Serotonin derivatives, major safflower (Carthamus tinctorius L.) seed antioxidants, inhibit Low-Density Lipoprotein (LDL) oxidation and atherosclerosis in apolipoprotein E-deficient mice

Article abstract of DOI:10.1021/jf060254p

The effects of defatted safflower seed extract and its phenolic constituents, serotonin derivatives, on atherosclerosis were studied. Ethanol-ethyl acetate extract of safflower seeds (SSE) inhibited low-density lipoprotein (LDL) oxidation induced in vitro

Full text of DOI:10.1021/jf060254p

4970 J. Agric. Food Chem. 2006, 54, 4970−4976
Serotonin Derivatives, Major Safflower (Carthamus tinctorius L.)
Seed Antioxidants, Inhibit Low-Density Lipoprotein (LDL)
Oxidation and Atherosclerosis in Apolipoprotein E-Deficient
Mice
,
†
†
†
NAOTO KOYAMA,* KANNA KURIBAYASHI, TETSUYA SEKI,
†
†
†
KATSUNORI KOBAYASHI, YASUFUMI FURUHATA, KATSUYA SUZUKI,
HARUMI ARISAKA, TAKASHI NAKANO, YUSUKE AMINO, AND KOICHI ISHII†
†
‡
‡
Research Institute for Health Fundamentals, Ajinomoto Co., Inc., Kawasaki 210-8681, Japan, and
Applied Research Department, AminoScience Laboratories, Ajinomoto Co., Inc.,
Kawasaki 210-8681, Japan
The effects of defatted safflower seed extract and its phenolic constituents, serotonin derivatives, on
atherosclerosis were studied. Ethanol-ethyl acetate extract of safflower seeds (SSE) inhibited low-
density lipoprotein (LDL) oxidation induced in vitro by an azo-containing free-radical initiator V70 or
copper ions. Two serotonin derivatives [N-(p-coumaroyl)serotonin, CS; N-feruloylserotonin, FS] and
their glucosides were identified as the major phenolic constituents of the extract. The study with
chemically synthesized materials revealed that a majority of the antioxidative activity of SSE was
attributable to the aglycones of these two serotonin derivatives. Orally administered CS and FS
4
suppressed CuSO -induced plasma oxidation ex vivo. Long-term (15 week) dietary supplementation
of SSE (1.0 wt %/wt) and synthetic serotonin derivatives (0.2-0.4%) significantly reduced the
atherosclerotic lesion area in the aortic sinus of apolipoprotein E-deficient mice (29.2-79.7%
reduction). The plasma level of both lipid peroxides and anti-oxidized LDL autoantibody titers
decreased concomitantly with the reduction of lesion formation. Serotonin derivatives were detected
as both intact and conjugated metabolites in the plasma of C57BL/6J mice fed on 1.0% SSE diet.
These findings demonstrate that serotonin derivatives of SSE are absorbed into circulation and
attenuate atherosclerotic lesion development possibly because of the inhibition of oxidized LDL
formation through their strong antioxidative activity.
KEYWORDS: Safflower; serotonin derivatives; antioxidants; LDL oxidation; apoE-deficient mice
INTRODUCTION
large-scale intervention trials have failed to demonstrate the
beneficial effects of antioxidant vitamins. Recently, researchers
have expressed a keen interest in the biological activity of plant
phenolics as a dietary source of antioxidants. In vivo antiathero-
genic activity in conjunction with antioxidative activity has been
demonstrated in green tea catechins (1), proanthocyanidins (2),
quercetin glucosides (3, 4), cacao liquor polyphenols (epicat-
echin and procyanidins) (5), and licorice extract (glabridin) (6).
Despite the wide diversity of phenolics in the plant kingdom,
to date, only a limited number of phenolics have been identified
as antiatherogenic principals, and in most cases, they are either
flavonoids or isoflavonoids. Detailed information on the anti-
atherogenic properties of plant phenolics from other subclasses
will provide a better and rational understanding of the relation-
ship between their antioxidant activity and prophylactic effects
on atherosclerosis.
Risk factors such as hyperlipidemia, hypertension, and
hyperglycemia are believed to increase oxidative stress in the
vascular wall by multiple mechanisms, thereby resulting in
upregulation of reactive oxygen species (ROS) production.
Proinflammatory cells that infiltrate into the vascular wall secrete
large amounts of ROS and oxidative enzymes and lead to
oxidative modification of low-density lipoprotein (LDL) and
foam cell deposition, the hallmark of early atherogenesis. Thus,
attenuation of oxidative stress is one of the major issues in the
prevention of initiation and progression of atherosclerosis. Many
epidemiological and clinical studies have suggested that anti-
oxidants play a preventive role in atherogenesis; however, many
*
To whom correspondence should be addressed: Research Institute for
Health Fundamentals, Ajinomoto Co., Inc. 1-1, Suzuki-cho, Kawasaki-ku,
Kawasaki 210-8681, Japan. Telephone: +81-44-210-5767. Fax: +81-44-
Safflower (Carthamus tinctorius L.) seeds have been pro-
duced worldwide for a long time period for edible oil production.
In addition, they have been used as an herbal medicine in Korea
2
11-7609. E-mail: naoto_koyama@ajinomoto.com.
†
Research Institute for Health Fundamentals.
Applied Research Department, AminoScience Laboratories.
‡
1
0.1021/jf060254p CCC: $33.50
© 2006 American Chemical Society
Published on Web 06/14/2006

Serotonin Derivatives Inhibit LDL Oxidation and Atherosclerosis
J. Agric. Food Chem., Vol. 54, No. 14, 2006 4971
carbodiimide hydrochloride (2.97 g, 15.49 mmol) were added to a
solution of serotonin hydrochloride (3.00 g, 14.11 mmol) and p-
coumaric acid (2.32 g, 14.13 mmol) in dimethylformamide (15 mL)
and dichloromethane (100 mL). After the mixture was stirred overnight
at room temperature, it was concentrated in vacuo. The residue was
treated with water (100 mL) and extracted with ethyl acetate (3 × 100
mL). The organic extract was washed successively with 5% citric acid
solution (2 × 100 mL), saturated sodium hydrogencarbonate solution
3
Figure 1. Structure of serotonin derivatives CS, RdH; FS, RdOCH .
for the promotion of bone formation and in the treatment of
osteoporosis and rheumatism. Serotonin derivatives (Figure 1)
were identified as the major and unique phenolic constituents
of defatted safflower seeds (7). These are members of a group
of indole hydroxycinnamic acid amides. Serotonin derivatives
have been reported to possess strong antioxidative activities in
vitro and exert various biological effects on plasma and liver
lipid status (8), viability and growth of cancer cell lines or
fibroblasts (9, 10), ischemia-reperfused Langendorff hearts (11),
and cellular proinflammatory cytokine or melanin production
(2 × 100 mL), and brine (100 mL). The extract was dried over MgSO
4
and concentrated to yield a solid that, upon crystallization from ethyl
acetate-ethanol (15:1), yielded 2.87 g of CS (8.89 mmol, 63.0%) as
-
crystals (ESI-MS m/z: 321.11 [M - H] ).
N-Feruloylserotonin (FS). FS was synthesized in 65.6% yield from
serotonin hydrochloride and trans-ferulic acid in a similar manner as
described above, with the exception that the crystallization was
performed using methanol-chloroform (1:15) (ESI-MS m/z: 351.14
-
[M - H] ).
1
The H NMR spectra of these compounds were identical to those
(12, 13). However, few reports ascribe a particular in vivo effect
reported in pertinent literary works (16). The purity of CS and FS was
analyzed by HPLC and was estimated to be greater than 96.0 and
to these little-known polyphenols.
9
7.5%, respectively.
The purpose of the present study was to investigate whether
the extract of defatted safflower seeds and serotonin derivatives
exert beneficial effects on atherosclerosis in vitro and in vivo.
For this purpose, safflower seed extract and synthetic serotonin
derivatives were prepared and their effects on the resistance of
LDL to in-vitro-induced oxidation and on aortic lesion develop-
ment in apoE-deficient mice were analyzed; in addition,
biological availability of serotonin derivatives was also exam-
ined.
LDL Oxidation Assay. LDL was isolated from healthy male
volunteers in accordance with the procedure of Chung et al. (17). In-
vitro-induced oxidation of the isolated LDL was performed in ac-
cordance with the procedure of Hirano et al. (18) with a slight
modification. In brief, LDL (final concentration of 100 µg/mL) was
incubated for 5 h at 37 °C with freshly prepared V70 (600 µmol/L) or
CuSO (10 µmol/L) in a spectrophotometer (DU640; Beckman Coulter,
4
Fullerton, CA) that was equipped with a 12-position automatic cell
changer. LDL oxidation was monitored by recording the change at 234
nm (formation of conjugated dienes) at 5 min intervals. LDL protein
was determined using the BCN protein assay kit (Pierce, Rockford,
IL). The time required to yield 50% of the maximum conjugated diene
formation was plotted as a function of the standard antioxidant
concentration and used as a standard curve to determine the antioxi-
dative capacity of a test sample relative to the standard antioxidant.
Gallic acid and Trolox were used as the standard for V70- and copper-
induced LDL oxidation, respectively.
MATERIALS AND METHODS
Defatted safflower seeds were provided by J-Oil Mills, Inc. (Tokyo,
Japan). V70 (2,2′-azobis(4-methoxy-2,4-dimethylvaleronitrile) was
purchased from Wako Pure Chemical Industries, Ltd. (Osaka, Japan).
Gallic acid, Trolox (6-hydroxy-2,5,7,8-tetramethylchroman-2-carboxylic
acid), p-coumaric acid, Folin-Ciocalteu’s phenol reagent, â-glucu-
ronidase (type X-A), and sulfatase (type H5) were purchased from
Sigma-Aldrich (St. Louis, MO). Serotonin hydrochloride and trans-
ferulic acid were purchased from Tokyo Kasei Kogyo Co., Ltd. (Tokyo,
Japan), and â-glucosidase was purchased from Oriental Yeast Co., Ltd.
Animals and Diets. A total of 48 male apoE-deficient mice
tm1Unc
(B6.129P2-Apoe
/J; Jackson Laboratories, Bar Harbor, ME) aged
6
-7 week were divided into 5 groups with 9-10 mice in each group.
These mice were fed one of the following diets: (1) control diet,
control” group (n ) 9); (2) control diet supplemented with 0.1% (wt/
(Tokyo, Japan).
“
Preparation of Defatted Safflower Seed Extract (SSE) and
Analysis of Major Phenolics. A total of 100 g of defatted safflower
seeds was extracted twice in 500 mL of 90% (v/v) ethanol for 3 h at
wt) CS and 0.1% FS, “CS + FS (0.2)” group (n ) 10); (3) control diet
supplemented with 0.2% CS and 0.2% FS, “CS + FS (0.4)” group
(
(
n ) 10); (4) control diet supplemented with 0.4% FS, “FS” group
n ) 9); and (5) control diet supplemented with 1.0% SSE, “SSE”
6
0 °C. The extracted solvent was then concentrated in vacuo and diluted
with water to a final volume of 200 mL, followed by n-hexane washing
twice). The aqueous layer (100 mL) was then extracted twice with
group (n ) 10). The control diet comprised the semisynthetic AIN-
3G formula with slight modifications (e.g., L-cystine and tert-
(
9
ethyl acetate and vacuum-dried. The term “SSE” in this paper refers
to this ethanol-ethyl acetate extract of defatted safflower seeds. Total
polyphenol content of the extract was determined by the Folin-Ciocalteu
method (14).
butylhydroquinone were omitted, and soybean oil and casein were
replaced with corn oil and vitamin-free casein, respectively). The diet
and drinking water were provided ad libitum. The mice were maintained
on these diets for 15 weeks, during which their body weight and feed
consumption were monitored at weekly intervals. All animal procedures
were approved by the institutional ethical review board for animal
research.
The extract was analyzed by high-performance liquid chromatog-
raphy (HPLC) equipped with a photodiode array detector (Alliance;
Waters, Milford, MA). The conditions for HPLC were as follows:
column, Capcell Pak ODS UG-120 (4.6 × 250 mm, Shiseido Co.,
Tokyo, Japan); mobile phase, linear gradient of water-acetonitrile
Analysis of Aortic Atherosclerotic Lesions. Quantification of
atherosclerosis was performed by calculating the lesion area in the aortic
sinus in accordance with the method of Paigen et al. (19). In brief,
after 15 weeks of dietary intervention, the mice were fasted overnight,
and blood was then collected from the inferior vena cava. The heart
and entire aorta were perfused with saline; subsequently, the heart along
with the ascending aorta was dissected and embedded in the OCT
compound. Cross sections of the heart (thickness of 3 µm) passing
through the aortic sinus were prepared, and a section where the aortic
valves were most prominent was analyzed. The section was stained
with Oil red O and counterstained with hematoxylin. Images of each
section were directly captured using a CCD camera attached to a
binocular stereomicroscope, and the lesioned area (intimal thickening)
was manually determined in accordance with the method of Tangirala
(containing 0.1% TFA), 0-35% acetonitrile over 35 min; flow rate, 1
mL/min; detection, 290 nm for CS, FS, CS-monoglucoside, and FS-
monoglucoside, 279 nm for 2-hydroxyarctiin; temperature, room
temperature. To identify glucosides from the serotonin derivatives,
enzymatic hydrolysis of the glucosides was performed with â-glucosi-
-
1
dase [1 unit/mL in 0.1% BSA (0.1 mol) (L of sodium phosphate
-
1
buffer) at pH 7.0] (24 h, room temperature) and the products were
compared with authentic serotonin derivatives by HPLC. 2-Hydroxy-
arctiin was separated and detected in accordance with the method of
Lyon et al. (15).
Chemical Synthesis of Serotonin Derivatives. N-(p-Coumaroyl)-
serotonin (CS). Triethylamine (2.20 mL, 15.78 mmol), 1-hydroxyben-
zotriazole (2.10 g, 15.54 mmol), and 1-ethyl-3-(3-dimethylaminopropyl)-

4972 J. Agric. Food Chem., Vol. 54, No. 14, 2006
Koyama et al.
Figure 2. SSE and serotonin derivatives inhibit V70- and CuSO
Effects of SSE on V70-induced oxidation (A) and CuSO
mg/mL SSE ( ). (B) Control ( ), 0.3 mg/mL SSE ( ), and 0.5 mg/mL SSE (
CuSO -induced oxidation (D). (C and D) Control ( ), 1 M CS ( ), 10 M CS (
4
-induced LDL oxidation (conjugated diene formation) in a dose-dependent manner.
4
-induced oxidation (B). (A) Control (
). Effects of serotonin derivatives on V70-induced oxidation (C) and
), 1 M FS ( ), 10 M FS ( ).
0), 0.2 mg/mL SSE (2), 0.6 mg/mL SSE (4), and 1.2
+
0
2
4
4
0
µ
2
µ
b
µ
4
µ
O
et al. (20) by using WinROOF image-analysis software (Mitani-Corp,
Tokyo, Japan).
HPLC were the same as described above, and the electrode potentials
were set between 0 and 700 mV in 100 mV steps.
Statistical Analysis. Statistical analyses were performed by using
ANOVA for post-hoc analysis when applicable. Data shown are mean
( SEM.
Plasma Lipid and Oxidation Biomarker Measurements. Overnight-
fasted blood samples were obtained as described above, mixed with
EDTA, and centrifuged to obtain plasma. The mouse non-HDL fraction
was obtained from the plasma by the single-spin gradient ultracen-
trifugation method (17), and the lipoprotein composition of the fraction
was confirmed by nonlinear-gradient polyacrylamide gel electrophoresis
RESULTS
(
Multigel-Lipo 2.4/15.2; Daiichi Pure Chemicals, Tokyo, Japan)
Inhibition of In-Vitro-Induced LDL Oxidation by Defatted
Safflower Seed Extract. Defatted safflower seeds were ex-
tracted with 90% ethanol and further fractionated into five
fractions by liquid-liquid extraction as described in the
Materials and Methods. The antioxidative activity of each
fraction was evaluated in a V70-induced LDL oxidation system.
The activity was highly correlated with the total polyphenol
content in each fraction (r ) 0.941). The fraction obtained by
extraction with ethanol followed by ethyl acetate (designated
as SSE) was the most potent in inhibiting LDL oxidation. For
a more careful examination of the SSE activity, LDL (100 µg
of protein/mL) was subjected to azo-containing radical initiator-
or metal-ion-dependent oxidation by incubation with V70 and
CuSO4, respectively. The addition of SSE markedly prolonged
the lag phase in both the oxidation systems (parts A and B of
Figure 2) in a dose-dependent manner. When the antioxidant
activity of SSE was compared with that of other antioxidants
with respect to the time required to yield 50% of the maximum
conjugated diene formation, 1 g of SSE was found to be almost
equivalent to 0.2 g of gallic acid or Trolox in the V70- or
copper-induced oxidation system, respectively (Table 1).
Analysis of Major Phenolic Components in SSE. To
identify the antioxidative principals in SSE, the extract was
fractionated using reverse-phase HPLC (Figure 3). Two major
peaks (Rt 31.8 and 32.3 min) were detected at 290 nm. These
constituents were identified as CS and FS, respectively, by using
ultraviolet spectrum [λmax ) 292.4 nm (CS) and 293.6 and 312.8
followed by Sudan black staining. Plasma triglyceride (TG), total
cholesterol (TC), and HDL cholesterol (HDL-C) were determined
enzymatically using an automated analyzer (Drychem; FujiFilm Medi-
cal, Tokyo, Japan). Thiobarbituric acid reactive substances (TBARS)
in the non-HDL fraction and plasma were determined in accordance
with the procedure of Ohkawa et al. (21). The plasma LPO (lipid
peroxide) concentration was determined using an LPO assay kit
(Determinar LPO; Kyowa Medex, Tokyo, Japan). Plasma autoantibody
titers to oxidized LDL (anti-oxLDL autoantibody) were determined
using oLAB ELISA kit (Biomedica, Vienna, Austria) with a slight
modification. In brief, aliquots (200 µL/well) of diluted (1:500) mouse
plasma samples were added to the manufacturer-provided microtiter
plate coated with human copper-oxLDL and incubated overnight at
4
°C. After washing, diluted (1:20 000) secondary antibody solution
(goat anti-mouse IgG peroxidase conjugate; Sigma-Aldrich) was added
at an amount of 100 µL/well. After extensive washing, 100 µL of the
substrate solution was added to each well, and the reaction was
terminated after 15-25 min. Absorbance was measured at 450 nm by
using a Wallac ARVO SX 1420 Multilabel Counter (Perkin-Elmer
Life Sciences, Boston, MA).
Analysis of Serotonin Derivatives in Mouse Plasma. Nine C57BL/
6
J mice (Charles River Japan, Inc., Yokohama, Japan) were housed in
light-controlled boxes with an inverted 12-h light-dark cycle (07:00-
9:00, dark). The mice were fed with the 1.0% SSE diet (described
above) for 10 days and sacrificed at 10:00 (n ) 3), 14:00 (n ) 3), and
7:00 (n ) 3) on the last day of the feeding period. Plasma (40 µL)
1
1
obtained from the inferior vena cava was mixed with or without 10 µL
of enzyme solution (100 units of â-glucuronidase and 20 units of
sulfatase in 0.5 mol/L sodium acetate buffer at pH 5.0) and then
incubated at 37°C for 1 h. Each mixture was extracted with 60 µL of
methanol/TCA (100:20), incubated on ice for at least 30 min, and
centrifuged (15 000 rpm, 10 min) subsequently. The supernatant was
collected and subjected to HPLC equipped with a coulometric detector
nm (FS)] and mass spectrometric analyses [m/z 323 (CS) and
+
3
53 (FS) [M + H] ]. These constituents were further confirmed
by coelution with synthetic materials. Two other constituents
that were eluted at 25.6 and 26.4 min were observed to possess
UV absorption patterns similar to those of CS and FS. Because
(CoulArray system; ESA, Inc., Chelmsford, MA). The conditions for

Serotonin Derivatives Inhibit LDL Oxidation and Atherosclerosis
J. Agric. Food Chem., Vol. 54, No. 14, 2006 4973
Table 1. Contents of Serotonin Derivatives in SSE, Their Antioxidative
Activity against V70- and Copper-Induced LDL Oxidation, and Their
Contribution to SSE Activity
V70-induced LDL oxidation Cu2+-induced LDL oxidation
content in
SSE (wt %)
specific
activity
contribution
to SSE (%)
specific
activity
contribution
to SSE (%)
a
b
c
SSE
0.2
0.9
1.0
0.05
0.05
0.2
1.0
1.4
c
CS
8.8
7.1
6.7
1.5
37.4
34.4
1.5
44.0
49.7
c
FS
Figure 4. Effect of orally administered serotonin derivatives (100 mg/kg)
on the susceptibility of plasma oxidation ex vivo. Plasma oxidation was
c
c
c
CS
FS
−gluc
NT
NT
c
c
c
−
gluc
0.3
NT
NT
induced by 100
4
µM CuSO and was measured as the TBARS formation
after 120 min of incubation. Bars are mean
Fisher LSD.
± SEM. p value; ANOVA,
a
b
Expressed as gallic acid equivalent (g of GAE/g). Expressed as Trolox
c
equivalent (g of TE/g). SSE, ethanol
−
ethyl acetate extract of defatted safflower
seeds; CS, N-(p-coumaroyl)serotonin; FS, N-feruloylserotonin; CS
−gluc, CS−
Table 2. Effects of Long-Term (15 Week) Dietary Supplementation of
Serotonin Derivatives and SSE on the Plasma Lipid Status in
ApoE-Deficient Mice
monoglucoside; FS gluc, FS monoglucoside; NT, not tested.
−
−
plasma lipid (mmol/L)a
groups
control
TGb
TCb
HDL-Cb
1.21
±
±
±
±
±
0.07
18.2
±
±
±
±
±
1.3
0.59
±
±
±
±
±
0.06
CS
CS
FS
+
+
FS (0.2)
FS (0.4)
1.14
1.09
0.88
1.13
0.11
0.09
0.08
0.09
15.0
11.9
12.6
12.5
1.4
0.9
1.1
1.1
0.54
0.74
0.43
0.53
0.06
0.07
0.06
0.11
c
d
c
SSE
Figure 3. HPLC
ethanolic extract of defatted safflower seeds (SSE). Detection
Peaks: 1, CS monoglucoside; 2, FS monoglucoside; 3, 2-hydroxyarctiin;
, CS; 5, FS.
−
UV chromatogram of the ethyl acetate fraction of the
)
290 nm.
a
SEM. b TG, triglyceride; TC, total cholesterol; HDL-C, HDL
p < 0.05 versus “control”
Values are mean
cholesterol. p < 0.01 versus “control” (ANOVA, Dunnett).
(ANOVA, Dunnett).
±
−
−
c
d
4
each constituent yielded a peak with the same retention time as
that of CS or FS when hydrolyzed with â-glucosidase (and not
with R-glucosidase), the two constituents were considered as
monoglucosides of the serotonin derivatives. This conclusion
was further supported by the mass spectrometric (CS-mono-
CS, FS (100 mg/kg each), and vehicle [50% poly(ethylene
glycol) 400] were orally administered to C57BL/6 mice (n )
, each). Plasma collected 30 min after ingestion was 10-fold
diluted with saline and then subjected to CuSO4 (100 µM)
induced oxidation for 120 min at 37 °C. As shown in Figure
, TBARS formation in plasma was significantly suppressed in
5
+
+
glucoside, m/z 485 [M + H] , 507 [M + Na] , 323 [M -
4
+
+
C6H10O5 + H] ; FS-monoglucoside, 515 [M + H] , 537
both serotonin-derivative-administered mice.
+
+
1
[
M + Na] , 353 [M - C6H10O5 + H] ) and H NMR (DMSO-
d6, 300 MHz, CS-monoglucoside, anomeric proton at δ 4.73
1H, d, J ) 7.9 Hz) and four methines and one methylene
between δ 3.1-3.8; FS-monoglucoside, anomeric proton at δ
.73 (1H, d, J ) 6.9 Hz) and four methines and one methylene
Effects of SSE and Serotonin Derivatives on Aortic Lesion
Formation in ApoE-Deficient Mice. Because SSE and sero-
tonin derivatives strongly inhibited in-vitro-induced LDL oxida-
tion, we investigated their antiatherosclerotic activity in apoE-
deficient mice. At the end of the intervention, no significant
difference was observed in the body weights of mice in different
groups [29.6 ( 0.6 g in “control”, 28.0 ( 0.9 g in “CS + FS
(
4
between δ 3.1-3.8) analyses of isolated compounds. An
additional constituent that was eluted at 29.8 min was identified
as 2-hydroxyarctiin (lignan) by ultraviolet absorption spectrum
(
0.2)”, 28.2 ( 0.4 g in “CS + FS (0.4)”, 27.9 ( 0.6 g in “FS”,
(
λmax ) 238.0 and 278.1 nm) and mass spectrometric analyses
and 28.1 ( 0.8 g in “SSE”]. Similarly, no significant differences
were observed in plasma triglyceride and HDL cholesterol
levels. However, the total cholesterol was significantly lower
in animals treated with SSE or serotonin derivatives (“safflower-
treated mice”) (Table 2). The extent of atherosclerosis in the
aortic sinus of the control and safflower-treated mice was
evaluated. The mean lesioned area was significantly smaller in
all of the safflower-treated mice; 29.2% (p < 0.05), 53.3%
(p < 0.01), 43.5% (p < 0.01), and 79.7% (p < 0.01) reduction
was observed in mice of the “CS + FS (0.2)”, “CS + FS (0.4)”,
“FS”, and “SSE”, respectively (Figure 5A). Although the total
content of the serotonin derivatives in the “SSE” diet (0.21%;
aglycone basis) was similar to that of the “CS + FS (0.2)” diet,
SSE-treated mice showed significantly less extensive lesions
than the mice fed with the diet containing pure serotonin
derivatives.
+
(
m/z 551 [M + H] ). The contents of CS, FS, CS-monoglu-
coside, FS-monoglucoside, and 2-hydroxyarctiin in the extract
were 8.8, 7.1, 6.7, 1.5, and 6.1% (wt/wt of SSE), respectively.
In Vitro and ex Vivo Antioxidant Properties of Serotonin
Derivatives. Because the serotonin derivatives were considered
to be the major phenolic components of SSE, their antioxidative
properties were examined in vitro with synthetic compounds.
Both CS and FS inhibited V70- or CuSO4-induced LDL
oxidation in a dose-dependent manner (parts C and D of Figure
2). The specific activities of both substances were approximately
5
times higher than that of SSE in both oxidation systems. The
glucosides of both of the serotonin derivatives were isolated
by HPLC and subjected to V70-induced oxidation. Their specific
activities were approximately 1/20th of that of their aglycones.
The total contribution of the serotonin derivative aglycones to
the antioxidative activity of SSE calculated on the basis of their
specific activities and their contents was 71.8% in V70-induced
oxidation and 93.7% in copper-induced oxidation systems
Effects of SSE and Serotonin Derivatives on Plasma Lipid
Peroxidation and Anti-oxLDL Autoantibody Titers. To study
the antiatherosclerotic mechanism of SSE or serotonin deriva-
tives, some oxidative biomarkers were examined in the saf-
(Table 1).

4974 J. Agric. Food Chem., Vol. 54, No. 14, 2006
Koyama et al.
rest of safflower-treated groups. The plasma level of LPO
showed the same tendency; a significant reduction (p < 0.05
versus “control”) was observed in the “CS + FS (0.4)” and
“
SSE” diets (Figure 5C).
Biological Availability of Serotonin Derivatives. We ques-
tioned whether the antiatherogenic and antioxidative effects
observed by the oral intake of SSE or serotonin derivatives could
be attributed to the absorption of serotonin derivatives into the
circulation. At 3, 7, and 10 h after the onset of the dark period,
blood was collected from the C57BL/6J mice fed a diet
containing 1% SSE ad libitum. Plasma was separated from the
blood and treated with or without â-glucuronidase and sulfatase
and then analyzed for serotonin derivatives and their metabolites
by using the HPLC-ECD system. Both serotonin derivatives
were observed to be present in the plasma in both intact and
conjugated forms (Table 3). The total concentration of the intact
aglycones (CS + FS) was maintained above 1 µmol/L at least
from 3 to 7 h; however, a majority of the serotonin derivatives
were detected as conjugated metabolites (glucuronides and/or
sulfates). Both aglycones accounted for approximately 10% of
the total detectable plasma serotonin derivatives and further
decreased to 5% toward the end of the feeding period.
DISCUSSION
The present study demonstrated that the ethanol-ethyl acetate
extract of defatted safflower seeds (SSE) inhibited both V70-
2
+
and Cu -induced LDL oxidation. Serotonin derivatives such
as CS, FS, and their glucosides were identified as the major
phenolics in the extract. Chemically synthesized serotonin
derivatives strongly inhibited in-vitro-induced LDL oxidation
and decreased plasma susceptibility to oxidation ex vivo. SSE
and serotonin derivatives attenuated the atherosclerosis develop-
ment in apoE-deficient mice and concomitantly reduced the total
cholesterol, LPO, and anti-oxLDL autoantibody titers in their
plasma. These serotonin derivatives were detectable in both
intact and conjugated forms in the plasma of C57BL/6J mice
fed with a diet containing SSE. To the best of our knowledge,
this is the first paper that demonstrates the pharmacological
action as well as the active compounds of safflower seed extract
in spontaneous disease model animals.
Both SSE and serotonin derivatives inhibited LDL oxidation
induced by either V70 (a free-radical-generating system) or
CuSO4 (a metal-ion-dependent lipid peroxidation system)
(Figure 2). In either oxidation system, the antioxidative activity
of the aglycones of the serotonin derivatives accounted for
>70% of that of SSE (Table 1). The glucosides exhibited
considerably lower specific activity (1/20th) than the aglycones.
These results indicate that the majority of the antioxidative
activity against in-vitro-induced LDL oxidation in SSE is
derived from aglycones of the serotonin derivatives. Consistent
with our findings, Zhang et al. (22) reported the absence of
significant DPPH-scavenging activity in both glucosides in
Figure 5. Effects of the long-term (15 week) dietary supplementation of
SSE and serotonin derivatives on atherosclerotic lesion formation (A),
plasma anti-oxLDL autoantibody (B), and plasma lipid peroxides (LPO)
(
C) in apoE-deficient mice. Bars are mean
p < 0.01 versus “control”, and (#) p < 0.05 versus “CS
Tukey Kramer). (B and C) (*) p < 0.05 versus “control” (ANOVA, Dunnett).
±
SEM. (A) (*) p < 0.05, (**)
+
FS (0.2)” (ANOVA,
−
flower-treated apoE-deficient mice. When the non-HDL frac-
tions (VLDL-LDL) separated from the safflower-treated mice
were subjected to V70-induced oxidation, no prolonged or even
slightly shorter lag time was observed [33.0 ( 2.6 min in
“
control”, 28.9 ( 1.4 in “CS + FS (0.2)”, 25.6 ( 1.2 in “CS +
FS (0.4)” (p < 0.05 versus “control”), 26.2 ( 1.3 in “FS”, and
7.6 ( 2.1 in “SSE”]. However, the mean TBARS values in
2
the lipoprotein fraction showed a tendency to decrease in all of
the safflower-treated groups except in the “CS + FS (0.2)” diet;
however, this decrease was not significant because of consider-
able variations [0.21 ( 0.11 nmol/mg protein in “control”,
0
.23 ( 0.08 in “CS + FS (0.2)”, 0.11 ( 0.07 in “CS + FS
(0.4)”, 0.18 ( 0.11 in “FS”, and 0.13 ( 0.05 in “SSE”]. There
was a significant reduction (p < 0.05 versus “control”) in the
plasma anti-oxLDL autoantibody titers in the “CS + FS (0.4)”
diet (Figure 5B). Although not significant, relatively lower auto-
antibody titers were also observed in the plasma of mice in the
Table 3. Appearance of Serotonin Derivatives in the Plasma of C57BL/6J Mice When Fed on a Diet Containing 1% SSEa
mol/L)b mol/L)b
7 h 7 h
CS (µ
FS (µ
3
h
10 h
3 h
10 h
intact (A)c
0.60
±
±
±
0.13
1.36
1.42
0.81
7.53
8.33
9.7
±
±
±
0.23
1.30
1.10
0.25
±
±
±
0.08
1.56
1.62
0.56
±
±
±
0.09
0.94
1.01
0.46
4.60
5.05
9.0
±
±
±
0.08
0.66
0.72
0.19
±
±
±
0.06
0.86
0.91
d
conjugates
6.57
7.17
8.4
5.35
5.60
4.5
4.91
5.47
10.3
3.77
3.96
4.7
e
total (B)
A/B (%)
a
Nine C57BL/6J mice were fed with 1% (wt/wt) SSE diet ad libitum; the mice were sacrificed at 3 h (n
)
3), 7 h (n
)
3), and 10 h (n
)
3) after the onset of the dark
b
c
period; and plasma was obtained. The plasma was treated with or without deconjugating enzymes prior to HPLC analysis. Values are mean
digestion. Total (B) minus intact (A) (glucuronides and/or sulfates). e Digested with
±
SEM. Without enzymatic
d
â
-glucuronidase and sulfatase.

Serotonin Derivatives Inhibit LDL Oxidation and Atherosclerosis
J. Agric. Food Chem., Vol. 54, No. 14, 2006 4975
which the 5-OH groups of the indole ring were glycosylated.
Further, the result is suggestive of the major contribution of
the serotonin moiety with the free 5-OH group with respect to
the antioxidative activity of the serotonin derivatives.
effect of SSE; however, we do not exclude the possibility of
involvement of the activities of serotonin derivatives other than
antilipoprotein oxidation because effects on inflammatory
cytokine production (12) or nitric oxide generation (11) have
been suggested. Future studies are needed to unveil other
potential beneficial activities in serotonin derivatives and to
elucidate the in vivo relevance of them.
Because the synthetic serotonin derivatives themselves sig-
nificantly attenuated lesion formation in apoE-deficient mice,
serotonin derivatives were considered, at least in part, if not
all, antiatherosclerotic principals in SSE. When almost equal
concentrations of SSE and pure serotonin derivatives (CS +
FS) were administered to the mice on a total serotonin derivative
aglycone basis (0.2%), the effect of reducing the lesion in the
synthetic serotonin derivative groups was significantly lower
than that in the SSE group (Figure 5A). This may suggest that
SSE contains other antiatherogenic compounds besides the
serotonin derivatives. The serotonin derivatives reduced the
lesion area in a dose-dependent manner. Administration of 0.4%
of the serotonin derivatives resulted in almost comparable lesion
reduction irrespective of whether the serotonin derivatives were
administered as a mixture (CS + FS) or alone (FS). CS and FS
may be equipotent in antiatherogenicity as well as in their in
vitro antioxidative activity described above.
We demonstrated for the first time the biological availability
of the serotonin derivatives by their detection in the plasma of
SSE-fed mice (Table 3). The plasma level of total serotonin
derivative aglycones (CS + FS) reached over 1 µmol/L, in
which the concentration was shown to inhibit LDL oxidation
in vitro (parts C and D of Figure 2). Thus, it seems presumable
that the serotonin derivatives might have prevented LDL
oxidation in vivo by scavenging the ROS present in the aqueous
milieu of plasma and/or interstitial fluid of the arterial wall.
The majority of the absorbed serotonin derivatives was likely
to be metabolized to conjugated forms. The biological properties
of some flavonoid conjugates have been shown to depend upon
the position of the substitution (28). At present, the position of
the substitution in the serotonin derivatives has not been
determined. The biological activity of these metabolites re-
mained to be evaluated.
The lesion-reducing effects were associated with parallel
reductions in plasma total cholesterol (Table 2), LPO (Figure
5
C), and anti-oxLDL autoantibody titers (Figure 5B). In
In conclusion, the ethanol-ethyl acetate extract of defatted
safflower seeds is preventive against atherosclerotic develop-
ment, and its major and novel subclass of polyphenols, serotonin
derivatives, are responsible for its antioxidative and, at least in
part, antiatherosclerotic activities. SSE or the serotonin deriva-
tives may be beneficial in preventing human cardiovascular
diseases.
conjunction with the result of this study, the reported hypolipi-
demic effects of safflower seed extracts or their phenolic
constituents on cholesterol-fed rats (8, 23) may enable the
deduction of some beneficial effects of safflower seeds on lipid
metabolism and atherogenesis; however, in three other experi-
ments that we performed, neither SSE nor the serotonin
derivatives reproduced the hypocholesterolemic effect, while
reduction in the atherosclerotic lesion area was similarly
observed (data not shown). Thus, the hypocholesterolemic effect
observed in this study may not be essential for the expression
of antiatherosclerotic action. The association of hypocholester-
olemic activity in safflower seeds with its antiatherosclerotic
activity seems to require a more careful assessment.
ACKNOWLEDGMENT
The authors are grateful to Narihide Matsuzaki of J-Oil Mills,
Inc. (Tokyo, Japan) for supplying defatted safflower seeds and
to Yoshiko Yamazaki for providing help with HPLC analysis.
The technical assistance provided by Tomoka Yamano with
respect to the biochemical measurements and animal experi-
ments is greatly appreciated.
A number of studies have suggested that higher titers of
autoantibodies to oxidative-modified LDL are found in patients
with cardiovascular diseases and also in apoE-deficient mice
(24). Circulating autoantibody titers have been shown to
correlate with oxidized LDL content in atherosclerotic lesions
of LDL-receptor-deficient mice (25). Our data on plasma lipid
peroxide and anti-oxLDL autoantibody titers in conjunction with
the ex vivo antioxidative activity of CS and FS provide an
indication of the suppressive effect of the serotonin derivatives
on the formation of oxidized LDL through their in vivo
antioxidative action. However, we failed to demonstrate the
prolongation of the lag time in the non-HDL fraction obtained
from the safflower-treated apoE-deficient mice when subjected
to V70-induced oxidation. This may suggest that serotonin
derivatives are not incorporated into lipoprotein particles and
removed from the lipoprotein fraction by the fractionation
procedures or metabolized and cleared from plasma after
overnight fasting.
LITERATURE CITED
(1) Miura, Y.; Chiba, T.; Tomita, I.; Koizumi, H.; Miura, S.;
Umegaki, K.; Hara, Y.; Ikeda, M.; Tomita, T. Tea catechins
prevent the development of atherosclerosis in apoprotein E-
deficient mice. J. Nutr. 2001, 131, 27-32.
(
2) Yamakoshi, J.; Kataoka, S.; Koga, T.; Ariga, T. Proanthocyan-
idin-rich extract from grape seeds attenuates the development
of aortic atherosclerosis in cholesterol-fed rabbits. Atherosclerosis
1
999, 142, 139-149.
(
3) Kamada, C.; da Silva, E. L.; Ohnishi-Kameyama, M.; Moon, J.
H.; Terao, J. Attenuation of lipid peroxidation and hyperlipidemia
by quercetin glucoside in the aorta of high cholesterol-fed rabbit.
Free Radical Res. 2005, 39, 185-194.
(4) Enkhmaa, B.; Shiwaku, K.; Katsube, T.; Kitajima, K.; Anuurad,
E.; Yamasaki, M.; Yamane, Y. Mulberry (Morus alba L.) leaves
and their major flavonol quercetin 3-(6-malonylglucoside) at-
tenuate atherosclerotic lesion development in LDL receptor-
deficient mice. J. Nutr. 2005, 135, 729-734.
Stocker and O’Halloran (26) and Waddington et al. (27)
reported the lack of an effect of red wine polyphenols on lipid
oxidation, while development of atherosclerosis in apoE-
deficient mice was partially inhibited. They suggested that
arterial lipoprotein oxidation could be dissociated from athero-
genesis, and activities other than inhibition of lipid peroxidation
might be more important. Antioxidative activity against lipo-
protein oxidation of serotonin derivatives is the only mechanism
examined here, and we do consider it relevant to the protective
(
5) Kurosawa, T.; Itoh, F.; Nozaki, A.; Nakano, Y.; Katsuda, S.;
Osakabe, N.; Tsubone, H.; Kondo, K.; Itakura, H. Suppressive
effects of cacao liquor polyphenols (CLP) on LDL oxidation
and the development of atherosclerosis in Kurosawa and Ku-
sanagi-hypercholesterolemic rabbits. Atherosclerosis 2005, 179,
237-246.

4976 J. Agric. Food Chem., Vol. 54, No. 14, 2006
Koyama et al.
(
6) Fuhrman, B.; Buch, S.; Vaya, J.; Belinky, P. A.; Coleman, R.;
Hayek, T.; Aviram, M. Licorice extract and its major polyphenol
glabridin protect low-density lipoprotein against lipid peroxida-
tion: in vitro and ex vivo studies in humans and in atherosclerotic
apolipoprotein E-deficient mice. Am. J. Clin. Nutr. 1997, 66,
(17) Chung, B. H.; Segrest, J. P.; Ray, M. J.; Brunzell, J. D.;
Hokanson, J. E.; Krauss, R. M.; Beaudrie, K.; Cone, J. T. Single
vertical spin density gradient ultracentrifugation. Methods En-
zymol. 1986, 128, 181-209.
(18) Hirano, R.; Kondo, K.; Iwamoto, T.; Igarashi, O.; Itakura, H.
Effects of antioxidants on the oxidative susceptibility of low-
density lipoprotein. J. Nutr. Sci. Vitaminol. 1997, 43, 435-444.
(19) Paigen, B.; Morrow, A.; Holmes, P. A.; Mitchell, D.; Williams,
R. A. Quantitative assessment of atherosclerotic lesions in mice.
Atherosclerosis 1987, 68, 231-240.
(20) Tangirala, R. K.; Rubin, E. M.; Palinski, W. Quantitation of
atherosclerosis in murine models: correlation between lesions
in the aortic origin and in the entire aorta, and differences in the
extent of lesions between sexes in LDL receptor-deficient and
apolipoprotein E-deficient mice. J. Lipid Res. 1995, 36, 2320-
2
67-275.
(
7) Sakamura, S.; Terayama, Y.; Kawakatsu, S.; Ichihara, A.; Saito,
H. Conjugated serotonins related to cathartic activity in safflower
seeds (Carthamus tinctorius L.). Agric. Biol. Chem. 1978, 42,
1
805-1806.
(
8) Cho, S. H.; Lee, H. R.; Kim, T. H.; Choi, S. W.; Lee, W. J.;
Choi, Y. Effects of defatted safflower seed extract and phenolic
compounds in diet on plasma and liver lipid in ovariectomized
rats fed high-cholesterol diets. J. Nutr. Sci. Vitaminol. 2004, 50,
3
2-37.
(
9) Bae, S.; Shim, S.; Park, Y.; Lee, J.; Chang, E.; Choi, S.
Cytotoxicity of phenolic compounds isolated from seeds of
safflower (Carthamus tinctorius L.) on cancer cell lines. Food
Sci. Biotechnol. 2002, 11, 140-146.
2
328.
(21) Ohkawa, H.; Ohishi, N.; Yagi, K. Assay for lipid peroxides in
animal tissues by thiobarbituric acid reaction. Anal. Biochem.
1
979, 95, 351-358.
(
10) Takii, T.; Hayashi, M.; Hiroma, H.; Chiba, T.; Kawashima, S.;
Zhang, H. L.; Nagatsu, A.; Sakakibara, J.; Onozaki, K. Serotonin
derivative, N-(p-coumaroyl)serotonin, isolated from safflower
(22) Zhang, H. L.; Nagatsu, A.; Watanabe, T.; Sakakibara, J.;
Okuyama, H. Antioxidative compounds isolated from safflower
(Carthamus tinctorius L.) oil cake. Chem. Pharm. Bull. 1997,
(Carthamus tinctorius L.) oil cake augments the proliferation
4
5, 1910-1914.
of normal human and mouse fibroblasts in synergy with basic
fibroblast growth factor (bFGF) or epidermal growth factor
(
23) Moon, K.; Back, S.; Kim, J.; Jeon, S.; Lee, M.; Choi, M.
Safflower seed extract lowers plasma and hepatic lipids in rats
fed high-cholesterol diet. Nutr. Res. 2001, 21, 895-904.
(EGF). J. Biochem. 1999, 125, 910-915.
(
11) Hotta, Y.; Nagatsu, A.; Liu, W.; Muto, T.; Narumiya, C.; Lu,
X.; Yajima, M.; Ishikawa, N.; Miyazeki, K.; Kawai, N.;
Mizukami, H.; Sakakibara, J. Protective effects of antioxidative
serotonin derivatives isolated from safflower against postischemic
myocardial dysfunction. Mol. Cell. Biochem. 2002, 238, 151-
(
24) Palinski, W.; Horkko, S.; Miller, E.; Steinbrecher, U. P.; Powell,
H. C.; Curtiss, L. K.; Witztum, J. L. Cloning of monoclonal
autoantibodies to epitopes of oxidized lipoproteins from apo-
lipoprotein E-deficient mice. Demonstration of epitopes of
oxidized low-density lipoprotein in human plasma. J. Clin. InVest.
1
62.
1
996, 98, 800-814.
(
12) Kawashima, S.; Hayashi, M.; Takii, T.; Kimura, H.; Zhang, H.
L.; Nagatsu, A.; Sakakibara, J.; Murata, K.; Oomoto, Y.;
Onozaki, K. Serotonin derivative, N-(p-coumaroyl)serotonin,
inhibits the production of TNF-R, IL-1R, IL-1â, and IL-6 by
endotoxin-stimulated human blood monocytes. J. Interferon
Cytokine Res. 1998, 18, 423-428.
13) Nagatsu, A.; Zhang, H.; Mizukami, H.; Okuyama, H.; Sakakibara,
J.; Tokuda, H.; Nishino, H. Tyrosinase inhibitory and anti-tumor
promoting activities of compounds isolated from safflower
(
(
(
(
25) Tsimikas, S.; Palinski, W.; Witztum, J. L. Circulating auto-
antibodies to oxidized LDL correlate with arterial accumulation
and depletion of oxidized LDL in LDL receptor-deficient mice.
Arterioscler. Thromb. Vasc. Biol. 2001, 21, 95-100.
26) Stocker, R.; O’Halloran, R. A. Dealcoholized red wine decreases
atherosclerosis in apolipoprotein E gene-deficient mice inde-
pendently of inhibition of lipid peroxidation in the artery wall.
Am. J. Clin. Nutr. 2004, 79, 123-130.
27) Waddington, E.; Puddey, I. B.; Croft, K. D. Red wine polyphe-
nolic compounds inhibit atherosclerosis in apolipoprotein E-
deficient mice independently of effects on lipid peroxidation.
Am. J. Clin. Nutr. 2004, 79, 54-61.
(
(Carthamus tinctorius L.) and cotton (Gossypium hirsutum L.)
oil cakes. Nat. Prod. Lett. 2000, 14, 153-158.
(
14) Singleton, V.; Orthofer, R.; Lamuela-Ravent o´ s, R. M. Analysis
of total polyphenols and other oxidation substrates and antioxi-
dants by means of Folin-Ciocalteu reagent. Methods Enzymol.
28) Williamson, G.; Barron, D.; Shimoi, K.; Terao, J. In vitro
biological properties of flavonoid conjugates found in vivo. Free
Radical Res. 2005, 39, 457-469.
1
999, 299, 152-178.
(
15) Lyon, C. K.; Gumbmann, M. R.; Betschart, A. A.; Robbins, D.
J.; Saunders, R. M. Removal of deleterious glucosides from
safflower meal. J. Am. Oil Chem. Soc. 1979, 56, 560-564.
16) Sakamura, S.; Terayama, Y.; Kawakatsu, S.; Ichihara, A.; Saito,
H. Conjugated serotonins and phenolic constituents in safflower
seed (Carthamus tinctorius L.). Agric. Biol. Chem. 1980, 44,
Received for review January 27, 2006. Revised manuscript received
May 8, 2006. Accepted May 10, 2006.
(
2
951-2954.
JF060254P