Effect of a medicinal extract from Agaricus blazei Murill on gene
expression in a human monocyte cell line as examined by
microarrays and immuno assays
, G. Hetland
, E. Johnson
, B. Grinde
Division of Environmental Medicine, Norwegian Institute of Public Health, Oslo, Norway
Division of Infectious Disease Control, Norwegian Institute of Public Health, Oslo, Norway
Department of Immunology and Transfusion Medicine, Ulleva˚l University Hospital, Oslo, Norway
Department of Gastroenterological Surgery, Ulleva˚ l University Hospital, Oslo, Norway
Received 21 June 2005; received in revised form 6 July 2005; accepted 18 July 2005
Extracts from the edible mushroom Agaricus blazei Murill (AbM) are used extensively as a non-prescription remedy against
cancer, infections, and immune related diseases. The presumed effect is to activate certain parts of the immune system. In order
to investigate the effect, we examined the changes of gene expression caused by the extract on a human monocyte cell line
(THP-1). Changes in the levels of mRNA transcripts were measured using 35 k microarrays, and the changes in select cytokine
gene products by immuno assays. Lipopolysaccharide (LPS) was included for comparison. Both AbM and LPS had drastic
effects on gene expression. Genes related to immune function were selectively up-regulated, particularly proinflammatoric
genes such as the interleukins IL1B and IL8. Although most genes induced by AbM were also induced by LPS, AbM produced
a unique profile, e.g., as to a particular increase in mRNA for the cytokines IL1A,CXCL1,CXCL2 and CXCL3, as well as
D2005 Elsevier B.V. All rights reserved.
Keywords: Agaricus blazei Murill; Microarray; Cytokines; Lipopolysaccharide; Immune function
1567-5769/$ - see front matter D2005 Elsevier B.V. All rights reserved.
Abbreviations: AbM, Agaricus blazei Murill; BG, background;
FG, foreground; GO, gene ontology; HUGO, human genome orga-
nisation; LPS, lipopolysaccharide; MIAME, minimum information
about a microarray experiment; NAE, not appreciably expressed;
TLR, Toll-like receptor.
* Corresponding author. Fax: +47 22 042447.
E-mail address: email@example.com (B. Grinde).
The Agaricus blazei mushroom belongs to the
family Basidiomycetes and is native of Brazil. It
was brought to Japan due to alleged health effects,
and is today widely used both in the form of extracts,
and as an edible, medicinal mushroom. According to
tradition, it helps against a variety of diseases, includ-
ing cancer, diabetes, arteriosclerosis and hepatitis. In
International Immunopharmacology 6 (2006) 133 – 143
recent years, considerable research has been carried
out to investigate the putative effects.
An immunomodulating impact of extracts from
this mushroom seems to be well documented [1–6].
Moreover, several reports suggest a positive effect on
cancer [7–10], possibly mediated either via stimula-
tion of natural killer cells by fungal glucans and
subsequent apoptosis [1,3,7–9,11–14], or by an anti-
angiogenic effect mediated by ergosterol and pyroglu-
tamate [15,16]. As to antimicrobial activity, AbM has
been reported to inhibit western equine encephalitis
virus in cell culture . Extracts may also be used as
an adjuvant to vaccines [18,19].
h-Glucans from other fungi have been shown to
have both antitumor and anti-infection effects [20–23].
The AbM extracts are rich in h-glucans that presum-
ably contribute to the observed activity [9,24], but
other substances are probably involved as well.
In the present study, a bacterial lipopolysaccharide
(LPS) preparation was used for comparison, as the
response to LPS has been well characterized [25–28].
LPS binds to Toll-like receptor 4 (TLR4), and thereby
activates mitogen-activated protein (MAP) kinases and
nuclear factor kappa B. The net result is to induce
proinflammatory cytokines such as IL8 and tumor
necrosis factor (TNF). The h-glucans are assumed to
exert their effect via Toll-like receptor 2 (TLR2), dec-
tin-1 (CLECSF12), and by binding to the lectin site of
complement receptor 3 (ITGAM) . Myeloid differ-
entiation protein-88 (MYD88) appears to be involved in
signalling in the case of both LPS and h-glucans .
Previous reports have established that AbM can
induce certain cytokines, e.g., IL1B and IL6 in mouse
peritoneal macrophages upon oral administration ,
and TNF and IL8 in rat bone marrow macrophages
upon in vitro administration . A down-regulation of
IL2, IL4 and IFNG in human peripheral blood mono-
nuclear cells has also been documented . Microar-
rays offer a powerful tool for investigating the global
gene expression effect of modulating agents. The
arrays used in the present study presumably cover the
vast majority of mRNAs made by human cells, as they
contain oligos reflecting 34580 known or putative
transcripts. Thus the present results offer a panorama
on how A. blazei preparations may act on human
As exemplified above, the HUGO gene symbols
will be used when discussing genes and gene pro-
ducts. Alternative names, e.g., those typically applied
to the corresponding proteins, will be included when
2. Materials and methods
2.1. Chemicals and cell culture
THP-1 (American Type Culture Collection, TIB-202)
cells were grown in plastic culture flasks (NunclonkSur-
face) in RPMI medium supplemented with 10% Foetal
Bovine Serum (Gibco, UK) and 1% Penicillin/Streptomycin
(Paa Laboratories GmbH, Austria) at 37 8C with 5% CO
The cells were seeded at a concentration of 7.510
ml (totally 10 10
THP-1 cells per flask), and stimulated
with either a 9% (v/v) solution of an aqueous extract of A.
blazei Murill (AbM) (Gold label, obtained from ACE Co.,
Ltd, Gifu-ken, Japan) containing 300 Ag/ml of h-glucans
according to the manufacturer, or 10 Ag/ml of E. coli
lipopolysaccharide (type L-2654 from Sigma, St.Louis,
MO, USA) for 24 h. Three independent experiments were
performed with each addition. The AbM extract was kept at
48C in dark glass bottles. Opened bottles were used for no
more than two weeks. The LPS content of the extract was
examined using the PyrogenekRecombinant Factor C
Endotoxin Detection System (BioWhittaker, Maine, USA)
and found to be negligible (i.e., 0.15 ng per ml).
2.2. RNA extraction
To stop the experiments, the cells were centrifuged
(800 g, 5 min), and the pellets immediately resuspended
in 600 Al of the lysis buffer supplied with the Agilent Total
RNA Isolation Mini Kit (Agilent Technologies, Palo Alto,
CA) used for RNA extraction. The lysed cells were kept at
70 8C until extraction according to manufacturer’s recom-
mendations. The RNA was eluted in 50 Al RNase-free water.
RNA concentration and quality were determined using UV
spectrophotometry as well as microcapillary electrophoresis
on a Bioanalyzer 2100 (Agilent Technologies, Palo Alto,
CA) according to manufacturer recommendations. The yield
was between 30 and 60 Ag of RNA per flask, with an UV
260 / 280 ratio between 2.016 and 2.050, and electrophoretic
profiles suggesting intact RNA.
RNA was converted to fluorescence-labelled cDNA with
the FairPlaykaminoallyl kit (Amersham International,
UK), according to the manufacturer’s recommendations,
using 20 Ag of total RNA per sample and Cy5 or Cy3 as
L.K. Ellertsen et al. / International Immunopharmacology 6 (2006) 133–143134
fluorescent dyes. The labelled cDNA was hybridized with
35 k human 70-mer oligonucleotide-based microarrays pro-
vided by the Norwegian Microarray Consortium (NMC,
http://www.mikromatrise.no). The microarrays were spotted
on Corning UltraGAPS glass-slides, using the Operon
Human Genome Oligo Set Version 3.0, with 34580 70-
mer probes representing 24,650 known or putative genes
and 37,123 gene transcripts (Operon Biotechnologies,
Huntsville, AL, USA). Controls in the form of Lucidea
Microarray ScoreCard v1.1 (Amersham Biosciences) were
included. Automated hybridization was performed for 12 h
at 45 8C with the ChipMap80 kit (including the Chi-
pHybe80 buffer) on the Ventana Discovery system (Ventana
Medical Systems, Tucson, AZ, USA). The arrays were
scanned on an Agilent fluorescence scanner (Axon Instru-
ments, Foster City, CA, USA) with a pixel size of 10 Am.
2.4. Microarray analyses
Quality control of the scanned images (tiff-files), and
conversion to numerical values were performed using the
GenePix Pro 4.1 (Axon Instruments). The gpr-files obtained
were further analysed using the freely available BASE BioAr-
ray Software Environment ; available at http://
alba.uio.no/base/. Analyses were based on mean FG–median
BG intensities. Flagged spots and controls were filtered away.
Next, the plug-in application Transformation Simsalabim (a
weak spot filter) was run with default settings, and then
normalization using Lowess. Subsequently, spots with both
channels mean FG–median BG b300 were removed, as well
as spots were log2 ratios were between 1 and 1. Only spots
where at least two of the three parallel experiments had passed
the previous criteria were retained, but for these spots data for
all three parallel experiments were included for further pro-
cessing. The files were transferred to Excel where the average
ratios for the three parallels were calculated. The following
filtering was performed: Reporters with average up- or down-
regulation above respectively 2or 4were selected, as
were reporters with average intensities above 1000. Table 1
shows the number of reporters retained as up- or down-
regulated at various stringency of filtering, as well as the
number of annotated genes these reporters corresponded to.
The gene ontology program eGOn (http://nova2.idi.ntnu.no/
egon/) and the Norwegian Microarray Consortium Dataware-
house (http://nova2.idi.ntnu.no/annotdb/) were used to exa-
mine gene function (based on gene annotations available as of
May 2004). The gene ontology analyses concentrated on
genes that were more than 4up- or down-regulated and
had mean intensities above 1000. Comparisons of up- or
down-regulated genes were performed with either the
bPaired-target testQ(for comparing AbM with LPS) or the
bTarget-Master testQ(for comparing either AbM or LPS up-
regulated genes with all expressed genes). Both tests are
available in the eGOn pack. MIAME compatible microarray
protocols and data are available in the ArrayExpress databank
2.5. Determination of cytokine concentrations
The concentrations of IL-6, IL-8, IL-10 and IL-12p40 in
cell culture supernatants were measured using sandwich
enzyme immuno assays (ELISA), with antibodies from
R&D Systems (UK). Briefly, polystyrene plates (Costar
EIA/RIA Plate, Sigma-Aldrich Norway) were coated with
100 Al/well of anti-hIL-6 (1 Ag/ml) anti-hIL-8 (1 Ag/ml),
anti-hIL-10 (4 Ag/ml), or anti-hIL-12 p40 (4 Ag/ml) antibo-
dies in carbonate–bicarbonate buffer (50 mM, pH 9.6). Serial
two-fold dilutions (in 1% of BSA/Tris–Tween20) of recom-
binant cytokines were used as standards (R&D Systems).
The standards were distributed in triplicates (100 Al/well),
while culture supernatants were used undiluted and distrib-
uted in duplicates (100 Al/well). The incubation period was 2
h at room temperature. Between incubation steps the plates
were washed five times (Skan Washer 400, Model 12019,
Skatron, Norway) with 0.01 M of Tris/HCL-buffer (pH 7.4)
containing 0.05% Tween20. Biotinylated anti-human IL-6
Number of up- or down-regulated transcripts at various stringencies of analysis
more than 2
more than 4
More than 4up/down and average
Transcripts Annotated genes
AbM 5993–4195 1876 423/280 215/111 94/13 84/13
LPS 7261–5436 2360 768/539 371/233 177/13 155/12
The arrays contained 34,580 probes of which 24,897 reflected gene transcripts with annotations. Transcripts expressed denote reporters
where the intensity of at least one channel in one experiment had mean FG–median BG above 300 after normalization. The subset that had
annotations in the Datawarehouse (http://nova2.idi.ntnu.no/annotdb/) is indicated.
The number of transcripts where the average intensity (three experiments) of either red or green channel was above 1000.
The following columns indicate the number of up- or down-regulated transcripts based on increasing stringencies of analyses, as subsets of
the transcripts referred to in the first column.
L.K. Ellertsen et al. / International Immunopharmacology 6 (2006) 133–143 135
(50 ng/ml), IL-8 (0.04 Ag/ml), IL-10 (500 ng/ml), or IL-12
(0.3 Ag/ml) antibodies (R&D Systems) were added (100 Al/
well) as a third layer, and the plates were incubated for 1 h at
room temperature. StreptABComplex/AP (Dako A/S, Den-
mark) was added to the plates according to manufacturer’s
protocol. After 1 h at room temperature, the plates were
developed with 1 mg/ml p-nitrophenyl phosphate (Sigma
104R, St. Louise, MO, USA) in 10% diethanolamine buffer
(pH 9.8). Optical density (OD) was measured at 405 nm on a
MRX Microplate Reader connected to a PC using Revelation
software for instrument operation and calculations (Dynatech
laboratories, Chantilly, VA, USA).
Commercial ELISA kits were used to analyse IL-1h
(DuoSet, R&D Systems, UK) and TNF-a(TNF-a
CytoSetsk, Biosource International, USA). The assays
were performed as recommended by the manufacturers,
and OD measured as described above, but at 450 nm.
2.6. DNA and protein synthesis
Cells from two of the three parallel experiments
described above were used to examine total DNA and
protein synthesis by the incorporation of, respectively,
[H]leucine. Cell suspensions (1 10
cells/well), either unstimulated or stimulated with AbM or
LPS, were added to 96-well plates in triplicates. In the
assay for thymidine incorporation, 1.25 ACi of
dine (Amersham Pharmacia Biotech, Buckinghamshire,
UK) was added to each well and the cells incubated for
21 h. A Skatron cell harvester (Skatron A/S, Lier, Norway)
transferred the cells onto filter papers (FilterMAT 11731)
that were subsequently washed and dried. In the assay for
[H]leucine incorporation, the cells were incubated for 21
h with 1 ACi/well (Amersham Pharmacia Biotech) in a
leucine free medium. The cells were subsequently centri-
fuged at 1000 rpm for 5 min and the pellets dissolved in
0.1 M NaOH. 100 Al of 20% trichloroacetic acid (TCA)
was added in order to precipitate the proteins, and the
precipitate transferred to the filterpaper using the Skatron
harvester (FilterMAT 11734). The filterpapers were mixed
with scintillation fluid, and the incorporated
[H]leucine measured in a scintillation counter (Tri-
Carb, Model 1500, Packard Instruments Co., Inc. Meridan,
3.1. Microarray analyses
Of the 34,580 transcript related reporters on the arrays,
5993 (corresponding to 4195 annotated transcripts) were
appreciably expressed (i.e., at least one channel in one
experiment had mean FG–median BG N300) in the experi-
ments with AbM, compared to 7261 (corresponding to 5436
annotated transcripts) in the experiments with LPS (Table
1). The difference between the AbM and the LPS experi-
ments was probably due partly to arbitrary variations in the
Fig. 1. Scatter plots showing spot intensities before normalization.
One representative experiment is shown for, respectively: A, AbM;
and B, LPS. The control cells were labelled green and the treated
cells red. Flagged spots and controls are not included. The regres-
sion lines F2SD are indicated. (For interpretation of the references
to colour in this figure legend, the reader is referred to the web
version of this article.)
L.K. Ellertsen et al. / International Immunopharmacology 6 (2006) 133–143136
Immune related genes up-regulated by AbM/LPS or otherwise of interest
HUGO Gene description AbM LPS
Selectively induced by AbM
CXCL3 Chemokine (CXC motif) ligand 3 60 (F18) 2230 NAE
PTGS2 Prostaglandin-endoperoxide synthase 2 (COX2) 41 (F3) 1756 15 (F2) 661
CXCL2 Chemokine (CXC motif) ligand 2 39 (F16) 1890 21 (F7) 831
CXCL1 Chemokine (CXC motif) ligand 1 31 (F6) 3470 12 (F4) 507
RGS1 Regulator of G-protein signalling 1 21 (F11) 1179 15 (F6) 930
IL1A Interleukin 1, alpha 13 (F4) 2201 4.0 (F1) 917
Induced by both AbM and LPS
IL8 Interleukin 8 972 (F416) 35,084 353 (F75) 19,018
IL1B Interleukin 1, beta 579 (F208) 36,846 407 (F43) 32,387
SCYA4 Small inducible cytokine A4 (=CCL4) 161 (F67) 6366 338 (F91) 27,989
CCL3 Chemokine ligand 3 (MIP-1a) 115 (F46) 6594 187 (F32) 13,675
IL23A Interleukin 23, asubunit p19 82 (F35) 3279 99 (F22) 8703
CCL3L1 Chemokine ligand 3-like 1 72 (F29) 6741 105 (F20) 16,831
SOD2 Superoxide dismutase 2, mitochondrial 67 (F17) 17,091 56 (F7) 27,435
SCYA20 Small inducible cytokine A20 (=CCL20) 48 (F22) 1963 37 (F9) 1742
NCF1 Neutrophil cytosolic factor 1 35 (F18) 2979 22 (F3) 9071
NFKBIA NFKB B-cells inhibitor, alpha 14 (F2) 4555 13 (F1) 9106
IFI30 Interferon, gamma-inducible protein 30 12 (F1) 1572 14 (F5) 3657
ALOX5AP Arachidonate 5-lipoxygenase-activating protein 11 (F6) 4560 6.1 (F0) 3113
FTH1 Ferritin, heavy polypeptide 1 11 (F4) 47,139 4.8 (F1) 47,140
NCF1 Neutrophil cytosolic factor 1 11 (F1) 2975 15 (F1) 10,000
IFNGR2 Interferon gamma receptor 2 11 (F3) 1317 6.6 (F1) 2141
CEBPB CCAAT/enhancer binding protein h11 (F2) 12,787 7.5 (F1) 20,342
PBEF Pre-B-cell colony-enhancing factor 10 (F3) 2996 7.4 (F0) 2724
G0S2 Putative lymphocyte G0/G1 switch gene 7.5 (F1) 2318 5.1 (F1) 1927
SCYA5 Small inducible cytokine A5 (=RANTES or CCL5) 7.0 (F1) 1525 4.6 (F1) 9876
NFKB1 NFK light polypeptide gene enhancer in B-cells 1 (p105) 6.1(F1) 1007 11 (F1) 2118
NFKBIE NFKB enhancer in B-cells inhibitor, epsilon 6.1 (F2) 1119 4.2 (F0) 1628
S100A9 S100 calcium binding protein A9 5.5 (F2) 4147 13 (F1) 11,606
CASP1 Caspase 1 (IL1B convertase) 4.4 (F0.1) 1075 4.8 (F0) 2232
Selectively induced by LPS
SCYA2 Small inducible cytokine A2 (=CCL2) 5.6 (F1) 384 211 (F43) 19,389
SCYB10 Small inducible cytokine B10 (=CXCL10) NAE 71 (F12) 3205
IFI27 Interferon, a-inducible protein 27 NAE 60 (F4) 7377
SERPINA1 Serine proteinase inhibitor A 1 8.4 (F1) 317 56 (F3) 2689
CYBB Cytochrome b-245, hpolypeptide 17 (F1) 783 49 (F6) 2773
TNFAIP6 TNF, a-induced protein 6 17 (F5) 744 48 (F9) 2128
ISG20 Interferon stimulated gene 20kDa NAE 40 (F7) 1600
EBI3 Epstein–Barr virus induced gene 3 13 (F2) 565 39 (F14) 3358
CD83 CD83 antigen 9.0 (F3) 367 37 (F7) 1449
IFIT1 Interferon-induced protein with tetratricopeptide repeats 1 NAE 27 (F2) 1503
TNF Tumor necrosis factor (=TNFA or TNFASF2) 10 (F6) 605 26 (F2) 5627
G1P2 Interferon, alpha-inducible protein NAE 21 (F2) 5838
IFI44L Interferon-induced protein 44 NAE 20 (F1) 2401
IL10RA Interleukin 10 receptor, alpha 8.0 (F0) 260 19 (F3) 1257
BCL6 B-cell CLL/lymphoma 6 11 (F0) 505 18 (F3) 1140
IL6 Interleukin 6 (interferon, beta 2) NAE 17 (F7) 1011
(continued on next page)
L.K. Ellertsen et al. / International Immunopharmacology 6 (2006) 133–143 137
intensity of the different arrays. However, as can be seen
from Table 1, LPS caused the induction of more genes (even
when correcting for differences in array intensities), an
effect that contributed to the bias in the number of genes
Both AbM and LPS caused considerable changes in gene
expression (Table 1 and Fig. 1). As much as 12% (703/
5993) and 18% (1307/7261), respectively, of the expressed
genes were either up- or down-regulated at least 2. Even at
a high stringency of filtering, i.e., 4changes and average
intensities above 1000, respectively, 107 and 190 genes
were retained, here referred to as considerably up- or
down-regulated genes. As expected for substances that acti-
vate the cells, many more genes were up-regulated com-
pared to down-regulated.
The scatter plots obtained when analysing the effect of
AbM displayed a curious split, where approximately half the
reporters ended up above the regression line, and the other
HUGO Gene description AbM LPS
Selectively induced by LPS
LY6E Lymphocyte antigen 6 complex, locus E NAE 17 (F2) 2529
FOS V-fos FBJ murine osteosarcoma viral oncogene homolog NAE 15 (F5) 2453
MX1 Myxovirus resistance 1, interferon-inducible protein p78 NAE 14 (F2) 3562
SPP1 Secreted phosphoprotein 1 6.1 (F1) 717 12 (F3) 2063
SCYA8 Small inducible cytokine subfamily A8 (=CCL8) NAE 11 (F9) 1854
OAS2 2V-5V-oligoadenylate synthetase 2 1.2 (F0) 425 11 (F1) 5013
IL1RN Interleukin 1 receptor antagonist 2.6 (F0) 495 11 (F1) 2940
NFKB2 NFK light polypeptide gene enhancer in B-cells 2 5.9 (F0) 941 9.8 (F2) 1191
OAS3 2V-5V-oligoadenylate synthetase 3 3.3 (F0) 297 9.4 (F3) 1852
IFITM2 Interferon induced transmembrane protein 2 1.0 (F0) 1577 9.0 (F1) 24,115
OAS1 2V,5V-oligoadenylate synthetase 1 NAE 8.3 (F1) 1516
TNFAIP3 TNFa-induced protein 3 9.0 (F3) 778 8.1 (F3) 1136
CD53 CD53 antigen 5.7 (F2) 654 7.7 (F0) 1551
TAP1 Transporter 1, ATP-binding cassette, B 3.5 (F1) 501 6.7 (F0) 1727
IRF7 Interferon regulatory factor 7 1.8 (F0) 810 6.6 (F1) 3243
G1P3 Interferon, alpha-inducible protein 3.0 (F0) 3683 5.9 (F0) 47,559
S100A8 S100 calcium binding protein A8 3.4 (F2) 1526 5.8 (F1) 3890
IFI35 Interferon-induced protein 35 NAE 5.3 (F1) 1154
IFI16 Interferon, gamma-inducible protein 16 NAE 4.9 (F1) 1004
IFIT3 Interferon induced protein with tetratricopeptide repeats 1 NAE 4.7 (F0) 2054
Other immuno related genes of interest
TLR2 Toll-like receptor 2 3.6 (F0) 881 3.5 (F1) 996
TLR8 Toll-like receptor 8 NAE 3.7 (F1) 636
CLECSF12 C-type lectin, super-family 12 (Dectin-1) 1.4 (F0) 12,139 1.0 (F0) 7372
MYD88 Myeloid differentiation primary response gene 1.9 (F0) 240 0.9 (F0) 665
MAL T-cell differentiation protein 0.5 (F0) 478 0.5 (F0) 566
TRIF TIR domain containing adaptor inducing interferon-beta NAE 4.6 (F1) 321
TRAM1 Translocation associated membrane protein 1 1.1 (F0) 1179 1.3 (F0) 1694
CD14 Monocyte differentiation antigen NAE 23 (F3) 875
PTGS1 Prostaglandin-endoperoxide synthase 1 (COX1) 5.8 (F1) 374 4.1 (F1) 502
CCL13 Chemokine ligand 13 (SCYA13) NAE 11 (F2) 388
CCL22 Chemokine ligand 22 (SCYA22) NAE 2.0 (F0) 563
CXCL13 Small inducible cytokine B13 8.8 (F2) 276 22 (F3) 740
The average channel (red/ green) ratios, as well as the average red channel intensities, of three experiments are offered. The two data sets were
filtered for considerably up-regulated genes, i.e., reporters with ratios above 4 and intensities above 1000. The table is divided into genes
selectively induced by AbM, induced by both AbM and LPS, selectively induced by LPS, and other genes of interest. Data that do not satisfy the
filtering criteria are added in cursive. NAE stands for dnot appreciably expressedT, meaning that none of the channels were above 300 (mean
FG–median BG) in intensity.
Table 2 (continued)
L.K. Ellertsen et al. / International Immunopharmacology 6 (2006) 133–143138
half below (Fig. 1A). The effect was present in all three
parallels, but neither in those with LPS, nor in any other
arrays from our laboratory. Closer inspection of the arrays
gave no indications that the effect was due to artefacts.
Thus, the split apparently reflected an actual consequence
on gene expression of AbM treatment, presumably in the
form of either a large number of genes being somewhat
repressed, or a different sizeable subset of genes being
3.2. Up- and down-regulated genes
Both treatments caused considerable down-regulation
of 13 transcripts, of which 5 (CAS2, CTSG, MS4A6A,
MYB and SCGF) were shared. The down-regulated genes
did not appear to converge on particular Gene Ontology
In the case of up-regulated transcripts, however, genes
involved in immune processes were highly over-repre-
sented. In Table 2 is listed fold induction, as well as spot
intensities, of relevant immune related genes. Interestingly,
AbM proved to be a particularly potent activator of chemo-
kines, with a 30- to 60-fold increase in the mRNAs for
CXCL1, 2 and 3. Several other cytokines were considerably
up-regulated by both treatments, including IL8,IL1B ,
SCYA4,CCL3 (MIP1-a), IL23A,CCL3L1 and SCY15
(Rantes), while a number of cytokines were selectively
induced by LPS (such as SCYA2,SCYB10,IL6 and
SCYA8). Particularly in the case of LPS, there was also a
tendency towards up-regulation of genes involved in cell
death and apoptosis (data not shown).
The concentration of key cytokines was measured using
ELISA methods (Table 3). As can be seen, in the case of the
two most up-regulated transcripts, IL8 and IL1B, both
showed considerably increased levels of proteins. In the
case of TNF and IL12B, the protein levels appeared to
have increased more than the mRNA levels, in fact, there
was no appreciable expression of IL12B mRNA. IL6
appeared to be slightly induced, particularly in the case of
LPS, while IL10 was not appreciably detected either as
protein or mRNA.
3.3. Cytostatic effects
In order to evaluate the general cytostatic or cytotoxic
effect on the cells, protein and DNA synthesis were mea-
sured (incorporation of respectively [
H] leucine and [
thymidine). AbM caused a 63% inhibition of DNA synth-
esis, compared to 69% for LPS, and a 30% inhibition of
protein synthesis, compared to 23% for LPS (all values are
mean of two experiments).
Both LPS and AbM induced a Th1-related inflam-
matory response, as indicated by the increased expres-
sion of IL8,IL1B,IL12B, and TNF (Tables 2 and 3).
None of the typical Th2 response cytokines (IL4, IL5,
IL9, and IL13) were expressed. The results contrast
with a previous report where LPS, but not a hemi-
cellulase-treated agaricus extract, was found to induce
proinflammatory cytokines in murine bone-marrow-
derived dendritic cells . Both the nature of the
extract, and the type of cell, may help explain the
discrepancy. Yet, in order to rule out the possibility
that the present effect of AbM could be due to con-
tamination with LPS, the LPS concentration was
assayed. It was found to be miniscule, i.e., 0.15 ng/
ml of LPS equivalents. In contrast, the LPS prepara-
tion used contained 10 Ag/ml. Moreover, our findings,
as to the capacity of agaricus extracts to induce key
Comparison of microarray data with data on expressed proteins
HUGO Gene description AbM LPS
Cytokine vs. control
Cytokine vs. control
IL8 Interleukin 8 39,467 (F664) vs. b1500 972 (F416) 24,900 (F2558) vs. b1500 353 (F75)
IL1B Interleukin 1h2154 (F96) vs. 18 (F2) 579 (F208) 1676 (F315) vs. 13 (F4) 407 (F43)
TNF Tumor necrosis factor a7234 (F1580) vs. 54 (F16) 10 (F6) 46,852 (F32466) vs. 46 (F8.8) 26 (F2)
IL12B Interleukin 12 subunit p40 1254 (F582,2) vs. b125 NAE 4525 (F1053) vs. b125 NAE
IL6 Interleukin 6 (interferon, h2) 72 (F20) vs. b32 NAE 178 (F40) vs. b32 17 (F7)
IL10 Interleukin 10 V490 vs. b490 NAE b490 vs. b490 NAE
Both protein and mRNA data are based on changes in expression compared to control (non-treated cells), and offered as the average of three
independent experiments. NAE stands for dnot appreciably expressedT. The symbol bimplies that the result was below the (indicated) detection
limit of the assay.
L.K. Ellertsen et al. / International Immunopharmacology 6 (2006) 133–143 139
Th1 cytokines, are consistent with those of other
reports [5,31,33]. We therefore assume that the pre-
sent effect of AgM is due to other components than
LPS, of which h-glucans (300 Ag/ml) may be of
While LPS binds primarily to TLR4 (which presum-
ably was present on the cells, although not appreciably
expressed as mRNA), glucans bind to TLR2, dectin-1
(CLECSF12)[30,34,35] and CD11b (ITGAM) .
TLR2 was appreciably induced by both AbM (3.3)
and LPS (3.2). CLECSF12 was highly expressed in
these cells, and induced 1.5by AbM while remaining
unchanged in the case of LPS. ITGAM was not appre-
ciably expressed or induced. The reason why TLR4 did
not appear on the arrays may be that the receptor has a
low turnover and therefore does not require continual
The microarray approach presumably allowed for
an evaluation of possible changes in most of the genes
expressed by the cells. A reasonably large fraction of
the mRNAs changed considerably upon treatment
with AbM, and even more genes displayed altered
expression upon the addition of LPS (Table 1). The
more drastic effect of the latter is in line with the
general understanding that TLR4 ligands activate two
or more pathways, including the MYD88 signalling
pathway used by TLR2; and that macrophages
respond more strongly to TLR4 ligands compared to
TLR2 ligands .
As may be expected from substances that cause
cellular activation, many more genes were up-regu-
lated compared to down-regulated. Genes belonging
to both groups were analysed as to molecular func-
tion, biological process and cellular component using
the gene ontology based program eGOn (http://
nova2.idi.ntnu.no/egon/). While the down-regulated
genes did not appear to cluster, both AbM and LPS
up-regulated genes clustered highly significantly ( p-
values from 10
) to immune related pro-
cesses such as immune response, inflammatory
response, response to pathogen, and response to exter-
nal biotic stimulus. These results substantiate the
validity of the use of microarrays to discern the effect
of immune modulators on gene expression, at least
when looking at genes with a certain level of tran-
script concentration. The main genes involved in the
microbial product based activation of the cells will be
MYD88 is assumed to be essential for the signal-
ling of all TLRs with the exception of TLR3 and TLR4
.MYD88 was expressed in the cells (Table 2), but
the mRNA concentration was neither high nor appre-
ciably altered. Several MAP kinases typically
involved in the downstream events were also detected
on the arrays, but at low levels and with no obvious
induction. Genes such as these may reflect a limitation
of the microarray approach. That is, genes that need
only to be expressed at low levels even when induced,
e.g., due to a signalling cascade type of effect, may
fall below the mRNA concentration range where
microarrays perform best. Moreover, genes that are
induced only temporarily and have short-lived
mRNAs, are also easily missed. The point is reflected
in the data on the cytokines IL12B and IL6, which
displayed a considerable increase in protein level in
the absence of detectable mRNA expression (Table 3).
As to the NF-kappa-B transcription regulator,
both NFKB1 and NFKB2 were induced by AbM
and LPS, as were the related inhibitors NFKBIA
and NFKBIE (Table 2). The induction of both the
regulator and its inhibitors suggest a fine-tuned reg-
ulation of NFKB-mediated transcriptional control.
Other negative regulators of the TLR-response,
such as IRAKM, SOCS1, NOD2 and TOLLIP
, were not appreciably expressed or induced.
However, as will be discussed below, apoptosis
related genes were induced, particularly in the case
IL8 and IL1B were the dominant cytokines to
come out of the activation with either AbM or LPS.
These genes were up-regulated several hundred
folds to high expression levels, and the correspond-
ing proteins were found at high concentrations in
the cell supernatants. Both AbM and LPS also
caused considerable increase in three interesting
chemokine receptor ligand mRNAs: CCL3 (MIP-
1a), CCL3L1, and SCYA5 (Rantes). These ligands
may inhibit HIV replication, possibly by binding to
the HIV uptake receptor CCR5 [37,38], suggesting a
potential role for AbM in the treatment of HIV
patients. It is also interesting to note that AbM
was particularly potent in inducing CXC chemo-
kines, while both treatments induced several CC-
type chemokines. The C-C and CXC chemokines
are well known stimulators of, respectively, mono-
cytes and neutrophil granulocytes.
L.K. Ellertsen et al. / International Immunopharmacology 6 (2006) 133–143140
Although several additional immune related genes
were induced by both preparations (Table 2), there were
notable differences in the effect of the two preparations.
CD14, which is involved in the TLR4 response ,
was induced 23-fold by LPS, but, as expected, not by
AbM. Other cytokines (e.g., SCYA2,SCYA8,SCYB10,
and IL6) and interferon related genes (e.g., IFI27,
ISG20,IFIT1,IFI44L,IFITM2,IFI35 ,IFI16, and
IFIT3) were also selectively induced by LPS. Notably,
the interferons themselves were not detectably induced.
TRIF, another gene involved in TLR4 signalling ,
appeared, as expected, to be induced by LPS and not
AbM, but again the level of expression was very low.
Interestingly, AbM was more potent as to the induction
of the key prostaglandin synthesis enzyme cox2
(PTGS2). PTGS1 increased upon both treatments, but
was expressed at low levels.
AbM, but to a less extent LPS, induced RGS1,a
regulator of G-protein signalling. RGS1 is crucial for
the activity of the G-protein-coupled rhodopsin-like
chemokine/cytokine receptor super-family. This family
includes the chemotaxis related receptors for IL8
(CD128), complement C5a anaphylatoxin (CD88),
bacteria-derived formyl peptides (fMLP) and leuko-
trienes (LTB4) [40–43]. Combined with the drastic
up-regulation of IL8, this may suggest that AbM
is involved in eliciting the directional migration of
The peculiar scatter plots observed in the case of
cells treated with AbM (Fig. 1) testify further to the
different nature of the two treatments. The split in the
scatter plots suggests that most genes were either
(slightly) up- or down-regulated. Alternatively (if one
of the two legs reflects status quo), that half the genes
were either up- or down-regulated. If the split in the
scatter plots was caused by a broad down-regulation of
genes, one might speculate that the effect reflected an
induction of apoptosis or other forms of cell death.
In order to assess a possible cytostatic or cytotoxic
effect, protein synthesis ([
and DNA synthesis ([
were measured. Both AbM and LPS caused an
approximately 25% inhibition of protein synthesis
and 65% inhibition of DNA synthesis. Moreover,
LPS, but to a less extent AbM, significantly induced
genes involved in programmed cell death and cell
growth regulation (as indicated by gene ontology
analyses); and LPS, but not AbM, caused an appreci-
able increase in the number of cells being stained with
the vital dye trypan blue (data not shown). Thus,
cytotoxic or cytostatic effects do not appear to explain
the peculiar microarray scatter plots.
An alternative explanation might be that AbM
caused the cells to differentiate, in which case one
might expect a massif change in gene expression.
However, markers associated with monocyte differ-
entiation, i.e., macrophage receptors such as CD23,
CD25 and CD69, were not induced.
In addition to the annotated genes, several tran-
scripts with no suggested functions were induced to
high levels either by one or by both treatments (for
details, consult data submitted to ArrayExpress). The
elucidation of the function of these genes would be
The Th1 response includes both proinflammatory
and anti-inflammatory cytokines, and plays a role in
fighting off infections caused by viruses, bacteria and
fungi, as well as combating cancer. The Th2 response
is presumably designed for parasites, but its activation
is also associated with allergy. The Th1/Th2 paradigm
postulates that stimulation of the Th1 response should
down-regulate the Th2 response . The recent iden-
tification of regulatory cells as a third arm of the
immune response has modified the paradigm ,
but the Th1/Th2 balance is still of interest. Conse-
quently, AbM may theoretically be used not only to
fight infections and cancer, but also to reduce allergy.
This option is currently being investigated.
The present study was supported by The National
Programme for Research in Functional Genomics in
Norway (FUGE) under The Research Council of Nor-
way, and by the Norwegian Asthma and Allergy
Foundation. We would like to thank Marc Gayorfar,
Jan Inge Helseth and A
˚se Eikeset for excellent tech-
 Ahn WS, Kim DJ, Chae GT, Lee JM, Bae SM, Sin JI, et al.
Natural killer cell activity and quality of life were improved by
consumption of a mushroom extract, Agaricus blazei Murill
L.K. Ellertsen et al. / International Immunopharmacology 6 (2006) 133–143 141
Kyowa, in gynecological cancer patients undergoing che-
motherapy. Int J Gynecol Cancer 2004;14:589 – 94.
 Kawamura M, Kasai H, He L, Deng X, Yamashita A, Ter-
unuma H, et al. Antithetical effects of hemicellulase-treated
Agaricus blazei on the maturation of murine bone-marrow-
derived dendritic cells. Immunology 2005;114:397– 409.
 Kuo YC, Huang YL, Chen CC, Lin YS, Chuang KA, Tsai WJ.
Cell cycle progression and cytokine gene expression of human
peripheral blood mononuclear cells modulated by Agaricus
blazei. J Lab Clin Med 2002;140:176 – 87.
 Mizuno M, Morimoto M, Minato K, Tsuchida H. Polysacchar-
ides from Agaricus blazei stimulate lymphocyte T-cell subsets
in mice. Biosci Biotechnol Biochem 1998;62:434 – 7.
 Nakajima A, Ishida T, Koga M, Takeuchi T, Mazda O, Takeu-
chi M. Effect of hot water extract from Agaricus blazei Murill
on antibody-producing cells in mice. Int Immunopharmacol
2002;2:1205 – 11.
 Shimizu S, Kitada H, Yokota H, Yamakawa J, Murayama T,
Sugiyama K, et al. Activation of the alternative complement
pathway by Agaricus blazei Murill. Phytomedicine 2002;9:
536 – 45.
 Ito H, Shimura K, Itoh H, Kawade M. Antitumor effects of a
new polysaccharide–protein complex (ATOM) prepared from
Agaricus blazei (Iwade strain 101) bHimematsutakeQand its
mechanisms in tumor-bearing mice. Anticancer Res 1997;17:
277 – 84.
 Lee YL, Kim HJ, Lee MS, Kim JM, Han JS, Hong EK, et al.
Oral administration of Agaricus blazei (H1 strain) inhibited
tumor growth in a sarcoma 180 inoculation model. Exp Anim
2003;52:371 – 5.
 Oshiman K, Fujimiya Y, Ebina T, Suzuki I, Noji M. Orally
administered beta-1,6-d-polyglucose extracted from Agaricus
blazei results in tumor regression in tumor-bearing mice.
Planta Med 2002;68:610 – 4.
 Takimoto H, Wakita D, Kawaguchi K, Kumazawa Y. Potentia-
tion of cytotoxic activity in naive and tumor-bearing mice by
oral administration of hot-water extracts from Agaricus blazei
fruiting bodies. Biol Pharm Bull 2004;27:404– 6.
 Ebina T, Fujimiya Y. Antitumor effect of a peptide-glucan
preparation extracted from Agaricus blazei in a double-grafted
tumor system in mice. Biotherapy 1998;11:259– 65.
 Fujimiya Y, Suzuki Y, Katakura R, Ebina T. Tumor-specific
cytocidal and immunopotentiating effects of relatively low
molecular weight products derived from the basidiomycete,
Agaricus blazei Murill. Anticancer Res 1999;19:113 – 8.
 Kawagishi H, Inagaki R, Kanao T, Mizuno T, Shimura K, Ito
H, et al. Fractionation and antitumor activity of the water-
insoluble residue of Agaricus blazei fruiting bodies. Carbo-
hydr Res 1989;186:267 – 73.
 Fujimiya Y, Suzuki Y, Oshiman K, Kobori H, Moriguchi K,
Nakashima H, et al. Selective tumoricidal effect of soluble
proteoglucan extracted from the basidiomycete, Agaricus bla-
zei Murill, mediated via natural killer cell activation and
apoptosis. Cancer Immunol Immunother 1998;46:147 – 59.
 Takaku T, Kimura Y, Okuda H. Isolation of an antitumor
compound from Agaricus blazei Murill and its mechanism
of action. J Nutr 2001;131:1409 – 13.
 Kimura Y, Kido T, Takaku T, Sumiyoshi M, Baba K. Isolation
of an anti-angiogenic substance from Agaricus blazei Murill:
its antitumor and antimetastatic actions. Cancer Sci 2004;95:
758 – 64.
 Sorimachi K, Ikehara Y, Maezato G, Okubo A, Yamazaki S,
Akimoto K, et al. Inhibition by Agaricus blazei Murill frac-
tions of cytopathic effect induced by western equine encepha-
litis (WEE) virus on VERO cells in vitro. Biosci Biotechnol
Biochem 2001;65:1645 – 7.
 Chen L, Shao HJ, Su YB. Coimmunization of Agaricus blazei
Murill extract with hepatitis B virus core protein through DNA
vaccine enhances cellular and humoral immune responses. Int
Immunopharmacol 2004;4:403 – 9.
 Jennemann R, Bauer BL, Bertalanffy H, Selmer T, Wiegandt
H. Basidiolipids from Agaricus are novel immune adjuvants.
Immunobiology 1999;200:277 – 89.
 Taguchi T, Furue H, Kimura T, Kondo T, Hattori T, Ogawa N.
Clinical efficacy of lentinan on neoplastic diseases. Adv Exp
Med Biol 1983;166:181 – 7.
 Ohno N, Kurachi K, Yadomae T. Antitumor activity of a
highly branched (1–3)-beta-d-glucan, SSG, obtained from
Sclerotinia sclerotiorum IFO 9395. J Pharmacobiodyn
1987;10:478 – 86.
 Reynolds JA, Kastello MD, Harrington DG, Crabbs CL, Peters
CJ, Jemski JV, et al. Glucan-induced enhancement of host
resistance to selected infectious diseases. Infect Immun
1980;30:51 – 7.
 Hetland G, Ohno N, Aaberge IS, Lovik M. Protective effect
of beta-glucan against systemic Streptococcus pneumoniae
infection in mice. FEMS Immunol Med Microbiol 2000;27:
111 – 6.
 Itoh H, Ito H, Amano H, Noda H. Inhibitory action of a (1–
N6)-beta-d-glucan–protein complex (F III-2-b) isolated from
Agaricus blazei Murill (bhimematsutakeQ) on Meth A fibro-
sarcoma-bearing mice and its antitumor mechanism. Jpn J
Pharmacol 1994;66:265 – 71.
 Bjorkbacka H, Fitzgerald KA, Huet F, Li X, Gregory JA, Lee
MA, et al. The induction of macrophage gene expression by
LPS predominantly utilizes Myd88-independent signaling cas-
cades. Physiol Genomics 2004;19:319 – 30.
 Jiang H, Van De Ven C, Satwani P, Baxi LV, Cairo MS.
Differential gene expression patterns by oligonucleotide
microarray of basal versus lipopolysaccharide-activated mono-
cytes from cord blood versus adult peripheral blood. J Immu-
nol 2004;172:5870 – 9.
 Malcolm KC, Arndt PG, Manos EJ, Jones DA, Worthen GS.
Microarray analysis of lipopolysaccharide-treated human neu-
trophils. Am J Physiol, Lung Cell Mol Physiol 2003;284:
L663 – 70.
 Schmitz F, Mages J, Heit A, Lang R, Wagner H. Transcrip-
tional activation induced in macrophages by Toll-like receptor
(TLR) ligands: from expression profiling to a model of TLR
signaling. Eur J Immunol 2004;34:2863 – 73.
 Vetvicka V, Thornton BP, Ross GD. Soluble beta-glucan poly-
saccharide binding to the lectin site of neutrophil or natural
killer cell complement receptor type 3 (CD11b/CD18) gene-
rates a primed state of the receptor capable of mediating
L.K. Ellertsen et al. / International Immunopharmacology 6 (2006) 133–143142
cytotoxicity of iC3b-opsonized target cells. J Clin Invest
1996;98:50 – 61.
 Roeder A, Kirschning CJ, Rupec RA, Schaller M, Weindl G,
Korting HC. Toll-like receptors as key mediators in innate
antifungal immunity. Med Mycol 2004;42:485– 98.
 Sorimachi K, Akimoto K, Ikehara Y, Inafuku K, Okubo A,
Yamazaki S. Secretion of TNF-alpha, IL-8 and nitric oxide by
macrophages activated with Agaricus blazei Murill fractions
in vitro. Cell Struct Funct 2001;26:103 – 8.
 Saal LH, Troein C, Vallon-Christersson J, Gruvberger S, Borg
A, Peterson C. BioArray Software Environment (BASE): a
platform for comprehensive management and analysis of
microarray data. Genome Biol 2002;3 [SOFTWARE0003].
 See D, Mason S, Roshan R. Increased tumor necrosis factor
alpha (TNF-alpha) and natural killer cell (NK) function using
an integrative approach in late stage cancers. Immunol Invest
2002;31:137 – 53.
 Gantner BN, Simmons RM, Underhill DM. Dectin-1 mediates
macrophage recognition of Candida albicans yeast but not
filaments. EMBO J 2005.
 Parker LC, Whyte MK, Vogel SN, Dower SK, Sabroe I. Toll-
like receptor (TLR)2 and TLR4 agonists regulate CCR
expression in human monocytic cells. J Immunol 2004;172:
4977 – 86.
 Liew FY, Xu D, Brint EK, O’neill LA. Negative regulation of
Toll-like receptor-mediated immune responses. Nat Rev
Immunol 2005;5:446 – 58.
 Gonzalez E, Kulkarni H, Bolivar H, Mangano A, Sanchez R,
Catano G, et al. The influence of CCL3L1 gene-containing
segmental duplications on HIV-1/AIDS susceptibility. Science
2005;307:1434 – 40.
 Xiang J, George SL, Wunschmann S, Chang Q, Klinzman D,
Stapleton JT. Inhibition of HIV-1 replication by GB virus C
infection through increases in RANTES, MIP-1alpha, MIP-
1beta, and SDF-1. Lancet 2004;363:2040 – 6.
 Miller SI, Ernst RK, Bader MW. LPS, TLR4 and infectious
disease diversity. Nat Rev Microbiol 2005;3:36– 46.
 Gerard C, Gerard NP. C5A anaphylatoxin and its seven trans-
membrane-segment receptor. Annu Rev Immunol 1994;12:
775 – 808.
 Kehrl JH. G-protein-coupled receptor signaling, RGS pro-
teins, and lymphocyte function. Crit Rev Immunol 2004;24:
409 – 423.
 Mahalingam S, Karupiah G. Chemokines and chemokine
receptors in infectious diseases. Immunol Cell Biol 1999;77:
469 – 75.
 Vauquelin G, Van LI. G protein-coupled receptors: a count
of 1001 conformations. Fundam Clin Pharmacol 2005;19:
45 – 56.
 Romagnani S. The Th1/Th2 paradigm. Immunol Today 1997;
18:263 – 6.
 McGuirk P, Higgins SC, Mills KH. Regulatory cells and the
control of respiratory infection. Curr Allergy Asthma Rep
2005;5:51 – 5.
L.K. Ellertsen et al. / International Immunopharmacology 6 (2006) 133–143 143