Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2017 Oct 12.
Published in final edited form as: Food Funct. 2016 Oct 12;7(10):4213–4221. doi: 10.1039/c6fo00562d

Bacterial components are the major contributors to the macrophage stimulating activity exhibited by extracts of common edible mushrooms

Heather L Tyler a,b,1, Mona H Haron c, Nirmal D Pugh c, Jin Zhang c, Colin R Jackson b, David S Pasco c,d,*
PMCID: PMC5455775  NIHMSID: NIHMS860154  PMID: 27711863

Abstract

Recent studies have indicated that a major contributor to the innate immune enhancing properties of some medicinal plants is derived from the cell wall components of bacteria colonizing these plants. The purpose of the current study was to assess if the bacteria present within edible and medicinal mushrooms substantially contribute to the innate immune stimulating potential of these mushrooms. Whole mushrooms from thirteen types of edible fungi and individual parts from Agaricus bisporus were analyzed for in vitro macrophage activation as well as bacterial lipopolysaccharides (LPS) content, cell load, and community composition. Substantial variation between samples was observed in macrophage activation (over 500-fold), total bacterial load (over 200-fold), and LPS content (over 10 million-fold). Both LPS content (ρ=0.832, p <0.0001) and total bacterial load (ρ=0.701, p <0.0001) correlated significantly with macrophage activation in the whole mushroom extracts. Extract activity was negated by treatment with NaOH, conditions that inactivate LPS and other bacterial components. Significant correlations between macrophage activation and total bacterial load (ρ= 0.723, p=0.0001) and LPS content (ρ=0.951, p<0.0001) were also observed between different tissues of Agaricus bisporus. Pseudomonas and Flavobacterium were the most prevalent genera identified in the different tissue parts and these taxa were significantly correlated with in vitro macrophage activation (ρ=0.697, p<0.0001 and ρ=0.659, p=0.0001, respectively). These results indicate that components derived from mushroom associated bacteria contribute substantially to the innate immune enhancing activity exhibited by mushrooms and may result in similar therapeutic actions as reported for ingestion of bacterial preparations such as probiotics.

Keywords: Fungi, mushroom, macrophage activation, lipopolysaccharide, bacterial load, bacterial community structure

Introduction

A growing body of evidence supports the theory that bacterial communities within immune-enhancing botanicals are the principal source of components responsible for the activation of innate immune cells by extracts from these plants. Our prior research has demonstrated that bacterial lipopolysaccharides (LPS) and Braun-type lipoproteins were responsible for 97% of the in vitro macrophage activation exhibited by extracts of eight botanicals traditionally used to enhance immune function.1 Variations in the levels of these two bacterial components are also responsible for the up to 200-fold difference in in vitro monocyte/macrophage activation potential exhibited by Echinacea purpurea and Echinacea angustifolia bulk commercial material.2 We have found that E. purpurea samples contain bacterial loads ranging from 106 to 108 bacterial cells g−1 of dry material and that both bacterial loads and LPS content are strongly correlated with in vitro macrophage activity.3 Other groups have reported similar findings. Endophytic bacteria contributed to the in vitro pro-inflammatory effects of E. purpurea crude extracts4 and to the immunostimulatory activity of Angelica sinensis, one of the most potent herbs in the traditional East Asian formula Juzen-taiho-to.5

While plants are important as a source of immune-stimulating preparations, mushrooms also have a long history of medicinal use for enhancing immune function. This is supported by studies that indicate that dietary ingestion of mushrooms or their extracts stimulates immune cells.6 As with plants, mushrooms are colonized by communities of bacteria. However, most studies on mushroom-colonizing bacteria have focused on their presence from a food safety perspective7 or on the bacteria that cause visible disease symptoms that can impact retail value.8 Although reducing or eliminating potentially pathogenic bacteria associated with mushrooms is important, not all mushroom-colonizing bacteria represent a food safety issue. It is our hypothesis that the naturally occurring bacterial communities on the surface of and within mushrooms contribute substantially to their innate immune enhancing potential.

The aim of the current study was to assess the in vitro macrophage activation potential of extracts from an assortment of commercially obtained common edible mushrooms and to determine if differences in activity correlate with the bacterial load and LPS levels present in these products. In addition, bacterial community composition within each sample was analyzed using 16S rRNA gene pyrosequencing to determine if specific bacterial taxonomic groups correlated with in vitro macrophage activation exhibited by the mushroom extracts.

Materials and Methods

Mushroom material

Thirteen types of edible mushroom samples (Table 1) were obtained from local supermarkets in Oxford, MS and Memphis, TN, USA. Mushroom samples were analyzed as whole material or dissected into different parts. For dissection, mushroom material (portobello and white button) were separated into four parts: stem tissue, gills, cap peel, and internal (represents mushroom cap with gills and cap peel removed). To mimic the culinary use of mushrooms, the fresh mushroom samples were rinsed with tap water according to the label instructions except for oyster mushrooms (label indicated not to wash prior to use). All samples were freeze-dried, ground into a fine powder and stored at −20°C until analysis.

Table 1.

Total bacterial load, lipopolysaccharide (LPS) levels and in vitro macrophage activation potential exhibited by extracts of 13 different types of edible mushrooms.

Sample # Mushroom Material TNF-alpha (pg ml−1)a LPS levelb Bacterial load (×106 cells g−1)
1 Maitake (Grifola frondosa) Fresh 19.6 7 0.22
2 Fresh 22.0 4 0.73
3 Bunashimeji (Hypsizygus marmoreus) Fresh 18.3 3 0.34
4 Fresh 23.2 2 0.44
5 Bunapi (Hypsizygus marmoreus) Fresh 24.5 2 0.29
6 Fresh 26.9 0.4 0.47
7 Oyster (Pleurotus ostreatus) Fresh 70.0 1174 1.4
8 Fresh 85.5 112 27
9 Dried 50.8 225 4.2
10 Dried 333.9 3658 7.8
11 Crimini (Agaricus bisporus) Fresh 134.4 9 24
12 Fresh 158.4 7 25
13 Shiitake (Lentinula edodes) Fresh 150.0 19 17
14 Fresh 211.6 26 0.94
15 White button (Agaricus bisporus) Fresh 1,075.0 1,626 17
16 Woodear (Auricularia auricula-judae) Dried 1,302.0 6,647,669 1.1
17 Dried 2,165.6 14,265,954 1.2
18 Enoki (Flammulina velutipes) Fresh 2,112.8 668 17
19 Fresh 2,610.2 2,445 21
20 Morel (Morchella esculenta) Dried 1,864.4 89 4.4
21 Dried 5,810.8 13,952 0.85
22 Lobster (Hypomyces lactifluorum) Dried 4,830.0 867,965 60
23 Dried 4,881.9 3,412,805 55
24 Chanterelle (Cantharellus cibarius) Fresh 4,639.3 3,557,130 27
25 Fresh 6,938.1 428,795 35
26 Portobello (Agaricus bisporus) Fresh 9,945.7 401,635 59
Open in a new tab
a

In vitro macrophage activation potential measured by the amount of TNF-alpha secreted by RAW 264.7 cells exposed to extract from 40 μg of mushroom material per ml.

b

LPS level expressed as endotoxin units per g of dried mushroom material.

Extract Preparation

Ground mushroom material (50 mg) was extracted four times with 95% ethanol (1.0 mL fresh solvent added and incubated at 75°C for 45–60 minutes for each extraction) to remove non-polar substances that may be anti-inflammatory (inhibitors of macrophage activation). Ethanol extracted mushroom material was then dried overnight at 50–55°C.

Crude extracts were prepared for assessment of macrophage enhancing activity by further extracting the dried ethanol extracted material with 0.5 mL of water containing 4% SDS at 98°C for 1 hour. SDS was then removed using SDS-out reagent (Life Technologies, Grand Island, NY) in the presence of 1% octylglucoside. For assessment of LPS (endotoxin) content, crude extracts were prepared by further extracting the dried ethanol extracted material with 0.5mL of water containing 2% SDS at 98°C for 1 hour, followed by removal of SDS using SDS-out reagent.

Macrophage activation assay and limulus amebocyte lysate (LAL) assay

RAW 264.7 mouse macrophages (American Type Culture Collection) were cultured in RPMI 1640 medium supplemented with 10% v/v FBS (Hyclone Laboratories, Inc., Logan UT) and amikacin (60 mg/L) at 37°C, under 5% CO2. Actively growing cells were suspended in culture medium at 0.5 million cells/mL and plated at a density of 1×105 cells/well in 96-well plates. Test samples were added to cells and after incubation for 18 h, the level of TNF-alpha in the cell culture supernatants was determined using ELISA (R&D Systems). Macrophage activation is reported as TNF-alpha production (pg/ml). Ultra pure E. coli LPS from E. coli 0111:B4 and Pam3CSK4 were purchased from InvivoGen (San Diego, CA). S. cerevisiae particulate beta-glucan (WGP® dispersible) was from Biothera (St. Paul, MN).

Endotoxin content of mushroom extracts was determined using the Pyrochrome® LAL assay with Glucashield® (1→3)-β-D-Glucan Inhibiting Buffer (Associates of Cape Cod, Inc., East Falmouth, MA). Data is reported as endotoxin units (EU) per g of dried material and represents the average of determinations that were performed at least twice.

Determination of total bacterial load

Fifty milligrams of ground, freeze-dried mushroom material was rehydrated with 150 μL sterile water and DNA extracted using PowerPlant® Pro DNA isolation Kits (MoBio Laboratories, Carlsbad, CA), followed by a secondary clean-up using PowerClean® DNA cleanup kits (MoBio). The bacterial 16S rRNA gene from each sample was then amplified using primers 799f and 1492r as described previously.3 Bacterial loads were estimated by comparing the intensity of the 735 bp 16S bacterial band from mushroom extracts to a standard curve of DNA extracted and amplified from known quantities of bacteria.3

Determination of bacterial community structure

Bacterial community structure was determined by bacterial tag-encoded FLX amplicon 454 pyrosequencing (bTEFAP)9 conducted on the amplified 16S rRNA product from each sample, through a dedicated sequencing facility (MR DNA, Shallowater, TX). Bacterial primers 939f and 1392r10,11 were used in the sequencing reaction, following conditions described previously.12 Following PCR, products from different samples were mixed in equal concentrations and purified using Agencourt AMPure XP beads (Agencourt Bioscience Corporation, Danvers, MA). Samples were sequenced using Roche 454 FLX titanium instruments and reagents following the manufacturer’s guidelines.

Raw pyrosequence data was transferred into FASTA files for each sample, along with sequencing quality files. Files were processed with the bioinformatics software Mothur13 and analyzed following the procedures recommended by Schloss et al.14 Briefly, sequences were denoised and trimmed to remove barcodes and primers. Chimeric sequences were removed and sequences were then aligned and classified according to those in the SILVA rRNA database,15 after which any sequences classified as mitochondrial or chloroplast were removed from the dataset. The cell load of each bacterial taxon in mushroom samples were subsequently estimated from the proportion of sequencing reads of each taxon and the total bacterial load in each sample. Selected samples had their bacterial classification verified by subsampling 50–100 sequences from FASTA files and comparing them (BLAST) to sequences in the NCBI Genbank 16S rRNA database.

Statistical analysis

Statistical analyses were performed in JMP®11.2.0. Because data may not meet the requirements for parametric analyses, Spearman’s Rank Correlation was performed to examine relationships between macrophage activation stimulatory activity exhibited by the mushroom extracts and both bacterial load (total and for each individual taxon) and LPS content. Differences in estimated total bacterial load between tissue parts for portobello and white button mushroom samples were determined using ANOVA followed by Tukey’s Honestly Significant Difference test on log10 transformed data.

Results

The 13 types of edible mushrooms exhibited dramatic variation in the levels of in vitro macrophage activation, content of bacterial LPS and bacterial load. Extract activity (TNF-alpha production by RAW 264.7 cells) and LPS levels varied by over 500-fold and more than 10-million fold, respectively (Table 1). Bacterial load also varied between samples, with numbers ranging from 2.2 × 105 cells g−1 mushroom to 6.0 × 107 (Table 1). There was a significant relationship between total bacterial load and macrophage activation exhibited by the mushroom extracts (Fig. 1a, ρ=0.701, p <0.0001). Macrophage activation exhibited by extracts was also significantly related to levels of LPS (Fig. 1b; ρ=0.832, p<0.0001). Treatment of extracts with 0.5M NaOH for 1 h at 98°C reduced TNF-alpha production to the same levels as observed with NaOH treatment of Pam3CSK4 and ultra pure E. coli LPS (Fig. 2a). None of the NaOH treated extracts contained components that collaborated with Toll-like receptor 4 (TLR4) signaling to enhance TNF-alpha production in RAW 264.7 cells (Fig. 2b). The positive control, particulate beta-glucan from S. cerevisiae, is not inactivated by NaOH treatment and both untreated and NaOH treated beta glucan material collaborated with the TLR4 agonist ultra pure E. coli LPS to synergistically enhance TNF-alpha production (Fig. 2b).

Figure 1.

Figure 1

Open in a new tab

Correlation of macrophage stimulatory activity exhibited by extracts from 13 different types of edible mushrooms with total bacterial load (a) and LPS content (b). Types of mushrooms and number of samples for each mushroom are detailed in Table 1.

Figure 2.

Figure 2

Open in a new tab

In vitro macrophage stimulatory activity of extracts from 13 different types of edible mushrooms is due to components that are inactivated by treatment with NaOH. TNF-alpha production was evaluated in RAW 264.7 macrophages exposed to untreated and 0.5M NaOH treated extracts (a), and NaOH treated extracts plus ultra pure LPS at 10ng/ml (b). Pam3CSK4 (P) was tested at 100ng/ml and ultra pure LPS at 10ng/ml (L10) and 100ng/ml (L100). S. cerevisiae particulate beta glucan (G) and NaOH treated beta glucan (NG) were evaluated at 100μg/ml alone and together with ultra pure LPS at 10ng/ml (G+L, NG+L). Numbers below bars correspond to sample numbers in Table 1. Concentrations of both untreated and NaOH treated samples in the culture medium represent extract from 100 (samples 1–15), 40 (samples 16–25) and 16 (sample 26) μg of mushroom material/ml. (C) refers to untreated cells.

Of the 26 mushroom samples analyzed, nine (from the bunapi, bunashimeji, maitake, and woodear varieties) did not yield sufficient bacterial DNA so that only 17 samples were analyzed by pyrosequencing. These samples yielded a total of 50,243 16S rRNA gene sequences ≥ 200 bp. None of these sequences were classified as mitochondrial or chloroplast. Following the removal of chimeric sequences (20.8% of total), 39,810 partial 16S rRNA gene sequences (mean 2,342 sequences per sample, but ranging from 767 to 7,884) remained for subsequent analyses.

Representatives of at least seven distinct bacterial phyla were identified, as well as members of four of the major subphyla of Proteobacteria (Table 1S). Proteobacteria were by far the most dominant phylum, accounting for over 80% of the recovered sequences. Of these, the majority of sequences were identified as belonging to the Gammaproteobacteria. Of the remaining phyla, only Bacteroidetes and Firmicutes accounted for > 1% of the total number of sequences obtained, although some specific samples had greater proportions of a specific phylum. When the estimated cell load of individual phyla was related to macrophage activation, none of the phyla correlated significantly with TNF-alpha production (p>0.1 for all).

At a finer taxonomic level, Pseudomonas (Gammaproteobacteria) was the only bacterial genus detected that composed at least 1% of the community in all 17 samples, and was the most abundant genera in 12 samples (Fig. 3 and Table 2S). Sequences classified as Pseudomonas often dominated the community, and accounted for >90% of the bacterial community in four samples. Other abundant genera included Flavobacterium (Bacteroidetes), which was the most abundant genera detected in two samples (morel and one of the dried oyster mushroom samples), and unclassified members of the Enteric Bacteria cluster (Gammaproteobacteria), which was the most abundant genera detected in both enoki samples (Fig. 3 and Table 2S). A total of 32 genera were identified that accounted for at least 1% of the sequences recovered from at least one sample (Table 2S). Proportions of none of the abundant (>1% total) phyla or genera correlated significantly with macrophage activation (Table 2), although this was likely skewed by the lack of sequence data from nine samples, many of which showed low macrophage activation.

Figure 3.

Figure 3

Open in a new tab

Average bacterial community structure within various edible mushrooms. Mushroom material and number of samples for each mushroom is detailed in Table 1. Data represent most abundant taxa identified that accounted for over 10% of the sequences recovered from at least one sample.

Table 2.

Statistics for Spearman’s Rank Correlations between cell load of various bacterial groups and in vitro macrophage activation potential exhibited by mushroom samples. Details for the whole mushrooms and number of samples for each mushroom are summarize in Table 1.

Bacterial group (total, phylum or genus) Whole mushrooms Tissue parts for Agaricus bisporus
ρ p n ρ p n
Total bacteria 0.701 <0.0001 26 0.723 <0.0001 28
Proteobacteria 0.370 0.1437 17 0.694 <0.0001 28
Bacteroidetes 0.378 0.1353 17 0.652 0.0002 28
Pseudomonas 0.238 0.3582 17 0.697 <0.0001 28
Flavobacterium −0.067 0.7983 17 0.659 0.0001 28
Open in a new tab

In addition to whole mushrooms, the variation in bacterial load in different tissue parts of Agaricus bisporus (white button and portobello mushrooms) was also evaluated. For portobello mushrooms the mean bacterial load in the different parts varied less than two-fold (Fig. 4; 2.8 × 107 to 4.8 × 107 cells g−1) and no statistically significant difference in load was observed between the parts. In white button mushrooms, the mean bacterial load in stem, internal tissue and peel varied from 1.5 × 107 to 3.6 × 107 cells g−1, whereas the bacterial load within the gills was almost two orders of magnitude lower (8.8 × 105 g−1), a difference that was statistically significant (p=0.0108; Fig. 4).

Figure 4.

Figure 4

Open in a new tab

Levels of bacterial load present within internal and external parts of white button (n=3 for each part) and portobello mushrooms (n=4 for each part). Internal part represents mushroom cap with external parts (gills and cap peel) removed. Stem tissue represents both internal and external tissue. Total bacterial load (cells/g dry material) was estimated using a PCR-based quantification method that we previously developed (Pugh et al., 2013).3 *Gills of white button had a significantly lower bacterial load than other white button tissues, p =0.0104 (determined by ANOVA followed by Tukey’s HSD test on log10 transformed data).

As with the whole mushrooms, there was a significant correlation between total bacterial load and macrophage activation exhibited by extracts of the portobello and white button tissue parts (Fig. 5a, ρ= 0.723, p=0.0001). LPS content within the tissue parts was also significantly correlated with in vitro macrophage activation potential (Fig. 5b, ρ=0.951, p<0.0001). Correlations between bacterial load and activation potential were slightly stronger when only external mushroom parts (gills and peel) were included in the analysis (ρ=0.776, p=0.0011), but there were no significant correlations between activity and bacterial load for internal tissue or stem tissue samples alone (ρ = 0.464 and ρ = 0.250, respectively).

Figure 5.

Figure 5

Open in a new tab

Correlation of total bacterial load (a) and LPS content (b) with macrophage activation exhibited by extracts of white button (n=12, indicated by white circles) and portobello parts (n=16, indicated by black circles).

Individual parts of both portobello and white button were dominated by the phylum Proteobacteria (especially Gammaproteobacteria), representing >96% of the sequences identified (Table 3S). Of the remaining phyla that could be identified, only Bacteroidetes and Firmicutes, made up >1% of the bacterial sequences in portobello (2.87% Bacteroidetes) or white button (1.27% Firmicutes) samples. Overall, correlations with macrophage activation were higher within portobello and white button samples than when all 13 mushroom types were compared (Table 2). Correlations between macrophage activation and total bacterial load were similar in all 13 mushroom types and the A. bisporus samples specifically (Table 2). However, some bacterial taxa identified in A. bisporus parts correlated significantly with macrophage activation (Table 2).

At the phylum level, both Proteobacteria (ρ=0.694, p <0.0001) and Bacteroidetes (ρ=0.652, p=0.0002) correlated significantly with macrophage activation. Pseudomonas was the dominant genus in A. bisporus, ranging from 70–100% of the sequences identified in each sample (Table 3S). The next most prevalent genus, Flavobacterium, made up 0–12% and 0–0.7% of the bacterial sequences in portobello and white button mushrooms, respectively (Table 3S). As the most prevalent member of the Proteobacteria, Pseudomonas correlated highly with macrophage activation (ρ=0.697, p<0.0001) in A. bisporus samples, as did Flavobacterium (ρ=0.659, p=0.0001; Table 2).

Discussion

We evaluated 13 types of culinary mushrooms, seven of which (maitake, shiitake, oyster, bunashimeji, bunapi, woodear and enoki) have also been used traditionally for medicinal purposes. Maitake, shiitake and oyster mushrooms have been used to enhance immune function, and studies support this therapeutic effect.16–18 Bunashimeji, bunapi and enoki mushrooms contain polysaccharides19 or proteoglycans20 that activate macrophages in vitro. Although Agaricus bisporus (white button, portobello, crimini) is generally used solely for culinary purposes, dietary supplementation of this mushroom enhances NK cell activity in mice.21 We hypothesized that bacterial products derived from the bacteria that colonize these fungi are major contributors to the innate immune enhancing activity exhibited by these medicinal and culinary mushrooms. The strong correlations between in vitro macrophage activation (as determined by TNF-alpha production by RAW 264.7 macrophages) and LPS content and bacterial load of these samples support this hypothesis. These results are consistent with our previous research on immune enhancing plant botanicals showing that immune enhancing activity is strongly correlated with LPS content and bacterial load within E. purpurea.3

The substantial reduction (97 – 100%) in in vitro macrophage activation exhibited by the mushroom extracts after NaOH treatment indicates that the activity of these samples is due to bacterial components. The bacterial components LPS, Braun-type lipoproteins and lipoteichoic acid are inactivated by the base-dependent removal of ester- and amide-linked fatty acids. The low level of base-resistant activity exhibited by some of the NaOH-treated mushroom extracts could be due to beta glucans since these polysaccharides retain dectin-1 dependent activity following treatment with strong base (e.g. boiling in 10M NaOH for 0.5 h).22 NaOH treatment converts the triple-helix structure of high molecular weight (1,3)-beta-D-glucans to a partially opened triple-helix conformation23 or single-helix conformation24, 25. Since most research demonstrates that the partially opened triple-helix and single-helix conformations exhibit enhanced macrophage activation in vitro23–25 compared to the triple-helix structure, the NaOH treatment of the mushroom extracts would not have decreased activity due to beta glucans. Although one study reported that TNF-alpha production from RAW 264.7 cells by Grifola frondosa (1,3)-beta-D-glucan was abolished by treatment with NaOH26, it is likely that the observed reduction in activity was due to inactivation of bacterial components derived from contamination of the test materials during the extraction/isolation procedure.

The combination of ultra pure LPS and a beta glucan result in a collaborative interaction that can be detected as an enhancement in TNF-alpha production by macrophages.27 The lack of enhancement in TNF-alpha production observed with the NaOH treated mushroom samples evaluated in combination with ultra pure LPS may be due to the concentration of extract tested (appropriately 25μg/ml and lower). By comparison, published studies typically assess activity of purified beta glucans at 100μg/ml24, 25. Even at 100μg/ml of purified S. cerevisiae 1,3-beta glucan we only observed a moderate increase in TNF-alpha production (19% of levels achieved by ultra pure LPS tested at 10ng/ml). Since a major portion of the activity detected in the mushroom extracts appears to be due to LPS and other bacterial substances, it will be important to determine whether contaminating bacterial components are partially or solely responsible for the activity of some of the mushroom derived polysaccharide preparations reported in the literature.

Production of TNF-alpha by RAW 264.7 cells was used to measure macrophage activation since this system detects the additive effect of bacterial LPS and Braun-type lipoproteins. Our previous research demonstrated that LPS and Braun-type lipoproteins are the most potent monocyte/macrophage activating components of bacteria and both contribute substantially to the in vitro monocyte/macrophage activation exhibited by extracts of Echinacea.2 In animal models and human clinical trials we have demonstrated the therapeutic potential of one of these bacterial components using an extract enriched for Braun-type lipoproteins. In mice, oral ingestion of this extract exhibited a protective effect against influenza A (H1N1) viral infection28 and enhanced ex vivo production of IgA and IL-6 from Peyer’s patch cells as well as interferon-gamma production from spleen cells.29 In two separate clinical trials dietary supplementation with this extract enhanced NK cell activity by 40 and 54%.30 Similar therapeutic benefits on enhancing immune function are observed following oral ingestion of both live and heat killed probiotic bacteria.31 The above studies suggest that consumption of bacterial components, whether derived from a bacterial preparation or from bacteria colonizing plants or mushrooms, can have therapeutically relevant effects on enhancing immune function.

Mushrooms used in the current study were purchased either fresh or dried from local supermarkets and therefore factors such as post-harvest events are unknown. It is conceivable that the high level of estimated total bacterial load in some of the mushroom samples could have originated from post-harvest contamination. However, any bacteria introduced post-harvest are likely to have been removed from the external surfaces of the fresh mushrooms as these samples were washed prior to use (the exception being oyster mushrooms, as label instructions indicated not to wash prior to consumption). For Agaricus bisporus the internal tissue contained high concentrations of bacteria that were comparable to, or higher than that observed within the external surfaces (peel and gills). These internally colonizing bacteria would have contributed substantially to the total bacterial load that we detected in the mushroom samples.

Gram-negative taxa were the most prevalent bacteria detected within mushroom samples, and these bacteria exhibited the strongest correlations with in vitro macrophage activation. Abundances of the phyla Proteobacteria and Bacteroidetes exhibited significant correlations with in vitro macrophage activation by white button and portobello mushrooms, whereas the next most abundant phyla detected, the Gram-positive Firmicutes and Actinobacteria, showed no significant relationships to in vitro macrophage activation. We have found more than 8000-fold variation in in vitro macrophage activation between different bacterial isolates obtained from plants32, so variation in the correlation between immune enhancing activity and the abundance of different bacterial taxa is not surprising. Plant-derived isolates identified as members of the Proteobacteria show the highest level of in vitro macrophage stimulatory activity, supporting the findings of the current study, and suggesting that members of this taxon may be important contributors to both botanical and mushroom immune stimulation.

Pseudomonas was the most commonly detected genus associated with mushrooms. A previous culture-based study also reported high numbers of Pseudomonas on mushroom samples.7 Leff and Fierer33 used 16S rRNA gene pyrosequencing to examine the bacterial communities on the surface of mushrooms, and also reported Pseudomonas sp. to be one of the most abundant bacterial genera present, accounting for 11% of the sequences obtained. Of the 20 most prevalent bacterial genera detected in the current study, only Pseudomonas and Flavobacterium displayed significant correlations to macrophage activity exhibited by portobello and white button mushroom. Although this indicates that Pseudomonas and Flavobacterium are important taxa that contribute to the in vitro macrophage activation potential of the mushroom extracts, none of the bacterial taxa showed significant correlations when all the sequenced mushroom types were analyzed together. However, the lack of sufficient quantities of DNA for sequencing from nine of the mushroom samples resulted in a lack of community composition data from several low activity samples, which may have limited the correlation results. The lack of significance also suggests that other taxa are also important contributors to the activity exhibited by the mushroom extracts. There may also be significant differences in activity between bacterial species of the same genus; a potential variable that could not be evaluated since the bioinformatics pipeline and short read length of pyrosequencing only allowed absolute identification of taxa to the genus level. BLAST searches of some sequences suggested that those identified as Pseudomonas or Flavobacterium could actually represent six or more species of each genus, with the more common species varying between samples, supporting the idea that there could be species level variation across different samples. Different macrophage activity between different species or even strains of a bacterial genus, might therefore explain some of the variation between samples.

Although Pseudomonas was present in all samples analyzed (and was the most predominant taxa in most samples), there was also variation in the bacterial community composition between the different mushrooms. One factor that may contribute to the differences in community structure between mushrooms is that some fungi contain antimicrobial substances. Extracts of Shiitake mushrooms have been reported to possess antimicrobial activity against species of Pseudomonas34 and this may explain why the bacterial community in the Shiitake samples examined in this study was composed of only 43–59% Pseudomonas, compared to more than 90% of the community in some of the other mushroom types that were analyzed. Oyster and morel mushroom varieties have also been found to display antimicrobial activity against several bacteria, including Pseudomonas.35 Sequences identified as Pseudomonas accounted for <50% of those derived from morel samples and <11% of two out of three sequenced oyster samples. The presence and type of naturally occurring antimicrobial agents is likely to differ between mushroom varieties, and this could influence the composition of the bacterial community in different mushroom types and the level of bacterial components contributing to extract activity.

Conclusion

The presence of bacteria in mushrooms is usually investigated from the standpoint of freshness and food safety. However, the current study examined the entire bacterial community composition within mushrooms, not just potential pathogens or organisms involved in spoilage. As with our previous research on plant botanicals,1–3 the results reported here indicate that components derived from mushroom associated bacteria contribute substantially to the innate immune enhancing activity exhibited by extracts of common medicinal and culinary mushrooms. This link between bacterial components in edible mushrooms and enhanced immune function provide new insight into the significance of bacterial colonization of food products.

Supplementary Material

Supplemental

Acknowledgments

This research was partly funded by Grant Number R01AT007042 from the National Center for Complementary and Integrative Health (NCCIH) and the Office of Dietary Supplements (ODS). The contents of this manuscript are solely the responsibility of the authors and do not necessarily represent the official views of the NCCIH, ODS or the National Institutes of Health. Additional funding of this research was also provided by a grant from the USDA, Agricultural Research Service Specific Cooperative Agreement No. 58-6408-1-603.

Abbreviations

LPS

Lipopolysaccharide

rRNA

ribosomal ribonucleic acid

LAL

limulus amebocyte lysate

ELISA

enzyme-linked immunosorbent assay

EU

endotoxin units

DNA

deoxyribonucleic acid

ANOVA

analysis of variance

Footnotes

Conflict of interest statement

Authors declare that there are no conflicts of interest.

Bibliographic References

  • 1.Pugh ND, Tamta H, Balachandran P, Wu X, Howell J, Dayan FE, Pasco DS. The majority of in vitro macrophage activation exhibited by extracts of some immune enhancing botanicals is due to bacterial lipoproteins and lipopolysaccharides. Int Immunopharmacol. 2008;8:1023–1032. doi: 10.1016/j.intimp.2008.03.007. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Tamta H, Pugh ND, Balachandran P, Moraes R, Sumiyanto J, Pasco DS. Variability in in vitro macrophage activation by commercially diverse bulk Echinacea plant material is predominantly due to bacterial lipoproteins and lipopolysaccharides. J Agric Food Chem. 2008;56:10552–10556. doi: 10.1021/jf8023722. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Pugh ND, Jackson CF, Pasco DS. Total bacterial load within Echinacea purpurea, determined using a new PCR-based quantification method, is correlated with LPS levels and in vitro macrophage activity. Planta Med. 2013;79:9–14. doi: 10.1055/s-0032-1328023. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Todd DA, Gulledge TV, Britton ER, Oberhofer M, Leyte-Lugo M, Moody AN, Shymanovich T, Grubbs LF, Juzumaite M, Graf TN, Oberlies NH, Faeth SH, Laster SM, Cech NB. Ethanolic Echinacea purpurea extracts contain a mixture of cytokine-suppressive and cytokine-inducing compounds, including some that originate from endophytic bacteria. PLoS One. 2015;10:e0124276. doi: 10.1371/journal.pone.0124276. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Montenegro D, Kalpana K, Chrissian C, Sharma A, Takaoka A, Iacovidou M, Soll CE, Aminova O, Heguy A, Cohen L, Shen S, Kawamura A. Uncovering potential ‘herbal probiotics’ in Juzen-taiho-to through the study of associated bacterial populations. Bioorg Med Chem Lett. 2015;25:466–469. doi: 10.1016/j.bmcl.2014.12.036. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Wasser SP. Medicinal mushroom science: Current perspectives, advances, evidences, and challenges. Biomed J. 2014;37:345–356. doi: 10.4103/2319-4170.138318. [DOI] [PubMed] [Google Scholar]
  • 7.Venturini ME, Reyes JE, Rivera CS, Oria R, Blanco D. Microbiological quality and safety of fresh cultivated and wild mushrooms commercialized in Spain. Food Microbiol. 2011;28:1492–1498. doi: 10.1016/j.fm.2011.08.007. [DOI] [PubMed] [Google Scholar]
  • 8.Abou-Zeid MA. Pathogenic variation in isolates of Pseudomonas causing the brown blotch of cultivated mushroom, Agaricus bisporus. Braz J Microbiol. 2012;43:1137–1146. doi: 10.1590/S1517-838220120003000041. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Dowd SE, Callaway TR, Wolcott RD, Sun Y, McKeehan T, Hagevoort RG, Edrington TS. Evaluation of the bacterial diversity in the feces of cattle using 16S rDNA bacterial tag-encoded FLX amplicon pyrosequencing (bTEFAP) BMC Microbiol. 2008;8:125. doi: 10.1186/1471-2180-8-125. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Jackson CR, Langner HW, Donahoe-Christiansen J, Inskeep WP, McDermott TR. Molecular analysis of microbial community structure in an arsenite-oxidizing acidic thermal spring. Environ Microbiol. 2001;3:532–542. doi: 10.1046/j.1462-2920.2001.00221.x. [DOI] [PubMed] [Google Scholar]
  • 11.Baker GC, Smith JJ, Cowan DA. Review and re-analysis of domain-specific 16S primers. J Microbiol Meth. 2003;55:541–555. doi: 10.1016/j.mimet.2003.08.009. [DOI] [PubMed] [Google Scholar]
  • 12.Jackson CR, Randolph KC, Osborn SL, Tyler HL. Culture dependent and independent analysis of bacterial communities associated with commercial salad leaf vegetables. BMC Microbiol. 2013;13:274. doi: 10.1186/1471-2180-13-274. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Schloss PD, Westcott SL, Raybin T, Hall JR, Hartmann M, Hollister EB, Lesniewski RA, Oakley BB, Parks DH, Robinson CJ, Sahl JW, Stres B, Thallinger GG, Van Horn DJ, Weber CF. Introducing mothur: Open-source, platform-independent, community-supported software for describing and comparing microbial communities. Appl Environ Microbiol. 2009;75:7537–7541. doi: 10.1128/AEM.01541-09. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Schloss PD, Gevers D, Westcott SL. Reducing the effects of PCR amplification and sequencing artifacts on 16S rRNA-based studies. PLoS One. 2011;6:e27310. doi: 10.1371/journal.pone.0027310. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Pruesse E, Quast C, Knittel K, Fuchs BM, Ludwig W, Peplies J, Glockner FO. SILVA: a comprehensive online resource for quality checked and aligned ribosomal RNA sequence data compatible with ARB. Nucleic Acids Res. 2007;35:7188–7196. doi: 10.1093/nar/gkm864. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Mayell M. Maitake extracts and their therapeutic potential. Altern Med Rev. 2001;6:48–60. [PubMed] [Google Scholar]
  • 17.Bisen PS, Baghel RK, Sanodiya BS, Thakur GS, Prasad GB. Lentinus edodes: a macrofungus with pharmacological activities. Curr Med Chem. 2010;17:2419–2430. doi: 10.2174/092986710791698495. [DOI] [PubMed] [Google Scholar]
  • 18.Kong F, Li FE, He Z, Jiang Y, Hao R, Sun X, Tong H. Anti-tumor and macrophage activation induced by alkali-extracted polysaccharide from Pleurotus ostreatus. Int J Biol Macromol. 2014;69:561–566. doi: 10.1016/j.ijbiomac.2014.05.045. [DOI] [PubMed] [Google Scholar]
  • 19.Wu M, Luo X, Xu X, Wei W, Yu M, Jiang N, Ye L, Yang Z, Fei X. Antioxidant and immunomodulatory activities of a polysaccharide from Flammulina velutipes. J Tradit Chin Med. 2014;34:733–740. doi: 10.1016/s0254-6272(15)30089-3. [DOI] [PubMed] [Google Scholar]
  • 20.Bao HH, Tarbasa M, Chae HM, You SG. Molecular properties of water-unextractable proteoglycans from Hypsizygus marmoreus and their in vitro immunomodulatory activities. Molecules. 2011;17:207–226. doi: 10.3390/molecules17010207. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Wu D, Pae M, Ren Z, Guo Z, Smith D, Meydani SN. Dietary supplementation with white button mushroom enhances natural killer cell activity in C57BL/6 mice. J Nutr. 2007;137:1472–1477. doi: 10.1093/jn/137.6.1472. [DOI] [PubMed] [Google Scholar]
  • 22.Gantner BN, Simmons RM, Canavera SJ, Akira S, Underhill DM. Collaborative induction of inflammatory responses by dectin-1 and Toll-like receptor 2. J Exp Med. 2003;197:1107–1117. doi: 10.1084/jem.20021787. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Young SH, Dong WJ, Jacobs RR. Observation of a partially opened triple-helix conformation in 1-->3-beta-glucan by fluorescence resonance energy transfer spectroscopy. J Biol Chem. 2000;275(16):11874–11879. doi: 10.1074/jbc.275.16.11874. [DOI] [PubMed] [Google Scholar]
  • 24.Tsuzuki A, Ohno N, Adachi Y, Yadomae T. Interleukin 8 production by human leukocytes stimulated by triple or single helical conformer of an antitumor (1-->3)-beta-D-glucan., 1999, preparation, Sonifilan. Drug Dev Res. 48:17–25. [Google Scholar]
  • 25.Ohno N, Hashimoto T, Adachi Y, Yadomae T. Conformation dependency of nitric oxide synthesis of murine peritoneal macrophages by beta-glucans in vitro. Immunol Lett. 1996;52:1–7. doi: 10.1016/0165-2478(96)02538-2. [DOI] [PubMed] [Google Scholar]
  • 26.Ishibashi K, Miura NN, Adachi Y, Ohno N, Yadomae T. Relationship between solubility of grifolan, a fungal 1,3-beta-D-glucan, and production of tumor necrosis factor by macrophages in vitro. Biosci Biotechnol Biochem. 2001;65(9):1993–2000. doi: 10.1271/bbb.65.1993. [DOI] [PubMed] [Google Scholar]
  • 27.Ferwerda G, Meyer-Wentrup F, Kullberg BJ, Netea MG, Adema GJ. Dectin-1 synergizes with TLR2 and TLR4 for cytokine production in human primary monocytes and macrophages. Cell Microbiol. 2008;10:2058–2066. doi: 10.1111/j.1462-5822.2008.01188.x. [DOI] [PubMed] [Google Scholar]
  • 28.Pugh ND, Edwall D, Lindmark L, Kousoulas KG, Iyer AV, Haron MH, Pasco DS. Oral administration of a Spirulina extract enriched for Braun-type lipoproteins protects mice against influenza A (H1N1) virus infection. Phytomedicine. 2015;22:271–276. doi: 10.1016/j.phymed.2014.12.006. [DOI] [PubMed] [Google Scholar]
  • 29.Balachandran P, Pugh ND, Ma G, Pasco DS. Toll-like receptor 2-dependent activation of monocytes by Spirulina polysaccharide and its immune enhancing action in mice. Int Immunopharmacol. 2006;6:1808–1814. doi: 10.1016/j.intimp.2006.08.001. [DOI] [PubMed] [Google Scholar]
  • 30.Nielsen CH, Balachandran P, Christensen O, Pugh ND, Tamta H, Sufka KJ, Wu X, Walsted A, Schjørring-Thyssen M, Enevold C, Pasco DS. Enhancement of natural killer cell activity in healthy subjects by Immulina®, a Spirulina extract enriched for Braun-type lipoproteins. Planta Med. 2010;76:1802–1808. doi: 10.1055/s-0030-1250043. [DOI] [PubMed] [Google Scholar]
  • 31.Adams CA. The probiotic paradox: live and dead cells are biological response modifiers. Nutr Res Rev. 2010;23:37–46. doi: 10.1017/S0954422410000090. [DOI] [PubMed] [Google Scholar]
  • 32.Haron MH, Tyler HL, Pugh ND, Moraes RM, Maddox VL, Jackson CR, Pasco DS. Activities and prevalence of Proteobacteria members colonizing Echinacea purpurea fully account for macrophage activation exhibited by extracts of this botanical. Planta Med. 2016 Jun 10; doi: 10.1055/s-0042-108590. Epub ahead of print. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Leff JW, Fierer N. Bacterial communities associated with the surfaces of fresh fruits and vegetables. PLoS One. 2013;8:e59310. doi: 10.1371/journal.pone.0059310. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Hearst R, Nelson D, McCollum G, Millar BC, Maeda Y, Goldsmith CE, Rooney PJ, Loughrey A, Rao JR, Moore JE. An examination of antibacterial and antifungal properties of constituents of Shiitake (Lentinula edodes) and Oyster (Pleurotus ostreatus) mushrooms. Complement Ther Clin Pract. 2009;15:5–7. doi: 10.1016/j.ctcp.2008.10.002. [DOI] [PubMed] [Google Scholar]
  • 35.Ahmad N, Mahmood F, Khalil SA, Zamir R, Fazal H, Abbasi BH. Antioxidant activity via DPPH, gram-positive and gram-negative antimicrobial potential in edible mushrooms. Toxicol Ind Health. 2014;30:826–834. doi: 10.1177/0748233712463775. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplemental
NIHMS860154-supplement-Supplemental.pdf (81.5KB, pdf)

RESOURCES