The inhibitory effects of mangiferin, a naturally occurring

glucosylxanthone, in bowel carcinogenesis of male F344 rats

Naoki Yoshimi
a,
*
, Kengo Matsunaga
a
, Masaki Katayama
a
, Yasuhiro Yamada
a
,
Toshiya Kuno
a
, Zheng Qiao
a
, Akira Hara
a
, Johji Yamahara
b
, Hideki Mori
a
a
Department of Pathology, Gifu University School of Medicine, 40 Tsukasa-machi, Gifu 500-8705, Japan
b
Natural Medical Resources Division, The Foundation of Research Institute for Production Development, Kyoto, Japan
Received 23 August 2000; received in revised form 1 November 2000; accepted 1 November 2000
Abstract
Mangiferin, 1,3,6,7-tetrahydroxyxanthone-C2-beta-d-glucoside, is one of xanthone derivatives and C-glucosylxanthones, is
widely distributed in higher plants and is one of constituents of folk medicines. Recent studies showed that mangiferin has a
potential as an anti-oxidant and an anti-viral agent. In this study, we examined the effects of mangiferin in rat colon carcino-
genesis induced by chemical carcinogen, azoxymethane (AOM). We performed two experiments: a short-term assay to
investigate the effects of mangiferin on the development of preneoplastic lesions by AOM, aberrant crypt foci (ACF), and
the following long-term assay for the inuence of mangiferin on tumorigenesis induced by AOM. In the short-term assay, 0.1%
mangiferin in a diet signicantly inhibited the ACF development in rats treated with AOM compared to rats treated with AOM
alone (64:6 ^22:0 vs. 108:3 ^43:0). In the long-term assay, the group treated with 0.1% mangiferin in initiation phase of the
experimental protocol had signicantly lower incidence and multiplicity of intestinal neoplasms induced by AOM (47.3 and
41.8% reductions of the group treated with AOM alone for incidence and multiplicity, respectively). In addition, the cell
proliferation in colonic mucosa was reduced in rats treated with mangiferin (6585% reductions of the group treated with AOM
alone). These results suggest that mangiferin has potential as a naturally-occurring chemopreventive agent. q 2001 Elsevier
Science Ireland Ltd. All rights reserved.
Keywords: Mangiferin; Xanthone derivatives; Chemoprevention; Rat colon carcinogenesis
1. Introduction
Cancer chemoprevention is a very promising means
to control cancer [1,2]. The minor dietary constituents
are very important as chemopreventive agents [2]. In
our search for effective cancer chemopreventive
agents, we have focused on an Asian-transitional
medicine using herbs with an anti-oxidant activity.
We have previously reported the inhibitory effects
of some oriental herbs in rat colon carcinogenesis
[3]. In this study, we now demonstrate the inhibition
of colon tumorigenesis in rats by mangiferin, which is
extracted from the bark of Mangifera indica.
Mangiferin, 1,3,6,7-tetrahydroxyxanthone-C2-bet-
a-d-glucoside, is one of xanthone derivatives and C-
glucosylxanthones [4]. This chemical is widely
distributed in higher plants such as Anacardiaceae
and Gentianaceae families, especially in the leaves
and the bark. In the Philippines, the mangos leaves
are prepared as a tea, and the juice of the leaf is
Cancer Letters 163 (2001) 163170
0304-3835/01/$ - see front matter q 2001 Elsevier Science Ireland Ltd. All rights reserved.
PII: S0304-3835(00)00678-9
www.elsevier.com/locate/canlet
* Corresponding author. Tel.: 181-58-267-2235, fax; 181-58-
265-9005.
E-mail address: yoshimi@cc.gifu-u.ac.jp (N. Yoshimi).
considered to be useful in bleeding dysentery [5]. The
bark and seeds are astringent. In the traditional Indian
system of medicine, mangiferin is also used for
melancholia and nervous debility [6]. Furthermore,
one of traditional Chinese medicines, sann-Joong-
kuey-jian-tang, also includes the mangiferin [7].
Recently, Sato et al. [8] reported that mangiferin
had a strong antioxidative potency as scavengers of
free radicals.
Aberrant crypt foci (ACF) were identied as
preneoplastic markers in rat colon carcinogenesis
[9]. This has allowed ACF to be used as short-term
markers in chemoprevention studies [10]. Therefore,
to determine whether mangiferin might work as a
chemopreventive agent, we rst performed a short-
term assay, which examines ACF induced by azoxy-
methane (AOM) in rat colonic mucosa. Next we
conducted a long-term experiment to investigate the
inhibitory effects of mangiferin in AOM-induced
colon carcinogenesis in rats.
2. Materials and methods
2.1. Chemicals
Mangiferin was extracted from Mangifera indica
Linn in our laboratory. In brief, the bark of mango
was extracted in hot water for 3 h. After the ltered
solution was cooled down and dried up, it was
dissolved in methanol. Mangiferin in the solution
was puried by silica gel column chromatography.
The extracted mangiferin was identied by compari-
son to an authentic standard by nuclear magnetic reso-
nance. The purity was more than 99.5%. The
molecular structure of mangiferin is shown in Fig. 1.
AOM was purchased from Sigma (St. Louis, MO).
2.2. Animal treatment
The experimental designs for the two different
experiments we performed are shown in Fig. 2.
2.2.1. Animals
Fischer 344 rats (4 weeks old), were purchased
from Japan SLC Inc. (Hamamatsu, Japan). Rats
were housed at 23 ^28C, 50 ^10% humidity and a
12 h light/dark cycle, and fed a basal diet (CE-2,
CLEA Japan Inc., Tokyo, Japan).
2.2.2. Experiment I
Thirty-four male Fischer 344 rats were divided into
four groups. Groups 1 and 2 were treated with AOM,
15 mg/kg b.w., s.c., at 6, 7, and 8 weeks of age. Group
2 was administered mangiferin, 0.1% in the basal diet
during the experiment (from 5 to 12 weeks of age).
Group 3 was treated with 0.1% mangiferin in the basal
diet, and rats in group 4 were used as a negative
control (basal diet only). At 12 weeks of age, all rats
were killed, the colons were removed, ushed with
saline, and opened from anus to cecum. The opened
colon was xed at on a paper lter in 10% buffered
formalin for 24 h to observe ACF.
2.2.3. Experiment II
We performed the second experiment to examine
the effect of mangiferin for long duration in rat colon
carcinogenesis. One hundred and seventy-ve rats
were divided into seven groups. Groups 59 were
treated with AOM, 15 mg/kg b.w., s.c., at 6, 7, and
8 weeks of age. Mangiferin was administered in two
different stages of carcinogenesis (Fig. 1). Groups 6
and 7 were treated with 0.02% and 0.1% mangiferin,
respectively, for 4 weeks, in the initiation phase of
AOM treatment. Groups 8 and 9 were treated with
0.02% and 0.1% mangiferin, respectively, for 1
week after the last treatment of AOM, in the post-
initiation phase of AOM treatment. Rats in group 10
were fed 0.1% mangiferin during the experiment, and
group 11 was used as an untreated control. All rats
were sacriced at 40 weeks after the rst AOM treat-
ment. The intestines were removed, ushed with
saline, and opened from anus to duodenum. After
xation in 10% buffered formalin, the whole intes-
N. Yoshimi et al. / Cancer Letters 163 (2001) 163170 164
Fig. 1. The molecular structure of mangiferin.
N. Yoshimi et al. / Cancer Letters 163 (2001) 163170 165
Fig. 2. Experimental protocol. # AOM, 15 mg/kg, s.c. injection; O killed, A, basal diet (CE-2); B, 0.1% mangiferin in basal diet; g, 0.02%
mangiferin in basal diet.
tines were observed for neoplasms, and examined
histopathologically according to the criteria of Ward
[11]. The tumor size was calculated as a 1b 1c
p=3 (a, n and c are length, breadth and height, respec-
tively).
2.3. Identication of ACF
In Experiment I, the xed colons were stained with
0.5% methylene blue in saline. ACF were recorded
according to the procedure of Bird [9] and our labora-
tory [12]. ACF were distinguished from the surround-
ing normal crypts by their swelling size and
discernible pericryptal zone. In this study, we
analyzed the mucosa of whole colon for occurrence
and multiplicity of ACF. Crypt multiplicity means the
number of constituted aberrant crypts in each focus,
and categorized as those containing up to three, four,
or more aberrant crypts/focus. The scores were
checked by two observers in a double-blind fashion.
2.4. Immunohistochemistry for cell proliferation
In Experiment I, all rats were treated with 5-bromo-
deoxyuridine (BrdU) 1 h before sacrice. After the
colons were analyzed for ACF, they were embedded
in parafn for immunohistochemical analysis of
BrdU. In Experiment II, the colonic mucosa without
neoplasms or the surrounding mucosa of neoplasms
were also examined for staining of proliferative cell
nuclear antigen (PCNA) instead of BrdU. The immu-
nohistochemical stains were performed according to
the method in our previous papers [13,14]. The
embedded tissues were cut into 4 mm thin sections,
and then stained by using a cell proliferation assay kit
(Amersham, Aylesbury, UK) for BrdU or anti-PCNA
antibody and LSAB kit (DAKO, Carpinteria, CA).
The number of BrdU or PCNA-positive nuclei in
crypts per section was counted as described in
previous papers [14]. BrdU or PCNA index was deter-
mined by measuring the number of the BrdU or
PCNA-positive nuclei as a proportion of the total
nuclei in a half crypt, which is vertically cut in colonic
mucosa.
2.5. Statistical analysis
Data were presented as mean ^SD. Student's t-
test, Welch's method and the x
2
test were used to
determine the signicance of the difference between
groups, and the criterion of the signicance was taken
as P , 0:05.
3. Results
Table 1 summarizes body weight, liver weight, and
the relative ratio of liver weight to body weight in
both experiments at the termination of the experi-
ments. There were no differences of those among
groups in Experiment I (Table 1). In Experiment II,
N. Yoshimi et al. / Cancer Letters 163 (2001) 163170 166
Table 1
The summary of the body weight, liver weight, and the relative ratio of liver weight to body weight in both experiments
Experiment/group Treatment No. of rats Body weight (g) Liver weight (g) Relative liver weight
(g/100 g body weight)
I/1 AOM alone 10 225 ^11
a
10.7 ^0.9 4.7 ^0.4
2 AOM10.1% mangiferin 10 224 ^11 10.4 ^0.5 4.6 ^0.4
3 0.1% mangiferin alone 7 232 ^12 11.0 ^1.0 4.7 ^0.5
4 Non treatment 7 233 ^10 11.5 ^0.6 4.8 ^0.2
II/5 AOM alone 22 383 ^27 15.9 ^1.0 4.2 ^0.5
6 AOM10.02% mangiferin 30 370 ^24 15.5 ^1.8 4.2 ^0.4
7 AOM10.1% mangiferin 29 381 ^22 16.9 ^2.1 4.4 ^0.4
8 AOM ! 0.02% mangiferin 26 337 ^21
b
12.7 ^2.5
c
3.8 ^ 0.9
9 AOM ! 0.1% mangiferin 28 322 ^22
b
12.6 ^2.0
c
3.9 ^0.6
10 0.1% mangiferin alone 20 353 ^18
b
15.1 ^1.3
c
4.3 ^0.3
11 Non-treatment 20 398 ^24 15.9 ^1.0 4.4 ^0.5
a
Mean ^SD.
b
Signicant different from the corresponding group (group 5 or 11) by Student's t-test (P , 0:01, respectively).
c
Signicant different from the corresponding group (group 5 or 11) by Welch's method (P , 0:01, respectively).
although the body and liver weights in groups 810
treated with mangiferin for a long duration were lower
than those in other groups, the relative ratios were not
signicantly different among groups (Table 1). Food
consumption in all groups was about the same (data
not shown).
The effect of mangiferin on AOM-induced ACF
formation was evaluated in Experiment I. Mangiferin
signicantly (P , 0:01 or P , 0:05 by Welchs
method) inhibited the total number of ACF induced
by AOM by 40%, the total number of aberrant crypts
in ACF per colon by 43%, and the number of foci
containing four or more crypts by 52% (Table 2).
In Experiment II, the incidence and multiplicity in
each group is shown in Table 3. Incidences of total
tumors were 73% in group 5 treated with AOM alone.
Those in groups 6 and 7 treated with mangiferin
during the initiation phase were reduced to 60 and
34.5%, and those in groups 8 and 9 treated with
mangiferin in post-initiation phase were reduced to
42 and 46%. Tumor incidences in groups 7 and 8
were signicantly inhibited, compared to group 5
(P 0:007 and P 0:0447 by Fisher exact test,
respectively). The colonic neoplasms in the above
mangiferin groups tended to be inhibited, but the
differences were not statistically signicant (Table
3). Furthermore, the multiplicity of total tumors
induced by AOM was inhibited in each group treated
with mangiferin, and was statistically signicant in
group 7 (P , 0:01 by Student's t-test) (Table 3).
Histological observations detected no differences
among groups; approximately half the neoplasms
were adenomas and other half were adenocarcino-
mas, although the size of neoplasms in group 5
(338 ^598 mm
3
) seemed to be slightly bigger than
those in other groups (174 ^251, 208 ^232,
132 ^159, 244 ^281 mm
3
, respectively) (There
were no signicant differences among groups). The
tumor invasion to a muscular layer also could not
differ among groups.
N. Yoshimi et al. / Cancer Letters 163 (2001) 163170 167
Table 2
Occurrence of AOM-induced ACF formation in Experiment I
Group/treatment No. of rat Total number of ACF Total number of aberrant crypt Number of foci containing
four or more crypts
1/AOM alone 10 108.3 ^43.0 244.3 ^111.3 11.3 ^7.8
2/AOM10.1% mangiferin 10 64.6 ^22.0
a
138.7 ^50.4
a
5.4 ^3.6
b
3/0.1% mangiferin alone 7 0 0 0
4/Non treatment 7 0 0 0
a
Signicant difference from group 1 (AOM alone) by Welch's method (P , 0:01.
b
Signicant difference from group 1 (AOM alone) by Welch's method (P , 0:05.
Table 3
The incidence and multiplicity of the intestinal neoplasms in Experiment II
Group Treatment No.of rats Incidence (%) Multiplicity
a
Small intestine Colon Total Small intestine Colon Total
5 AOM alone 22 6 (27) 12 (54) 16 (73) 0.32 ^0.55 0.59 ^0.58 0.91 ^0.67
6 AOM 10.02% mangiferin 30 8 (27) 12 (40) 18 (60) 0.32 ^0.64 0.63 ^0.91 0.97 ^0.98
7 AOM 10.1% mangiferin 29 0 10 (34.5) 10 (34.5)
b
0 0.38 ^0.56 0.38 ^0.56
c
8 AOM ! 0.02% mangiferin 26 4 (15) 8 (30) 11 (42)
d
0.19 ^0.48 0.50 ^0.84 0.69 ^1.03
9 AOM ! 0.1% mangiferin 28 6 (21) 10 (36) 13 (46) 0.21 ^0.41 0.54 ^0.82 0.75 ^0.87
10 0.1% mangiferin alone 20 0 0 0
11 Non- treatment 20 0 0 0
a
Mean ^SD.
b
Signicantly different from group 5 by Fisher exact test (P 0:0107).
c
Signicantly different from group 5 by Student's t-test (P , 0:01).
d
Signicantly different from group 5 by Fisher exact test (P 0:0447).
The effects of mangiferin on cell proliferation are
summarized in Table 4. In Experiment I, treatment of
rats with AOM increased the BrdU-labeling index in
colonic crypts by 2.3 times compared to untreated
controls (Table 4). Rats that received mangiferin in
addition to AOM had signicantly lower BrdU label-
ing indices compared to AOM alone (P , 0:01)
(Table 4). In Experiment II, AOM increased PCNA-
staining index 2.25 times compared to the untreated
control. Treatment with mangiferin either at the initia-
tion phase (group 7) or the post-initiation phase
(groups 8 and 9) signicantly reduced the PCNA-
staining index compared to treatment with AOM
alone (P , 0:05 or P , 0:01, Table 4).
4. Discussion
Xanthone derivatives, including mangiferin, are
present in common used folk medical preparations
in the world. However, little is known about the scien-
tic and medical mechanisms and even toxicological
properties. Mangiferin is one of xanthone derivatives
and consists of glucose as C-glucosylxanthone. Some
xanthone derivatives are mutagenic in Salmonella
typhimurium TA100 and TA98 [15,16], but mangi-
ferin is non-mutagenic [17]. Mangiferin is also used
in traditional alternative medicine but its mechanism
of action is unclear. Battar [6] reported that the
mechanisms of this compound in the central nervous
system is related to its ability to inhibit monoamine
oxidase activity. Mangiferin also has anti-viral effects
on herpes simplex [18] and anti-oxidant action [8]. In
addition, mangiferin activates lymphocytes in tumor-
bearing mice [19], and inhibits the growth of ascitic
brosarcomas in Swiss mice [20]. Mangiferin also
enhances tumor cell cytotoxicity of lymphocytes and
macrophages, and antagonizes in vitro the cytopathic
effect of HIV [20]. These activities are slightly similar
to other chemopreventive agents such as 1
0
-acetoxy-
chavicol acetate [21,22]. Therefore, we examined the
chemopreventive effects of this compound, mangi-
ferin, in colon carcinogenesis of rats.
In this study, we found the chemopreventive effects
of mangiferin in rat colon carcinogenesis, with treat-
ment of mangiferin in both the initiation and post-
initiation phase. The higher dose treatment in the
initiation phase was the most effective of the treat-
ments tested to inhibit the incidence and multiplicity
of neoplasms induced by AOM. Although the
mechanisms of chemoprevention by mangiferin are
unclear, one possibility might be to inhibit the produc-
tion of DNA adducts by AOM as well as its anti-viral
and anti-oxidant effects [8,18]. Furthermore, mangi-
ferin inhibited cell proliferation induced by AOM, as
shown by inhibition of BrdU-labeling in Experiment I
N. Yoshimi et al. / Cancer Letters 163 (2001) 163170 168
Table 4
The BrdU-labeling indecies in the colonic crypts in Experiment I and PCNA-staining indecies in Experiment II
Experiment/group Treatment No. of rats BrdU-labeling or PCNA-staining index (%)
a
I/1 AOM alone 10 10.81 ^1.15
b
2 AOM10.1% mangiferin 10 7.08 ^1.07
b, c
3 0.1% mangiferin alone 7 4.75 ^0.11
4 Non-treatment 7 4.83 ^0.15
II/5 AOM alone 10 18.71 ^4.36
6 AOM10.02% mangiferin 10 15.90 ^4.18
7 AOM10.1% mangiferin 10 13.88 ^2.85
d
8 AOM ! 0.02% mangiferin 10 14.21 ^4.17
e
9 AOM ! 0.1% mangiferin 10 13.19 ^3.37
d
10 0.1% mangiferin alone 5 9.24 ^1.51
11 Non-treatment 5 8.29 ^2.74
a
Mean ^SD, BrdU for groups 14, PCNA for groups 511.
b
Signicantly different from group 3 or 4 by Welch's method (P , 0:01).
c
Sigcant difference between groups 1 and 2 by Student's t-test (P , 0:01).
d
Signicantly different from group 5 by Student's t-test (P , 0:01).
e
Signicantly different from group 5 by Student's t-test (P , 0:05).
and PCNA-staining indices in Experiment II. This
effect may explain the inhibitory action of mangiferin
in the post-initiation phase. It is similar to the acts of
other several chemopreventive agents reported
previously [23,24]. In addition, since mangiferin has
a potential on the activation of lymphocytes [19], the
pathway on apoptosis through some cytokines
induced by the lymphocytes might be related to the
inhibitory results [25].
In conclusions, mangiferin showed a potential as
one of the chemopreventive agents. However, the
mechanisms of mangiferin remain unclear, even
though it is widely used in traditional alternative
medicine. In this study, mangiferin inhibited the
body weight, although the relative ratio of liver
weight to body weight was not changed and there
were no effects on other organs. Especially, the
body weights in groups 810 treated with mangiferin
for a long time were lower than those in other groups.
Recently, it has been reported that mangiferin has a
potential to cure diabetes mellitus [26]. Its potential
might be related to the inhibitory effects for body
weight and tumorigenicity in this study. Therefore,
it is necessary to investigate more mechanisms of
mangiferin at the molecular level.
Acknowledgements
We thank K. Takahashi, K. Sato, C. Usui, and T.
Kajita in our laboratory for technical support, and Dr
H. Kleiner, MD Anderson Cancer Center, for a good
advice and editing a manuscript. This work was
supported by a Grant-in-Aid from the Ministry of
Education, Science, Sports and Culture, Japan, a
Grant-in-Aid fromthe Ministry of Health and Welfare,
Japan, and the Program for Promotion of Fundamental
Studies in Health Science from the Organization for
Pharmaceutical Safety and Research (OPSR), Japan.
References
[1] I. Weinstein, Cancer prevention: recent progress and future
opportunities, Cancer Res. 51 (1991) 50805085.
[2] L. Wattenberg, Inhibition of carcinogenesis by minor dietary
constituents, Cancer Res. 52 (1992) 20852091.
[3] N. Yoshimi, A.J. Wang, Y. Morishita, T. Tanaka, S. Sugie, K.
Kawai, J. Yamahara, H. Mori, Modifying effects of fungal and
herb metabolites on azoxymethane-induced intestinal carcino-
genesis in rats, Jpn. J. Cancer Res. 83 (1992) 12731278.
[4] M. Aritomi, T. Kawasaki, A new xanthone C-glucoside, posi-
tion isomer of mangiferin, from anemarrhena asphodeloides
bunge, Tetrahedron Lett. 12 (1969) 941944.
[5] E. Quisumbing, Medicinal Plants of the Philippines, Katha
Publishing Co & JMC Press, Quezon, 1978, pp. 538541.
[6] S.K. Bhattacharya, A.K. Sanyal, S. Ghosal, Monoamine
oxidase-inhibiting activity of mangiferin isolated from
canscora decussata, Naturwissenschaften 59 (1972) 651.
[7] S.J. Lin, H.H. Tseng, K.C. Wen, T.T. Suen, Determination of
gentiopicroside, mangiferin, palmatine, berberine, baicalin,
wogonin and glycyrrhizin in the traditional Chinese medicinal
preparation sann-joong-kuey-jian-tang by high-performance
liquid chromatography, J. Chromatogr. A 730 (1996) 1723.
[8] T. Sato, A. Kawamoto, A. Tamura, Y. Tatsumi, T. Fujii,
Mechanism of antioxidant action of pueraria glycoside
(PG)-1 (an isoavonoid) and mangiferin (a xanthonoid),
Chem. Pharm. Bull. 40 (1992) 721724.
[9] R.P. Bird, Observation and quantication of aberrant crypts in
the murine colon treated with a colon carcinogen: preliminary
ndings, Cancer Lett. 37 (1987) 147151.
[10] S. Olivo, M.J. Wargovich, Inhibition of aberrant crypt foci by
chemopreventive agents, In Vivo 12 (1998) 159166.
[11] J.M. Ward, Morphogenesis of chemically induced neoplasms
of the colon and small intestine in rats, Lab. Invest. 30 (1974)
505513.
[12] N. Yoshimi, K. Kawabata, A. Hara, K. Matsunaga, Y.
Yamada, H. Mori, Inhibitory effect of NS-398, a selective
cyclooxygenase-2 inhibitor, on azoxymethane-induced aber-
rant crypt foci in colon carcinogenesis of F344 rats, Jpn. J.
Cancer Res. 88 (1997) 10441051.
[13] A. Wang, N. Yoshimi, T. Tanaka, H. Mori, Inhibitory effects
of magnesium hydroxide on c-myc expression and cell prolif-
eration induced by methylazoxymethanol acetate in rat colon,
Cancer Lett. 775 (1993) 7378.
[14] H. Mori, Y. Mori, T. Tanaka, N. Yoshimi, S. Sugie, T. Kawa-
mori, T. Narisawa, Cell kinetic analysis of the mucosal epithe-
lium and assay of ornithine decarboxylase activity during the
process of 1-hydroxyanthraquinone-induced large bowel
carcinogenesis in rats, Carcinogenesis 13 (1992) 22172220.
[15] I. Morimoto, T. Nozaka, F. Watanabe, M. Ishino, Y. Hirose, T.
Okitsu, Mutagenic activities of gentisin and isogenitisin from
Gentianae radix (Gentianaceae), Mutat. Res. 116 (1983) 103
117.
[16] T. Nozaka, I. Morimoto, F. Watanabe, T. Okitsu, Mutagenic
activities of bellidifolin, methylbellidifolin, and methylswer-
tianin the methanol extract from Swertiae Herbe (Gentiana-
ceae), Shoyakugaku Zasshi 38 (1984) 96101.
[17] T. Matsushima, A. Araki, O. Yagame, M. Muramatsu, K.
Koyama, K. Ohsawa, S. Natori, H. Tomimori, Mutagenicities
of xanthone derivatives in salmonella typhimurium TA100,
TA98, TA97, and TA2637, Mutat. Res. 150 (1985) 141
146.
[18] M.S. Zheng, Z.Y. Lu, Antiviral effect of mangiferin and
isomangiferin on herpes simplex virus, Chin. Med. J. 103
(1990) 160165.
N. Yoshimi et al. / Cancer Letters 163 (2001) 163170 169
[19] U. Chattopadhyay, S. Das, S. Guha, S. Ghosal, Activation of
lymphocytes of normal and tumor bearing mice by mangi-
ferin, a naturally occurring glucosylxanthone, Cancer Lett.
37 (1987) 293299.
[20] S. Guha, S. Ghosal, U. Chattopadhyay, Antitumor, immuno-
modulatory and anti-HIV effect of mangiferin, a naturally
occurring glucosylxanthone, Chemotherapy 42 (1996) 443
451.
[21] A. Murakami, S. Jiwajinda, K. Koshimizu, Screening for in
vitro anti-promoting activities of edible plants from Thailand,
Cancer Lett. 95 (1995) 139146.
[22] A. Murakami, S. Ohura, Y. Nakamura, K. Koshimizu, H.
Ohihashi, 1
0
-Acetoxychavicol acetate, a superoxide anion
generation inhibitor, potently inhibits tumor promotion by
12-O-tetradecanoylphorbol-13-acetate in ICR mouse skin,
Oncology 53 (1996) 386391.
[23] T. Tanaka, H. Mori, Inhibition of colon carcinogenesis by
non-nutritive constituents in foods, J. Toxicol. Pathol. 9
(1996) 139149.
[24] H. Mori, S. Sugie, N. Ino, R. Mahidor, T. Tanaka, A. Hara, Y.
Morishita, Inhibitory effects of naturally-occurring and related
synthetic organosulfur compounds on genotoxicity in hepato-
cytes and digestive organs carcinogenesis, J. Environ. Pathol.
Toxicol. Oncol. 16 (1997) 281285.
[25] A. Hara, N. Yoshimi, M. Niwa, N. Ino, H. Mori, Apoptosis
induced by NS-398, a selective cyclooxygenase-2 inhibitor, in
human colorectal cancer cell lines, Jpn. J. Cancer Res. 88
(1997) 600604.
[26] H. Ichiki, T. Miura, M. Kubo, E. Ishihara, Y. Komatsu, K.
Tanigawa, M. Okada, New antidiabetic compounds, mangi-
ferin and its glucoside, Biol. Pharm. Bull. 21 (1998) 1389
1390.
N. Yoshimi et al. / Cancer Letters 163 (2001) 163170 170