All mushrooms were obtained from V.O.F. de Rijk-Stenssen (Rossum, The Netherlands). Mushrooms were harvested, directly transported to our laboratory, and frozen at −20°C. The frozen mushrooms were lyophilized and ground to a fine powder that was used as a complete homogenate (CH) in the experiments described. Part of the CH was alcohol precipitated to obtain a fraction enriched in polysaccharides according to Ng and Yap (15). Briefly, hot water extraction was performed for 20 min at 121°C in an autoclave on CH followed by overnight precipitation (4°C) with 70% ice-cold ethanol. The ethanol precipitate was removed by centrifugation, lyophilized, and subjected to a second hot water extraction. The second hot water extract was centrifuged and the supernatant was subjected to overnight precipitation with 50% ethanol at 4°C. The precipitate was collected by centrifugation (3500 × g, 30 min), lyophilized and ground to a fine powder, stored at −20°C, and used as an alcohol-precipitated polysaccharide fraction in the experiments described. For stimulations, CH and AP were dissolved in PBS in a concentration of 25 mg/ml.

Bone marrow was obtained from the femur and tibia of 6- to 12-week-old C57BL/6 mice. Bone marrow-derived dendritic cells (BMDCs) were differentiated for 10 days as previously described (16) using 10% spent medium from murine GM-CSF-transfected X63 cells (17). X63-GM-CSF cells were kindly provided by Dr. M. Lutz (University of Erlangen-Nuremberg, Germany) with approval of Dr. B. Stockinger (Francis Crick Institute, London, UK). Bone marrow cells were plated at 2 × 10⁵ cells/ml in bacteriological petri dishes (Greiner Bio-One, Alphen aan de Rijn, The Netherlands) and incubated at 37°C/5% CO₂. At days 3, 6, and 8, medium was refreshed, and at day 10, both adherent and non-adherent cells were harvested. At this time, typically 90–95% of the cells expressed the DC markers CD11c and MHC class II as assessed by flow cytometry.

T cells were obtained from spleens of 6- to 12-week-old C57BL/6 mice. Isolation of the CD4+ T cells was done via the MagCellect Mouse CD4+ T Cell Isolation kit (R&D systems, Minneapolis, MN, USA) according to the manufacturer’s protocol.

Bone marrow-derived dendritic cells (5 × 10⁵ cells/ml) were stimulated in a 96-well plate (Greiner Bio-One) for 24 h upon addition CH or AP of different mushroom species to a final concentration of 250 μg/ml. Curdlan and zymosan were used as control stimulants in a final concentration of 50 μg/ml and 1 mg/ml, respectively. In CLR neutralization studies, 5 mM EGTA (Sigma-Aldrich, Zwijndrecht, The Netherlands) or 250 μg/ml laminarin (Sigma) was added 30 min prior to stimulation with the CH of different mushroom species. In a coculture of BMDCs (10⁵ cells/ml) and total T cells (5 × 10⁵ cells/ml) in 96-well plates, BMDCs were stimulated for 24 h where after T cells were added. Costimulation was done for 7 days.

The coculture in ThinCerts using 24-well plates (Greiner Bio-One) was done with 200 μg/ml CH, 250 μl DCs (2 × 10⁵ cells/ml) and 250 μl T cells (1 × 10⁶ cells/ml). During this experiment, there were two different setups. The control setup allowed the physical interaction between DCs and T cells, whereas the test setup did not. In both cases, a semipermeable membrane (ThinCert, 3 μm pore size, Greiner Bio-One), that allowed the exchange of medium and cytokines but not cells, was used. T cells were added 24 h after DC stimulation. In the test setup, the T cells were put into the “membrane cup” instead of in the well to avoid migration of DCs toward the T cells.

Mushroom α- and β-glucan composition was determined using a β-Glucan (yeast and mushroom) Assay Kit (Megazyme, Bray, Ireland) according to the manufacturer’s protocol. The amount of protein in the mushroom preparations was determined using a BCA protein assay (Life Technologies Europe BV, Bleiswijk, The Netherlands). Polyphenols were determined using a Folin Ciocalteu assay. Thereto, 30 μl Folin Ciocalteu’s phenol reagent (Sigma) was added to 15 μl sample. This was incubated for 7–8 min where after 75 μl of a solution of 20% anhydrous Na₂CO₃ was added. The amount of polyphenol was determined by measuring the absorbance at 550 nm using a standard line of gallic acid. Chitin content was determined as reported (18). In short, freeze dried mushrooms were heated using a water bath for 2 h in 1 M NaOH under continuous stirring. The sample was filtered and washed with distilled water. The procedure was repeated for 3 h using saturated KOH. The samples were centrifuged at 8000 × g, the pellet washed with distilled water, lyophilized and dissolved in 10% acetic acid, and heated at 100°C for 1 h. Of each sample, 2 μl was spotted on a thin layer plate (Polygram Sil G 0.2 mm, Macherey Nagel, Düren, Germany). The spots were detected with 1% Lugol’s solution, after color formation the plates were dried with hot air. The density was analyzed from an image using Image J. A dilution series of chitin (Sigma) was used as reference.

TNF-α, Interleukin (IL)-6, IL-10, IL-12, IL-17, and IFN-γ levels in cell culture supernatants were determined using sandwich enzyme linked immunosorbent assay (ELISA) with ELISA Ready-Set-Go! Sets (eBioscience, Vienna, Austria) according to the supplier’s protocols.

Bone marrow-derived dendritic cells were stained in FACS buffer (PBS containing 0.1% bovine serum albumin and 5 mM EDTA) using the following monoclonal antibodies in the indicated concentration for cell surface markers (all obtained from eBioscience): PE-conjugated anti-CD86 (2 μg/ml), FITC conjugated anti-CD-40 (5 μg/ml), PE-conjugated anti-CD11c (1 μg/ml), and APC-conjugated MHC-II (0.28 μg/ml). Cells were first incubated with Fc receptor block, 5 μg/ml (eBioscience) for 10 min to block any non-specific binding and subsequent staining steps were performed for 20 min at 4°C, followed by washing with FACS buffer. An isotype-matched control staining was done for each antibody to determine a specific background staining. Cells were acquired using a Cyan-ADP cytometer (Beckman Coulter, Woerden, The Netherlands) and analyzed with FlowJo software (FlowJo LLC, Ashland, OR, USA).

Polysaccharides from alcohol-precipitated preparations were labeled with Alexa Fluor 488 hydrazide (Invitrogen, Life Technologies) according to the manufacturer’s protocol. Binding of the labeled mushroom compounds to BMDCs was first assessed using flow cytometry using 5 × 10⁵ cells/ml and 5 μg/ml-labeled polysaccharides. Cells were acquired using a Cyan-ADP cytometer (Beckman Coulter). Next, endocytosis of polysaccharides was visualized by seeding BMDCs on borosilicate chambered coverglasses at a density of 5 × 10⁴ cells/chamber followed by incubation with Alexa Fluor 488-labeled stimulants (10 μg/ml-labeled polysaccharides of Agaricus subrufescens or Lentinula edodes). As a positive control, 10 μg/ml FITC-labeled dextran (Sigma) was used. Unstimulated cells were used as a negative control. After 6 h, the stimulants were removed by refreshing the medium. Confocal images were taken 24 h after the start of stimulation. Plasma membranes were stained with Cellmask Deep Red plasma membrane stain (Invitrogen, Life Technologies) and lysosomes with Lysotracker Red DND-99 (Invitrogen, Life Technologies). Imaging was done using an Axiovert LSM510 confocal microscope (Carl Zeiss, Sliedrecht, The Netherlands) using a water objective C-Apochromat 40×. Images were processed using the LSM Image browser (Carl Zeiss), release 4.2.

For statistical analysis, independent-samples T test was used. Analysis was carried out using SPSS version 22 (IBM corporation, Armonk, NY, USA).

Before assessing immunomodulatory effects of CHs and alcohol precipitates (APs) of the seven mushroom species under investigation the amounts of protein, α- and β-glucan, chitin, and polyphenol were assessed as percentage dry weight (Table 1). The obtained results emphasize a great variety in composition between the different mushroom species. The chitin and polyphenol contents remained below 5% (w/w) for all preparations, whereby chitin was never detected in AP. Because chitin is insoluble and the procedure included a centrifugation step, this result was to be expected. The polyphenol contents increased upon ethanol precipitation for A. subrufescens, L. edodes, and Hypsizygus tessellatus. This result may indicate that these three species contain relatively high amounts of polyphenol/polysaccharide complexes or that the composition of the polyphenols differs. In the other four species, such complexes may still be present but to a much lower extent.

The amounts of protein, α-, and β-glucan varied between preparations and between species. Pholiota nameko contained the highest amounts of β-glucan in the CH; however after ethanol precipitation, both the α- and β-glucan content was reduced. This is in contrast to what was expected as upon ethanol precipitation the glucan fraction is often enlarged which was the case for all other preparations. The most striking effect of the ethanol precipitation was seen for H. tessellatus and Pleurotus eryngii where the total amount of glucan increased from, respectively, 45 and 40% to around 80% of the dry weight. Upon ethanol precipitation, the protein content decreased in five of the seven species; only for P. nameko and Grifola frondosa, the protein content increased. This may be caused by a relative high amount of saccharopeptides present in these species. In the AP of L. edodes and H. tessellatus, no protein was detected. This result could indicate that in these two species no protein is bound to polysaccharides that are precipitated with ethanol. All in all, protein as well as α- and β-glucan contents vary greatly between species.

To determine the effect of mushroom compounds on cytokine expression by DCs, CH or AP of each of the seven mushroom species was added to DCs. ELISA was used to measure the effect on the secretion of cytokines TNF-α, IL-6, IL-10, and IL-12 (Figures 1A–D, respectively). In the control experiments, the expression of TNF-α, IL-6, and IL-12 remained below detection level. All preparations induced expression of the pro-inflammatory cytokines TNF-α and IL-6. H. tessellatus can be considered as the worst inducer of these cytokines. The CH of H. tessellatus induced TNF-α secretion in low amounts, which was almost absent upon stimulation with AP. H. tessellatus CH and AP hardly induced secretion of any of the other cytokines measured. It is worth noting that alcohol precipitation of H. tessellatus CH lead to the largest increase of glucan content, notably β-glucans (Table 1), whereas protein was absent in AP. Apparently, the H. tessellatus glucans cannot be considered as strong immunomodulators. Secretion of the anti-inflammatory cytokine IL-10 was significantly induced by CH of A. subrufescens, G. frondosa, and Pleurotus ostreatus. Of the AP preparations, only A. subrufescens significantly induced IL-10. Strikingly, the three species that induce IL-10 secretion show the highest protein content of all CH preparations (Table 1). Expression of the Th-1-inducing cytokine IL-12 was induced by A. subrufescens (CH and AP), G. frondosa (CH and AP), and P. nameko (AP). All other preparations induced IL-12 to concentrations <100 pg/ml. Considering all cytokines, alcohol precipitation, to enrich the polysaccharide fraction, affects the secretion of cytokines only to a minor extent. Only for G. frondosa, H. tessellatus, and P. nameko, the induction of TNF-α expression, and for P. ostreatus, the induction of IL-12 expression differed significantly between the CH and AP preparations (Figure 1). The seven species investigated have all been reported to immunomodulate. It was surprising to observe that L. edodes and P. eryngii, both species that have been reported to be strong immunomodulators, only moderately induced TNF-α and IL-6 secretion. In general, the extent to which the seven species investigated affect secretion of TNF-α, IL-6, IL-10, and IL-12 by DCs varies greatly and likely depends on the structure and composition of polysaccharide complexes either or not in combination with peptides/protein.

The differences in cytokine responses and notably the low responses induced by several species prompted us to study the maturation of the DCs. Thereto, DCs were stimulated with AP of each of the seven mushrooms. After 24 h the cells were labeled with anti-CD11c-PE, anti-MHC-II-APC, anti-CD40-FITC, or anti-CD86-PE antibodies and the fluorescence intensity was measured with flow cytometry. The average mean fluorescence of three independent experiments is depicted in Figure 2. Untreated cells, incubated in tissue culture medium, showed a typical immature BMDC phenotype, i.e., CD11c⁺-CD40^low-CD86^low-MHC-II^low. Striking differences in the maturation state of the DCs stimulated with the AP from different mushrooms were observed. All mushrooms significantly upregulated MHC-II expression (Figure 2A). The expression of the costimulatory molecule CD40 was only significantly affected by A. subrufescens AP and curdlan (Figure 2B). In the case of the costimulatory molecule CD86, all mushrooms except G. frondosa lead to significant upregulation (Figure 2C). It was noted that the extent of CD86 upregulation differed considerably between species. Upon maturation CD11c is expected to be downregulated. However, considerable downregulation of CD11c was only observed for A. subrufescens AP and curdlan, although not significant (Figure 2D). Taken together, these data reveal that compounds from different mushrooms do affect DCs as revealed by the MHC-II and CD86 expression. However, CD11c and CD40 expression is much less affected. A. subrufescens is the most potent inducer of DC maturation.

Because of the variation in the extent of cytokine secretion and maturation, and notably the weak responses observed for H. tessellatus, L. edodes, and P. eryngii, we investigated whether or not mushroom compounds are recognized by DCs in all cases. To investigate this, Alexa Fluor 488-labeled mushroom compounds and FITC-labeled dextran as control were added to DCs. Because Alexa Fluor 488 hydrazide was used, it is expected that only polysaccharides and glycoproteins were labeled. Using flow cytometry we confirmed that labeled compounds from all mushrooms investigated as well as dextran bind DCs; however, the amounts vary considerably (Figure 3A). Using confocal microscopy, we confirmed that compounds of all mushrooms investigated are indeed endocytosed (Figure 3B). We next investigated endocytosis by DCs in more detail using A. subrufescens as the best responder and L. edodes as weak responder. The labeled mushroom compounds showed indeed to ultimately reside in the (endo-)lysosomes (Figure 3C). Both protein and glucan composition vary considerably as do the amounts that bind DCs (Figure 3A). Nevertheless, endocytosis was observed in all cases (Figure 3B). We therefore next investigated the role of CLRs as carbohydrate binding PRRs.

Because mushrooms contain a substantial amount of carbohydrates that may function as PAMP the involvement of carbohydrate-specific PRRs in endocytosis seems plausible. CLRs and CR3 are major endocytosis receptors as well as an important carbohydrate binding class of PRRs that function in a calcium-dependent manner. The addition of a chelating agent inhibits carbohydrate binding to CLRs and CR3. To investigate CLR/CR3 involvement, DCs were therefore pre-incubated with the chelating agent EGTA, whereupon CH of six mushroom species (A. subrufescens, H. tessellatus, L. edodes, P. eryngii, P. nameko, and P. ostreatus) was added. After 24 h, the TNF-α response was measured. EGTA inhibits the TNF-α response with 90–95% (Figure 4). The inhibition of the cytokine response by EGTA suggests that mushroom compounds do indeed target CLRs and/or CR3.

β-Glucans are polymeric carbohydrates assumed to play an important role in immune stimulation by mushrooms through the CLR dectin-1 (2, 6). Laminarin is a low molecular weight water-soluble β-glucan that binds dectin-1 without stimulating downstream signaling, yet blocking binding of other β-glucans to dectin-1 (19). Pre-incubation of DCs with laminarin inhibits the TNF-α response with 25–60% (Figure 4). Therefore, we conclude that dectin-1 binding of β-glucans plays an important role in the stimulation of DCs by mushrooms. Dectin-1 is, however, not the only carbohydrate binding PRR involved as can be inferred from the differences in inhibition by laminarin.

Since mushroom compounds are endocytosed, induce cytokine secretion, MHC-II expression and in some cases expression of CD40 and CD86 molecules, we next assessed the effect of mushroom-stimulated DCs on T cells. Thereto, T cells were added to DCs that were stimulated for 24 h with CH of different mushrooms. After 7 days, the presence of the T-cell cytokines IFN-γ, IL-4, and IL-17 was assessed using ELISA. The incubation of T cells with mushroom extracts for 7 days never resulted in the secretion of IFN-γ, IL-4, and IL-17 (data not shown). In the cocultures, detectable levels of IL-4 were never measured (data not shown). However, secretion of IL-17 and IFN-γ (Figure 5) was stimulated by A. subrufescens, G. frondosa, L. edodes, P. nameko, and P. ostreatus, although secretion of IL-17 induced by L. edodes was low. IFN-γ and IL-17 are the signature cytokines for a Th-1 and Th-17 response, respectively. These results suggest that mushrooms are capable of stimulating a mixed T-cell response, as seen for many pathogens. As mushrooms are (non-pathogenic) fungi, it was expected that a Th-17 response would prevail alike what has been shown for pathogenic fungi.

Although mushroom compounds can stimulate T cells through DCs, the question that remains is whether or not DCs process endocytosed mushroom compounds and present fragments to T cells via MHC-II molecules. To date, only the zwitterionic polysaccharide A of Bacteroides fragilis has been shown to activate T cells in a MHC-II dependent manner (20). Other polysaccharides stimulate the immune system in a T cell-independent manner. Since the preparations of mushroom compounds also contain protein, these may be presented to T cells via MHC-II molecules. If presentation through MHC-II molecules takes place, physical contact between DCs and T cells is required. An experiment was carried out wherein DCs and T cells were physically separated by using ThinCert™ membrane supports. DCs were stimulated with A. subrufescens CH for 24 h in a 24-well plate, when a ThinCert™ membrane support holding T cells was placed in the well. Simultaneously, DCs were incubated together with T cells in a 24-well plate. A. subrufescens was chosen for this experiment as this mushroom showed the strongest effect on DCs and T cells with regard to cytokine secretion as well as on the upregulation of the costimulatory molecules CD40 and CD86. After 7 days, the secretion of IFN-γ and IL-17 was determined using ELISA. Physical contact between DCs and T cells was not necessary to stimulate T cells to secrete IFN-γ and IL-17 (Figure 5C). Hence, fragments of mushroom proteins likely do not mediate the stimulation of T cells as this would be mediated by MHC-II molecules. This result implies that (polymeric) mushroom carbohydrates activate DCs resulting in cytokine secretion that stimulate T cells.