THE JOURNAL OF ALTERNATIVE AND COMPLEMENTARY MEDICINE
Volume 13, Number 1, 2007, pp. 97–102
Mary Ann Liebert, Inc.
The Mushroom Agaricus blazei
Murill in Combination
with Metformin and Gliclazide Improves Insulin Resistance
in Type 2 Diabetes: A Randomized, Double-Blinded,
CHUNG-HUA HSU, M.D., Ph.D.,1–4 YANG-LI LIAO, M.D.,2 SU-CHING LIN, M.D.,2
KUNG-CHANG HWANG, M.D.,3 and PESUS CHOU, Dr.P.H.4
Complementary and alternative medicine use in adults with type 2 diabetes is popular. Al-
though most of the herbs and supplements appear to be safe, there is still insufficient evidence that demon-strates their definitive beneficial effects. This study was done to determine whether the supplement of Agari-cus blazei
Murill extract improves insulin resistance in type 2 diabetes.
Materials and Methods:
This study was a clinical randomized, double-blind, placebo-controlled trial. Of a
population of 536 registered diabetes patients with 72 subjects (1) aged between 20 and 75 years, (2) beingChinese, (3) having type 2 diabetes for more than 1 year, and (4) having been taking gliclazide and metforminfor more than 6 months were enrolled in this study. The enrolled patients were randomly assigned to either re-ceiving supplement of Agaricus blazei
Murill (ABM) extract or placebo (cellulose) 1500 mg daily for 12 weeks.
Homeostasis model assessment for insulin resistance (HOMA-IR) was used as the major outcome measure-ment.
At the end of the study, subjects who received supplement of ABM extract (n
ϭ 29) showed sig-
nificantly lower HOMA-IR index (3.6[standard deviation, 2.5] versus 6.6[standard deviation, 7.4], p
ϭ 0.04)than the control group (n
ϭ 31). The plasma adiponectin concentration increased 20.0(standard deviation,40.7)% in the ABM group after 12 weeks of treatment, but decreased 12.0(20.0)% among those taking theplacebo (p
Supplement of ABM extract improves insulin resistance among subjects with type 2 diabetes.
The increase in adiponectin concentration after taking AMB extract for 12 weeks might be the mechanism thatbrings the beneficial effect. Studies with longer periods of follow-up should be conducted in the future.
able to find a natural herbal medicine that can help boostinsulin resistance but has less undesirable side effects.
ype 2 diabetes represents a heterogeneous group of dis- The mushroom Agaricus blazei
Murill (ABM) is a nat-
orders characterized by increased insulin resistance.1
ural food, which has been used as a health care product for
Lifestyle, diet, obesity, and family history of diabetes have
the prevention of a range of illness including cancer, dia-
been associated with the development of insulin resistance,
betes, arteriosclerosis, and chronic hepatitis. It has been re-
although the molecular pathway remains unknown.2 Thia-
ported that ABM has beneficial effects in fighting can-
zolidinediones (TZD) have been found to improve periph-
cer,6–10 virus,11 and Streptococcus
pneumonia infection in
eral insulin resistance and has been employed for treating
mice12 as well as enhancing antibody production of vac-
type 2 diabetes.3,4 However, some adverse side effects on
cine.13,14 Rich polysaccharides such as ␤
-glucans are found
liver function have also been reported.5 Hence, it is desir-
to be the main compounds of ABM.14–17 ␤
1Department of Chinese Medicine, Taipei Hospital, Taiwan.
2Department of Endocrinology, Taipei Hospital, Taiwan.
3Department of Pediatrics, Taipei Hospital, Taiwan.
4Community Medicine Research Center and Institute of Public Health, National Yang-Ming University, Taipei, Taiwan.
HSU ET AL.
ABM have demonstrated antidiabetic activity.17 An animal
dom number between 0.50 and 0.99 would be assigned to
model study and a pilot study have both indicated its bene-
the placebo group with cellulose given. The same opaque
ficial effect in type 2 diabetes. Hence, we conducted this
capsules containing either dried powdered ABM extract or
clinical trial to examine whether ABM extract given as a
placebo (cellulose) were administered to the subjects by a
supplement can improve insulin resistance in type 2 dia-
research assistant blinded to the contents in the capsules. All
patients were treated in the same fashion.
RESEARCH DESIGN AND METHODS
Homeostasis model assessment for insulin resistance
(HOMA-IR) [fasting glucose (mmol/L) ϫ fasting insulin
(UI/L)/22.5] was used as the major outcome measure-
The trial was conducted from July 1, 2005 through De-
ment.18,19 At baseline and after 12 weeks of treatment, an-
cember 31, 2005 in the Taipei Hospital, Taiwan. Of a pop-
thropometric measurements, blood pressure, fasting glucose,
ulation of 536 registered patients with diabetes, 72 subjects
hemoglobulin A1C% (HbA1C), insulin, adiponectin and
met with the inclusion criteria: (1) aged between 20 and 75
plasma lipoproteins (triglyceride, cholesterol, choles-
years; (2) being Chinese; (3) having type 2 diabetes for more
terol–high-density lipoprotein, and cholesterol–low-density
than 1 year; and (4) having been taking gliclazide and met-
lipoprotein) of both groups were measured. The change (%)
formin for more than 6 months were enrolled in this study;
in concentration of adiponectin after the 12-week treatment
and exclusion criteria were as follows: (1) aminotransferases
aspartate or aminotransferases alanine Ͼ80 IU/L, serum cre-
All measurements were made at 8 AM to 9 AM after an
atinine Ͼ2.0 mg/dL; (2) prolacting or pregnant women, heart
overnight fast using standardized methods. A sample of
failure, acute myocardic infarct, stroke, and heavy injuries;
whole blood was drawn and centrifuged at 4°C, and a 1-mL
and (3) any other conditions not suitable for trial as evalu-
aliquot of serum was rapidly frozen (Ϫ80°C) for subsequent
ated by the physician. The protocol was approved by the
hormone analysis. The plasma adiponectin concentration
Human Ethics Committee of our hospital. Informed consent
was measured by a radioimmunoassay kit (Linco Research,
was obtained from all the enrolled patients. The patients
Inc., St. Charles, Missouri). This kit employs the double-an-
were instructed to maintain an isocaloric diet and their pre-
tibody/polyethylene glycol technique using 125I-lableled
vious eating habits during the study period. All subjects were
adiponectin and a multispecies adiponectin rabbit antis-
free to withdraw at any time during the course of the study.
erium. Plasma insulin levels were measured using a com-mercially available radioimmunoassay (Linco ResearchInc.). The intra- and interassay coefficients of variation were
Preparation of sample and treatment
3.1% and 4.9%, respectively. The limit of sensitivity is 0.5
ABM is a health care product popularly used in Taiwan.
Our ABM extract samples, obtained from Eng Chiao Bio-Technology Co. Ltd., Taiwan, were extracted from dried
fungal bodies of ABM according to the preset standard pro-cedures with certificate of analysis given. The placebo given
The data were analyzed with SPSS software (version
to the control group comprised pure microcrystalline cellu-
11.5, SPSS Software, Inc., Chicago, IL). Paired t
lose. The subjects were asked to take one capsule contain-
used to examine differences within-group at 0 to 12 weeks.
ing 500 mg of either ABM extract or cellulose three times
test was used to examine the main outcome, de-
each day for 12 weeks. The capsule was taken with gli-
mographic data, and other measurements between group
clazide 30 minutes before eating and metformin was taken
means. Chi-square test was employed for gender compari-
30 minutes after eating. During the study period, the sub-
son between groups. All p
values were two-tailed, and the
jects should keep taking the same dose of gliclazide and met-
level of significance was set at 0.05. We estimated in
formin except when hypoglycemia occurs, in which case the
power 0.8 that each group required 28 subjects.
dose of gliclazide or metformin should be reduced immedi-ately.
All subjects were randomly assigned to one of the two
groups. A random number between 0.0 and 0.99 would be
Seven subjects of the ABM extract group and five sub-
generated by the computer for each patient. Patients with a
jects of the placebo group withdrew because of personal rea-
random number between 0.0 and 0.49 were assigned to the
sons. In the end, 60 patients completed the study. Table 1
group with ABM extract given, whereas those with a ran-
shows the demographic data, fasting serum glucose meta-
AGARICUS BLAZEI SUPPLEMENT FOR INSULIN RESISTANCE
TABLE 1. CHARACTERISTICS OF SUBJECTS AT BASELINE
HbA1c, hemoglobin A1c; HOMA-IR, homeostasis model assessment for insulin resistance; HDL, high-density lipoprotein; LDL,
bolic factors, plasma lipoprotein, and clinical profiles of the
groups at the time of entry. As can be seen, there were no
Three subjects of the ABM extract group had hypo-
significant differences in all the baseline measurements be-
glycemia-like symptoms; 2 patients had the dose of oral hy-
tween the ABM extract group and the placebo group.
poglycemia agent (OHA) reduced, whereas 1 patient with-drew. One subject of the placebo group also developed
Within-group comparison at 12 weeks
hypoglycemia-like symptoms. The dose of OHA was re-
Table 2 shows the within-group comparisons at baseline
duced and he stayed on in the study.
and after the 12-week treatment. According to the analysis
Two subjects of the ABM extract group developed skin
of the fasting serum glucose metabolic factors, the ABM ex-
itching, 1 subject of the placebo group had skin allergy with
tract group showed a significant difference in reduced
papules, and another had nausea and fullness sensation. All
HbA1C, insulin concentration, and homeostasis model as-
4 of them withdrew. No major adverse effects were noticed.
sessment for insulin resistance (HOMA-IR) index as well asincreased concentration of adiponectin, compared with theinitial values, whereas the placebo group showed significantdifference only in reduced adiponectin level compared with
This study shows the benefits of ABM extract supple-
ment for reducing the HOMA-IR index in subjects with type
Between-group comparison at 12 weeks
2 diabetes treated with gliclazide and metformin.
At the end of 12 weeks (Table 2), there were no signifi-
Insulin resistance, a common cause of type 2 diabetes,
cant differences between the two groups in all the mea-
implies impairment of insulin signaling in target tissues. It
surements except insulin concentration and HOMA-IR in-
has been reported that some OHA might influence insulin
ϭ 0.05 and 0.04). Figure 1 shows a significant
resistance.4,5,17,20–22 To avoid existing confounders and bias,
difference between the ABM extract group and the placebo
we examined a homogenous Chinese cohort who had had
Ͻ 0.001) in terms of change (%) in plasma
type 2 diabetes for more than 1 year and had been taking
adiponectin concentrations after 12 weeks of treatment.
gliclazide and metformin for more than 6 months.
HSU ET AL.
TABLE 2. CHARACTERISTICS OF SUBJECTS AFTER 12-WEEK TREATMENT
Data are means (SD).
Ͻ 0.05 from baseline to the end (12 weeks) with paired t
HbA1c, hemoglobin A1c; HOMA-IR, homeostasis model assessment for insulin resistance; HDL, high-density lipoprotein; LDL,
Many animal model studies have demonstrated the anti-
diabetic activity of ABM.17 After an animal model study in2003 and a pilot study had showed its beneficial effect inpatients with type 2 diabetes, we conducted this clinical trialto examine whether the supplement of ABM extract im-proves insulin resistance in patients with type 2 diabetes.
We also evaluated the adiponectin concentration using manyfasting serum glucose metabolic factors to further examinethe possible mechanism of its beneficial effects.
In this study, subjects who received the supplement of
ABM extract showed a significant reduction in HOMA-IRindex (from 4.8 to 3.6), which was much lower than that ob-served among the control subjects (from 6.3 to 6.6).
It is interesting to note that the subjects who had taken a
supplement of ABM extract for 12 weeks showed a signif-icant increase in plasma adiponectin concentrations com-pared with the placebo (cellulose) group (p
Ͻ 0.001) (Fig.
1). Many previous studies have reported that the adiponectinlevel might be a major modulator of insulin resistance23 andpredict the development of type 2 diabetes.24–26 Circulatingadiponectin levels in human is positively correlated with in-sulin sensitivity.23,27 The level of adiponectin appeared to
Change (%) in plasma adiponectin concentrations after
play a very important role in the regulation of insulin act-
12-week treatment in Agaricus blazei
Murill extract (ABME) groupand placebo (cellulose) group (p
Ͻ 0.001). CI, confidence interval.
ing and energy homeostasis.27,28 We observed a 20% in-crease in adiponectin level after 12 weeks of treatment with
AGARICUS BLAZEI SUPPLEMENT FOR INSULIN RESISTANCE
ABM extract supplement. This might account for the effect
tion with ABM extract improves insulin resistance in type
of improving insulin resistance in the ABM extract group.
2 diabetes. The increase in adiponectin concentration after
Our finding reveals that after the 12-week treatment, sub-
12-week treatment with ABM extract supplement might be
jects receiving ABM extract had a 6.7% reduction of HbA1C
the mechanism that brings beneficial effect. Studies with a
(from 8.9 to 8.3) and those taking cellulose had a 2.4% re-
longer period of follow-up should be made in the future.
duction (from 9.1 to 8.9). Although both supplementsshowed a similar effect on reducing HbA1C with no statis-tical difference, the results still had clinical implications. Inthe AMB extract group, the decrease in HbA1C could be at-
tributed to the improved insulin resistance, whereas that inthe control group might be caused by loss of appetite or re-
We thank all colleagues in Taipei Hospital, Taiwan for
duced food intake. Although the initial results showed no
helping with this study. This study was supported by grants
statistical difference in HbA1C, treatment of longer dura-
from the Taipei Hospital and Eng Chiao Bio-Technology
tion might result in significant statistical difference.
As mentioned in our study design, all the subjects should
maintain the same dose of gliclazide and metformin duringthe study period unless hypoglycemia occurs. During the 12-
week treatment, 3 subjects of the ABM extract group de-veloped hypoglycemia-like symptoms; 2 patients had the
1. Tuomilehto J, Lindstrom J, Eriksson JG, et al. The Finnish Di-
dose of OHA reduced and stayed in the study whereas the
abetes Prevention Study. Group prevention of type 2 diabetes
other patient withdrew. Although at the end of study there
mellitus by changes in lifestyle among subjects with impaired
was no significant difference in the dose of OHA received,
glucose tolerance. N Engl J Med 2001;344:1343–1350.
the supplement of ABM extract might have the effect of re-
2. Lawlor DA, Smith GD, Ebrahim S. Life course influences on
ducing the current OHA dose for type 2 diabetes, which is
insulin resistance: Findings from the British women’s heart
beneficial to diabetes control. Hence, future studies on ef-
and health study. Diabetes Care 2003;26:97–103.
fects of ABM should be alert for hypoglycemia-like symp-
3. Fujiwara T, Yoshioka S, Yosioka T, et al. Characterization of
new oral antidiabetic agent CS-045: Studies in KK and ob/ob-
mice and Zucker fatty rats. Diabetes 1988;37:1549–1558.
Although many positive effects of ABM have been re-
4. Nolan JJ, Lodvik B, Beerdsen P, et al. Improvement in glu-
ported, most of them were from in vitro
or animal stud-
cose tolerance and insulin resistance in obese subjects treated
ies.6–17 To our knowledge, this is the first randomized clin-
with troglitazone. N Eng J Med 1994;331:1188–1193.
ical ABM study. Our results reveal improvement in insulin
5. Forman LM, Simmons DA, Diamond RH. Hepatic failure in
resistance by ABM. Although TZD can enhance insulin sen-
a patient taking rosiglitazone. Ann Intern Med 2000;132:
sitivity,3,4,29 its toxicity affects the liver.5 Hence, ABM can
be a possible supplement that has no major adverse effects
6. Guterrez ZR, Mantovani MS, Eira AF, et al. Variation of the
and does not result in liver function impairment, as noted in
antimutagenicity effects of water extracts of Agaricus blazei
this study. Attention must be paid to the 2 subjects of the
Murrill in vitro. Toxicol In Vitro 2004;18:301–309.
ABM extract group who developed skin itching. Clinical ob-
7. Kimura Y, Kido T, Takaku T, et al. Isolation of an anti-an-
giogenic substance from Agaricus blazei
Murill: Its antitumor
servation of other possible side effects should be made in
and antimetastatic actions. Cancer Sci 2004;95:758–764.
8. Ahn WS, Kim DJ, Chae GT, et al. Natural killer cell activity
Complementary and alternative medicine use in adults
and quality of life were improved by consumption of a mush-
with type 2 diabetes is popular.30–32 Although most of the
room extract, Agaricus blazei
Murill Kyowa, in gynecologi-
herbs and supplements appear to be safe, there is still in-
cal cancer patients undergoing chemotherapy. Int J Gynecol
sufficient evidence that demonstrates their definitive bene-
ficial effects.31 The ABM extract was obtained from dried
9. Luiz RC, Jordao BQ, da Eira AF, et al. Mechanism of anti-
fungal bodies of ABM, which has been widely taken as a
clastogenicity of Agaricus blazei
Murill mushroom organic ex-
health care product in Taiwan and Japan for years. The ef-
tracts in wild type CHO (K(1)) and repair deficient (xrs5) cells
fect of insulin resistance improvement attributed to taking
by chromosome aberration and sister chromatid exchange as-
supplement of ABM extract was demonstrated in our stud-
says. Mutation Res 2003;528:75–79.
10. Giacomini NL, Eira AF, Ribeiro LR, Mantovani MS. Anti-
ies. The beneficial effects of AMB are worth further explo-
clastogenic effect of aqueous extracts of Agaricus blazei
ration in detailed clinical studies.
CHO-k1 cells, studying different developmental phases of the
Despite the encouraging results, our work still had limi-
mushroom. Toxicol In Vitro. 2003;17:465–469.
tations, such as short period of study. Future research should
11. Sorimachi K, Ikehara Y, Maezato G, et al. Inhibition by Agar-
be performed with a longer follow-up period and other well-
Murill fractions of cytopathic effect induced by
western equine encephalitis (WEE) virus on VERO cells in
In conclusion, this study demonstrated that supplementa-
vitro. Biosci Biotechnol Biochem 2001;65:1645–1647.
HSU ET AL.
12. Bernardshaw S, Johnson E, Hetland EG. An extract of the
23. Weyer C, Funahashi T, Tanaka S, et al. Hypoadiponectinne-
mushroom Agaricus blazei
Murill administered orally protects
mia in obesity and type 2 diabetes: Closed association with in-
against systemic streptococcus pneumoniae infection in mice.
sulin resistance and hyperinsulinemia. J Clin Endocrinol
Expert Panel on the Identification, Evaluation, and Treatment
of Overweight in Adults. Scand J Immunol 2005;62:393–398.
24. Spranger J, Kroke A, Mohlig M, et al. Adiponectin and pro-
13. Chen L, Shao HJ, Su YB. Coimmunization of Agaricus blazei
tectin against type 2 diabetes. Lancet 2003;361:226–228.
Murill extract with hepatitis B virus core protein through DNA
25. Lindsay RS, Funahahi T, Hanson RL, et al. Adiponectin and
vaccine enhances cellular and humoral immune responses. Int
development of type 2 diabetes in the Pima Indian population.
14. Nakajima A, Ishida T, Koga M, et al. Effect of hot water ex-
26. Daimon M, Oizumi T, Saitoh T, et al. Decreased serum lev-
tract from Agaricus blazei
Murill on antibody-producing cells
els of adiponectin are a risk factor for progression to type 2
in mice. Int Immunopharmacol 2002;2:1205–1211.
diabetes in Japanese population: The Funagata study. Diabetes
15. Kobayashi H, Yoshida R, Kanada Y, et al. Suppressing effects
of daily oral supplementation of beta-glucan extracted from
27. Havel PJ. Control of energy homeostasis and insulin action by
Murill on spontaneous and peritoneal dissem-
adipocyte hormones: Leptin, acylation stimulating protein and
inated metastasis in mouse model. J Cancer Res Clin Oncol
adiponectin. Lipidology 2002;13:51–59.
28. Mattthias BS, Eris S, Nader R, Frank BH. Relationship be-
16. Fujimiya Y, Suzuki Y, Katakura R, Ebina T. Tumor-specific
tween adiponectin and glycemic control, blood lipids and in-
cytocidal and immunopotentiating effects of relatively low
flammatory markers in men with type 2 diabetes. Diabetes
molecular weight products derived from the basidiomycete,
Murill. Anticancer Res 1999;19:113–118.
29. Henry RR. Thiazolidinediones. Endocrinol Metab Clin North
17. Kim YW, Kim KH, Choi HJ, Lee DS. Anti-diabetic activity
of beta-glucans and their enzymatically hydrolyzed oligosac-
30. Yeh GY, Eisenberg DM, Kaptchuk TJ, Phillips RS. System-
charides from Agaricus blazei. Biotechnology Lett 2005;27:
atic review of herbs and dietary supplements for glycemic con-
trol in diabetes. Diabetes Care 2003;26:1277–1294.
18. Matthews D, Hosker J, Rudenski A, et al. Homeostasis model
31. Garrow D, Egede LE. Association between complementary
assessment: Insulin resistance and ␤
-cell function from fast-
and alternative medicine use, preventive care practices, and
ing plasma glucose and insulin concentractions in man. Dia-
use of conventional medical services among adults with dia-
betes. Diabetes Care 2006;29:15–19.
19. Bonora E, Formentini G, Calcaterra F, et al. HOMO-estimated
32. Egede LE, Ye X, Zheng D, Silverstein MD. The prevalence
insulin resistance is an independent predictor of cardiovascu-
and pattern of complementary and alternative medicine use in
lar disease in type 2 diabetic subjects: Prospective data from
individuals with diabetes. Diabetes Care 2002;25:324–329.
the Verona Diabetes Complications Study. Diabetes Care2002;25:1135–1141.
20. Pistrosch F, Passauer J, Fischer S, et al. In type 2 diabetes,
rosiglitazone therapy for insulin resistance ameliorates en-
dothelial dysfunction independent of glucose control. Diabetes
21. Bailey CJ. Treating insulin resistance in type 2 diabetes with
metformin and thiazolidinediones. Diabetes Obes Metab2005;7:675–691.
22. Noble J, Baerlocher MO, Silverberg J. Management of type 2
diabetes mellitus. Role of thiazolidinediones. Can Fam Physi-cian 2005;51:683–687.
Interviewee: Ellen Arlene Smith Interviewee: Manuel Zax Date of Interview: October 26, 2006 Location: Worcester, Massachusetts Transcriber: Manuel Zax Abstract: Ellen A. Smith was born in Worcester and has lived here almost all 92 years of her life. She attended the public schools and Salter’s, where she learned more about shorthand and typing, leading to work as a secretary for Templ
Welcome to “Hot News”, a section of AIDS Reviews written by the editors and invited experts which focuses on recently reported information believed to be of both impact and higher interest to the readership. Update on the Teratogenicity of Efavirenz efavirenz-containing regimens, most of which were In March 2005, Bristol-Myers Squibb and the FDA first-trimester exposures. Birth