Effect of an Extract Based on the Medicinal Mushroom Agaricus blazei Murill on Release of Cytokines, Chemokines and Leukocyte Growth Factors in Human Blood Ex Vivo and In Vivo

Effect of an Extract Based on the MedicinalMushroom Agaricus blazei Murill on Release ofCytokines, Chemokines and Leukocyte Growth Factorsin Human Blood Ex Vivo and In Vivo

E. Johnson*,�, D. T. Førland*, L. Sætre�, S. V. Bernardshaw*, T. Lyberg� & G. Hetland§

Introduction

The medicinal mushroom Agaricus blazei Murill (AbM) ofthe Basidiomycetes family, grows in the wild in the Pied-ade region outside of Sao Paulo, Brazil, and has for cen-turies been used there as a food ingredient. Because ofthe alleged lesser frequency of serious diseases in Piedadethan in the neighbouring regions, AbM has traditionallybeen employed to treat common serious diseases like ath-erosclerosis, hepatitis, hyperlipidaemia, diabetes mellitus,dermatitis and cancer [1]. In the mid-1960s the mush-room was taken to Japan and cultivated for the health

food market, and from the beginning of the 1980s sub-jected to an expanding experimental research effort. AbMis rich in b-glucans [2] with a b-1,6-backbone andb-1-3-side branches (ratio of 1:2), which have immuno-modulatory properties on monocytes ⁄ macrophages andnatural killer (NK) cells of the native immune system[3–7], and anti-tumour effects in rodents [8].

As to AbM itself, it has been shown to promotein vitro macrophage [9] secretion of interleukin (IL)-8,tumour necrosis factor (TNF)-a and nitric oxide, andhuman monocyte [10] synthesis of pro-inflammatorycytokines (IL-1b, IL-6, IL-8 and TNF-a) but not IL-12

*Department of Gastroenterological Surgery;

�Faculty of Medicine, University of Oslo;

�Center for Clinical Research; and

§Department of Immunology and Transfusion

Medicine, Ulleval University Hospital, Oslo,

Norway

Received 3 October 2008; Accepted in revisedform 14 November 2008

Correspondence to: E. Johnson, Department of

Gastroenterological Surgery, Ulleval University

Hospital, Kirkeveien 166, 0407 Oslo, Norway.

E-mail: [email protected]

Abstract

An immunostimulatory extract based on the medicinal mushroom Agaricusblazei Murill (AbM) has been shown to stimulate mononuclear phagocytesin vitro to produce pro-inflammatory cytokines, and to protect against lethalperitonitis in mice. The present aim was to study the effect of AbM on releaseof several cytokines in human whole blood both after stimulation ex vivo andin vivo after oral intake over several days in healthy volunteers. The 17 signalsubstances examined were; T helper 1 (Th1) cytokines [interleukin (IL)-2,interferon (IFN)-c and IL-12], T helper 2 cytokines (IL-4, IL-5 and IL-13),pleiotropic (IL-7, IL-17), pro-inflammatory [IL-1b, IL-6, tumour necrosis fac-tor (TNF)-a (mainly produced by Th1 cells)] – and anti-inflammatory (IL-10)cytokines, chemokines [IL-8, macrophage inhibitory protein (MIP)-1b andmonocyte chemoattractant protein (MCP)-1] and leukocyte growth factors[granulocyte colony-stimulating factor (G-CSF), granulocyte ⁄ macrophagecolony stimulating factor]. After stimulation of whole blood ex vivo with0.5–5.0% of a mushroom extract, AndoSan� mainly containing AbM, therewas a dose-dependent increase in all the cytokines studied, ranging from twoto 399-fold (TNF-a). However, in vivo in the eight volunteers who completedthe daily intake (60 ml) of this AbM extract for 12 days, a significant reduc-tion was observed in levels of IL-1b (97%), TNF-a (84%), IL-17 (50%) andIL-2 (46%). Although not significant, there was a trend towards reduced levelsfor IL-8, IFN-c and G-CSF, whilst those of the remaining nine cytokinestested, were unaltered. The discrepant results on cytokine release ex vivo andin vivo may partly be explained by the antioxidant activity of AbM in vivo andlimited absorption of its large, complex and bioactive b-glucans across theintestinal mucosa to the reticuloendothelial system and blood.

C L I N I C A L I M M U N O L O G Y doi: 10.1111/j.1365-3083.2008.02218.x..................................................................................................................................................................

� 2009 The Authors

242 Journal compilation � 2009 Blackwell Publishing Ltd. Scandinavian Journal of Immunology 69, 242–250

or the anti-inflammatory cytokine IL-10. Recently, it hasbeen shown that transcription factor NF-jB is activatedby binding of AbM to TLR-2 but not to TLR-4 [11],which facilitates increased production of pro-inflamma-tory cytokines. By demonstrating selective upregulationof genes for IL-1b and IL-8, but not for IL-10 and IL-12,gene expression microarray analysis of promonocyticTHP-1 cells (the THP-1 cell line is a monocytic leuke-mia cell line obtained from the American Type CultureCollection (ATCC TIB 202)) [12] supported previousresults with the AbM extract [10]. Moreover, by stimula-tion of whole blood with the AbM extract ex vivo [13], amirroring effect of expression of adhesion moleculesCD62L (L-selectin) (decrease) and CD11b (increase),which also is important for phagocytosis, was demon-strated both on monocytes and granulocytes.

As to in vivo effects, in mice given pneumococcal [14]

or polymicrobial fecal peritonitis [15], a single oral doseof the current aqueous AbM extract (AndoSan�, ACECo., Ltd, Gifu, Japan) prior to bacterial challenge, signifi-cantly reduced the lethality of the animals from septicae-mia as compared with the saline controls. In addition,when such unchallenged mice were given the AbMextract perorally, an increase in cytokines macrophageinhibitory protein (MIP)-2 (mouse IL-8 analogue) andTNF-a occurred, which probably contributed to increasedphagocytosis by mononuclear phagocytes, as demon-strated for uptake of Staphylococcus aureus and Mycobacte-rium tuberculosis [16, 17]. However, although it isprobable but unknown whether the protective effect ofAbM extract against infection, as demonstrated in mice,may also be valid in humans.

Our aim was to study the effect of this particular AbMextract on release of a wide range of cytokines, chemokinesand leukocyte growth factors in human whole blood, bothafter stimulation ex vivo, and in vivo after oral intake overdays by healthy volunteers. Among 17 analytes studied,some were T helper 1 (Th1)-specific- [IL-2, interferon(IFN)-c, IL-12], T helper 2 (Th2)-specific- (IL-4, IL-5 andIL-13), pleiotropic- (IL-7, IL-17), pro-inflammatory-(IL-1b, IL-6 and TNF-a) or anti-inflammatory (IL-10)cytokines, chemokines [IL-8, MIP-1b, monocyte chemo-attractant protein (MCP)-1], or growth factors [granulocytecolony-stimulating factor (G-CSF), granulocyte ⁄ macro-phage colony stimulating factor (GM-CSF)].

Materials and Methods

Reagents. The mushroom extract (AndoSan�) used in ourexperiments was obtained from ACE Co., Ltd. It was storedat 4 �C in dark bottles and used under sterile conditionsex vivo and kept sterile until taken by volunteers for in vivoexperiments. This mushroom extract is a commercialproduct and its exact content a business secret, part ofwhich has not been revealed until very recently. The AbM

mixed powder contains per 100 g the followingconstituents: moisture 5.8 g, protein 2.6 g, fat 0.3 g,carbohydrates 89.4 g of which b-glucan constitutes 2.8 g,and ash 1.9 g. The AndoSan� extract contains 82.4% ofBasidiomycetes mushroom derived from AbM (jap.: Him-ematsutake), 14.7% from Hericium erinaceum (Yamabushi-take) [18] and 2.9% from Grifola frondosa (Maitake) [19],and its final concentration was 340 g ⁄ l. The amount perlitre of the extract for sodium was 11 mg, phosphorus254 mg, calcium 35 mg, potassium 483 mg, magnesium99 mg and zinc 60 mg. In addition, experiments wereperformed with an extract based only on AbM powder at afinal concentration of 340 g ⁄ l as well, and this extract isdefined as pure AbM extract. Accordingly, it was possibleto compare the effects of AndoSan� versus pure AbM inthe experiments on cytokine release in whole blood ex vivo.The LPS content of AndoSan� and pure AbM were found,using the Limulus amebocyte lysate test (COAMATICChromo-LAL; Chromogenix, Falmouth, MA, USA) withdetection limit 0.005 EU ⁄ ml (1 EU = 0.1 ng ⁄ ml), to be aminiscule concentration of <0.5 pg ⁄ ml. The results fromtests for heavy metals were conformable with strictJapanese regulations for health foods. AndoSan� had beenheat-sterilized (124 �C for 1 h) by the producer. LPS wasfrom Escherichia coli (E. coli 026:B6) (Sigma Co., St. Louis,MO, USA).

Experimental design

Ex vivo experiments. Heparinized and blood was col-lected from six healthy volunteers (four men) with med-ian age 38.5 years (range: 37–61 years) who deniedsmoking and medication. This blood was immediatelyincubated with 0.1%, 0.5%, 2.0%, 5.0%, 10.0% or15.0% of sterile AndoSan� at 37 �C for 6 h. Plasmawas then harvested by centrifugation at 2000 g for15 min and frozen at )70 �C until analysed for cyto-kines, chemokines and growth factors.

In vivo experiments. Five healthy men and five healthywomen, of median age 30.5 years (range: 26–51 years),volunteered to participate in the study of oral intake oflow dose AndoSan�; 20 ml thrice daily for 12 days. Thedose of a total of 60 ml of the mushroom extract per daywas chosen as this was the dose recommended by themanufacturer and regularly used for AndoSan� as ahealth product. The time interval between each doseshould be from 6 to 10 h. Participants were asked toavoid mushroom-containing foods for 3 days prior to andduring the experimental period. Three women used con-traceptive drugs, one of whom also 100 mg · 1 of ironon days 0, 7, 8 and 9, and another woman ate a pizzawith some mushroom on day 1 and used medicationagainst migraine (rizatriptan, rapitab, ibuprofen, paraceta-mol and phenazone-caffeine) on days 0, 5, 6 and 7. Onefemale participant stopped the intake of AndoSan� due

E. Johnson et al. AbM and Cytokine Release Ex Vivo and In Vivo 243..................................................................................................................................................................


Journal compilation � 2009 Blackwell Publishing Ltd. Scandinavian Journal of Immunology 69, 242–250

to reoccurrence of localized labial eruption, probably herknown herpes simplex labialis. One male participant wasfor practical reasons unable to give the last blood sample.Accordingly, two of the 10 participants were notincluded in the complete analysis of the results. None ofthe participants reported trouble with intake of Ando-San� due to taste or volume.

Ethylenediaminetetraacetate (EDTA) and heparininizedvenous blood (total 15 ml) was collected by puncture of anantecubital vein on day 0 immediately prior to intake of themushroom extract and on days 1, 2, 5, 8 and 12. Heparin-ized plasma was harvested and frozen at )70 �C untilanalysed for cytokines, chemokines and growth factors atday 0 prior to intake of AndoSan� and at day 2 and 12subsequent to intake of the extract. The heparinized bloodharvested from the participants was in one set ofexperiments also stimulated ex vivo with LPS (1 ng ⁄ ml) for6 h at 37 �C before storing plasma samples at )70 �C untilanalysed. The EDTA blood samples were each time analysedfor haemoglobin, haematocrite, mean cellular volume, meancellular haemoglobin, reticulocytes, immature reticulo-cytes, leukocytes including a differential count of neutroph-ils, basophils, eosinophils, lymphocytes and monocytesas well as C-reactive protein (CRP), thrombocytes, urea,creatinine, bilirubin, aspartate aminotransferase, alanineaminotransferase, lactate dehydrogenase, c-glutaminetransferase, alkaline phosphatase and pancreatic amylase.

Same type of in vivo experiment (except from incuba-tion with LPS ex vivo) was performed in 15 healthypersons (eight women) of median age 36 years (range:23–55 years), the dose of AndoSan� was high (120 mlthrice daily) and the experiment limited to 2 days. Onewoman used daily a 20 mg tablet of esomeprazole asanti-secretory medication. Six of these 15 persons alsoparticipated in the experiment using low-dose Ando-San�. By giving 720 ml of the extract to individuals ofweight 60–80 kg during 2 days, the amount of Ando-San� taken was comparable to that given orogastrically(0.2 ml AndoSan�) to mice [15] (18–20 g) prior toexperimentally infecting the animals with fecal bacteriafor lethal peritonitis, which was significantly reduced byAndoSan� pretreatment. Thus, the notion was that useof comparable amounts of the mixed mushroom extractrelative to body weight (9–12 and 10–11 ml ⁄ kg in manand mouse respectively), would be a rational strategy forobserving beneficial or possible side (toxic) effects inhumans. All the participants in the in vivo and ex vivoexperiments denied regular smoking or intake of medica-tion, unless otherwise stated.

Multiplex cytokine assay. We used the multiplexbead-based sandwich immunoassay technology (Luminex,Austin, TX, USA) and a human cytokine 17-plex kit(Bio-Rad laboratories, Hercules, TX, USA), strictlyfollowing the manufacturer’s instructions, to measure theconcentrations in individual heparinized plasma samples

of the following cytokines, chemokines and growth fac-tors [lower detection limits (in pg ⁄ ml) in parentheses];IL-1b (2.0), IL-2 (1.2), IL-4 (0.3), IL-5 (2.3), IL-6 (2.1),IL-7 (3.0), IL-8 (1.6), IL-10 (1.8), IL-12 (3.0), IL-13(0.9), IL-17 (2.5), G-CSF (1.9), GM-CSF (0.8), IFN-c(2.0), MCP-1 (1.7), MIP-1b (2.0) and TNF-a (5.4).

Statistics. All measurements were based on duplicatesfrom six donors when blood was sampled for the ex vivoexperiments. For the in vivo experiments blood wassampled prior to and twice after intake of AndoSan�after 2 and 12 days, or 1 and 2 days, in low- and high-dose experiments respectively. Data are presented asmean ± SEM. Differences in cytokine levels in bloodex vivo prior to and after stimulation with different con-centrations of the AbM extract were after passage ofnormality test, assessed with parametric ANOVA forpaired data with Dunn’s multiple comparisons testusing the Instat for WindowsTM statistics softwarepackage (Graphpad Software, San Diego, CA, USA).Non-parametric repeated measures ANOVA (Friedmantest) with Dunn’s multiple comparisons test was usedwhen measuring the effect prior to and after AndoSan�intake in volunteers on expression in vivo of cytokinesin whole blood. P-values of or below 0.05 were consid-ered statistically significant.

Ethics. The study was approved by the regional ethicscommittee and followed the guidelines of the Helsinkideclaration. The participants were informed also in writ-ten form and signed an agreement of consent for partici-pation in the study.

Results

Effect of AbM on cytokine release in whole blood ex vivo

After stimulation of whole blood with 0.1%, 0.5%,2.0% and 5.0% of AndoSan�, there were an increase inplasma levels of all cytokines studied, ranging from twoto 399-fold (Fig. 1A–D), and a maximal plateau atAndoSan� concentrations of 2% and 5%. The increasewas most pronounced for pro-inflammatory-cytokinesTNF-a (399-fold), IL-6 (271-fold) (Fig. 1A) and IL-1b(86-fold) (Fig. 1B), and the anti-inflammatory cytokineIL-10 (129-fold) (Fig. 1C). Chemokines IL-8, MIP-1band MCP-1 were the cytokines with the highest concen-tration in untreated blood, and Th2 cytokines IL-4, IL-5and IL-13 the ones with the lowest initial values.

Use of pure AbM extract in the same concentrationsas above, gave similar results (data not shown) as themixed AndoSan� extract, which contained 82% ofAbM, 15% of H. erinaceum and 3% of G. frondosa. Thissupports the notion that AbM is the main biologicallyactive component of AndoSan� and responsible for gen-erating the effects on release of cytokines, chemokinesand leukocyte growth factors.

244 AbM and Cytokine Release Ex Vivo and In Vivo E. Johnson et al...................................................................................................................................................................



Effect of AbM in vivo on general blood parameters

We obtained complete data from eight of the 10 volun-teers who ingested AndoSan� in doses of 20 ml thrice aday for 12 days. The purpose was to examine prospectivedeleterious effects of AbM on parameters in blood. Bloodsamples testing haematological, kidney, liver and pancre-atic function were obtained prior to (day 0) and afterintake of AndoSan� (days 1, 2, 5, 8 and 12). There wereno significant differences whatsoever at any time-pointfrom normal range values in the following parameterstested; haemoglobin, haematocrite, mean cellular volume,mean cellular haemoglobin, reticulocytes, immature reti-culocytes, differential count of leukocytes, CRP, throm-bocytes, urea, creatinine, bilirubin, aspartateaminotransferase, alanine aminotransferase, lactate dehy-

drogenase, c-glutamine transferase, alkaline phosphataseand pancreatic amylase (data not shown).

Effect of AbM in vivo on cytokine release in whole blood

In the eight volunteers who completed the intake ofAndoSan� for 12 days, a significant reduction occurredfrom day 0 to day 12 in five of the 16 cytokines studied(Fig. 2). This reduction was most pronounced for pro-inflammatory cytokines IL-1b (97%) and TNF-a (84%),and least for IL-17 (50%) and IL-2 (46%). Although notsignificant, there was a trend from day 0 to 12 towards areduction in the levels (pg ⁄ ml) of IL-8 (935 ± 163 versus614 ± 140), IFN-c (8.7 ± 8.7 versus 0) and G-CSF(11 ± 3 versus 5 ± 4). The levels for four of the remain-ing eight cytokines at day 0 and day 2 were 331 ± 46

C

A

D

B

Figure 1 (A–D) Levels (pg ⁄ ml) of specified cytokines in whole blood ex vivo after incubation without (first left bar graph) or with different concentra-

tions of the AbM extract (AndoSan�) at 0.1%, 0.5%, 2% and 5% (second to fifth bar graph from the left respectively). The results are from six

donors and given as mean ± SEM. The P-values compare with the signals in unstimulated blood control. *P < 0.05, **P < 0.01, ***P < 0.001, xP-va-

lue not determined because of too few data [MCP-1 (n = 4), MIP-1b (n = 5)].




and 234 ± 40 for MCP-1, 29 ± 13 and 40 ± 13 forGM-CSF, 14 ± 5 and 8 ± 11 for IL-12 and 6.0 ± 0.8and 4.2 ± 0.9 for IL-10, whilst IL-4, IL-5, IL-7 and IL-13 were detected at negligible levels of 0.1–1.6 pg ⁄ ml.Thus, in vivo intake of the mixed mushroom extract by

healthy volunteers resulted in either decreased or mainlyunaltered cytokine blood levels (Fig. 2). This is quite dif-ferent from the markedly increased signal levels detectedby direct stimulation with AndoSan� in whole bloodex vivo (Fig. 1).

In parallel experiments, cytokine levels were also anal-ysed after enhancement by stimulation ex vivo with LPS(1 ng ⁄ ml) for 6 h in sampled blood from the AndoSan�consuming volunteers. Compared with day 0 prior tointake of AbM, there was at day 12 a significant andapproximately 50% reduction in levels of IL-1b, IFN-c(Fig. 3A), IL-4 and IL-12, and a less pronounced reduc-tion in G-CSF (32%) (Fig. 3B). Moreover, from day 0 today 12 after intake of AndoSan� there was a similar butnon-significant trend towards reduced levels (pg ⁄ ml) ofTNF-a (27,866 ± 10,052 versus 13,111 ± 22) and IL-2(49 ± 18 versus 22 ± 15). Except from a significant andslight increase of IL-6 from day 0 (42,215 ± 8041) today 2 (55,585 ± 7331) which returned to baseline levelat day 12, the levels of the remaining eight cytokineswere unaltered at days 0, 2 and 12. Their initial concen-trations (pg ⁄ ml) were 17,339 ± 1530 for IL-8,5414 ± 810 for MCP-1, 569 ± 74 for GM-CSF,256 ± 57 for IL-10 and 173 ± 22 for IL-17, whilst IL-5,IL-7 and IL-13 were detected at negligible levels from0.4 to 3.7 pg ⁄ ml. Accordingly, LPS-stimulation ex vivoof blood from healthy volunteers given AndoSan� for12 days, also confirmed a profile of reduced or unalteredcytokine levels.

We then studied the dose–response effect on cytokinerelease in vivo by increasing the intake of AbM sixfold to120 ml thrice daily, but limiting the time period to2 days only. This was tested in 15 healthy volunteers,and there were no alterations in the general haematologi-cal parameters during the experiment (data not shown).

Figure 2 Significantly reduced levels (pg ⁄ ml) of five cytokines in whole

blood in vivo after intake of AbM (AndoSan�) at doses of 20 ml thrice

per day for 12 days. Days 0, 2 and 12 are depicted by the first, second

and third bar graph from the left respectively. The results are from

eight donors and given as mean ± SEM. The P-values compare with the

cytokine levels at day 0 prior to intake of AndoSan�. wP = 0.05,

*P < 0.05, **P < 0.01 respectively. The chemokine MIP-1 was not

measured.

A B

Figure 3 (A and B) Significantly reduced levels (pg ⁄ ml) of five cytokines in whole blood after (days 2 and 12) intake of AbM (AndoSan�) at doses

of 20 ml thrice per day for 12 days and after stimulation ex vivo with LPS (1 ng ⁄ ml). Days 0, 2 and 12 are depicted by the first, second and third

bar graph from the left respectively. The results are from eight donors and given as mean ± SEM. The P-values compare with the cytokine levels at

day 0 prior to intake of AndoSan�. *P < 0.05, **P < 0.01. The chemokine MIP-1b was not measured.




Except for a significant increase in IL-8 (pg ⁄ ml) fromday 0 (405 ± 91) to day 1 (569 ± 109) and then adecline to baseline values at day 2, there were no signifi-cant alterations in the levels of the remaining 16 analytes(data not shown for 12 of these analytes). However, fortwo chemokines (MCP-1 and MIP-1b) and two cytokines(IL-6, and IFN-c) a falling trend was evident (Fig. 4),but statistically non-significant partly due to great varia-tion in their baseline concentrations. When focusing onthe seven volunteers in the group with the highest base-line cytokine levels, significant decrease from day 0 today 2 was revealed for these cytokines (Fig. 5).

Discussion

In this study, we demonstrate that when the extractAndoSan�, based on Basidiomycetes mushrooms andmainly AbM, was incubated with whole blood ex vivo, itinduced release of different types of cytokines includingTh1, Th2, pro-inflammatory and anti-inflammatory, aswell as chemokines and leukocyte growth factors. On theother hand, in vivo release of these immunological sub-stances after oral intake of AndoSan� for several days byhealthy volunteers, exhibited a different pattern; the lev-els were mainly reduced or stabilized, and as a rule notincreased, which points to a net anti-inflammatory effect.

Not surprisingly, the most abundant cytokines afterAbM stimulation ex vivo were the pro-inflammatory ones,

IL-6, IL-8 and TNF-a, which have also been demon-strated after AbM stimulation of human monocytes andmacrophages in vitro [9, 10]. The results on AbM stimu-lation of whole blood on signal substance release hereinreported, are also supported by an in vitro study [12]:A human pro-monocytic cell line (THP-1) was stimu-lated with �10% of the AbM extract AndoSan�, whichresulted in upregulation of genes (mRNA) for IL-1a and-b and IL-8, as well as a unique increase in mRNA forCXL1-3. Interestingly, there was also a considerable rela-tive increase (129-fold) of IL-10 but also of IL-4 (16-fold), which points to a potent anti-inflammatory effect[20] exerted by AbM. Whereas IL-10 is known to inhibitTNF-a, and IL-6 release from mononuclear cells [21],IL-4 may act anti-inflammatory by additionally inhibit-ing synthesis of IL-1b and IL-8 [22]. We conclude thatthe cytokines released in abundant amounts in plasma(Fig. 1) were mainly derived from monocytic cells. Webelieve this holds true even though endothelial cells,which outnumber monocytes in the body, also have beenshown to synthesize cytokines [10]. The contribution ofGM-CSF [23], TNF-a [24] and IL-17 [25] were largelyfrom T-lymphocytes, Th1 cells and Th17 cells respec-tively. On the other hand, the level of IL-7 which is syn-thesized by stromal cells [26] was largely unaltered. Thepure AbM extract revealed similar ex vivo effects as themixed mushroom extract. This implied that AbM (Him-ematsutake) per se and not H. erinaceum (Yamabushitake)

A B

Figure 4 (A and B) Non-significant reduction in levels (pg ⁄ ml) of four cytokines in whole blood in vivo from before (day 0) compared with after

intake of AbM (AndoSan�) at doses of 120 ml thrice per day for 2 days. Days 0, 1 and 2 are depicted by the first, second and third bar graph from

the left respectively. The results are from 15 donors and given as mean ± SEM.




(relative amount �15%) and G. frondosa (Maitake) (rela-tive amount �3%) is the crucial biological mediator ofthe mixed mushroom extract, AndoSan�, which also wasused for the in vivo experiments. However, synergisticeffects are probable.

As AbM is rich in biologic response modulators suchas proteoglucans [27, 28] and b-glucans [8], the stimula-tory effect is proposedly mediated mainly via patternrecognizing receptors like the lectin-binding site forb-glucan in complement receptor 3 (CR3) (CD11b ⁄ 18)[29, 30], toll-like receptor 2 [11, 31] and the dectin-1receptor [32, 33] on these cells. As LPS was not detectedin the mixed mushroom extract (detection limit0.5 pg ⁄ ml), possible contamination by LPS must be neg-ligible and beyond any impact on the results, especiallyas 200 pg ⁄ ml was the lowest concentration needed for amodest cytokine release in monocytes [34].

Intake of AbM as AndoSan� in a standardized dailydose of 60 ml [35] for 12 days or a high daily dose of360 ml for each of 2 days, revealed no pathological effectswhatsoever on haematological parameters including thosefor liver, pancreatic and renal function. The high dose of

AndoSan� was chosen because it is equivalent per kgbody weight to that given to mice in the bacterial sepsisstudies [14, 15]. One volunteer was excluded due to re-occurrence of herpes labialis when using the low dose ofthe AndoSan�, but other potentially adverse effects werenot seen. There are conflicting reports regarding the effectof AbM on liver function. Whereas use of AbM for up toseveral months has been suggested as possible cause forsevere hepatic dysfunction in three cancer patients [36],intake of AbM normalized liver function in four patientswith chronic hepatitis B virus infection [37]. Moreover,our study on AbM intake in patients with chronic hepati-tis C virus infection [38], did not reveal any negativeeffects on liver function, rather there was a little, but sta-tistically not significant reduction in HCV load in serum.Also, the current study in healthy volunteers, showed nochange in liver function. From this we must concludethat AndoSan� is not hepatotoxic.

After intake of the low dose of AbM (AndoSan�60 ml per day) there actually was among the analytestested a significant reduction after 12 days in levels ofthe three most abundant analytes on day 0, i. e. the pro-inflammatory cytokines IL-1b, IL-6 and TNF-a, as wellas for IL-2 and the pleiotropic IL-17, which also mediateshost defensive mechanisms to especially extracellular bac-terial infections [25]. The stimulation ex vivo with LPS ofwhole blood sampled from these volunteers who hadtaken AndoSan� extract (Fig. 3), substantiated theresults by demonstrating a similar pattern for IL-1b,IFN-c, G-CSF and IL-4, which increase antibody synthe-sis and cytotoxic activity of lymphocytes [39]. Moreover,there was also a reduction of the major Th1 cytokine, IL-12, which stimulates T and NK cells, regulates cell-med-iated immunity and induces production of IFN. As bothanti-infection and anti-tumour effects, which are pro-moted by AbM, are due to Th1 responses, one wouldhave anticipated increased levels of cytokines IL-2, IL-12and INFc, as well as of IL-17, after AbM intake in vivo.However, although we might have kinetically missedpossible higher levels of these cytokines, most probablythe Th1 responses induced by AbM have been promotedotherwise (ADCC by T cells or direct cell destruction bycytotoxic T cells and NK cells or primed granulocytes[2] in the case of tumour). During infection there proba-bly also was direct and possibly AbM-primed [2] bacte-riophagocytosis by granulocytes and possiblyopsonization of bacteria by complement C3b and deriva-tives thereof, facilitated by activation of the alternativecomplement pathway by AbM [40]. Nevertheless, thisapparent discrepancy between anti-tumour and anti-infec-tion properties of AbM extract and its current anti-inflammatory effect, is intriguing. The only cytokineswith a temporary significant increase were IL-6 from day0 to 2 (low dose AndoSan�) and IL-8 from day 0 to 1(high dose). Generally, although not significant, the

Figure 5 Significantly reduced levels (pg ⁄ ml) of four cytokines in

whole blood in vivo before and after intake of AbM (AndoSan�) at

doses of 120 ml thrice per day for 2 days. Days 0, 1 and 2 are depicted

by the first, second and third bar graph from the left respectively. The

results are from seven donors with the highest initial cytokine levels

prior to intake of AbM and given as mean ± SEM. *P < 0.05,

**P < 0.01. The P-values compare with the cytokine levels at day 0

prior to intake of AbM.




tendency was reduced concentrations of the remainingcytokines, chemokines and growth factors 12 days afterAndoSan� intake by the volunteers (Figs 2 and 3).When the high dose of AbM extract (AndoSan�) wasused for only 2 days, the cytokine levels were generallyunaltered. However, examination of the seven volunteerswith the highest initial analyte levels, revealed a signifi-cant decrease in four (IL-6, IFN-c, MCP-1 and MIP-1b)(Fig. 5) of the 17 mediators. These results support thenotion that high initial cytokine levels, as seen duringinflammation, are more prone to a decline after intake ofAbM. Two important conclusions can be made concern-ing the intake of AbM; the low dose was sufficient but a2 days time span too short for obtaining significantresults with this dose.

The discrepant results with AbM on immunologicalsignal substance release ex vivo and in vivo are alsointriguing and demand careful consideration because theabsorption of AbM from the intestinal mucosa is not wellstudied. Presumably, there is a limited absorption oflarge and bioactive b-glucan fragments across theintestinal mucosa to the reticuloendothelial system andblood, which precludes their potential stimulatory effecton cytokine release in vivo as demonstrated ex vivo andin vitro [10]. Regarding uptake of saccharides it is com-monly believed that hardly anything larger than mono-saccharides are absorbed from the human gut. However,in murine models uptake of b-1,3-glucans by gastrointes-tinal macrophages and shuttling to reticuloendothelialsystem and bone marrow has been demonstrated(reviewed by Firenzuoli [2]). Concerning anti-tumouractivity [2, 41] high-molecular-weight polysaccharidefractions (100–200 kDa) were most active in vivo,whereas small molecular fractions (0.5–10 kDa) showedno activity. Increased uptake of AbM from the gut inmice and increased uptake and antitumour effect of themushroom has been shown after encapsulation of AbMwith its b-glucans in marine phospholipids [42]. How-ever, b-glucans could also have had an indirect effectlocally in the intestines by stimulating immune cells inPeyer’s patches. In addition, low molecular antioxidantsubstances [43], which may easily be absorbed, could con-tribute to the stabilizing or reducing effect on cytokinerelease in vivo (Figs 2 and 5). This assumption is sup-ported by the finding that inhibitors of reactive oxygenspecies (ROS) (such as N-acetylcysteine) block IL-1b pro-duction in macrophages [44]. In fact, very little ROS wasproduced in blood ex vivo by granulocytes, and not at allby monocytes [13] upon stimulation with AbM. If thisalso was true for the in vivo situation, it would contributeto the explanation for the reduced IL-1b levels in indi-viduals taking AbM. In addition, the differential effecton gene expression of pro-inflammatory cytokines byAbM stimulation in vitro (stimulatory) on a pro-mono-cytic cell line [12] and in vivo (no effect) after intake for

1 week in patients with hepatitis C [38], also supportsthe present results. However, it remains to be elucidatedwhether effects of AbM observed ex vivo on expression ofL-selectin and b2-integrins [13] necessary for phagocyosis,rolling and adhesion to endothelial cells, also will trans-late into an anti-inflammatory effect in vivo.

In conclusion, AbM induced a general stimulatoryeffect on the release of immunological signal substancesex vivo. In vivo, however, there is a reduced or unchangedlevels of these substances, which points to a stabilizingand an anti-inflammatory effect of AbM when given theoral route.

Acknowledgment

This work was supported by grants from the ResearchCouncil at the Oncology and Surgical Division, UllevalUniversity Hospital and the Faculty of Medicine, Univer-sity of Oslo.

References

1 Wasser SP, Weis AL. Medicinal properties of substances occurring

in higher Basidiomycetes mushrooms. Current perspectives

(Review). Int J Med Mush 1999;1:31–62.

2 Firenzuoli F, Gori L, Lombardo G. The medicinal mushroom Agari-

cus blazei Murill: review of literature and pharmaco-toxicological

problems. Evid Based Complement Alternat Med 2008;5:3–15.

3 Fujimiya Y, Suzuki Y, Oshiman K-I et al. Selective tumoricidal

effect of soluble proteoglucan extracted from the basidiomycete,

Agaricus blazei Murill, mediated via natural killer cell activation and

apoptosis. Cancer Immunol Immunother 1998;46:147–59.

4 Riggi SJ, Di Luzio NR. Identification of a reticuloendothelial stim-

ulating agent in zymosan. Am J Physiol 1961;200:297–300.

5 Bøgwald J, Gouda I, Hoffmann J, Larm O, Larsson R, Seljelid R.

Stimulatory effect of immobilized glucans on macrophages in vitro.

Scand J Immunol 1984;20:355–60.

6 Hetland G, Sandven P. b-1,3-glucan reduces growth of Mycobacte-

rium tuberculosis in macrophage cultures. FEMS Immunol Med Micro-biol 2002;33:41–5.

7 Hetland G, Johnson E, Lyberg T, Bernardshaw S, Tryggestad

AMA, Grinde B. Effects of the medicinal mushroom Agaricus blazeiMurill on immunity, infection and cancer. Review. Scand J Immunol

2008;68:363–70.

8 Ohno N, Furukawa M, Miura NN, Adachi Y, Motoi M, Yadomae

T. Antitumor b-glucan from the cultured fruit body of Agaricusblazei. Biol Pharm Bull 2001;24:820–8.

9 Sorimachi K, Akimoto K, Ikehara Y, Inafuku K, Okubo A, Yama-

zaki S. Secretion of TNF-a, IL-8 and nitrite oxide by macrophages

activated with Agaricus blazei Murill fractions in vitro. Cell StructFunct 2001;26:103–8.

10 Bernardshaw S, Hetland G, Ellertsen LK, Aaland Tryggestad AM,

Johnson E. An extract of the medicinal mushroom Agaricus blazei

Murill differentially stimulates production of pro-inflammatory

cytokines in human monocytes and human vein endothelial cells in

vitro. Inflammation 2005;29:147–53.

11 Tryggestad AMA, Espevik T, Førland DT, Ryan L, Hetland G. TheMedical Mushroom Agaricus Blazei Murill Activates NF-jB via

TLR2. 13th International Congress of Immunology, Rio de Janeiro,

2007: P2.23 INI-02 signalling pathways of innate immune recep-

tors; P1193.




12 Ellertsen LK, Hetland G, Johnson E, Grinde B. Effect of a medici-

nal extract from Agaricus blazei Murill on gene expression in a

human monocytic cell line as examined by microassays and immu-

noassays. Int Immunopharmacol 2006;6:133–43.

13 Bernardshaw SV, Lyberg T, Hetland G, Johnson E. The effect of an

extract of the mushroom Agaricus blazei Murill on expression of

adhesion molecules and production of reactive oxygen species in

monocytes and granulocytes in whole blood ex vivo. APMIS

2007;115:719–25.

14 Bernardshaw S, Johnson E, Hetland G. An extract of the mushroom

Agaricus blazei Murill administered orally protected against systemic

Streptococcus pneumonia infection in mice. Scand J Immunol 2005;

62:393–8.

15 Bernardshaw S, Hetland G, Grinde B, Johnson E. An extract of the

mushroom Agaricus blazei Murill protects against lethal septicaemia

in a mouse model for fecal peritonitis. Shock 2006;25:420–5.

16 Rainard P, Riollet C, Poutrel B, Paape MJ. Phagocytosis and kill-

ing of Staphylococcus aureus by bovine neutrophils after priming by

tumor necrosis factor-alpha and the des-arginine derivative of C5a.

Am J Vet Res 2000;61:951–9.

17 Kisich KO, Higgins M, Diamond G, Heifets L. Tumor necrosis fac-

tor alpha stimulates killing of Mycobacterium tuberculosis by human

neutrophils. Infect Immun 2002;70:4591–9.

18 Lee EW, Shizuki K, Hosokawa S, Yadomae T. Two novel diterpe-

noids, erinaces H and I from the mycelia of Hericium erinaceum. Bio-sci Biotechnol Biochem 2000;64:2402–5.

19 Adachi Y, Okazaki M, Ohno N, Yadomae T. Enhancement of cyto-

kine production by macrophages stimulated with (1 to >3)-beta-D-

glucan, grifolan (GRN), isolated from Grifola frondosa. Biol PharmBul 1994;17:1553–60.

20 Burger D, Dayer JM. IL-1 and TNFa production in contact-medi-

ated activation of monocytes by T lymphocytes. Ann NY Acad Sci2002;966:464–73.

21 Trindade MC, Lind M, Nakashima Y et al. Interleukin-10 inhibits

polymethylmethacrylate particle induced interleukin-6 and tumor

necrosis factor-alpha release by human monocyte macrophages in vi-

tro. Biomaterials 2001;22:2067–73.

22 Ghoreschi K, Rocken M. Immune deviation strategies in the ther-

apy of psoriasis. Curr Drug Targets Inflamm Allergy 2004;3:193–8.

23 Kothari SS, Abrahamsen MS, Cole T, Hammond WP. Expression of

granulocyte colony stimulating factor (G-CSF) and granulo-

cyte ⁄ macrophage colony stimulating factor (GM-CSF) mRNA upon

stimulation with phorbol ester. Blood Cell Mol Dis 1995;21:192–

200.

24 Hou L, El-Omar EM, Chen J et al. Polymorphisms in Th1-type cell

mediated response genes and risk of gastric cancer. Carcinogenesis

2007;28:118–23.

25 Ouyang W, Kolls JK, Zheng Y. The biological functions of T

helper 17 cell effector cytokines in inflammation. Immunity 2008;

28:454–67.

26 Link A, Vogt TK, Favre S et al. Fibroblastic reticular cells in lymph

nodes regulate the homeostasis of naıve T cells. Nat Immunol

2007;8:1255–65.

27 Kawagishi H, Inagaki R, Kanao T et al. Fractionation of antitumor

activity of the water-soluble residue of Agaricus blazei fruiting

bodies. Carbohydr Res 1989;186:267–73.

28 Itoh H, Ito H, Amano H, Noda H. Inhibitory action of a (1 fi 6)-

beta-D-glucan- protein complex (F III-2-b) isolated from Agaricus

blazei. Jpn J Pharmacol 1994;66:265–71.

29 Czop JK, Valiante NM, Janusz MJ. Phagocytosis of particulate activa-

tors of the human alternative complement pathway through monocyte

beta-glucan receptors. Prog Clin Biol Res 1989;297:287–96.

30 Vetvicka V, Thornton BP, Ross GD. Soluble beta-glucan polysac-

charide binding to the lectin site of neutrophil of natural killer cell

complement receptor type 3 (CD11b ⁄ 18) generates a primed state

of the receptor capable of mediating cytotoxicity of iC3b-opsonized

target cells. J Clin Invest 1996;98:50–61.

31 Roeder A, Kirschning CJ, Rupec RA, Schaller M, Weindl G,

Koerting HC. Toll-like receptors as key mediators in innate anti-

fungal immunity. Med Mycol 2004;42:485–98.

32 Brown GD, Taylor PR, Reid DM et al. Dectin-1 is a major

b-glucan receptor on macrophages. J Exp Med 2002;196:407–12.

33 Ganthner BN, Simmons RM, Canavera SJ, Akira S, Underhill DM.

Collaborative induction of anti-inflammatory responses by dectin-1

and Toll-like receptor 2. J Exp Med 2003;197:1107–17.

34 Cavaillon JM. Cytokines and macrophages. Biomed Pharmacother

1994;48:445–53.

35 Takaku T, Kimura Y, Okuda H. Isolation of an antitumor com-

pound from Agaricus blazei Murill and its mechanism of action.

J Nutr 2001;131:1409–13.

36 Mukai H, Watanabe T, Ando M, Katsumata N. An alternative

medicine, Agaricus blazei, may have induced severe hepatic dysfunc-

tion in cancer patients. Jpn J Clin Oncol 2006;36:808–10.

37 Hsu C-H, Hwang K-C, Chang Y-H, Chou P. The mushroom Agar-

icus blazei Murill extract normalizes liver function in patients with

chronic hepatitis B. J Altern Compl Med 2008;14:299–301.

38 Grinde B, Hetland G, Johnson E. Effects on gene expression and

viral load of a medicinal extract from Agaricus blasie in patients with

chronic hepatitis C infection. Int Immunopharmacol 2006;6:1311–4.

39 Rankin JA. Biological mediators of acute inflammation. AACN Clin

Issues 2004;15:3–17.

40 Shimizu S, Kitada H, Yokota H et al. Activation of the alternative

complement pathway by Agaricus blazei Murill. Phytomedicine2002;9:536–45.

41 Whistler RL, Bushaway AA, Singh PP, Nakahara W, Tokuzen R.

Noncytotoxic, antitumor polysaccharides. Adv Carbohydr Chem Bio-chem 1976;32:235–75.

42 Murakawa K, Fukunaga K, Tanouchi M, Hosokawa M, Hossain Z,

Takahashi K. Therapy of myeloma in vivo using marine phospholip-

ids in combination with Agaricus blazei Murill as an immune

respond activator. J Oleo Sci ossain ZHH 2007;56:179–88.

43 Izawa S, Inoue Y. A screening system for antioxidants using thiore-

doxin-deficient yeast: discovery of thermostable antioxidant activity

from Agaricus blazei Murill. Appl Microbiol Biotechnol 2004;64:537–42.

44 Neill LAJ. How frustration leads to inflammation. Science

2008;320:619–20.




Documents

Effect of an Extract Based on the Medicinal Mushroom Agaricus blazei Murill on Release of Cytokines, Chemokines and Leukocyte Growth Factors in Human Blood Ex Vivo and In Vivo