Biochemical and Molecular Action of Nutrients

Isolation of an Antitumor Compound from Agaricus blazei Murill and Its

Mechanism of Action1
Takeshi Takaku,* Yoshiyuki Kimura†2 and Hiromichi Okuda†
†
Second Department of Medical Biochemistry and *Central Research Laboratory, School of Medicine,
Ehime University, Shigenobu-cho, Onsen-gun, Ehime 791-0295, Japan

Downloaded from https://academic.oup.com/jn/article/131/5/1409/4686970 by guest on 17 December 2020

ABSTRACT The Basidiomycete fungus Agaricus blazei Murill has traditionally been used as a health food for the
prevention of cancer, diabetes, hyperlipidemia, arteriosclerosis and chronic hepatitis. In the present study, we
examined the antitumor activities of various substances isolated from the lipid fraction of A. blazei. Tumor growth
was retarded by the oral administration of the lipid fraction extracted from A. blazei with a chloroform/methanol
mixture in sarcoma 180 – bearing mice. The substance with the antitumor activity in the lipid fraction was isolated
via silica gel column chromatography, eluted with an acetonitrile/methanol (3:2) mixture and identified as ergosterol
by direct comparison of the 1H NMR and mass spectrometry spectral data of an authentic sample. The oral
administration of ergosterol to sarcoma 180 – bearing mice significantly reduced tumor growth at doses of 400 and
800 mg/kg administered for 20 d without side effects, such as the decreases in body, epididymal adipose tissue,
thymus, and spleen weights and leukocyte numbers induced by cancer chemotherapy drugs. Ergosterol had no
cytotoxicity against tumor cells. To clarify the antitumor activity of ergosterol, we examined the effects of ergosterol
on tumor-induced angiogenesis using two in vivo models. Intraperitoneal administration of ergosterol at doses of
5, 10 and 20 mg/kg for 5 consecutive d inhibited the neovascularization induced by Lewis lung carcinoma
cell–packed chambers, suggesting that either ergosterol or its metabolites may be involved in the inhibition of
tumor-induced neovascularization. Therefore, we further examined the inhibitory effects of ergosterol on Matrigel-
induced neovascularization. Female C57BL/6 mice were subcutaneously inoculated with Matrigel containing acidic
fibroblast growth factor and heparin with or without ergosterol. Ergosterol inhibited the Matrigel-induced neovas-
cularization, suggesting that ergosterol directly inhibits Matrigel-induced neovascularization. From these results, it
seems likely that the antitumor activity of ergosterol might be due to direct inhibition of angiogenesis induced by
solid tumors. This is the first report of ergosterol as an antiangiogenic substance. J. Nutr. 131: 1409 –1413, 2001.

KEY WORDS: ● Agaricus blazei ● antitumor activity ● antiangiogenic activity ● antitumor substance
● mice

The Basidiomycete fungus Agaricus blazei Murill (Japanese stances. We first isolated ergosterol as an antitumor substance
name: Himematsutake or Agarikusutake) has been tradition- from the lipid fraction of A. blazei. Ergosterol is contained in
ally used as a health food source in Brazil for the prevention of various mushrooms, such as Lentinus edodes (Berk.) Sing. (Jap-
cancer, diabetes, hyperlipidemia, arteriosclerosis and chronic anese name: Shiitake) and Polyporus umbellatus Fries (Japanese
hepatitis. It has been reported that 100,000 –300,000 kg of the name: Chorei) (5), and ergosterol is a precursor of ergocalcif-
dried body of A. blazei is produced every year in Japan. A. erol. In addition, we studied the side effects (e.g., myelotox-
blazei is used by ⬃300,000 –500,000 persons for the prevention icity, immunotoxicity and reduction in body weight) of these
of cancer and/or as an adjuvant with cancer chemotherapy substances from A. blazei. To clarify the mechanism of anti-
drugs after the removal of a malignant tumor. The intake of tumor activity by an isolated active substance (ergosterol), we
the water extract of A. blazei is ⬃3–5 g three times daily. The examined the effects of ergosterol on splenic immunofunction
hot water extract of A. blazei has potent antitumor activity in and tumor-induced neovascularization.
sarcoma 180 – bearing mice (1– 4), and the antitumor sub-
stance was postulated to be the ␤-(1– 6)-glucan fraction. How-
ever, the antitumor effects of lipid fractions have not been well MATERIALS AND METHODS
studied. We examined the antitumor activities of various General experimental procedures. Melting points, which were
substances isolated from the lipid fraction of A. blazei via oral determined with a Yamato MO-21 capillary apparatus (Yamato Sci-
or intraperitoneal administration to identify the active sub- ence, Tokyo, Japan), are uncorrected. Infrared and ultraviolet spectra
were measured with a Shimadzu IR-400 spectrometer (Kyoto, Japan)
and a JASCO ORD/UV-5 spectrometer (Tokyo, Japan), respectively.
1
Supported by research grants from Bizen Chemical Ltd. (Okayama, Japan). The 1H NMR (499.83 MHz) spectra were recorded in CDCl3 with a
2
To whom correspondence should be addressed. E-mail: yokim@ Varian Unity Inova 500 spectrometer (TOSO, Tokyo, Japan). Mass
m.ehime-u.ac.jp spectra were measured with an Hitachi M-4000 H spectrometer

0022-3166/01 $3.00 © 2001 American Society for Nutritional Sciences.

Manuscript received 1 December 2000. Initial review completed 12 January 2001. Revision accepted 24 February 2001.

1409
1410 TAKAKU ET AL.

(Tokyo, Japan). Column chromatography was performed using silica The tumor volume was determined through direct measurement
gel 60 (70 –230 mesh; ASTM, Merck, Germany). Other chemicals with calipers and calculated by the formula [length (mm) ⫻ width
were of reagent grade. (mm2)]/2 every 2–3 d. On d 21, blood was obtained via venipuncture
Natural materials and isolation of antitumor substances from A. in mice with diethyl ether anesthesia, and then the tumor, epididy-
blazei. A. blazei was supplied by Bizen Chemical (Okayama, Japan). mal adipose tissue, spleen and thymus were removed and weighed for
Voucher samples are stored at the Second Department of Medical evaluation of antitumor activity and side effects. The blood samples
Biochemistry, School of Medicine, Ehime University, Japan. The were chilled in test-tubes containing heparin, and the number of
dried fungal bodies of A. blazei (1 kg) were directly extracted with leukocytes was measured using a Coulter Counter (Japan Scientific
chloroform/methanol (1:1, v/v) (2 L ⫻ 3) for 3 h under reflux. The Instruments Ltd., Tokyo, Japan).
chloroform/methanol extract was concentrated under reduced pres- Solid-type LLC cells were also prepared through subcutaneous
sure to provide a brown extract (250 g). The chloroform/methanol transplantation of 5 ⫻ 105 cells (1 mL) into the right abdomen of
extract (200 g) was divided into acetone-soluble (160 g) and -insol- mice on day 0. Ergosterol (50, 200 or 800 mg/kg) was administered
uble (40 g) fractions. The acetone-soluble fraction (35 g) was further orally once daily for 24 consecutive d, starting 12 h after the implan-
divided into n-hexane–insoluble (16 g) and –soluble (14 g) fractions. tation of tumor cells. Control mice were administered distilled water
The n-hexane–insoluble fraction (15 g) was chromatographed on a alone on the same schedule. On d 25, the mice were killed by cervical
silica gel column, and ergosterol (2.4 g) was isolated as an active dislocation, and their spleens, thymus and lungs were quickly re-

Downloaded from https://academic.oup.com/jn/article/131/5/1409/4686970 by guest on 17 December 2020

substance. Each fraction was tested for antitumor action in sarcoma moved and weighed.
180 – bearing mice. Measurement of lymphocyte number and T-cell population
Cells. The highly metastatic, drug-resistant mouse Lewis lung (CD4ⴙ, CD8ⴙ and NK1.1ⴙ T cells) in LLC-bearing C57BL/6
carcinoma (LLC)3 cells were obtained from Riken Gene Bank mice. The spleen was gently teased to release cells through dissec-
(Tukuba, Japan) and maintained in Dulbecco’s modified Eagle’s me- tion in cold PBS. The cell suspension (5 mL) was layered onto 5 mL
dium (DMEM) supplemented with 10% fetal bovine serum, penicillin of Lympholytes-Mouse and centrifuged at 1500 ⫻ g for 30 min. The
(1 ⫻ 105 U/L), streptomycin (100 mg/L) and amphotericin B (0.25 lymphocyte band at the interface was recovered, and the cells were
mg/L). Sarcoma 180 cells were also maintained in the laboratory of rinsed three times with PBS (pH 7.4). The number of lymphocytes
the Second Department of Medical Biochemistry, School of Medi- was measured using a Coulter Counter. The cell concentration was
cine, Ehime University. adjusted to 2 ⫻ 1010 cells/L, and then 10 ␮L of FITC-labeled
Materials. Mouse lymphocyte separation medium (Lym- anti-mouse CD8, FITC-labeled anti-mouse CD4 or phycoerythrin-
pholytes-Mouse) was purchased from Dainippon Pharmacy Ltd. labeled anti-mouse NK1.1 was added to 100 ␮L of the cell suspen-
(Osaka, Japan), and fluorescein isothiocyanate (FITC)-labeled anti- sion. After incubation for 30 min at 4°C, lymphocytes were rinsed
mouse CD4 and CD8 and phycoerythrin-labeled anti-mouse NK1.1 three times with 1 mL of PBS and centrifuged at 700 ⫻ g for 5 min.
antigen were purchased from Serotec (Oxford, U.K.). Matrigel base- Then CD4⫹, CD8⫹ and NK1.1⫹ T-cell populations were analyzed by
ment membrane (reduced growth factor) was obtained from Becton flow cytometry with an FACS Calibur (Becton Dickinson, Mountain
Dickinson Labware (Bedford, MA). DMEM were obtained from View, CA).
Nissui Pharmaceutical Ltd. (Tokyo, Japan) and used as culture me- Measurement of neovascularization induced by tumor cells. In
dium. Antibiotic and antimycotic solutions were purchased from vivo tumor-induced neovascularization was assayed according to the
Sigma Chemical (St. Louis, MO). Fetal bovine serum was purchased dorsal air-sac method. Briefly, 1.5 ⫻ 106 cultured LLC cells were
from ICN Biochemicals (Aurora, OH). Round nitrocellulose mem- suspended in DMEM, packed into a round nitrocellulose membrane
brane chambers (pore size 0.45 ␮m) were purchased from Millipore chamber with a diameter of 14 mm and implanted into dorsal air-sac
(Bedford, MA). Culture plates were purchased from Corning Glass mice on day 0. Ergosterol (5, 10 or 20 mg/kg) was administered once
Works (Corning, NY). Other chemicals were of reagent grade. ␤-Cy- daily on d 1–5. The mice were killed on d 6, and the hair on the skin
clodextrin was supplied by Ensui Sugar Refining Ltd. (Yokohama, in contact with the chamber was carefully shaved off. The formation
Japan). Ergosterol was suspended in 9 g NaCl/L solution or distilled of new blood vessels in the subcutaneous region was photographed.
water containing 20 g ␤-cyclodextrin/L. Measurement of Matrigel-induced neovascularization. In vivo
Animals. Male ICR strain mice (6 wk old) and female C57BL/6 Matrigel-induced neovascularization was assayed according to the
strain mice (5 wk old) were obtained from Clea Japan (Osaka, Japan). methods of Passaniti et al. (6) Briefly, female C57BL/6 mice were
They were housed for 1 wk in a room maintained at 25 ⫾ 1°C with each injected subcutaneously with 0.5 mL of Matrigel containing 1
60% humidity and had free access to standard nonpurified diet (8 g mg of acidic fibroblast growth factor (aFGF) and 64 ⫻ 103 U of
water, 51.3 g crude carbohydrate, 24.6 g crude protein, 5.6 g crude heparin per L in the presence or absence of ergosterol (400 or 800
lipid, 3.1 g crude fiber, 6.4 g mineral mixture and 1 g vitamin mixture mg/L). The mice were killed on d 5 with an overdose of pentobar-
per 100 g diet; Oriental Yeast Ltd., Osaka, Japan) and water. The bital, and the gels were removed and weighed. Then the hemoglobin
room was illuminated for 12 h/d starting at 0700 h. Animals were contents in the gels were determined using Hemoglobin-Test kits
treated according to the ethical guidelines of the Animal Center, (Wako Pure Chemical, Osaka, Japan).
School of Medicine, Ehime University. The experimental protocol Data and statistical analyses. All values are expressed as means
was approved by the Animal Studies Committee of Ehime University. ⫾ SEM. Data were analyzed by one-way ANOVA, and then differ-
Measurements of antitumor activities and side effects of various ences in means among groups were analyzed using Dunnett’s test or
fractions and active substances isolated from A. blazei in sarcoma Fisher’s protected LSD multiple comparison test (significantly differ-
180 – and LLC-bearing mice. Solid-type sarcoma 180 was prepared ent at P ⬍ 0.05).
through the subcutaneous transplantation of 1.0 ⫻ 106 or 2.5 ⫻ 106 The structure of the isolated substance. The isolated substance
cells into the right abdomen of mice on d 0. Various lipid fractions formed colorless needles with a melting point of 157°C. It was reddish
such as chloroform/methanol extract, acetone-soluble and -insoluble violet with concentrated H2SO4 and CH3COOH [fast atom bom-
fractions and n-hexane–soluble and –insoluble fractions were sus- bardment–mass spectrometry m/z 489 (M⫹ H⫹)]. The isolated sub-
pended in water containing 50 g gum arabic/L through sonication. stance was identified as ergosterol through direct comparison of the
These lipid fractions were administered orally for 20 consecutive d at 1
H NMR spectral data of an authentic sample. The yield was ⬃1.5
a dose of 800 mg/kg, starting 12 h after the implantation of tumor g/kg dried body of A. blazei.
cells. Ergosterol was administered intraperitoneally at doses of 10, 50,
100 or 200 mg/kg or orally at doses of 100, 200, 400 or 800 mg/kg for
20 consecutive d. Control mice were also fed 9 g NaCl/L or water RESULTS
containing 50 g gum arabic/L alone on the same schedule.
Antitumor and side effects of lipid fractions prepared from
A. blazei in sarcoma 180 – bearing mice. The chloroform/
3 methanol extract (800 mg/kg ⫻ 20 d) strongly inhibited tumor
Abbreviations used: aFGF, acidic fibroblast growth factor; DMEM, Dulbec-
co’s modified Eagle’s medium; FITC, fluorescein isothocyanate; LLC, Lewis lung growth (Table 1). Moreover, the oral administration of the
carcinoma. acetone-soluble fraction isolated from the chloroform/metha-
 ANTITUMOR AND ANTIANGIOGENIC ACTIONS OF ERGOSTEROL 1411

Downloaded from https://academic.oup.com/jn/article/131/5/1409/4686970 by guest on 17 December 2020

FIGURE 1 Effects of oral administration of ergosterol isolated from
Agaricus blazei for 20 d on tumor volume in sarcoma 180 – bearing mice.
Solid-type sarcoma 180 was prepared by subcutaneous transplantation of
2.5 ⫻ 106 cells into the right abdomen of mice on d 0. The indicated
amounts of ergosterol were administered orally for 20 consecutive d,
starting 12 h after the implantation of tumor cells. Control mice were also
administered water containing 50 g gum arabic/L alone on the same
schedule. The tumor volume was determined by direct measurement with
calipers and calculated by the formula [length (mm) ⫻ width (mm2)/2] every
2–3 d. Results are expressed as means ⫾ SEM, n ⫽ 10. Those at a time not
sharing a letter are significantly different, P ⬍ 0.05. FIGURE 3 Photographs of Matrigel gel 5 d after subcutaneous
injection with 0.5 mL of Matrigel alone (A), Matrigel supplemented with
1 mg/L acidic fibroblast growth factor and 64,000 U/L heparin (B) or
nol extract significantly inhibited tumor growth at a dose of Matrigel/acidic fibroblast growth factor/heparin mixture at 400 (C) or
800 mg/kg during the 20-d treatment. On the other hand, 800 (D) mg/L.
tumor growth was not affected by the oral administration of
the acetone-insoluble fraction on d 20. Therefore, the ace-
tone-soluble fraction, which had higher antitumor activity, Oral administration of the chloroform/methanol extract
was divided into two fractions through treatment with n- and the various fractions prepared from the chloroform/meth-
hexane. The n-hexane–soluble and –insoluble fractions (800 anol extract had no effect on food intakes and body weight
mg/kg ⫻ 20 d) also inhibited the tumor growth, and their gain for 20 consecutive d in sarcoma 180 – bearing mice (data
inhibitory ratios were 80.2 and 90.1%, respectively. not shown).

FIGURE 2 Effects of ergosterol on neovascularization in C57BL/6 mice bearing LLC cell–packed chambers. Chambers packed with Lewis lung
carcinoma (LLC) cells were subcutaneously implanted into a dorsal air-sac of C57BL/6 mice on day 0. Dulbecco’s modified Eagle’s medium alone
(normal, A), LLC cell–packed chamber (control, B) or 5 (C), 10 (D) or 20 (E) mg ergosterol/kg was intraperitoneally administered from d 1 to 5. The mice
were killed on d 6, and the hair of skin in contact with the chamber was carefully shaved. The formation of new blood vessels in the subcutaneous
region was photographed.
1412 TAKAKU ET AL.

TABLE 1
Effects of various lipid fractions of chloroform/methanol extract (1:1, v/v), acetone-soluble and -insoluble fractions and n-hexane–
soluble and –insoluble fractions on tumor volume at 20 d and tumor weight at 21 d in sarcoma 180 – bearing mice1

Tumor volume2 Inhibition2 Tumor weight3 Inhibition3

mm3 % mg %

Chloroform/methanol extract
Control 4826.9 ⫾ 1150.6a — 4470.0 ⫾ 870.6a —
Chloroform/methanol extract 1087.3 ⫾ 567.6b 77.5 844.2 ⫾ 425.1b 81.1
Acetone soluble and insoluble fractions
Control 928.8 ⫾ 250.9a — 827.7 ⫾ 381.2a —
Acetone-soluble fraction 124.1 ⫾ 83.6b 86.6 108.6 ⫾ 83.4b 86.8
Acetone-insoluble fraction 649.1 ⫾ 140.9ab 30.1 490.5 ⫾ 382.3ab 40.7

Downloaded from https://academic.oup.com/jn/article/131/5/1409/4686970 by guest on 17 December 2020

n-Hexane soluble and insoluble fractions
Control 766.9 ⫾ 302.9a — 812.0 ⫾ 277.2a —
n-Hexane–soluble fraction 152.0 ⫾ 74.6b 80.2 163.9 ⫾ 150.7b 79.8
n-Hexane–insoluble fraction 75.7 ⫾ 24.6b 90.1 54.6 ⫾ 21.4b 93.3

1 Various lipid fractions (800 mg/kg) were orally administered for 20 d in sarcoma 180 – bearing mice.
2 At 20 d.
3 At 21 d. The inhibition ratio (%) was measured as tumor volume or tumor weight of various lipid fraction–treated mice/tumor volume or tumor
weights of control mice. Each value represents the mean ⫾ SEM, n ⫽ 10. Those not sharing a letter are significantly different, P ⬍ 0.05.

Antitumor activity and side effects of ergosterol in sarcoma or immunotoxicity (reduction in thymus and spleen weights)
180 – and LLC-bearing mice. Oral administration of ergos- (data not shown). In addition, ergosterol reduced final tumor
terol for 20 d reduced tumor volume (Fig. 1). The inhibition weight and tumor growth only at the dose of 800 mg/kg on d
ratios for tumor growth with oral administration of ergosterol 20 –23 in LLC-bearing mice (data not shown).
at the doses of 100, 200, 400 and 800 mg/kg ⫻ 20 d were 0.0 Effects of ergosterol on immunofunction and metastasis to
⫾ 30.6, 62.3 ⫾ 8.8, 70.9 ⫾ 10.8 and 85.5 ⫾ 4.7%, respec- the lung in LLC-bearing mice, and cytotoxicity against sar-
tively. Intraperitoneal injection of ergosterol also inhibited coma 180 and LLC cells. Ergosterol had no effect on the
tumor growth at doses of 10, 50, 100 and 200 mg/kg for 20 d, numbers of splenic lymphocytes or CD4⫹, CD8⫹ or NK1.1⫹ T
with inhibition ratios of 20.6 ⫾ 20.5, 57.1 ⫾ 6.5, 65.8 ⫾ 4.7 cells (data not shown). Colony numbers in the lungs of LLC-
and 84.7 ⫾ 4.2%, respectively. bearing mice (control) were 4.80 ⫾ 1.24 ⫻ 106. The oral
Tumor weights were significantly reduced with both intra- administration of ergosterol had no effect on metastasis to the
peritoneal and oral administration of ergosterol (Table 2). lung at the doses of 50, 200 and 800 mg/kg for 23 d, with tumor
Neither oral nor intraperitoneal administration of ergosterol colony numbers (⫻106) of 3.80 ⫾ 1.62, 4.00 ⫾ 0.58 and 5.33
caused side effects such as a reduction in body or epididymal ⫾ 0.67, respectively. Ergosterol had no cytotoxicity against
adipose tissue, myelotoxicity (reduction in leukocyte number) sarcoma 180 and LLC cells (data not shown).
Effects of ergosterol on LLC-induced neovascularization.
At 5 d after implantation of the LLC cells, which were packed
TABLE 2 into the membrane chamber, neovascularization was evident
in the region in contact with the chamber containing LLC
Effects of intraperitoneal (IP) administration and oral cells. Intraperitoneally administered ergosterol (20 mg/kg) pre-
administration ergosterol on tumor weight at 21 d vented neovascularization induced by LLC cells (Fig. 2).
in sarcoma 180 – bearing mice1,2 Effects of ergosterol on Matrigel-induced neovasculariza-
tion. The gels that formed after subcutaneous implantation of
Tumor weight Matrigel alone were readily distinguished from surrounding
tissue and produced little or no local reaction or angiogenic
mg response (Fig. 3A). However, Matrigel supplemented with 1
mg aFGF and 64,000 U heparin per L produced gels that
IP dose, mg/kg showed an angiogenic reaction (Fig. 3B). The Matrigel/aFGF/
Control 2923.5 ⫾ 590.2a
10 2320.4 ⫾ 599.3a
heparin mixture significantly increased the weight of the gel
50 1254.3 ⫾ 192.0b and the hemoglobin contents in the gels at 6 d after implan-
100 998.4 ⫾ 136.9bc tation compared with mice treated with Matrigel alone (Table
200 447.4 ⫾ 124.0c 3). Ergosterol (400 and 800 mg/L) inhibited increases in the
Oral dose, mg/kg weight and hemoglobin concentration of the gels (Fig. 3,
Control 2728.2 ⫾ 913.9a Table 3).
100 2748.8 ⫾ 835.3a
200 1028.7 ⫾ 241.4ab
400 793.1 ⫾ 293.3bc
800 394.1 ⫾ 127.9c
DISCUSSION

1 Ergosterol was intraperitoneally or orally administered for 20 d in A. blazei has been used by ⬃300,000 –500,000 persons for
sarcoma 180 – bearing mice. the prevention of cancer and/or as an adjuvant with cancer
2 Each value represents the means ⫾ SEM, n ⫽ 10. Those not chemotherapy drugs after the removal of a malignant tumor.
sharing a letter are significantly different, P ⬍ 0.05. There have been a number of reports that various Basidio-
 ANTITUMOR AND ANTIANGIOGENIC ACTIONS OF ERGOSTEROL 1413

TABLE 3 metabolites (e.g., ergocalciferol) may be involved in the inhi-

bition of tumor-induced neovascularization. Therefore, we fur-
Effects of ergosterol on the weights and hemoglobin contents ther examined the inhibitory effects of ergosterol on Matrigel-
in the gels 5 d after implantation into mice of Matrigel induced neovascularization using an in vivo model. Female
supplemented with acidic fibroblast growth C57BL/6 mice were subcutaneously injected with Matrigel
factor (aFGF) and heparin1,2 containing aFGF and heparin with or without ergosterol.
Ergosterol inhibited the Matrigel-induced neovascularization
Hemoglobin at concentrations of 400 and 800 mg/L. This finding indicates
Treatment Matrigel weight content that ergosterol may directly inhibit Matrigel-induced neovas-
cularization and that the antitumor activity of ergosterol may
mg mg/Matrigel be due to direct inhibition of angiogenesis induced by solid
Matrigel alone 103.16 ⫾ 10.15b 2.6 ⫾ 0.68b tumors. This is the first report that ergosterol is an antiangio-
Matrigel ⫹ aFGF (1 mg/L) genic substance.
⫹ heparin (64,000 U/L)
371.60 ⫾ 39.75a 21.0 ⫾ 4.00a

Downloaded from https://academic.oup.com/jn/article/131/5/1409/4686970 by guest on 17 December 2020

(Control) ACKNOWLEDGMENT
Matrigel, aFGF, heparin ⫹
ergosterol (400 mg/L) 185.58 ⫾ 44.40b 6.4 ⫾ 1.86b The authors are grateful to K. Baba (Second Department of
Matrigel, aFGF, heparin ⫹ Pharmacognosy, Osaka University, Pharmaceutical Sciences) for 1H
ergosterol (800 mg/L) 108.84 ⫾ 9.69b 3.8 ⫾ 0.58b NMR, 13C NMR and mass spectra and for helpful discussions on the
chemical structures. We are also indebted to T. Fukuda, K. Maeyama
1 C57BL/6 mice were each injected subcutaneously with 0.5 mL of
and N. Maeda of School of Medicine, Ehime University for helpful
Matrigel supplemented 1 mg aFGF/L and 64 ⫻ 103 U heparin/L in the discussion.
absence or presence of ergosterol (400 or 800 mg/L).
2 Each value represents the means ⫾ SEM, n ⫽ 5. Those not sharing
a letter are significantly different, P ⬍ 0.05. LITERATURE CITED
1. Mizuno, T., Inagaki, R., Kanno, T., Hagiwara, T., Nakamura, T., Itoh, T.,
Shimura, K., Sumiya, T. & Asakura, A. (1990) Antitumor activity and some
mycetes have antitumor activity in sarcoma 180 – bearing properties of water-soluble polysaccharides from “Himematsutake,” the fruiting
mice. Chihara (7) reported a study in which the assessments of body of Agaricus blazei Murill. Agric. Biol. Chem. 54: 2889 –2896.
antitumor activity were performed by the intraperitoneal in- 2. Itoh, H., Ito, H., Amano, H. & Noda, H. (1994) Inhibitory action of
(1– 6)-␤-glucan-protein complex (FIII 020b) isolated from Agaricus blazei Murill
jection of water extracts of various Basidiomycetes. Mizuno et (“Himematsutake”) on Meth A fibrosarcoma-bearing mice and its antitumor
al. (1) also evaluated the antitumor activity in sarcoma 180 – mechanism. Jpn. J. Pharmacol. 66: 265–271.
bearing mice by intraperitoneal injection of a water extract of 3. Kawagishi, H., Kanao, T., Inagaki, R., Mituno, T., Shimura, K., Itoh, H.,
Hagiwara, T. & Nakamura, T. (1990) Formolysis of a potent antitumor (1– 6)-
A. blazei. Recently, Ito et al. (8) reported that the intraperi- ␤-glucan-protein complex from Agaricus blazei fruiting bodies and antitumor
toneal or oral administration of a polysaccharide/protein com- activity of the resulting products. Carbohydr. Polym. 12: 393– 403.
plex of A. blazei had antitumor activity in tumor-bearing mice. 4. Fujimiya, Y., Suzuki, T., Oshiman, K. Kobori, H., Moriguchi, K., Na-
However, the antitumor activities of lipid fractions are not yet kashima, H., Matumoto, Y., Takahara, S., Ebina, T. & Katakura, R. (1998)
Selective tumoricidal effect of soluble proteoglucan extracted from the Basidio-
known. In the present study, we found that the lipid fraction, mycete, Agaricus blazei Murill, mediated via natural killer cell activation and
as well as the ␤-glucan fraction, inhibited tumor growth. We apotosis. Cancer Immunol. Immunother. 46: 147–159.
isolated the antitumor substance ergosterol from the lipid 5. Yazawa, Y., Yokota, M. & Sugiyama K. (2000) Antitumor promoting
effect of an active component of polyporus ergosterol and related compounds on
fraction through repeated silica gel column chromatography. rat urinary bladder carcinogenesis in a short-term test with concanavalin A. Biol.
Furthermore, we found several compounds in addition to er- Pharm. Bull. 28: 1298 –1302.
gosterol. Studies of the isolated unknown compounds are in 6. Passaniti, A., Taylor, R. M., Pili, R., Guo, Y., Long, P. V., Haney, J. A.,
Pauly, R. R., Grant, D. S. & Martin, G. R. (1992) A simple, quantitative method
progress to determine their structures and to examine their for assessing angiogenesis and antiangiogenic agents using reconstituted base-
effects on tumor growth. Intraperitoneal and oral administra- ment membrane, heparin and fibroblast growth factor. Lab. Invest. 67: 519 –528.
tion of ergosterol inhibited the tumor growth without causing 7. Chihara, G. (1988) Current status and possible future of antitumor or
drugs: with special reference to pharmacology of oriental medicine. Yakugaku
any of the side effects that are usually caused by cancer Zasshi 108: 171–186.
chemotherapy drugs. Ergosterol had no cytotoxic effect against 8. Ito, H., Shimura, K., Itoh, H. & Kawase, M. (1997) Antitumor effects of
sarcoma 180 cells in vitro (data not shown). Ergosterol is a a new polysaccharide-protein complex (ATOM) prepared from Agaricus blazei
(Iwade strain 101) “Himematsutake” and its mechanisms in tumor-bearing mice.
precursor of ergocalciferol. It has been reported that cholecal- Anticancer Res. 17: 277–284.
ciferol inhibits angiogenesis (9 –11). The ergosterol metabolite 9. Fujioka, T., Hasegawa, M., Ishikura, K., Matsushita, Y., Sato, M. & Tanji,
ergocalciferol may also inhibit angiogenesis. Therefore, to S. (1998) Inhibition of tumor growth and angiogenesis by vitamin D3 agents in
murine renal cell carcinoma. J. Urol. 160: 247–251.
clarify the mechanism of the antitumor activity of ergosterol, 10. Iseki, K., Tatsuta, M., Uehara, H., Iishi, H., Yano, H., Sakai, N. & Ishiguro,
we examined the effects of ergosterol on tumor-induced neo- S. (1999) Inhibition of angiogenesis as a mechanism for inhibition by 1␣-
vascularization. The neovascularization induced by LLC cell– hydroxyvitamine D3 and 1.25-dihydroxyvitamine D3 of colon carcinogenesis in-
packed chambers was inhibited by the intraperitoneal admin- duced azomethane in Wistar rats. Int. J. Cancer 81 730 –733.
11. Van Den Bend, G. J., Pols, H. A. & Van Leeuwen, J. P. (2000). Antitumor
istration of ergosterol at doses of 5, 10 and 20 mg/kg for 5 effects of 1, 25-dihydroxyvitamine D3 and vitamin D analogs. Curr. Pharm. Des. 6,
consecutive d. This finding suggests that either ergosterol or its 717–732.