replacing the MTT-containing medium with 500
µL
of dimethyl sulfoxide (Sigma-Aldrich), and the formazan salts were dissolved by gentle shaking for
approximately 5 min at room temperature. The optical density (OD) of each well was measured at
595 nm using a microplate reader. Each assay was completed in triplicate wells, and each experiment
was repeated three times.
4.6. Cytoprotective Effects of CA Derivatives
HepG2 cells (5
was replaced containing various concentrations of caffeic acid and its derivatives (1, 5, 10, or 20
or 20 M silymarin (Sigma-Aldrich; as a positive control). Commercially available caffeic acid
×
104 cells/well) were seeded in 96-well plates. After 24 h, the culture medium
M)
µ
µ
(Sigma-Aldrich) was used as a reference. After 24 h, the culture medium containing the samples
was discarded, and the cells were treated for 20 h with 0.7 mM t-BHP (Sigma-Aldrich) to induce
oxidative stress. Cell viability, as an indication of the cytoprotective effects of each of the samples,
was then evaluated using the MTT assay.
4.7. Measurement of Lipid Peroxidation in HepG2 Cells
HepG2 cells were seeded in a 6-well plate at a density of 7.5
the culture medium was replaced with FBS-free medium containing different concentrations of CA and
its derivatives (1, 5, 10, or 20 M). After 24 h, the cells were treated for 24 h with 0.5 mM t-BHP to induce
×
105 cells/well. After 24 h,
µ
oxidative stress. After treatment, the c◦ells were harvested and lysed by sonication. The lysates were
centrifuged at 10,000
×
g for 10 min at 4 C and used for the lipid peroxidation assay using TBARS assay.
The absorbance was measured at 535 nm calibrated using solutions of authentic malondialdehyde
(Sigma-Aldrich). The protein concentration was determined using the Bradford dye-binding method
(Bio-Rad, Hercules, CA, USA). The results were expressed as nmol/mg protein.

Molecules 2017, 22, 702
13 of 16
4.8. Determination of Total GSH Level
HepG2 cells were seeded in 6-well plates at a density of 7.5
×
105 cells/well. After 24 h,
the culture medium was replaced with FBS-free medium containing various concentrations of CA
and its derivatives (1, 5, 10, and 20 M). After 24 h, the cells were treated for 24 h with 0.5 mM
t-BHP to induce oxidative stress. After treatment, the cells were harvested and lysed using a sonicator.
The lysates were centrifuged at 10,000 g for 10 min, and used for the determinations of protein and
µ
×
GSH. The GSH levels in HepG2 cells were determined by a commercially available GSH assay kit
(Cayman Chemical, Ann Arbor, MI, USA). The results were expressed as nmol GSH/mg of protein.
4.9. Measurement of Mitochondrial Oxygen Consumption Rate
HepG2 cells were cultured at 1.5
Billerica, MA, USA). After 42 h, cells were treated with 20
incubation, the media was replaced with 250
2 h. Next, the media was exchanged with 675
×
104 cells/well in XF24 cell culture plates (Seahorse Bioscience,
M of CA and its derivatives. After 6 h
L DMEM in the presence of 0.5 mM t-BHP for
L XF Assay Medium-modified DMEM (Seahorse
µ
µ
µ
Bioscience) incubated at 37 ◦C without CO2 for 1 h. Mitochondrial oxygen consumption rate (OCR)
was measured using an XF24 analyzer and software (Seahorse Bioscience). The cellular metabolic
capacity and mitochondrial coupling were evaluated by injecting the following metabolic probes
using the liquid injection ports on the XF instrument sensor plate: oligomycin, which inhibits the
mitochondrial F1F0-ATPase, FCCP (carbonyl cyanide 4-trifluoromethoxy-phenylhydrazone), which is
a protonophore that depolarizes the inner mitochondrial membrane, and the mitochondrial electron
transport chain (ETC) inhibitors of complex I (rotenone) and complex III (antimycin A). The final
concentrations of mitochondrial inhibitors were at 1
µM oligomycin, 1 µM FCCP, and 0.5 µM
antimycin/rotenone. Basal respiration is the baseline oxygen consumption reading before compounds
are injected. Maximal respiration represents the maximum OCR measurement value after FCCP
injection. Spare respiratory capacity is calculated by noting the OCR response to FCCP.
4.10. Immunoblotting Analysis
HepG2 cells were grown in 100 mm culture dishes (2
×
106 cells/dish) for 24 h, and then
incubated with silymarin, CA, and CA derivatives (20 M). After 24 h incubation, the cells were
µ
treated with 0.5 mM t-BHP to induce oxidative stress for 12 h. Cells were then harvested and
washed with PBS. Whole cell lysates were prepared in a lysis buffer (Cell Signaling, Beverly, MA,
USA) containing 10 mM phenylmethylsulfonyl fluoride (PMSF). The cell lysates were sonicated and
cleared by centrifugation at 4 ◦C, 14,000× g for 10 min. Whole cell lysates were examined for HIF-1α
expression. The nuclear proteins from harvested cells were prepared using the nuclear fractionation
kit (Imgenex; San Diego, CA, USA) and were examined for Nrf2 expression. Protein concentrations
were determined by DC protein assay (Bio Rad). Aliquots of the cell lysates (each containing 20
µg
of protein) or nuclear protein (10 g) were boiled for 5 min and electrophoresed on a 10% sodium
µ
dodecyl sulfate (SDS)-polyacrylamide gel. Following SDS-polyacrylamide gel electrophoresis, proteins
were transferred to PVDF membranes. Membranes were blocked by incubation in gelatin-NET
buffer at room temperature, and then incubated with 1:1000 dilution of primary antibodies of Nrf2
(GeneTex, Irvine, CA, USA), HIF-1α (GeneTex), lamin B (GeneTex) and β-actin (Sigma-Aldrich),
followed by horseradish peroxidase-conjugated secondary antibody according to the manufacturer’s
instructions. The immunoreactive proteins were detected using the enhanced ECL chemiluminescence
Western Blotting Detection System (ChemiDoc XRS, Bio-Rad). Signal strengths were quantified using
densitometric program (Image Lab, Bio-Rad).
4.11. Statistical Analysis
All data are presented as means
statistical package (Chicago, IL, USA). The data were evaluated for statistical significance with the
±
SD. Statistical analyses were performed using the SPSS 19.0

Molecules 2017, 22, 702
14 of 16
one-way ANOVA and Duncan multiple comparison test. A p value of <0.05 was considered statistically
significant. The concentration of tested samples required to inhibit 50% of the activity under the assay
conditions was determined from dose-response curves and defined as the IC50 value.
Acknowledgments: This work was supported by the grants from the Ministry of Science and Technology,
Taipei, Taiwan under contract number MOST104-2320-B-003-005 (to Po-Jung Tsai), the Taipei Tzu-Chi Hospital,
Buddhist Tzu-Chi Medical Foundation under grant number TCRD-TPE-103-RT-8 (to Chun-Hsien Yu), the China
Medical University under the Aim for Top University Plan of the Ministry of Education, Taiwan (CHM106-5-2),
and the Taiwan Ministry of Health and Welfare Clinical Trial Center, Taiwan (MOHW106-TDU-B-212-113004)
(to Yueh-Hsiung Kuo). Yu-Ping Chang was the recipient of the Undergraduate Research Scholarship of Ministry
of Science and Technology, Taiwan.
Author Contributions: Tzung-Hsun Tsai and Po-Jung Tsai designed the research and wrote the paper;
Chun-Hsien Yu designed the research and performed the experimental works. Yu-Ping Chang and Yu-Ting
Lin performed the experimental works; Yueh-Hsiung Kuo synthesized all of the caffeic acid derivatives;
Ching-Jung Huang contributed reagents/analysis tools and revised the manuscript. All authors read and approved
the final manuscript.
Conflicts of Interest: The authors declare no conflict of interest.
References
1.
2.
Muriel, P. Role of free radicals in liver diseases. Hepatol. Int. 2009, 3, 526–536. [CrossRef] [PubMed]
˙
Cichoz-Lach, H.; Michalak, A. Oxidative stress as a crucial factor in liver diseases. World J. Gastroenterol.
2014, 20, 8082–8091. [CrossRef] [PubMed]
3.
Ivanov, A.V.; Valuev-Elliston, V.T.; Tyurina, D.A.; Ivanova, O.N.; Kochetkov, S.N.; Bartosch, B.;
Isaguliants, M.G. Oxidative stress, a trigger of hepatitis C and B virus-induced liver carcinogenesis. Oncotarget
2017, 8, 3895–3932. [CrossRef] [PubMed]
4.
5.
6.
7.
8.
9.
Medvedev, R.; Ploen, D.; Hildt, E. HCV and Oxidative Stress: Implications for HCV life cycle and
HCV-associated pathogenesis. Oxid. Med. Cell. Longev. 2016, 2016, 9012580. [CrossRef] [PubMed]
Li, S.; Tan, H.Y.; Wang, N.; Zhang, Z.J.; Lao, L.; Wong, C.W.; Fen, Y. The role of oxidative stress and
antioxidants in liver diseases. Int. J. Mol. Sci. 2015, 16, 26087–26124. [CrossRef] [PubMed]
Olthof, M.R.; Hollman, P.C.; Katan, M.B. Chlorogenic acid and caffeic acid are absorbed in humans. J. Nutr.
2001, 131, 66–71. [PubMed]
Touaibia, M.; Jean-Francois, J.; Doiron, J. Caffeic Acid, a versatile pharmacophore: An overview. Mini Rev.
Med. Chem. 2011, 1, 695–713. [CrossRef]
Murtaza, G.; Karim, S.; Akram, M.R.; Khan, S.A.; Azhar, S.; Mumtaz, A.; Bin Asad, M.H.H. Caffeic acid
phenethyl ester and therapeutic potentials. BioMed Res. Int. 2014, 2014, 145342. [CrossRef] [PubMed]
Silva, T.; Oliveira, C.; Borges, F. Caffeic acid derivatives, analogs and applications: A patent review
(2009–2013). Expert Opin. Ther. Pat. 2014, 24, 1257–1270. [CrossRef] [PubMed]
10. Haidara, K.; Morel, I.; Abalea, V.; Gascon Barre, M.; Denizeau, F. Mechanism of tert-butylhydroperoxide
induced apoptosis in rat hepatocytes: Involvement of mitochondria and endoplasmic reticulum.
Biochim. Biophys. Acta 2002, 1542, 173–185. [CrossRef]
11. Martin, C.; Martinez, R.; Navarro, R.; Ruiz-Sanz, J.I.; Lacort, M.; Ruiz-Larrea, M.B. tert-Butyl
hydroperoxide-induced lipid signaling in hepatocytes: Involvement of glutathione and free radicals.
Biochem. Pharmacol. 2001, 62, 705–712. [CrossRef]
12. Drahota, Z.; Kriváková, P.; Cervinková, Z.; Kmonícková, E.; Lotková, H.; Kucera, O.; Houstek, J.
tert-Butyl hydroperoxide selectively inhibits mitochondrial respiratory-chain enzymes in isolated rat
hepatocytes. Physiol. Res. 2005, 54, 67–72. [PubMed]
13. Knasmuller, S.; Parzefall, W.; Sanyal, R.; Ecker, S.; Schwab, C.; Uhl, M.; Mersch-Sundermann, V.;
Williamson, G.; Hietsch, G.; Langer, T.; et al. Use of metabolically competent human hepatoma cells
for the detection of mutagens and antimutagens. Mutat. Res. 1998, 402, 185–202. [CrossRef]
14. Mersch-Sundermann, V.; Knasmuller, S.; Wu, X.J.; Darroudi, F.; Kassie, F. Use of a human-derived liver cell
line for the detection of cytoprotective, antigenotoxic and cogenotoxic agents. Toxicology 2004, 198, 329–340.
[CrossRef] [PubMed]

Molecules 2017, 22, 702
15 of 16
15. Kim, Y.; Choi, Y.; Ham, H.; Jeong, H.S.; Lee, J. Protective effects of oligomeric and polymeric procyanidin
fractions from defatted grape seeds on tert-butyl hydroperoxide-induced oxidative damage in HepG2 cells.
Food Chem. 2013, 137, 136–141. [CrossRef] [PubMed]
16. Deferme, L.; Briede, J.J.; Claessen, S.M.H.; Jennen, D.G.J.; Cavill, R.; Kleinjans, J.C.S. Time series analysis
of oxidative stress response patterns in HepG2: A toxicogenomics approach. Toxicology 2013, 306, 24–34.
[CrossRef] [PubMed]
17. Itoh, K.; Tong, K.I.; Yamamoto, M. Molecular mechanism activating Nrf2-Keap1 pathway in regulation of
adaptive response to electrophiles. Free Radic. Biol. Med. 2004, 36, 1208–1213. [CrossRef] [PubMed]
18. Lee, K.J.; Choi, J.H.; Khanal, T.; Hwang, Y.P.; Chung, Y.C.; Jeong, H.G. Protective effect of caffeic acid
phenethyl ester against carbon tetrachloride-induced hepatotoxicity in mice. Toxicology 2008, 248, 18–24.
[CrossRef] [PubMed]
19. Es¸refog˘lu, M.; Ara, C. Beneficial effect of caffeic acid phenethyl ester (CAPE) on hepatocyte damage induced
by bile duct ligation: An electron microscopic examination. Ultrastruct. Pathol. 2010, 34, 273–278. [CrossRef]
[PubMed]
20. Lee, K.J.; Choi, J.H.; Hwang, Y.P.; Chung, Y.C.; Jeong, H.G. Protective effect of caffeic acid phenethyl ester
on tert-butyl hydroperoxide-induced oxidative hepatotoxicity and DNA damage. Food Chem. Toxicol. 2008
,
46, 2445–2450. [CrossRef] [PubMed]
21. Shen, H.; Yamashita, A.; Nakakoshi, M.; Yokoe, H.; Sudo, M.; Kasai, H.; Tanaka, T.; Fujimoto, Y.; Ikeda, M.;
Kato, N.; et al. Inhibitory effects of caffeic acid phenethyl ester derivatives on replication of hepatitis C virus.
PLoS ONE 2013, 8, e82299. [CrossRef] [PubMed]
22. Yang, J.; Kerwin, S.M.; Bowman, P.D.; Stavchansky, S. Stability of caffeic acid phenethyl amide (CAPA) in rat
plasma. Biomed. Chromatogr. 2012, 26, 594–598. [CrossRef] [PubMed]
23. Chiang, H.M.; Chen, C.W.; Lin, T.Y.; Kuo, Y.H. N-Phenethyl caffeamide and photodamage: Protecting skin
by inhibiting type I procollagen degradation and stimulating collagen synthesis. Food Chem. Toxicol. 2014
72, 154–161. [CrossRef] [PubMed]
,
24. Weng, Y.C.; Chuang, S.T.; Lin, Y.C.; Chuang, C.F.; Chi, T.C.; Chiu, H.L.; Kuo, Y.H.; Su, M.J. Caffeic acid
phenylethyl amide protects against the metabolic consequences in diabetes mellitus induced by diet and
streptozocin. Evid. Based Complement. Altern. Med. 2012, 2012, 984780. [CrossRef] [PubMed]
25. Yang, J.; Marriner, G.A.; Wang, X.; Bowman, P.D.; Kerwin, S.M.; Stavchansky, S. Synthesis of a series of
caffeic acid phenethyl amide (CAPA) fluorinated derivatives: Comparison of cytoprotective effects to caffeic
acid phenethyl ester (CAPE). Bioorg. Med. Chem. 2010, 18, 5032–5038. [CrossRef] [PubMed]
26. Vargas-Mendoza, N.; Madrigal-Santillán, E.; Morales-González, Á.; Esquivel-Soto, J.; Esquivel-Chirino, C.;
González-Rubio, M.G.L.; Gayosso-de-Lucio, J.A.; Morales-González, J.A. Hepatoprotective effect of silymarin.
World J. Hepatol. 2014, 6, 144–149. [CrossRef] [PubMed]
27. Bosisio, E.; Benelli, C.; Pirola, O. Effect of the flavanolignans of Silybum marianum L. on lipid peroxidation in
rat liver microsomes and freshly isolated hepatocytes. Pharmacol. Res. 1992, 25, 147–154. [CrossRef]
28. Dixit, N.; Baboota, S.; Kohli, K.; Ahmad, S.; Ali, J. Silymarin: A review of pharmacological aspects and
bioavailability enhancement approaches. Indian J. Pharmacol. 2007, 39, 172–179. [CrossRef]
29. Sharma, R.K.; Siddiqi, N.J.; Sharma, B. Protective effect of silymarin on human erythrocyte against tert-butyl
hydroperoxide induced oxidative stress in vitro. Am. J. Biochem. Mol. Biol. 2012, 2, 167–174. [CrossRef]
30. Malhi, H.; Gores, G.J.; Lemasters, J.J. Apoptosis and necrosis in the liver: A tale of two deaths? Hepatology
2006, 43, S31–S44. [CrossRef] [PubMed]
31. Wang, X.; Bowman, P.D.; Kerwin, S.M.; Stavchansky, S. Stability of caffeic acid phenethyl ester and its
fluorinated derivative in rat plasma. Biomed. Chromatogr. 2007, 21, 343–350. [CrossRef] [PubMed]
32. Wu, W.M.; Lu, L.; Long, Y.; Wang, T.; Liu, L.; Chen, Q.; Wang, R. Free radical scavenging and antioxidative
activities of caffeic acid phenethyl ester (CAPE) and its related compounds in solution and membranes:
A structure activity insight. Food Chem. 2007, 105, 107–115. [CrossRef]
33. Marí, M.; Morales, A.; Colell, A.; García-Ruiz, C.; Fernández-Checa, J.C. Mitochondrial glutathione, a key
survival antioxidant. Antioxid. Redox Signal. 2009, 11, 2685–2700. [CrossRef] [PubMed]
34. Wei, Y.; Rector, R.S.; Thyfault, J.P.; Ibdah, J.A. Nonalcoholic fatty liver disease and mitochondrial dysfunction.
World J. Gastroenterol. 2008, 14, 193–199. [CrossRef] [PubMed]

Molecules 2017, 22, 702
16 of 16
35. Wang, X.; Li, H.; Zheng, A.; Yang, L.; Liu, J.; Chen, C.; Tang, Y.; Zou, X.; Li, Y.; Long, J.; et al.
Mitochondrial dysfunction-associated OPA1 cleavage contributes to muscle degeneration: Preventative
effect of hydroxytyrosol acetate. Cell Death Dis. 2014, 5, e1521. [CrossRef] [PubMed]
36. Movafagh, S.; Crook, S.; Vo, K. Regulation of hypoxia-inducible factor-1α by reactive oxygen species: New
developments in an old debate. J. Cell Biochem. 2015, 116, 696–703. [CrossRef] [PubMed]
37. Lee, I.C.; Kim, S.H.; Baek, H.S.; Moon, C.; Kang, S.S.; Kim, S.H.; Kim, Y.B.; Shin, I.S.; Kim, J.C. The involvement
of Nrf2 in the protective effects of diallyl disulfide on carbon tetrachloride-induced hepatic oxidative damage
and inflammatory response in rats. Food Chem. Toxicol. 2014, 63, 174–185. [CrossRef] [PubMed]
38. Li, M.; Wang, X.F.; Shi, J.J.; Li, Y.P.; Yang, N.; Zhai, S.; Dang, S.S. Caffeic acid phenethyl ester inhibits liver
fibrosis in rats. World J. Gastroenterol. 2015, 21, 3893–3903. [CrossRef] [PubMed]
Sample Availability: Samples of the compounds caffeic acid derivatives are not available from the authors.
©
2017 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC BY) license (http://creativecommons.org/licenses/by/4.0/).