J Nat Med DOI 10.1007/s11418-014-0879-z
ORIGINAL PAPER
The TLR2 agonist in polysaccharide-K is a structurally distinct lipid which acts synergistically with the protein-bound b-glucan Kenneth Quayle • Catherine Coy Leanna Standish • Hailing Lu
•
Received: 28 May 2014 / Accepted: 19 November 2014 Ó The Japanese Society of Pharmacognosy and Springer Japan 2014
Abstract Protein-bound polysaccharide-K (Krestin; PSK) is a hot-water extract of Trametes versicolor with immune stimulatory activity. It has been used for the past 30 years and has demonstrated anti-tumor efficacy in multiple types of cancer. The ability of PSK to activate dendritic cells and T cells is dependent on its ability to stimulate Toll-like receptor 2 (TLR2), yet it remains unknown which structural component within PSK activates TLR2. The purpose of this study was to identify the TLR2 agonist within PSK and understand its role in the overall mechanism of PSK’s immunogenic activity. TLR2 activity was eliminated by treatment with lipoprotein lipase but not by trypsin or lyticase. Rapid centrifugation of PSK can separate the fraction with TLR2 agonist activity from the soluble b-glucan fraction. To study the potential interaction between the b-glucan component and the lipid component, we labeled the soluble b-glucan with fluorescein. Uptake of the labeled b-glucan by J774A macrophages and JAWSII dendritic cells was inhibited by anti-Dectin-1 antibody but not by anti-TLR2 antibody, confirming that Dectin-1 is the receptor for b-glucan. Interestingly, pre-treatment of JAWSII cells with the TLR2-active lipid fraction significantly enhanced the uptake of the soluble b-glucan, indicating the synergy between the TLR2 agonist component and the b-glucan component. Altogether, these results present evidence that PSK has two active components—the K. Quayle (&) C. Coy H. Lu Tumor Vaccine Group, Department of Medicine, University of Washington, 850 Republican Street, Seattle, WA 98109, USA e-mail:
[email protected] K. Quayle C. Coy L. Standish Bastyr University Research Institute, Bastyr University, 14500 Juanita Drive NE, Kenmore, WA 98028, USA
well-characterized protein-bound b-glucan and a previously unreported lipid—which work synergistically via the Dectin-1 and TLR2 receptors. Keywords Natural products Mushroom extracts Immunotherapy Toll-like receptors PSK Phagocytosis
Introduction Protein-bound polysaccharide-K (Krestin; PSK) is a hotwater extract of the Trametes versicolor mushroom (family Polyporaceae; also known as Coriolus versicolor or Polyporus versicolor, commonly known as Turkey tail) which is used as a prescription drug for cancer treatment in Japan due to its immunomodulatory effects [1, 2]. Specifically, PSK elicits anti-tumor activity by maturation of dendritic cells, activation of cytotoxic CD8? T cells and natural killer cells [3], and production of pro-inflammatory cytokines such as tumor necrosis factor alpha (TNF-a) and interleukins such as IL-1b and IL-12 [4]. As is the case with most natural products, PSK is not a pure compound and its precise mechanism of action on the molecular level is not fully understood. The material is a granular brown solid which is isolated by ammonium sulfate precipitation after mycelia are extracted with hot water [5, 6]. The main constituent of this solid is a mixture of large protein-bound b-glucan polysaccharides with an average molecular weight of *100 kDa [7] as determined by sizeexclusion chromatography and multi-angle laser light scattering. The polysaccharides consist mainly of b-(1,3), b(1,4), and b-(1,6)-glucopyranosyl residues with smaller amounts of mannose, galactose, xylose, and fucose. The protein portion is covalently linked to the polysaccharide and constitutes 25–38 % of the material by weight [8].
123
J Nat Med
A key finding in the study of PSK’s mechanism of action demonstrated that PSK is an agonist of Toll-like receptor 2 (TLR2) but none of the other known TLRs [3]. The TLRs represent a family of pattern recognition receptors (PRRs) expressed in immune cells such as macrophages and dendritic cells which recognize pathogen-associated molecular patterns (PAMPs) and trigger immune responses [9–11]. In addition to PSK, TLR2 agonist activity has been reported in other mushroom extracts [12, 13]. The major b-glucan component of PSK and other mushroom extracts are known to activate the PRR Dectin-1, the b-glucan receptor [14–16]. However, the component within PSK and other mushroom extracts which activates TLR2 remains to be identified. Immune-enhancing mushroom extracts have been popular remedies in Asia for centuries and are gaining popularity as dietary supplements in the Western world [17, 18]. The primary obstacles preventing their integration into mainstream medicine are the lack of a clearly identifiable single active ingredient and their nebulous mechanism of action [17, 19]. Therefore, the compound or structural motif responsible for each mode of bioactivity must be identified and its role in the overall mechanism of action must be understood for mushroom products to gain acceptance in Western medicine. Here we report evidence that the TLR2 agonist within PSK is not the known protein-bound b-glucan, but a previously unreported lipid or fatty acid ester. We also show evidence that phagocytosis of the protein-bound b-glucan is dependent on Dectin-1, but can be facilitated by the lipid component. The two ligands may thus act synergistically to generate an immune response.
Experimental procedures Cell cultures and reagents TLR2-expressing HEK293 cells (HEK-Blue hTLR2) were acquired from Invivogen (San Diego, CA, USA) and cultured in Dulbecco’s modified Eagle’s medium (DMEM) containing 10 % v/v fetal bovine serum (FBS), 0.1 mg/mL streptomycin, 100 U/mL penicillin, and 10 lg/mL blasticidin S. J774A.1 mouse macrophage cells were acquired from ATCC (Manassas, VA, USA) and cultured in DMEM containing 10 % v/v FBS, 0.1 mg/mL streptomycin, and 100 U/mL penicillin. JAWSII mouse immature dendritic cells were acquired from ATCC and cultured in alpha minimal essential medium (a-MEM) containing 20 % v/v FBS, 4 mM L-glutamine, 1 ng/mL GM-CSF, 0.1 mg/mL streptomycin and 100 U/mL penicillin. All cell cultures were maintained at 37 °C, 5 % CO2, and [95 % relative humidity. PSK was acquired from Kureha Corporation (Tokyo, Japan) as a granular solid in 1 g dose packets.
123
Lyticase digestion of PSK A 5 mg/mL stock solution of lyticase from Arthrobacter luteus (Sigma-Aldrich, 1330 units/mg solid) was prepared in sterile deionized (DI) water. PSK lot 63A was dissolved in sterile 19 phosphate buffered saline (PBS) at 5.08 mg/ mL, and 3 lL of lyticase stock solution per mg of PSK was added to afford a final concentration of 5 mg/mL PSK and 100 units/mL of lyticase. The mixture was shaken at 37 °C for 24 h, after which the enzyme was inactivated by heating to 70 °C for 10 min. The extent of TLR2 activation was measured in HEK-Blue hTLR2 cells as reported previously [3]. Lipoprotein lipase digestion of PSK A stock solution containing 600 units/mL of lipoprotein lipase from Pseudomonas sp. (Sigma-Aldrich, 1293 units/ mg solid) was prepared in sterile 19PBS. PSK lot 63A was dissolved in sterile 19PBS at 5.56 mg/mL, and 20 lL of lipoprotein lipase stock solution per mg of PSK was added to afford a final concentration of 5 mg/mL PSK and 60 units/mL of lipoprotein lipase. The mixture was shaken at 37 °C for 24 h, after which the enzyme was inactivated by heating to 70 °C for 10 min. The extent of TLR2 activation was measured in HEK-Blue hTLR2 cells as reported previously [3]. Trypsin digestion of PSK A 1 mg/mL stock solution of trypsin from porcine pancreas (Sigma-Aldrich proteomics grade bioreagent, 14,131 units/ mg solid) was prepared in sterile filtered 0.1 M HCl. PSK lot 63A was dissolved in sterile 19PBS at 5.13 mg/mL, and 5 lL of trypsin stock solution per mg of PSK was added to afford a final concentration of 5 mg/mL PSK and 85 units/mL of trypsin. The mixture was shaken at 37 °C for 24 h, after which the enzyme was inactivated by heating to 70 °C for 10 min. The extent of TLR2 activation was measured in HEK-Blue hTLR2 cells as reported previously [3]. Centrifugal fractionation of PSK PSK lot 11A (1.04 g) was reconstituted in 40 mL of sterile DI water and centrifuged at 3500g for 5 min. The clear, brown supernatant was removed and lyophilized to afford F1 as a solid. A fresh 40 mL portion of sterile DI water was added and vortexed until no visible solids remained. The mixture was centrifuged again at 3500g for 5 min and the clear supernatant was discarded. This process was repeated 2 more times. The off-white, gel-like wet solid was carefully removed (excluding the dark brown pellet at the
J Nat Med
bottom of the tube) and lyophilized to afford F2 as a solid. The remaining dark brown pellet was suspended in 5 mL of sterile DI water and lyophilized to afford F3 as a solid. Identification of fatty acids by derivatization and GC–MS PSK lot 63A was ground to a fine powder with a mortar and pestle, and 12.7 mg was suspended in 5 mL of anhydrous methanolic HCl (Sigma-Aldrich, 0.5 M). The suspension was heated to 65 °C for 1 h, and then quenched with 0.5 mL of DI water. Methyl esters were extracted with 3 mL of hexane. Half of the hexane phase (1.5 mL) was evaporated to dryness and taken up in 318 lL of hexane. This solution was analyzed on a Shimadzu QP2010 GC– MS with a Restek Rtx-225 fused silica column. The electron ionization mass spectrum of each peak in the GC trace was compared against the NIST 11 mass spectral library using the NIST mass spectral search program version 2.0 g. For each peak, a reference standard for the best matching compound was purchased from Sigma-Aldrich and a spiked sample was injected to verify co-elution. Preparation of PSK b-glucan fluorescein conjugate (F1-fluorescein) A 10 mg/mL stock solution of 5-(4,6-dichlorotriazinyl)aminofluorescein (5-DTAF, Life Technologies) was prepared in DMSO (Sigma-Aldrich, C99.5 %). PSK lot 63A (5.0 mg) was dissolved in 0.5 mL of sterile filtered 0.1 M pH 9.0 bicarbonate buffer. While gently vortexing, 50 lL of 5-DTAF solution was added dropwise and the mixture was shaken at 20 °C for 1 h. The entire reaction mixture was then filtered through a Zeba gel filtration spin column (Pierce, 7 kDa molecular-weight cut-off). The filtrate was then rapidly centrifuged to remove any remaining solids. Confocal microscopy J774A.1 mouse macrophage cells were allowed to adhere to 0.5 mm glass coverslips overnight, then 50 lg/mL of F1-fluorescein was added to the medium for 30 min at 37 °C. The cells were rinsed with sterile 19PBS and a chase of plain medium was added for 30 min or 24 h at 37 °C. The cells were then stained with LysoTracker Red (Life Technologies), fixed with 4 % p-formaldehyde in 19PBS, counterstained with DAPI, and mounted in sterile filtered glycerol containing 10 % v/v PBS. Images were acquired on a Nikon A1R confocal microscope with a 609 oil immersion objective.
TLR2/Dectin-1 blocking and flow cytometry JAWSII or J774A.1 cells were harvested by gentle scraping and transferred to test tubes (105 cells/tube). The tubes were centrifuged at 310g for 5 min and the medium discarded, then the cells were suspended in appropriate complete medium containing 10 lg/mL of rat anti mouse Dectin-1 (AbD Serotec, low endotoxin) or mouse antihuman/mouse TLR2 (eBioscience, Purified) for 1 h at 37 °C. F1-fluorescein (33 lg/mL) was then added to the medium and incubated for 30 min at 37 °C. Cells were rinsed three times with FACS buffer (19PBS, 1 % v/v FBS, 0.1 % w/v NaN3), suspended in 0.5 mL FACS buffer, and analyzed on a BD FACS Canto II flow cytometer.
Results The TLR2 agonist activity of PSK was significantly decreased by treatment with lipoprotein lipase To identify the specific chemical moiety or moieties responsible for TLR2 activation, PSK was treated with a panel of hydrolytic enzymes: trypsin, which specifically hydrolyzes peptide bonds on the carboxyl side of lysine and arginine residues; lyticase, which specifically hydrolyzes the b-(1,3)-glycosidic bonds that constitute the backbone of yeast and fungal b-glucan polymers; and lipoprotein lipase, which hydrolyzes fatty acid esters such as triglycerides. The potency of PSK as a TLR2 agonist was compared before and after 24 h digestion using the HEK-Blue hTLR2 system. As shown in Fig. 1a, b, at the dose of 100 lg/mL a significant difference between the untreated control and digested samples was observed [F(3,7) = 43.16, P \ 0.0001]. After applying Dunnett’s multiple comparisons test, the mean TLR2 activity in PSK decreased from 2.205 to 0.4277 (P \ 0.0001), or 81 %, following lipoprotein lipase treatment but only to 1.397 (P = 0.0059), or 37 %, following lyticase treatment. The difference in mean TLR2 activity was not significant (P = 0.9930) following trypsin treatment. These results indicate that the TLR2 activity of PSK is mainly mediated by a lipid component. PSK contains lipids which are primarily composed of linoleic acid To identify the fatty acids present in the lipid components of PSK, the material was treated with anhydrous methanolic HCl to convert all lipids into their fatty acid methyl ester (FAME) derivatives. Six peaks were present in the GC trace after derivatization, out of which five were positively identified as FAMEs (Table 1; Fig. 2). Linoleic acid
123
J Nat Med
A
Table 1 PSK fatty acid constituents Fatty acid derivative
Retention time (min)
NIST 11 library match (%)
Fatty acid/g PSK (mg)
Fatty acid (wt %)
Methyl myristate
7.9
1.6
0.08
0.01
Methyl pentadecanoate
9.6
53.9
0.09
0.01
Methyl palmitate
12.0
79.0
0.27
0.03
Methyl stearate
16.6
74.9
0.17
0.02
Methyl (Z,Z)-9,12octadecadienoate (methyl linoleate)
17.6
37.4
1.21
0.12
1.81
0.18
Total
B activity of F2 (off-white, gel-like solid) was 3.7 times more potent than control (unmodified PSK), 19.8 times more potent than F1 (soluble portion), and 3.8 times more potent than F3 (brown pellet). The soluble b-glucan component of PSK is internalized by macrophages and localizes in lysosomes
Fig. 1 TLR2 activity of PSK lot 63A decreased by 83 % after 24 h lipoprotein lipase digestion. a Representative dose–response curves plotted with Graph Pad Prism 5.0. Bars represent mean and standard error of triplicate wells. b Mean and standard error of three independent experiments at 100 lg/mL dose. A significant difference between the columns was found using one-way ANOVA [F(3,7) = 43.16, P \ 0.0001]. The P value summaries reported above each column are the adjusted P values from Dunnett’s multiple comparisons post-hoc test (ns P [ 0.05, **P B 0.01, ***P B 0.001)
was the predominant fatty acid, with minor amounts of palmitic acid, stearic acid, pentadecanoic acid, and myristic acid. The total fatty acid content in PSK is approximately 0.2 % by weight, as determined by interpolation of a threepoint calibration with reference standards of FAME derivatives of each fatty acid listed above. PSK can be separated into three fractions with the bulk of TLR2 activity in the insoluble fraction An aqueous solution of PSK was separated into three visually distinct fractions (F1–3, Fig. 3a, b) by rapid centrifugation. Each fraction was lyophilized and tested independently for TLR2 activity on a dry weight basis. As shown in Fig. 3c, d, at the dose of 64 lg/mL the TLR2
123
The F1 fraction is positive in the phenol/H2SO4 assay, indicating that it is the major protein-bound b-glucan component. This soluble glucan fraction was isolated by rapid centrifugation and conjugated to a fluorophore to determine whether it localizes in the lysosomes of immune cells in a manner consistent with phagocytosis. After dye conjugation, F1 was passed through a gel filtration cartridge to remove any unbound dye as well as any smaller molecular weight components that were possibly also labeled, such as polyphenols. Large aggregates such as melanin that were not removed by the cartridge were removed by rapid centrifugation. J774A.1 cells imaged after a 30-min incubation showed little colocalization between the F1-fluorescein (green) channel and LysoTracker (red) channel, with Pearson’s coefficient = 0.0370 (Fig. 4). Cells imaged after 24 h incubation showed good co-localization, with Pearson’s coefficient = 0.3374. Pearson’s coefficient measures the linearity of the two channels’ intensities, where 1 denotes a perfect linear relationship, -1 a perfect inverse relationship, and 0 no correlation [20]. These results indicate that the soluble glucan fraction of PSK is localized in the lysosome after phagocytosis by macrophages. Cellular uptake of the soluble b-glucan component of PSK is dependent on Dectin-1 receptor and facilitated by the TLR2 agonist component To determine whether the uptake observed by confocal microscopy was dependent on either TLR2 or Dectin-1,
methyl arachidate
methyl (Z,Z) 9,12-octadecadienoate
PSK spiked with FAME mix
methyl stearate
C
methyl (E,E) 9,12-octadecadienoate
FAME reference standard mix
methyl palmitate
B
methyl myristate
PSK fatty acid methyl esters
methyl dodecanoate
A
methyl pentadecanoate
J Nat Med
retention time (min)
Fig. 2 Four fatty acid methyl esters were identified by GC-MS after derivatization of PSK. a GC chromatograph of PSK lot 63A after reflux with 0.5 M methanolic HCl and extraction with hexane.
b Mixture of commercially available FAME reference standards at approximately 50 lg/mL each. c PSK lot 63A spiked with FAME reference standard mixture
immune cells were incubated with monoclonal antibodies against either TLR2 or Dectin-1 prior to F1-fluorescein exposure. Uptake of the fluorophore was then observed by flow cytometry. In all cases, two populations of cells were observed: one with relatively low fluorescein intensity and one with high intensity (Fig. 5). In both J774A.1 (macrophage) and JAWSII (dendritic) cell lines, Dectin-1 blocking caused a significant decrease in the high-intensity population, confirming that the uptake of F1 fraction is dependent on Dectin-1, the b-glucan receptor [21]. Interestingly, pre-treatment of the cells with anti-TLR2 antibody caused a small increase in uptake of the fluorescent b-glucan (Fig. 5). To further investigate the potential collaboration between the TLR agonist component and the b-glucan component, we pre-treated the JAWSII cells with increasing doses of the TLR2-active fraction F2 prior to treatment with F1-fluorescein. As shown in Fig. 6, there
was a dose-dependent increase in uptake of F1-fluorescein after treatment with the TLR2 agonist component, indicating that this component within PSK may facilitate the uptake of the soluble glucan.
Discussion It is now recognized that the immune system plays an active role in controlling cancer growth [22, 23]. With the recently demonstrated clinical success in cancer immunotherapy [24, 25], natural products with immune modulatory potential are also gaining attention. Medicinal mushrooms have been widely used for immune-enhancing effects, so it is critically important to understand their mechanism of action. The results reported here demonstrated that the TLR2 agonist activity within PSK is due to a previously
123
J Nat Med Fig. 3 TLR2 activity of F2 after centrifugation was 3.7 times greater than PSK lot 11A control. a Schematic depiction of the three PSK fractions F1–3. b Actual appearance of a PSK solution in DI water after centrifugation. c Representative dose–response curves plotted with Graph Pad Prism 5.0. Bars represent mean and standard error of triplicate wells. d Mean and standard error of three independent experiments at 64 lg/mL dose. Each fraction F1–3 was analyzed for TLR2 activation independently and compared to a control sample of PSK that was not separated, using the two-tailed Student’s t test (ns P [ 0.05, *P B 0.05, **P B 0.01, ***P B 0.001)
A
C
reconstitute, spin
F1
PSK powder F2 F3
D
B
30 min F1-fluorescein
lysotracker red
overlay
colocalization
24 hr Fig. 4 F1-fluorescein co-localizes with LysoTracker Red in J774A.1 macrophage cells after 24 h incubation. Cells were imaged with a Nikon A1R confocal microscope with a Plan Apo 609 oil immersion
123
objective. Images shown are maximum intensity projections of 10 optical sections with 0.4 lm step size created using ImageJ. The colocalization channel (far right) was generated using Imaris
J Nat Med
Fig. 5 Phagocytosis of F1-fluorescein by JAWSII dendritic cells and J774A.1 macrophage cells is inhibited by saturation of Dectin-1 but not TLR2. Cells were incubated with 10 lg/mL of anti-Dectin-1 or TLR2 antibody for 1 h before exposure to F1-fluorescein. The highintensity population represents uptake of F1-fluorescein and the low-
intensity population represents no uptake. Data were analyzed using FlowJo 9.6. Each histogram (top row) represents 20,000 analyzed events. Plots of side scatter (SSC) versus forward scatter (FSC) (bottom row) show the gates used to exclude dead cells and aggregates from the analysis
unrecognized lipid component distinct from the soluble bglucan, yet facilitates the uptake of b-glucan. This finding significantly improved our understanding of the immune modulatory mechanism of PSK. Although the TLR2 agonist activity in mushroom has been found in extracts from different species of medicinal mushrooms [12, 13, 26], it remained unknown whether the activity was mediated by b-glucan or other components. We originally hypothesized that the tertiary structure of the soluble 100-kDa protein-bound b-glucan was responsible for the TLR2-stimulating activity of PSK, as it is the only biologically active compound found in PSK that has been reported. Glucan polymers with (1?6)-b-glucopyranosyl branches depend on a specific triple helix conformation [27–29]. We studied how the TLR2 activity changed in response to enzymatic digestion of the (1?3)-b-glucan or
protein fragments individually. Since the majority of naturally occurring TLR2 agonists are lipoproteins or atypical lipopolysaccharides, we also studied the effect of digestion with lipoprotein lipase, which specifically hydrolyzes fatty acid esters to release free fatty acids. The nearly complete loss of TLR2 activity following lipoprotein lipase treatment led to the conclusion that the TLR2 agonist is not the major protein-bound b-glucan component but a trace lipidcontaining component. The presence of a lipid component or components was confirmed by lipid derivatization and GC–MS analysis of the FAMEs in PSK. Although lipids containing linoleic acid esters are present in the fruiting bodies of many common mushrooms [30], the discovery of linoleic acid and other trace fatty acids in PSK was surprising because PSK is prepared by extracting the mycelium of Trametes versicolor with hot water and purification
123
J Nat Med F1-fluorescein only (control)
100 µg/mL
1000 µg/mL
500
500
400
400
400
400
300
300
300
200
200
3
0
10
3
10
4
10
5
-10
fluorescein intensity
3
0
10
3
10
4
10
5
300 200 100
0
0 -10
200 100
100
0
Count
500
Count
500
100
-10
fluorescein intensity
3
0
10
3
10
4
10
0
5
-10
fluorescein intensity 250K
200K
200K
200K
200K
100K 50K
150K 100K
50K 100K 150K 200K 250K
FSC-A
100K
0
50K 100K 150K 200K 250K
FSC-A
0
10
3
10
4
10
5
150K 100K 50K
0
0 0
150K
50K
50K
0
SSC-A
250K
SSC-A
250K
150K
3
fluorescein intensity
250K
SSC-A
SSC-A
F1-fluorescein + F2
33 µg/mL
Count
Count
10 µg/mL
0 0
50K 100K 150K 200K 250K
FSC-A
0
50K 100K 150K 200K 250K
FSC-A
Fig. 6 Phagocytosis of F1-fluorescein by JAWSII cells is facilitated by F2 in a dose-dependent fashion. PSK lot 63A was suspended in sterile DI water at 10 mg/mL and centrifuged at 3500g for 10 min. The supernatant was removed and solids rinsed with an equal volume of sterile DI water and spun again. The solids were suspended in an equal volume of sterile DI water and added to cell cultures at the concentrations indicated for 30 min before exposure to F1-
fluorescein. The high-intensity population represents uptake of F1fluorescein and the low-intensity population represents no uptake. Data were analyzed using FlowJo 9.6. Each histogram (top row) represents 20,000 analyzed events. Plots of side scatter (SSC) versus forward scatter (FSC) (bottom row) show the gates used to exclude dead cells and aggregates from the analysis
of the protein-bound b-glucan by ammonium sulfate precipitation. Our attempts to isolate this trace TLR2-active component chromatographically were unsuccessful, but did lead to the observation that PSK is not completely water-soluble, contrary to what is regularly reported in the literature. Rapid centrifugation of an aqueous PSK solution reveals considerable heterogeneity, with a pellet of dark brown solid and a layer of tan, gel-like material above the clear, brown solution. When the supernatant was removed the remaining solids did not dissolve in 100 equivalents (v/v) of DI water even after extensive vortexing and heating to 90 °C, which confirms that the solids are chemically distinct and not simply the result of a saturated solution. Thus, three distinct fractions were isolated. The majority of TLR2 activity was found in the gel-like intermediate fraction (F2). The soluble fraction (F1) was almost completely inactive compared to unmodified PSK. We then asked: what exactly is the role of the soluble protein-bound b-glucan component if it is not the TLR2 agonist? According to the traditional model of antigen presentation, cells such as macrophages and dendritic cells recognize pathogens via interaction with extracellular PRRs such as TLR2. The pathogen is then engulfed via phagocytosis and
digested in a lysosome. The cell then presents an epitope fragment to T cells of the adaptive immune system. We were able to confirm by confocal microscopy that the dyeconjugated soluble b-glucan was internalized by J774A.1 cells (as opposed to surface binding only) and that it colocalized with a lysosome-specific probe after overnight incubation. This observation is consistent with phagocytosis of the soluble b-glucan by the innate immune system. While it is well known that Dectin-1 is the major bglucan receptor [15, 21, 31] and it has been shown that its activation controls maturation of b-glucan-containing phagosomes [32], it has been recently reported that particulate but not soluble b-glucans activate Dectin-1 [33]. Therefore, we studied whether or not phagocytosis of the PSK-soluble b-glucan was dependent on Dectin-1. We observed by flow cytometry a significant decrease in the amount of dye-conjugated b-glucan (F1-fluorescein) taken up by both J774A.1 macrophage and JAWSII dendritic cell cultures which were previously saturated with a Dectin-1specific antibody. This result suggests that phagocytosis is indeed dependent on Dectin-1. Surprisingly, we also observed a slight increase in uptake when identical cell cultures were pre-incubated with a TLR2-specific antibody. We theorized that this might be
123
J Nat Med
due to the potential synergy between TLR2 and Dectin-1: stimulation of TLR2 primes the innate immune system towards phagocytosis of the primary protein-bound b-glucan antigen via Dectin-1. It is known that TLR2 and Dectin-1 work in tandem to elicit an immunogenic response to zymosan [34], and synergistic effects on cytokine production have been observed when the two receptors are differentially stimulated [35, 36]. In a subsequent flow cytometry experiment, we observed a dosedependent increase in dye-conjugated b-glucan uptake when the insoluble fraction of PSK (not dye-conjugated) was re-introduced. These findings present evidence for a synergistic effect between the known protein-bound b-glucan component which interacts with the immune system via Dectin-1 and a chemically distinct lipid which stimulates TLR2. Under this paradigm, reduced uptake of the protein-bound bglucan would be expected under TLR2 knockdown conditions, which represents an interesting angle for further study. Though the exact structure of the TLR2 agonist is beyond the scope of this report, efforts to isolate and elucidate its structure are currently underway. Synergy between PSK and TLR4-active endotoxin (lipopolysaccharide) has been demonstrated [37]. This is interesting and relevant to the synergy between the TLR2 component and polysaccharide component reported here. Of note, PSK does not have TLR4 agonist activity, as reported in our previous publication [3]. We believe these findings have implications not only for PSK but for a broad spectrum of mushroom-derived immunopotentiating natural products. TLR2 is known to be involved in the biological response to other mushroom extracts. For example, extracts of Agaricus brasiliensis fruiting bodies were shown to contain TLR2 ligands which were responsible for IL-6 production in splenocytes [13]. It is important to note that PSK was selected for this study not because it is unique in its variable TLR2 activity, but rather because it is among the best immunomodulatory mushroom products available. It is one of the few mushroom products approved as a pharmaceutical-grade medication anywhere in the world and has been used for the past 30 years with consistent clinical efficacy. Furthermore, it and the closely related PSP (polysaccharide peptide) have a wealth of published clinical data [38–45] dating back to 1981. The logical next step is to elucidate the exact structure of the lipid TLR2 agonist and determine its source. One hypothesis is that the agonist is a lipoprotein or lipopolysaccharide due to an endosymbiotic bacterium. In this scenario, the bacterium would be an integral part of the Trametes versicolor organism, not a contaminant. It has been shown by Pasco et al. that variation in macrophage activation by immune-enhancing botanical preparations is largely due to such bacterial components [46, 47].
Of note, the synergy between TLR2 and Dectin-1 in stimulating immunity has only been recognized within the last decade [36, 48]. Interestingly, medicinal mushrooms have been used for thousands of years for their immuneenhancing potential [39, 49]. Our results indicate that natural products and alternative medicine are areas that need continuous investigation, because there might be a strong scientific rationale for the traditional remedy that remains unrecognized. Once the identity and source of the agonist is discovered, the next challenge is to understand how its expression depends on every stage of the production process for a given natural product: from cultivation conditions to harvest to extraction method. This is no simple task, but it is necessary if immunogenic mushroom products are to gain recognition as anything other than dietary supplements in Western medicine. Acknowledgments This work was supported by NIH NCCAM grant 3U19AT006028-02S1 and a faculty seed grant from Bastyr University Research Institute.
References 1. Tsukagoshi S, Hashimoto Y, Fujii G, Kobayashi H, Nomoto K, Orita K (1984) Krestin (PSK). Cancer Treat Rev 11:131–155 2. Maehara Y, Tsujitani S, Saeki H, Oki E, Yoshinaga K, Emi Y, Morita M, Kohnoe S, Kakeji Y, Yano T, Baba H (2012) Biological mechanism and clinical effect of protein-bound polysaccharide K (KRESTINÒ): review of development and future perspectives. Surg Today 42:8–28 3. Lu H, Yang Y, Gad E, Wenner CA, Chang A, Larson ER, Dang Y, Martzen M, Standish LJ, Disis ML (2011) Polysaccharide Krestin is a novel TLR2 agonist that mediates inhibition of tumor growth via stimulation of CD8 T cells and NK cells. Clin Cancer Res 17:67–76 4. Engel A L, Sun G-C, Gad E, Rastetter L R, Strobe K, Yang Y, Dang Y, Disis M L, Lu H (2013) Protein-bound polysaccharide activates dendritic cells and enhances OVA-specific T cell response as vaccine adjuvant. Immunobiology 218(12):1468-76 5. Hotta T, Enomoto S, Yoshikumi C, Ohara M, Ueno S (1981) Protein-bound polysaccharides. Kureha Chemical Industry Co. Ltd, Japan 6. Ohtsuka S, Ueno A, Yoshikumi C, Hirose F, Omura Y, Wada T, Fujii T, Takahashi E (1971) Polysaccharides with anticancer activity. Kureha Chemical Industry Co. Ltd, European patent office 7. Muller A, Pretus HA, McNamee RB, Jones EL, Browder IW, Williams DL (1995) Comparison of the carbohydrate biological response modifiers Krestin, schizophyllan and glucan phosphate by aqueous size exclusion chromatography with in-line argon-ion multi-angle laser light scattering photometry and differential viscometry detectors. J Chromatogr B Biomed Appl 666:283–290 8. Ren L, Perera C, Hemar Y (2012) Antitumor activity of mushroom polysaccharides: a review. Food Funct 3:1118–1130 9. Takeda K, Kaisho T, Akira S (2003) Toll-like receptors. Ann Rev Immunol 21:335–376 10. Engel AL, Holt GE, Lu H (2011) The pharmacokinetics of Tolllike receptor agonists and the impact on the immune system. Expert Rev Clin Pharmacol 4:275–289 11. Pasare C, Medzhitov R (2004) Toll-like receptors and acquired immunity. Semin Immunol 16:23–26
123
J Nat Med 12. Sheu F, Chien PJ, Hsieh KY, Chin KL, Huang WT, Tsao CY, Chen YF, Cheng HC, Chang HH (2009) Purification, cloning, and functional characterization of a novel immunomodulatory protein from Antrodia camphorata (bitter mushroom) that exhibits TLR2-dependent NF-kappaB activation and M1 polarization within murine macrophages. J Agric Food Chem 57:4130–4141 13. Yamanaka D, Motoi M, Ishibashi K, Miura NN, Adachi Y, Ohno N (2012) Effect of Agaricus brasiliensis-derived cold water extract on Toll-like receptor 2-dependent cytokine production in vitro. Immunopharmacol Immunotoxicol 34:561–570 14. Brown GD, Herre J, Williams DL, Willment JA, Marshall ASJ, Gordon S (2003) Dectin-1 mediates the biological effects of bglucans. J Exp Med 197:1119–1124 15. Fang J, Wang Y, Lv X, Shen X, Ni X, Ding K (2012) Structure of a beta-glucan from Grifola frondosa and its antitumor effect by activating dectin-1/Syk/NF-kappaB signaling. Glycoconj J 29:365–377 16. Qi C, Cai Y, Gunn L, Ding C, Li B, Kloecker G, Qian K, Vasilakos J, Saijo S, Iwakura Y, Yannelli JR, Yan J (2011) Differential pathways regulating innate and adaptive antitumor immune responses by particulate and soluble yeast-derived beta-glucans. Blood 117:6825–6836 17. Sullivan R, Smith JE, Rowan NJ (2006) Medicinal mushrooms and cancer therapy: translating a traditional practice into Western medicine. Perspect Biol Med 49:159–170 18. Normile D (2003) Asian medicine. The new face of traditional Chinese medicine. Science 299:188–190 19. Wasser SP (2011) Current findings, future trends, and unsolved problems in studies of medicinal mushrooms. Appl Microbiol Biotechnol 89:1323–1332 20. Manders EMM, Verbeek FJ, Aten JA (1993) Measurement of colocalization of objects in dual-colour confocal images. J Microsc 169:375–382 21. Brown GD, Taylor PR, Reid DM, Willment JA, Williams DL, Martinez-Pomares L, Wong SY, Gordon S (2002) Dectin-1 is a major beta-glucan receptor on macrophages. J Exp Med 196:407–412 22. Schreiber RD, Old LJ, Smyth MJ (2011) Cancer immunoediting: integrating immunity’s roles in cancer suppression and promotion. Science 331:1565–1570 23. Galon J, Costes A, Sanchez-Cabo F, Kirilovsky A, Mlecnik B, Lagorce-Pages C, Tosolini M, Camus M, Berger A, Wind P, Zinzindohoue F, Bruneval P, Cugnenc PH, Trajanoski Z, Fridman WH, Pages F (2006) Type, density, and location of immune cells within human colorectal tumors predict clinical outcome. Science 313:1960–1964 24. Wolchok JD, Kluger H, Callahan MK, Postow MA, Rizvi NA, Lesokhin AM, Segal NH, Ariyan CE, Gordon RA, Reed K, Burke MM, Caldwell A, Kronenberg SA, Agunwamba BU, Zhang X, Lowy I, Inzunza HD, Feely W, Horak CE, Hong Q, Korman AJ, Wigginton JM, Gupta A, Sznol M (2013) Nivolumab plus ipilimumab in advanced melanoma. N Engl J Med 369:122–133 25. Hamid O, Robert C, Daud A, Hodi FS, Hwu WJ, Kefford R, Wolchok JD, Hersey P, Joseph RW, Weber JS, Dronca R, Gangadhar TC, Patnaik A, Zarour H, Joshua AM, Gergich K, Elassaiss-Schaap J, Algazi A, Mateus C, Boasberg P, Tumeh PC, Chmielowski B, Ebbinghaus SW, Li XN, Kang SP, Ribas A (2013) Safety and tumor responses with lambrolizumab (anti-PD1) in melanoma. N Engl J Med 369:134–144 26. Lu H, Yang Y, Gad E, Wenner CA, Chang A, Larson ER, Dang Y, Martzen M, Standish LJ, Disis ML (2011) Polysaccharide Krestin is a novel TLR2 agonist that mediates inhibition of tumor growth via stimulation of CD8 T cells and NK cells. Clin Cancer Res 17:67–76 27. Falch BH, Espevik T, Ryan L, Stokke BT (2000) The cytokine stimulating activity of (1 ? 3)-b-D-glucans is dependent on the triple helix conformation. Carbohydr Res 329:587–596
123
28. Bohn JA, BeMiller JN (1995) (1 ? 3)-b-D-Glucans as biological response modifiers: a review of structure-functional activity relationships. Carbohydr Polym 28:3–14 29. Zhang L, Li X, Xu X, Zeng F (2005) Correlation between antitumor activity, molecular weight, and conformation of lentinan. Carbohydr Res 340:1515–1521 30. Gu¨nc¸ Ergo¨nu¨l P, Akata I, Kalyoncu F, Ergo¨nu¨l B (2013) Fatty acid compositions of six wild edible mushroom species. ScientificWorldJournal 2013:163964 31. Brown GD, Gordon S (2001) Immune recognition. A new receptor for beta-glucans. Nature 413:36–37 32. Mansour MK, Tam JM, Khan NS, Seward M, Davids PJ, Puranam S, Sokolovska A, Sykes DB, Dagher Z, Becker C, Tanne A, Reedy JL, Stuart LM, Vyas JM (2013) Dectin-1 activation controls maturation of beta-1,3-glucan-containing phagosomes. J Biol Chem 288:16043–16054 33. Qi C, Cai Y, Gunn L, Ding C, Li B, Kloecker G, Qian K, Vasilakos J, Saijo S, Iwakura Y, Yannelli J R, Yan J (2011) Differential pathways regulating innate and adaptive antitumor immune responses by particulate and soluble yeast-derived bglucans. Blood 117(25):6825-36 34. Brown GD, Gordon S (2003) Fungal b-glucans and mammalian immunity. Immunity 19:311–315 35. Timmermans K, Plantinga TS, Kox M, Vaneker M, Scheffer GJ, Adema GJ, Joosten LAB, Netea MG (2013) Blueprints of signaling interactions between pattern recognition receptors: implications for the design of vaccine adjuvants. Clin Vaccine Immunol 20:427–432 36. Gantner BN, Simmons RM, Canavera SJ, Akira S, Underhill DM (2003) Collaborative induction of inflammatory responses by dectin-1 and Toll-like receptor 2. J Exp Med 197:1107–1117 37. Pang Z-J, Zhou M, Chen Y, Wan J (1998) A protein-bound polysaccharide synergistic with lipopolysaccharide induces nitric oxide release and antioxidant enzyme activities in mouse peritoneal macrophages. Am J Chin Med 26:133–141 38. Toge T, Yamaguchi Y (2000) Protein-bound polysaccharide increases survival in resected gastric cancer cases stratified with a preoperative granulocyte and lymphocyte count. Oncol Rep 7:1157–1161 39. Kidd PM (2000) The use of mushroom glucans and proteoglycans in cancer treatment. Altern Med Rev 5:4–27 40. Kono K, Kawaguchi Y, Mizukami Y, Mimura K, Sugai H, Akaike H, Fujii H (2008) Protein-bound polysaccharide K partially prevents apoptosis of circulating T cells induced by anti-cancer drug S-1 in patients with gastric cancer. Oncology 74:143–149 41. Koda K, Miyazaki M, Sarashina H, Suwa T, Saito N, Suzuki M, Ogawa K, Watanabe S, Kodaira S, Nakazato H (2003) A randomized controlled trial of postoperative adjuvant immunochemotherapy for colorectal cancer with oral medicines. Int J Oncol 23:165–172 42. Shibata M, Nezu T, Kanou H, Nagata Y, Kimura T, Takekawa M, Ando K, Fukuzawa M (2002) Immunomodulatory effects of low dose cis-diaminedichloroplatinum (cisplatin) combined with UFT and PSK in patients with advanced colorectal cancer. Cancer Invest 20:166–173 43. Morimoto T, Ogawa M, Orita K, Sugimachi K, Toge T, Dohi K, Nomura Y, Monden Y, Ogawa N (1996) Postoperative adjuvant randomised trial comparing chemoendocrine therapy, chemotherapy and immunotherapy for patients with stage II breast cancer: 5-year results from the Nishinihon cooperative study group of adjuvant chemoendocrine therapy for breast cancer (ACETBC) of Japan. Eur J Cancer 32A:235–242 44. Nakazato H, Koike A, Saji S, Ogawa N, Sakamoto J (1994) Efficacy of immunochemotherapy as adjuvant treatment after curative resection of gastric cancer. Study group of immunochemotherapy with PSK for gastric cancer. Lancet 343:1122–1126
J Nat Med 45. Toi M, Hattori T, Akagi M, Inokuchi K, Orita K, Sugimachi K, Dohi K, Nomura Y, Monden Y, Hamada Y et al (1992) Randomized adjuvant trial to evaluate the addition of tamoxifen and PSK to chemotherapy in patients with primary breast cancer. 5-Year results from the Nishi–Nippon group of the adjuvant chemoendocrine therapy for breast cancer organization. Cancer 70:2475–2483 46. Pugh ND, Tamta H, Balachandran P, Wu X, Howell JL, Dayan FE, Pasco DS (2008) The majority of in vitro macrophage activation exhibited by extracts of some immune enhancing botanicals is due to bacterial lipoproteins and lipopolysaccharides. Int Immunopharmacol 8:1023–1032
47. Tamta H, Pugh ND, Balachandran P, Moraes R, Sumiyanto J, Pasco DS (2008) 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 56:10552–10556 48. Yadav M, Schorey JS (2006) The beta-glucan receptor dectin-1 functions together with TLR2 to mediate macrophage activation by mycobacteria. Blood 108:3168–3175 49. Fisher M, Yang LX (2002) Anticancer effects and mechanisms of polysaccharide-K (PSK): implications of cancer immunotherapy. Anticancer Res 22:1737–1754
123
