Aureobasidium pullulans produced β-glucan is effective to enhance Kurosengoku soybean extract induced Thrombospondin-1 expression

Black yeast, Aureobasidium pullulans is extracellularly produced β-(1,3), (1,6)-D-glucan (β-glucan) under certain conditions. In this study, using Glycine max cv. Kurosengoku (Kurosengoku soybeans), the production of β-glucan through fermentation of A. pullulans was evaluated, and the effects of A. pullulans cultured fluid (AP-CF) containing β-glucan made with Kurosengoku soybeans (kAP-CF) on a human monocyte derived cell line, Mono Mac 6 cells were investigated. Concentration of β-glucan in kAP-CF reached the same level as normal AP-CF. An anti-angiogenic protein, Thrombospondin-1 (THBS1) was effectively induced after the stimulation with kAP-CF for comparison with AP-CF. The THBS1 is also induced after stimulation with hot water extract of Kurosengoku soybeans (KS-E), while the combined stimulation of β-glucan with KS-E more effectively induced THBS1 than that with KS-E alone. These results suggest effects of A. pullulans-produced β-glucan on the enhancement of Kurosengoku soybean-induced THBS1 expression.

Kurosengoku soybeans activate type-1 immunity in a Toll-like receptor (TLR)4-and TLR2-dependent manner 22 . Activation of type-1 immunity is involved in the prevention of tumor 23 formation and allergic diseases 24 . These effects resemble the effects of β-glucan, and may be anticipated to be enhanced by the combined administration of A. pullulans together with Kurorengoku soybeans.
In the present study, to utilize the beneficial effects of Kurosengoku soybeans AP-CF, the production of β-glucan through fermentation of A. pullulans using Kurosengoku soybeans was evaluated, and the effects of the AP-CF produced using Kurosengoku soybeans (kAP-CF) on culture cells were investigated. The results show that A. pullulans also effectively produces β-glucan with Kurosengoku soybeans as a nitrogen source with similar efficacy as with conventional methods using rice bran. The AP-CF prepared with Kurosengoku soybeans (kAP-CF) exhibited significantly higher Thrombospondin-1 (THBS1) induction activity in the human monocyte-derived cell line Mono Mac 6 25 , higher than with AP-CF prepared by conventional methods. The THBS1 is known to be an endogenous angiogenic inhibitor, and is involved in the inhibition of tumor angiogenesis and growth 26,27 . Hot-water extracts of Kurosengoku soybeans (KS-E) have also indicated THBS1 induction activity, and the THBS1 induction after stimulation with KS-E was enhanced after co-stimulation with conventional AP-CF and purified β-glucans. The THBS1 induction activity of KS-E is mainly dependent on soy isoflavones, and the THBS1 induction activities of daidzein and genistein are higher than that of glycitein. These results suggest that β-glucan is effective in enhancing the effects of soy isoflavones on the cells.

Stimulation with Aureobasidium pullulans-cultured fluid (AP-CF) made with Kurosengoku soybeans (kAP-CF) effectively induces Thrombospondin-1 (THBS1) in Mono Mac 6 cells. Initially,
production of A. pullulans-fermented β-glucan using powdered Kurosengoku soybeans as the nitrogen source was investigated. Conventionally, A. pullulans-cultured fluid (AP-CF) containing β-glucan is prepared using rice bran as the nitrogen source. The concentration of β-glucan in AP-CF made with Kurosengoku soybeans as the nitrogen source (kAP-CF) was estimated to be 6 mg/ml, a level similar to conventional AP-CF produced with rice bran. The results indicate that the ground Kurosengoku soybeans may be substituted for and used as the nitrogen source in the fermentation of A. pullulans for β-glucan production.
Next, to assess the kAP-CF and compare it with conventional AP-CF made with rice bran, a human, acute monocytic leukemia derived cell line, Mono Mac 6 cells were stimulated with kAP-CF, and the expression of mRNAs was evaluated using primer arrays. The primer array analysis showed that there is significant expression of Thrombospondin-1 (THBS1) mRNA after stimulation with kAP-CF. As shown in Fig. 1A, the THBS1 mRNA expression was statistically significantly increased after stimulation with kAP-CF, with only weak THBS1 mRNA induction in conventionally stimulated AP-CF cells; the stimulation with hot water extracts of Kurosengoku soybeans (KS-E) also effectively induced THBS1 mRNA.
The kAP-CF contained 0.3% (W/V) powdered Kurosengoku soybeans, while KS-E was extracted with 30% (W/V) of Kurosengoku soybeans. These results indicate that the stimulation with kAP-CF is more effective in inducing THBS1 mRNA than KS-E, suggesting that the β-glucan in the AP-CF may be involved in the enhancement of THBS1 mRNA expression after stimulation with KS-E, and next the effects of β-glucan containing AP-CF on the enhancement of THBS1 mRNA expression after stimulation with KS-E were investigated. The results show that conventional AP-CF made without Kurosengoku soybean addition enhances the KS-E induced THBS1 mRNA expression (Fig. 1B). The induction of THBS1 mRNA after stimulation with AP-CF and KS-E are transient and peaked at 2 hours after the stimulation, suggesting that the induction of THBS1 mRNA is a primary response to the simulation. To confirm this, the THBS1 mRNA expression was analyzed with protein synthesis inhibited by a high concentration of cycloheximide (5 mM). Under this condition, new protein synthesis is almost completely blocked 28 . As shown in Fig. 1C, THBS1 mRNA was effectively induced after stimulation with KS-E, and the induction of THBS1 mRNA was enhanced after the combined stimulation with AP-CF. The results indicate that the THBS1 mRNA induction after stimulation with KS-E and the co-stimulation effects of AP-CF on THBS1 mRNA induction are the primary responses of the cells to the stimulations, and that new protein synthesis is not required for the THBS1 mRNA induction.
To confirm the induction of THBS1 after stimulation with KS-E at the protein level, the amount of THBS1 protein in the supernatant of the culture medium was monitored by ELISA. The results show that the THBS1 protein increased significantly after stimulation with KS-E, and the induction of THBS1 protein was statistically significantly enhanced after co-stimulation with AP-CF (Fig. 1D). These results show that the stimulation with KS-E induces THBS1 in Mono Mac 6 cells, and that AP-CF is involved in enhancing the THBS1 induction. β-glucan is involved in the enhancement of the THBS1 induction after stimulation with KS-E. To assess the THBS1 induction identified in Mono Mac 6 cells after stimulation in other cell lines, a human monocyte derived cell line, THP-1 cells 29 were used, and the responses to the stimulation with KS-E were investigated. As shown in Fig. 2A, stimulation with KS-E also effectively induces THBS1 mRNA expression in THP-1 cells, similar to that in the Mono Mac 6 cells. In addition, co-stimulation with AP-CF significantly enhanced THBS1 mRNA induction after the stimulation with KS-E.
β-glucan is the main component of AP-CF and thought to be the most important component providing most of the beneficial effects of AP-CF. To investigate whether the effects of AP-CF on the enhancement of the THBS1 induction after stimulation with KS-E depend on β-glucan, A. pullulans-produced purified β-glucan (AP-PG) was prepared from AP-CF, and the effects of β-glucan on KS-E induced THBS1 mRNA expression were examined. As shown in Fig. 2B, the results show that AP-CF and AP-PG, purified β-glucan from AP-CF, exhibit very similar activities after the enhancement by KS-E induced THBS1 mRNA. The results suggest that the enhancement of KS-E induced THBS1 expression depends on the β-glucan in AP-CF.
Kurosengoku soybeans are also ingested as a tea using roasted beans, and the activity of KS-E prepared from roasted beans as it affects the THBS1 mRNA induction was investigated. As shown in Fig. 2C, KS-E prepared from the roasted beans also induced THBS1 mRNA statistically significantly after the stimulation. Here, stimulation with KS-E prepared from the whole beans induced THBS1 mRNA more strongly than with KS-E prepared from ground Kurosengoku soybean powder. These results suggest that components contained in the shell or germ of the Kurosengoku soybeans may be involved in the induction of THBS1.

Isoflavones contained in Kurosengoku soybeans are involved in the induction of THBS1.
The isoflavones, saponin, and lecithin are soybean components where beneficial effects on health have been reported [10][11][12][13][14][15][16] . In addition to these components, the THBS1 induction activity of lunasin, a peptide derived from soybeans has been reported 32 . Lunasin, a peptide originally found in soybeans, has been shown to display antitumor effects through the inhibition of histone acetylation 33 . To determine the soybean component which is involved in the induction of THBS1, the expression of THBS1 mRNA in Mono Mac 6 cells after stimulation with purified products of these various components was investigated. The results show that the expression of THBS1 mRNA is significantly increased after stimulation with an isoflavone mixture isolated from soybeans (Fig. 3A), and THBS1 mRNA is weakly induced after stimulation with saponin. The involvement of lunasin in the induction of THBS1 in non-tumorigenic human prostate epithelial cells has been reported 32 , in the Mono Mac 6 cells the THBS1 expression is however not affected by stimulation with lunasin. These results suggest that isoflavones are the main components involved in the induction of THBS1.
The isoflavones used in this study were mixtures of isoflavones isolated from soybeans and included daidzein, glycitein, and genistein which are known to be the major components of isoflavones in soybeans. To establish which isoflavone is involved in the induction of THBS1, the induction activities of these isoflavones were monitored using purified genistein, glycitein, and daidzein. The results indicate that the THBS1 mRNA expression is significantly increased after stimulation with all of these three isoflavones (Fig. 3B). The THBS1 mRNA induction activity of glycitein is weaker than daidzein and genistein, and the induced activities of daidzein and genistein are at similar levels. These results suggest that genistein and daidzein are the main components involved in the THBS1 induction after stimulation with KS-E.
Isoflavones are contained in a number of bean varieties, the seeds of Fabaceae family plants 34 . To investigate whether the THBS1 induction also takes place with other Fabaceae family beans, a conventional soybean (Glycine max (L.) Merrill), the Adzuki bean (Vigna angularis (Willd.) Ohwi & H. Ohashi), and a cultivar of the white-colored runner bean, Shirohanamame (Phaseolus coccineus cv. Shirohanamame) were tested, and the THBS1 induction activities of hot water extracts prepared from these beans were measured. As shown in Fig. 3C, the THBS1 mRNA expression is significantly increased after stimulation with the hot water extracts of these Fabaceae family beans.
The kinases activated downstream of β-glucan receptors are involved in the enhancement of the THBS1 induction. Spleen tyrosine kinase (Syk) is known as a kinase which is activated downstream of the β-glucan receptors CR3 and dectin-1 35,36 . The p38 mitogen-activated protein kinase (p38 MAPK) and c-Jun N-terminal kinase (JNK) are known to be kinases which are activated downstream of the Syk-mediated signaling pathway 37 , and are involved in the cellular response to stimulation with β-glucan. To assess the involvement of these kinases in the induction of THBS1 after stimulation with KS-E, and the effect on the enhancement of the induction after co-stimulation with AP-CF, the effects of selective kinase inhibitors on the THBS1 mRNA induction were investigated. The results show that although the basal expression of THBS1 mRNA was increased after the treatment, the stimulation with KS-E induced THBS1 mRNA expression was not affected after the treatment with a Syk-selective inhibitor, piceatannol (Fig. 4A). However, the effect of co-stimulation with AP-CF on the enhancement of KS-E induced THBS1 mRNA expression was strongly inhibited by the piceatannol treatment. These results suggest that the Syk-mediated signaling pathway which is activated by β-glucan receptors is closely related in the AP-CF mediated enhancement of THBS1 induction after stimulation with KS-E.
Treatment with a selective inhibitor for p38 MAPK, SB203580, strongly inhibited THBS1 induction, while treatment with a selective inhibitor for JNK, SP600125 partially inhibited the induction of THBS1 mRNA expression after stimulation with KS-E ( Fig. 4B and C). The enhancement of the THBS1 mRNA expression after co-stimulation with AP-CF was strongly inhibited after treatment with both the selective inhibitors for p38 MAPK (SB203580) and JNK (SP600125). These results suggest that signaling molecules which are known to be activated downstream of β-glucan receptors are involved in the enhancement of THBS1 induction after stimulation with KS-E.

Discussion
The results of this study demonstrate that THBS1 is induced in a human monocyte derived cell line, Mono Mac 6 cells, after stimulation with KS-E, and this THBS1 induction is enhanced after co-stimulation with β-glucans including the β-glucan produced by the fermentation of A. pullulans. Several molecules have been reported as involved in recognition of β-glucans: CR3 (complement receptor 3) 38 , dectin-1 39 , langerin 40 , and lactosylceramide 41 . The data reported here using purified β-glucans demonstrates that the KS-E induced THBS1 expression is enhanced after co-stimulation with the purified β-glucans isolated from AP-CF, seaweed (laminarin), and Trametes versicolor (Krestin) (Fig. 2B). These β-glucans are structurally distinct 2,30,31 , and in particular laminarin is known to be an antagonist of dectin-1 that competitively inhibits activation of the dectin-1-mediated signaling pathway after stimulation with other β-glucans 42,43 . Therefore, the results here may suggest that the effects of β-glucan on the enhancement of THBS1 induction after stimulation with KS-E are involved in the dectin-1-independent signaling pathway.
The analysis using components isolated from soybeans indicates that the isoflavones contained in soybeans are involved in the THBS1 induction activity of KS-E (Fig. 3A). The THBS1 induction activity of isoflavone mixtures is clearly lower than that of KS-E, a crude extract of Kurosengoku soybeans (Fig. 3A). In addition, single compounds of soybean isoflavones: genistein, glycitein, and daidzein, show weaker THBS1 induction activities than that of the isoflavone mixture (Fig. 3B). The isoflavone mixture used in this study was extracted and purified from soybeans, while the genistein, glycitein, and daidzein were chemically synthesized products. The reason why the THBS1 induction activities of the purified components are weaker than that of the less fully described ones is not fully elucidated, however the observations here may suggest that other unidentified compound(s) contained in soybeans are involved in the induction of THBS1.
As THBS1 is known to be a protein exhibiting anti-angiogenic activity 26,27 , and anti-angiogenic activity has been established in the soybean isoflavone, genistein 44,45 . In addition, anti-angiogenic activity of soy saponins has also been reported 46 . As shown in Fig. 3A, saponin isolated from soybeans also indicated THBS1 induction activity although the activity was weaker than that of the isoflavones. Overall the results here indicating that the stimulation with KS-E induces THBS1 in Mono Mac 6 cells, could in part be explained to involve the mechanism of the anti-angiogenic activity of soybeans.
The results using specific kinase inhibitors indicate that Syk, the tyrosine kinase activated downstream of dectin-1 and the CR3 receptor 35,36 , is involved in the AP-CF-mediated enhancement activity of the KS-E induced THBS1 expression (Fig. 4A). However, the Syk-mediated signaling pathway is not involved in the induction of THBS1 after stimulation with KS-E. These results support the hypothesis that a β-glucan receptor-mediated pathway would be involved in the enhancement of KS-E induced in the THBS1 expression. Further, these results also suggest that activation of a different pathway from the β-glucan receptor-mediated signaling pathway is involved in the KS-E mediated THBS1 induction. The results using specific inhibitors for p38 MAPK, SB203580, and for JNK, SP600125 indicate that the KS-E induced THBS1 mRNA expression is mainly dependent on p38 MAPK, and both kinases are involved in the enhancement of the KS-E induced THBS1 mRNA expression ( Fig. 4B and C). JNK is known to be a stress induced kinase activated by various stress stimuli including cycloheximide treatment 28 . This may explain the reason why the KS-E induced THBS1 mRNA expression was strongly enhanced after treatment with cycloheximide (Fig. 1C). Many studies report that there are health beneficial effects of β-glucans and soybeans, and that several of these beneficial effects of β-glucan and soybeans overlap, including anti-tumor 3-5, 13, 14 and anti-atherosclerotic 9,47 effects. On the other hand, combinational effects of treatment with β-glucan together with soybeans have not been fully assessed, and the report here would be the first to report cooperative effects of β-glucan and soybeans on cells. However, whether the cooperative effects of AP-CF and Kurosengoku soybeans shown in this study are actually effective in vivo is unknown. Further investigation is required to understand and elucidate the significance of the effects of Kurosengoku soybeans and β-glucan produced by A. pullulans on the health of organisms (humans).
Preparation of the hot water soluble extracts from various bean species. The Kurosengoku soybeans (Glycine max cv. Kurosengoku) used in this study were obtained from Mr. Yukio Takada, board chairman, the Kurosengoku business cooperative association, Hokuryu town, Hokkaido, Japan. Conventional soybeans, Glycine max (L.) Merrill, Adzuki beans, Vigna angularis (Willd.) Ohwi & H. Ohashi, and a cultivar of the white-colored runner bean, Shirohanamame (Phaseolus coccineus cv. Shirohanamame) were purchased as commercially available products. The black soybeans, Kurosengoku, were ground at 16 °C. A 10% (w/v) suspension of Kurosengoku powder in distilled water was autoclaved at 120 °C for 30 min. After lipids and insoluble debris were removed using filter paper, the water soluble extracts of Kurosengoku were sterilized by further autoclaving, and used in this study. To prepare a hot water extract from whole beans, a mixture of 10% (w/v) of whole beans was placed in water, autoclaved at 120 °C for 30 min, and the supernatant was used for the study.
Preparation of Aureobasidium pullulans cultured fluid (AP-CF) containing β-glucan. The AP-CF was prepared as described elsewhere 6 . The AP-CF made with Kurosengoku soybeans (kAP-CF) was prepared in a manner similar to that of conventional AP-CF using Kurosengoku soybeans instead of rice bran as the nitrogen source. Briefly, A. pullulans was grown at 24.5 °C for 4 days, in a medium containing 2% Sucrose, 0.3% powdered Kurosengoku soybeans, 0.08% sodium L-ascorbate, and 0.02% L-ascorbic acid. This cultured medium was heated at 90 °C for 30 min and this heat-sterilized cultured medium was used for the study. The concentration of β-glucan in the kAP-CF was estimated to be 6 mg/ml. The purified β-glucan produced by A. pullulans (AP-PG) was prepared using ultrafiltration with a cut-off molecular weight of 20,000 followed by ethanol precipitation as previously described 6 . A protein conjugated β-glucan derived from Trametes versicolor CM-101, Krestin (Kureha Chemical Industry, Tokyo, Japan) 31 and a Laminaria digitata-derived β-glucan, laminarin (Sigma-Aldrich) 30 were purchased as commercially available products.

Real-time reverse transcription polymerase chain reaction (RT-PCR). To monitor the
Thrombospondin-1 (THBS1) mRNA expression, the total RNA was extracted from cultured cells using TRIzol reagent (Invitrogen, Carlsbad, CA, USA). The isolated total RNAs were treated with DNaseI (Takara, Shiga, Japan) and then subjected to oligo-dT-and random-primed reverse transcription using ReverTra Ace (Toyobo, Osaka, Japan). Real-time RT-PCR was performed using SYBR Premix Ex Taq II (Takara). The PCR reactions and analysis of mRNA expressions were performed using the CFX96 Real-Time PCR Detection System (Bio-Rad, Hercules, CA, USA). Each procedure was performed according to the manufacturer's protocol. The following specific primer set for THBS1 was used in this study: sense primer: 5′-ATGGAGAATGCTGTCCTCGC-3′, antisense primer: 5′-CCATTGCCACAGCTCGTAGA-3′.
Enzyme-linked immunosorbent assay (ELISA). The concentration of THBS1 in the supernatant of the culture medium was quantified using a commercially available ELISA kit (Quantikine ELISA human Thrombospondin-1, R&D Systems, Minneapolis, MN, USA) in accordance with the manufacturer's protocol.
Assays using selective kinase inhibitors. Selective protein kinase inhibitors, piceatannol, SB203580, and SP600125 were purchased as commercially available products (Abcam, Cambridge, MA, USA). Piceatannol (50 mM), and SB203580 (20 mM) and SP600125 (20 mM) were dissolved in dimethyl sulfoxide (DMSO) as stock solutions. The Mono Mac 6 cells were treated with piaceatannol, SB203580, or SP600125 for 1 hour. The concentrations of DMSO in the culture medium were equalized to 0.1%. After the treatment, the cells were stimulated with a 20-fold dilution of the hot water extracts of Kurosengoku soybeans (KS-E) and AP-CF at the concentration of 100 μg/ml β-glucan.

Statistical analysis.
To determine statistically significant differences between pairs of data, a two-tailed unpaired Student's t-test was performed in this study. A value of p < 0.05 was used to show statistical significance.