J Nutr Sci Vitaminol, 53, 133137, 2007

Effects of Dietary Angelica keiskei on Serum and Liver Lipid Profiles,

and Body Fat Accumulations in Rats

Junichi NAGATA, Tomoko MORINO and Morio SAITO

Food Function and Labeling Program, Incorporated Administrative Agency, National Institute of
Health and Nutrition, 1231 Toyama, Shinjuku-ku, Tokyo 1628636, Japan
(Received July 19, 2006)

Summary Angelica keiskei (Ashitaba) is a perennial plant belonging to the Umbelliferae

family. Recently, much attention has been focused on Ashitaba products as a so-called
health food for the breakdown of cellulite among various physiological benefits of Ashitaba.
The current study was carried out to investigate the physiological efficacy of dietary Ashi-
taba on serum and liver lipid profiles and body fat accumulation in rats. Rats were fed a
high-fat diet with various amounts of Ashitaba for 28 d. Perirenal adipose tissue weights of
rats fed the
10 (170 mg/100 g BW) Ashitaba diet were significantly higher (p0.05) than
those of the control group. Serum triacylglycerol concentrations of rats fed the
100
(1,700 mg/100 g BW) Ashitaba diet were significantly higher (p0.05) than those of the

1 (17 mg/100 g BW) group. Fecal weights and bile acid excretions of rats fed the
10 or

100 Ashitaba diet were significantly higher (p0.05) than those of the control group.
However, there were no significant differences in the body weight gain, epididymal adipose
tissue weight, serum cholesterol or liver lipid concentrations or other biochemical profiles in
the serum. Furthermore, even the excessive ingestion of Ashitaba had no significant patho-
logical impact on the liver or kidney. These results indicate that the large intake of Ashitaba
products may supply dietary fiber and thus improve gastrointestinal condition through the
increased excretion of feces containing high level of bile acids, although even excessive
intake of Ashitaba for a short period of 28 d did not show any impact on the decrease in
body fat or modification of lipid profiles in this study.
Key Words Angelica keiskei, lipid Profiles, body fat accumulation, rats

Angelica keiskei (Ashitaba) is a perennial plant
belonging to the Umbelliferae family, which grows nat-
urally in the Izu islands and the Izu, Bouso and Miura
peninsulas of Japan (1). The main bioactive constitu-
ents of this plant are flavonoids, coumarins and chal-
cones (2). In many studies it has been reported that this
plant has various physiological benefits including anti-
bacterial (3), antitumor (4, 5), antiulcer (6), and anti-
thrombotic activity (7), and vasodilative action (8).
Recently, much attention has been paid to the physio-
logical effects of Ashitaba or its chalcones, particularly
for the resolution of lipodystrophy of adipose tissue, oth-
erwise known as cellulite. Cellulite is defined as a local-
ized metabolic disorder of the subcutaneous tissue,
which provokes alteration of the female body shape (9).
The effective constituents of Ashitaba in the reduction
of cellulite are thought to be chalcones and coumarins.
For these reasons, a number of Ashitaba products in the
form of powder or tablets are widely marketed as so-
called health food for weight loss and the breakdown of
cellulite in Japan.
However, to date, there is little information concern-
ing the physiological effects of Ashitaba on body fat loss
or potential cellulite breakdown and lipid metabolism
E-mail: jnagata@nih.go.jp
including lipolysis. The relevance of the amount of chal-
cones contained in Ashitaba to their various effects are
also unclear, although it is considered that about 200
300 mg of chalcones/100 g raw leaves are contained in
the edible portion of the leaf. Moreover, absorption of
Ashitaba chalcones and coumarins from the digestive
tract is also not clear. Thus, much of the proposed effec-
tiveness of Ashitaba chalcones has little scientific basis.
In this study, we focused first on the physiological
impacts of dietary Ashitaba on body fat accumulation
and lipid profiles in rats. In order to investigate the rela-
tionship between the amount of Ashitaba ingested and
lipid profiles or body fat accumulation in rats, high fat
diets containing different amounts of Ashitaba powder
were fed to rats. In addition, the liver and kidney of rats
were also histopathologically examined after an exces-
sive intake of Ashitaba.
MATERIALS AND METHODS
Animals and diets. Male Wistar rats, weighing 60
80 g, were purchased from Japan SLC (Shizuoka,
Japan), and maintained under a controlled room tem-
perature (222C) and lighting cycle (lighting 08:00 to
20:00 h). They were fed commercial chow (CE-2; CLEA
Japan, Tokyo) for 1 wk. After the acclimation, rats were
randomly divided into four groups (n6/group). The

133
134
Table 1. Body weight gain, food intake and relative tissue weights of rats fed experimental diets.
Control (
0)
Ashitaba
(0 mg/100 g BW)
Body weight gain (g) 1214.2
Food intake (g/d) 11.50.2a
Tissue weight (g/100 g BW)
Liver 4.200.05
Kidney 0.730.02
Spleen 0.250.01a
Testis 1.180.02
White adipose tissue
Perirenal 1.710.06a
Epididymal 1.630.13
Values are meansSE, n6. Values not sharing a common letter differ significantly at p0.05.
experimental groups received high fat diets based on
AIN-93G formula containing Ashitaba powder at 17
(
1), 170 (
10) or 1,700 (
100) mg/100 g body
weight, while the control group received the diet with-
out Ashitaba powder. The daily amount of Ashitaba
intake in rats was estimated on a body weight-conver-
sion basis recommended for humans. Ashitaba powder
is commercially available as a so-called functional food,
which contains chalcones and dietary fiber at a concen-
tration of approximately 318.9 mg/100 g product and
31.2 g/100 g product, respectively. The composition of
this diet (g/kg) was 200 g milk casein, 100 g sucrose,
200 g soybean oil, 50 g cellulose, 35 g AIN-93G min-
eral mixture, 10 g AIN-93VX vitamin mixture, 3 g L-
cystine, 2.5 g choline bitartrate, 0.014 g tert-butylhyd-
roquinone and -corn starch to 1,000 g. All diets con-
tained 0.5% cholesterol and 0.125% sodium cholate.
Animals were housed individually and fed each experi-
mental diet ad libitum for 28 d. At the end of the exper-
imental period, after overnight fasting, rats were anes-
thetized and sacrificed for analysis. The care and use of
laboratory animals were in accordance with the guide-
lines laid down by Incorporated Administrative Agency,
National Institute of Health and Nutrition.
Blood sample collection and chemical analysis. Blood
samples were obtained from the abdominal aorta, and
centrifuged at 3,000 rpm for 15 min. Sera were stored
at 80C until analysis. Serum lipids, indices of liver
function and the other biochemical parameters were
enzymatically analyzed using an Auto Analyzer (Olym-
pus AU5232, Tokyo, Japan)
Analysis of liver lipids. The liver was excised from
each rat, weighed and stored at 80C until analysis.
Liver lipids were extracted with chloroformmethanol
(2:1 (v/v)) as described by Folch et al. (10). Liver choles-
terol and triacylglycerol concentrations were deter-
mined using a Wako enzyme assay kit (Cholesterol E-
test and Triglyceride E-test Wako, Osaka, Japan) as
described elsewhere, with minor modifications (11).
Analysis of fecal bile acids. Total fecal bile acids were
extracted with ethanol at 70C for 1 h, and analyzed
enzymatically by a modification of the method of Eaton
and Klassen (12). The standard used for this assay was
NAGATA J et al.

1
10
100
(17 mg/100 g BW) (170 mg/100 g BW) (1700 mg/100 g BW)
1194.8 1244.9 1204.2
11.40.2a 12.20.3ab 12.70.3b
4.190.17 4.300.15 3.970.09
0.730.02 0.740.01 0.740.02
0.250.01a 0.240.00ab 0.230.01b
1.210.01 1.170.03 1.200.03
1.800.07ab 1.990.09b 1.780.13ab
1.640.07 1.860.10 1.710.10
cholic acid.
Histopathological examination. The livers and kid-
neys were obtained from the rats fed the control and

100 Ashitaba diets, fixed with 10% formaldehyde
solution in phosphate-buffered saline (pH 7.4) and
embedded in paraffin. Paraffin-embedded specimens
were prepared, and stained with hematoxylin and
eosin. These samples were observed by optical micro-
scope.
Statistical analysis. The data presented are the
meansSE (standard error of the mean). Statistical sig-
nificance of any differences was evaluated by one-way
ANOVA followed by Fishers PLSD test. Differences at
p0.05 were considered to be significant.
RESULTS
Body weight gain, food intake and relative organ weights of
rats fed experimental diets
Body weight gain, food intake and relative organ
weights of rats fed experimental diets are shown in
Table 1. No significant differences were observed in
body weight gain, or relative weights of liver, kidney, tes-
tis or epididymal adipose tissue between any of the
treatment groups at the end of the feeding period. Food
intake of rats fed the experimental diets tended to
increase in an Ashitaba dose-dependent manner. In
particular, the food intake of the
100 Ashitaba group
was significantly higher (p0.05) than those of the
control and
1 groups. The relative spleen weight of
the
100 Ashitaba group was significantly lower
(p0.05) than those of the control and
1 groups.
However, the perirenal adipose tissue weight of the
10
Ashitaba group was significantly higher (p0.05) than
that of the control group.
Serum and liver lipid profiles
Serum triacylglycerol concentration of rats fed the

100 Ashitaba diets was significantly higher (p0.05)
than that of the
1 group although the Ashitaba
intake did not affect the serum total and high-density
lipoprotein (HDL) cholesterol concentrations. Moreover,
the Ashitaba intakes did not affect the liver cholesterol
or triacylglycerol concentrations in rats fed diets con-
taining various amounts of Ashitaba (Table 2).
 Effects of Ashitaba on Lipid Profiles and Body Fat Accumulation in Rats 135

Table 2. Serum and liver lipid concentrations of rats fed experimental diets.

Control (
0)
1
10
100

Ashitaba
(0 mg/100 g BW) (17 mg/100 g BW) (170 mg/100 g BW) (1,700 mg/100 g BW)

Serum (mmol/L)
Total cholesterol 1.600.07 1.570.07 1.530.73 1.710.04
HDL-cholesterol 0.480.02 0.510.04 0.470.02 0.510.02
Triacylglycerol 0.500.04ab 0.460.05a 0.510.05ab 0.640.06b
Liver (mol/g liver)
Total cholesterol 54.32.7 58.02.9 63.35.4 62.91.5
Triacylglycerol 53.42.3 62.44.8 62.12.8 55.71.9

Values are meansSE, n6. Values not sharing a common letter differ significantly at p0.05.

Table 3. Serum protein, indices of liver and kidney function, and blood glucose concentration of rats fed experimental
diets.

Control (
0)
1
10
100

Ashitaba
(0 mg/100 g BW) (17 mg/100 g BW) (170 mg/100 g BW) (1700 mg/100 g BW)

(g/dL)
Total protein 6.280.07 6.280.11 6.400.14 6.400.04
Albumin 3.270.02 3.230.02 3.270.06 3.330.03
Albumin/globulin 1.080.02 1.080.05 1.030.02 1.100.03
(IU/L)
AST 15219.5 15712.9 14832.1 13812.0
ALT 63.86.3ab 67.37.4a 66.89.7a 43.82.7b
ALP 1,41646.4a 1,33328.5a 1,21737.4b 1,10737.2b
(mg/dL)
Creatinine 0.220.00 0.220.00 0.230.01 0.220.01
Urea nitrogen 10.80.5 10.50.2 10.80.3 12.00.8
(mmol/L)
Glucose 5.430.73 4.360.48 5.940.62 5.160.45

Values are meansSE, n6. Values not sharing a common letter differ significantly at p0.05.

Serum biochemical parameters
100 Ashitaba-containing diet was approximately 1.6-

There were no significant differences in serum pro-
tein concentrations (total protein, albumin and the
ratio of albumin to globulin), indices of renal function
(creatinine clearance and urea nitrogen) or blood glu-
cose concentrations between any of the treatment
groups. Indices of liver function as judged by aspartate
aminotransferase (AST), alanine aminotransferase
(ALT) and alkaline phosphatase (ALP) tended to
decrease in response to an increase of Ashitaba inges-
tion. Serum ALT in the
100 Ashitaba group was sig-
nificantly lower (p0.05) than those of the
1 and

10 groups, and serum ALP in the
10 and
100
groups were significantly lower (p0.05) than that of
the
1 group (Table 3).
Fecal weight and bile acid excretion
Figure 1 shows the fecal weight and fecal bile acid
level of rats fed the experimental diets. Fecal weight and
bile acid level of rats fed Ashitaba-containing diets
increased in a dose-dependent manner. Fecal weight of
the
100 group was significantly higher than those of
the other groups. Similarly, the fecal bile acid levels of
the
10 and
100 groups were also significantly
higher (p0.05) than those of the control and
1
groups. Fecal weight and bile acid level of rats fed the
fold higher than the control group.
Histopathological examinations
Lipid droplet depositions and infiltrations of eosino-
philic granules were observed throughout the rat liver
and kidney of all the groups. The livers from rats fed the
control diet accumulated lipid droplets in the peripheral
area of the hepatic lobulus, while the livers of rats fed
the
100 Ashitaba diet showed deposition of lipid drop-
lets in the hepatic centrolobular area. In addition, infil-
tration of eosinophilic granules in the renal tubule epi-
thelium was observed in all the samples. However, there
were no significant differences in the severity of lipid
depositions in the liver or infiltrations of eosinophilic
granules in the renal tubules between any of the treat-
ment groups (data not shown).
DISCUSSION
In the present study, the large and excessive intakes of
Ashitaba significantly increased fecal weights and fecal
bile acid levels in rats, but the serum triacylglycerol
concentration of rats fed the excessive Ashitaba diet was
significantly increased. In addition, no significant
impacts of Ashitaba intake were observed in the body
weight gain, adipose tissue weights or liver lipid concen-
136 NAGATA J et al.
Fig. 1. Fecal weight and fecal bile acid level of rats fed
diets containig various amounts of Ashitaba. Data are
shown as meansSE of 6 rats. Values not sharing a
common letter differ significantly at p0.05.
trations. However, the excessive Ashitaba intake did not
appear to have adverse effects on hepatic or renal func-
tions of rats as estimated by serum biochemical indices
and histopathological examinations.
Cellulite is essentially an alteration to the topography
of the skin that occurs mainly in the pelvic region,
lower limbs and abdomens of women (13). The hyper-
trophy and hyperplasia observed in cellulite appear to
be responsible for hypothermia, which is a typical met-
abolic feature of cellulite, caused by stasis. Thus, cellu-
lite has several structural alterations in the dermis, in
the microcirculation and within adipocytes.
The physiological function of Ashitaba is thought to
be ascribed to the vasodilation caused by its chalcones
such as xanthoangerol, xanthoangerol E and 4-hydrox-
yderricin as active components (8, 14). Ashitaba intake
is, therefore, expected to improve the impaired circula-
tory system of the subcutaneous tissue. These effects
contribute to the elimination of cellulite and are advo-
cated as it is effective for weight loss. Thus, a number of
Ashitaba products are marketed extensively as a so-
called health food in Japan as if the physiological func-
tions of Ashitaba are connected with the eradication of
hypodermic cellulite. However, the effects of Ashitaba
on cellulite, and the mechanisms involved, have not yet
been elucidated scientifically.
From many previous investigations, the weight loss
thorough metabolic changes such as an increase in
lipolysis and a decrease in lipogenesis is considered to be
necessary to suppress the hypertrophy and/or hyperpla-
sia of adipocytes. For instance, an increase in -oxida-
tion of fatty acids induces not only a decrease in serum
and liver lipid levels but also a mobilization of fatty acids
from body fats (15, 16). If fatty acids are mobilized from
the body-fat mass and utilized for energy production,
body-fat mass will decrease accordingly. Unexpectedly,
however, the present data did not show a significant
decrease in the serum triacylglycerol concentration,
body weight or fat mass of rats fed Ashitaba diets. These
observations agree, in part, with those of Ogawa et al.
(17), who also showed that the serum lipid levels in
SHRSP rats were elevated after treatment with Ashitaba
extract. Moreover, they reported that mRNA expression
of hepatic enzymes involved in triacylglycerol metabo-
lism was not affected by Ashitaba ingestion. Thus, it is
unlikely that Ashitaba intake may directly affects lipid
metabolism such as an increase in fatty acid catabolism
though -oxidation in rats. Therefore, many questions
remain to be answered about physiological effects of
Ashitaba intake on lipolysis of adipocytes. In contrast,
we may have to pay attention to the increase in serum
triacylglycerol concentrations by the overconsumption
of Ashitaba because Ogawa et al. (17) also reported a
similar observation. Thus, further detailed examina-
tions on the relationship between the intake of Ashitaba
chalcones and the changes in serum triacylglycerol
concentration will be needed.
It is noteworthy in the present results that the signif-
icant excretions of fecal bile acids were observed in
response to an increase of Ashitaba intake, although
serum lipid levels were not influenced. These observa-
tions were thought to be due to the rich dietary fiber in
Ashitaba. The increased bile acid excretion is expected
to promote bile acid synthesis from cholesterol through
a decreased feedback inhibition of cholesterol 7-
hydroxylase, which is considered to catalyze the rate-
limiting step in the biosynthesis of bile acids. Conse-
quently, increased bile acid excretion results in the
decreased serum cholesterol concentration in the long
run. In this short-term study, however, we could not
observe such a relationship.
In general, dietary fiber has various physiological
functions and is helpful to those with life-style related
diseases (18). Therefore, diets rich in dietary fiber may
contribute not only to ameliorate lipid metabolism but
also to improve gastrointestinal condition as pre-biotics
(19) as well as to prevent colorectal cancer and cardio-
vascular diseases (20, 21). Ashitaba is rich in dietary
fiber and thus, may be effective to prevent diseases
involved in the digestive tract.
The results obtained here show that Ashitaba intake
promotes excretion of both feces and bile acids,
although there were no significant influences on the
lipid metabolism or body fat accumulations in rats.
Therefore regular daily intake of Ashitaba products may
keep gastrointestinal system healthy. In addition, even
 Effects of Ashitaba on Lipid Profiles and Body Fat Accumulation in Rats
excess intake of Ashitaba for a short period did not
appear to have any adverse effect on hepatic or renal
functions.
REFERENCES
1) Monma K, Kikutani N, Kasahara T, Iguchi M, Tomo-
matsu T, Murakami Y, Urano M. 1990. Proximate and
mineral composition of Ashitaba (Angelica keiskei) har-
vested in the Izu Islands. Ann Rep Tokyo Metr Res 41:
158161.
2) Kozawa M, Morita N, Baba K, Hata K. 1978. Chemical
components of the roots of Angelica keiskei Koidzumi. II.
The structure of the chalcone derivatives (authors
transl). Yakugaku Zasshi 98: 210214.
3) Inamori Y, Baba K, Tsujibo H, Taniguchi M, Nakata K,
Kozawa M. 1991. Antibacterial activity of two chal-
cones, xanthoangelol and 4-hydroxyderricin, isolated
from the root of Angelica keiskei Koidzumi. Chem Pharm
Bull 39: 16041605.
4) Kimura Y, Baba K. 2003. Antitumor and antimetastatic
activities of Angelica keiskei roots, part 1: Isolation of an
active substance, xanthoangelol. Int J Cancer 106: 429
437.
5) Kimura Y, Taguchi M, Baba K. 2004. Antitumor and
antimetastatic activities of 4-hydroxyderricin isolated
from Angelica keiskei roots. Planta Med 70: 211219.
6) Kyogoku K, Hatayama K, Yokomori S, Saziki R, Nakane
S, Sasajima M, Sawada J, Ohzeki M, Tanaka I. 1979.
Anti-ulcer effect of isoprenyl flavonoids. II. Synthesis
and anti-ulcer activity of new chalcones related to
sophoradin. Chem Pharm Bull 27: 29432953.
7) Fujita T, Sakuma S, Sumiya T, Nishida H, Fujimoto Y,
Baba K, Kozawa M. 1992. The effects of xanthoangelol
E on arachidonic acid metabolism in the gastric antral
mucosa and platelet of the rabbit. Res Commun Chem
Pathol Pharmacol 77: 227240.
8) Matsuura M, Kimura Y, Nakata K, Baba K, Okuda H.
1981. Artery relaxation by chalcones isolated from the
roots of Angelica keiskei. Planta Med 67: 230235.
9) Rossi AB, Vergnanini AL. 2000. Cellulite: a review. J Eur
Acad Dermatol Venereol 14: 251262.
10) Folch J, Lees M, Sloane Stanley GH. 1957. A simple
method for the isolation and purification of total lipids
137
from animal tissues. J Biol Chem 226: 497509.
11) Carr TP, Andresen CJ, Rudel LL. 1993. Enzymatic deter-
mination of triglyceride, free cholesterol, and total cho-
lesterol in tissue lipid extracts. Clin Biochem 26: 3942.
12) Eaton DL, Klassen CD. 1976. Effect of acute administra-
tion of taurocholic acid on biliary lipid excretion in the
rat. Proc Soc Exp Biol Med 151: 198202.
13) Binazzi M, Grilli-Cicioloni E. 1977. A proposito della
cosidetta cellulite e della dermato-panniculopatia edem-
ato fibrosclerotica. Ann It Derm Clin Sper 31: 121125.
14) Shimizu E, Hayashi A, Takahashi R, Aoyagi Y,
Murakami T, Kimoto K. 1999. Effects of angiotensin I-
converting enzyme inhibitor from Ashitaba (Angelica
keiskei) on blood pressure of spontaneously hypertensive
rats. J Nutr Sci Vitaminol 45: 375383.
15) Karmar R. 1986. The contribution of peroxisomes to
lipid metabolism. J Clin Chem Clin Biochem 24: 109
118.
16) Bebernitz GR, Schuster HF. 2002. The impact of fatty
acid oxidation on energy utilization: targets and ther-
apy. Curr Pharm Des 8: 11991227.
17) Ogawa H, Nakashima S, Baba K. 2003. Effects of dietary
Angelica keiskei on lipid metabolism in stroke-prone
spontaneously hypertensive rats. Clin Exp Pharmacol
Physiol 30: 284288.
18) Walker ARP, Trowell HC, Anderson JW, Jenkins AL,
Freeman HJ, Tsuji K. 2001. Section 5. Dietary fiber in
the prevention and treatment of disease. In: CRC Hand-
book of Dietary Fiber in Human Nutrition (Spiller GA,
ed), 3rd ed, p 363450. CRC Press, New York.
19) Dikeman CL, Murphy MR, Fahey GC Jr. 2006. Dietary
fibers affect viscosity of solutions and simulated human
gastric and small intestinal digesta. J Nutr 136: 913
919.
20) Nakaji S, Ishiguro S, Iwane S, Ohta M, Sugawara K,
Sakamoto J, Fukuda S. 2004. The prevention of colon
carcinogenesis in rats by dietary cellulose is greater
than the promotive effect of dietary lard as assessed by
repeated endoscopic observation. J Nutr 134: 935939.
21) Jensen MK, Koh-Banerjee P, Hu FB, Franz M, Sampson
L, Gronbaek M, Rimm EB. 2004. Intakes of whole
grains, bran, and germ and the risk of coronary heart
disease in men. Am J Clin Nutr 80: 14921499.