Scrodentoid A Inhibits Mast Cell–Mediated Allergic Response by Blocking the Lyn–FcεRIβ Interaction

Background: Mast cells are considered an attractive therapeutic target for treating allergic diseases, and the Lyn–FcεRIβ interaction is essential for mast cell activation. This study investigated the antiallergic effect of scrodentoid A (SA) on mast cells and mast cell–mediated anaphylaxis. Methods: For in vitro experiments, mast cells were treated with SA. Cell proliferation was tested using the XTT assay. The mRNA expression of various cytokines and chemokines was measured using qPCR. The levels of histamine, eicosanoids (PGD2, LTC4), and cytokines were measured using enzyme immunoassay kits. Signaling was investigated using Western blotting and immunoprecipitation. For in vivo experiments, the antiallergic activity of SA was evaluated using two mouse models of passive anaphylaxis as passive cutaneous and systemic anaphylaxis. The mechanism was investigated through immunohistochemistry and immunofluorescence. Results: SA considerably inhibited immunoglobulin (Ig) E-mediated mast cell activation, including β-hexosaminidase release, mRNA and protein expression of various cytokines, and PGD2 and LTC4 release. Oral administration of SA effectively and dose-dependently suppressed mast cell–mediated passive cutaneous and systemic anaphylaxis. SA significantly attenuated the activation of Lyn, Syk, LAT, PLCγ, JNK, Erk1/2, and Ca2+ mobilization without Fyn, Akt, and P38 activation by blocking the Lyn–FcεRIβ interaction. Conclusions: SA suppresses mast cell–mediated allergic response by blocking the Lyn–FcεRIβ interaction in vitro and in vivo. SA may be a promising therapeutic agent for allergic and other mast cell–related diseases.


INTRODUCTION
Allergy prevalence has been increasing worldwide, which has contributed directly or indirectly to health and economic burdens (1,2). Mast cells are central players in both the development and maintenance of allergic diseases. Mast cell activation releases various mediators, including preformed granule-associated chemical mediators, lipid mediators, and de novo synthesized cytokines, which are essential in allergies (3). Immunoglobulin (Ig) E and the high-affinity receptor FcεRI are vital in mast cell activation in an allergy context (4). FcεRI, present on mast cell surface, is a tetrameric complex comprising an IgE-binding α, signal-modulating β, and two signal-transducing γ subunits. The signaling cascades elicited by FcεRI aggregation start with Lyn phosphorylation, which transphosphylate the immunoreceptor tyrosine-based activation motif (ITAM) within FcεRIβ and FcεRIγ. Subsequently, another tyrosine kinase, Syk, is recruited and binds to phosphorylated ITAM, resulting in the phosphorylation of the adaptor proteins (LAT) and phospholipase Cγ (PLCγ). The principal axis pathway is then initated, activating the downstream signaling pathways, including the mitogen-activated protein kinase (MAPK), protein kinase C (PKC), and calcium flux pathways (5)(6)(7).
Binding of FcεRIα to the allergen-IgE complex initiates mast cell activation. FcεRIβ accelerates FcεRIα maturation in mast cells, thereby increasing FcεRI receptor expression on the cell membrane (8). FcεRIβ also amplifies the cell activation signal by enhancing the FcεRIγ signal by five to seven times, accelerating mast cell activation (9). Lyn is critical for ITAM phosphorylation on FcεRIβ (10), and a weak Lyn-FcεRIβ interaction is noted before FcεRI aggregation (11)(12)(13). Lyn next binds to the phosphorylated Y219 site of ITAM within FcεRIβ, and the interaction between Lyn and FcεRI increases considerably (14,15). The Lyn-FcεRIβ interaction is essential for human mast cell activation (16). Thus, we hypothesized that blocking the Lyn-FcεRIβ interaction may be a new direction for allergic disease treatment.
In our previous studies, five 19(4→ 3)-abeo-abietane diterpenoids (scrodentoids A-E) were firstly isolated from the whole plant of Scrophularia dentata. Among them, Scrodentoid B (SB) was considered as a potential immunosuppressive agent (17), and the other biological activities of these compounds have not been reported. Recently, it was suggested that some diterpenoid compounds have antiallergic activity, particularly in mast cell-mediated allergies (18,19). In our further antiallergic screening, only Scrodentoid A (SA) could modify IgE/Ag-stimulated mast cell activation and mast cell mediated anaphylaxis in vitro. However, the content of SA in S. dentata is very low, and its mechanism of action as an anti-allergic compound remains unknown. This is the first study to describe the chemical conversion from SB to SA and the inhibitory effects of SA on mast cell activation, evidenced by suppressed degranulation, lipid mediator and cytokine production, and reduced downstream signaling pathway activation. In agreement with this, SA considerably inhibits mast cell-mediated anaphylactic response by blocking the Lyn-FcεRIβ interaction. Therefore, SA might serve as a novel therapeutic target for allergic diseases.

Preparation of SA
Extracts from the dried whole plant of S. dentata (8 kg) were obtained with 95% EtOH (3×, each 80 L), using a reflux apparatus for 1.5 h. The extract (1,080 g) was obtained after in vacuo removal of the solvent. The extract was suspended in water and sequentially extracted using CH 2 Cl 2 , The CH 2 Cl 2 extract was evaporated to dryness in vacuo, and the resultant CH 2 Cl 2 fraction (243.6 g) was subjected to silica gel column chromatography (CC), eluted with EtOAc in petroleumether (PE) (0-100%, stepwise) to yield 12 fractions (Fr. 1-Fr. 12). Fr. 5 was separated repeatedly using CC and was further separated through reversed-phase HPLC by using CH 3  Trifluoroacetic acid (0.3 mL) was added into a solution of SB (300 mg, 1.007 mmol) in CH 2 Cl 2 (20 mL), then the mixture was stirred at 25 • C for 12 h. The reaction mixture was poured into ice water, adjusted to pH = 7 with NaHCO 3 (a.q.), then the residue was extracted with CH 2 Cl 2 (40 mL×3), dried over Na 2 SO 4 , and concentrated to give the product of compound 3 (238 mg, yield: 80%). 1

Viability Analysis Using XTT Assay
BMMCs were dispensed into 96-well plates (5 × 10 4 cells/well), and the cells were treated with SA (1, 10, 25, and 50 µmol/L) for 24 h. Cell viability was evaluated using a XTT-based cell proliferation assay kit, according to the manufacturer's protocol. Cell viability was expressed as the percentage OD test /OD control , where OD test and OD control indicate the optical densities of cells exposed to the test and control compounds, respectively.
RBL-2H3 cells were dispensed into a 96-well plate and incubated for 5 h to allow cells to adhere. The cells were sensitized with 0.1 µg/mL anti-DNP-IgE for 20 h. The next day, the medium was discarded and the cells were washed two times with HEPES buffer. The cells were then treated with SA for 30 min, followed by stimulation with DNP-HSA for 30 min. Finally, mast cell degranulation was assessed by measuring β-hex release.
β-hex in the supernatants and pellet was quantified theough the hydrolysis of p-NAG in 0.1 M sodium citrate buffer (pH 4.5) for 90 min at 37 • C. After adding 50 µL glycine (pH 10.7) to each fraction, the absorbance was measured at A 405 and A 570 in a microplate reader. The amount of total β-hex released was calculated using the following formula: % release = 100 × supernatant absorbance/(0.5 × supernatant absorbance + pellet absorbance).

Histamine Assay
IgE-sensitized BMMCs were washed, resuspended in HEPES at 2.5 × 10 6 cells/mL, and incubated with SA for 30 min at 37 • C and 5% CO 2 . The BMMCs were then stimulated using 0.1 µg/mL DNP-HSA for 30 min at 37 • C and 5% CO 2 . Samples were centrifuged at 300 × g for 5 min to obtain the supernatant. Histamine (working standards of 7.8-500 ng/mL) was freshly prepared. Histamine standards and supernatants were transferred to a 96-well flat-bottom plate and mixed with 12 µL of 1 M NaOH and 3 µL of 10 mg/mL o-phthalaldehyde. After 4 min at room temperature, 6 µL of 3 M HCl was added to stop the reaction. The fluorescence intensity was read using a 360-nm excitation filter and a 450-nm emission filter.

Cytokine Enzyme-Linked Immunosorbent Assay (ELISA)
BMMCs were dispensed into 24-well plates (2.5 × 10 5 cells per well) and sensitized with 0.5 µg/mL anti-DNP-IgE for 20 h. SA was dissolved in a medium containing DMSO and added to the cells to a final concentration of 5, 10, and 25 µmol/L for 30 min. Then cells were stimulated with DNP-HSA for 24 h for TNF, IL-6, and IL-13 or 6 h for LTC 4 and PGD 2 .
The supernatant was collected, centrifuged at 300 × g for 5 min to remove cell debris, and stored at −80 • C for ELISA. The supernatant was analyzed for TNF, IL-6, and IL-13 release using ELISA kits and for LTC 4 and PGD 2 with an ELISA kit, according to the manufacturer's protocol.

Intracellular Ca 2+ Mobilization
BMMC Ca 2+ mobilization was measured using a Fluo-4 NW Calcium Assay kit. IgE-sensitized BMMCs were washed and resuspended in assay buffer at 2.5 × 10 6 cells/mL. A 50 µL suspension was transferred to a 96-well plate, loaded with dye for 30 min at 37 • C, and treated with various concentrations of SA and PP2 for 30 min at room temperature. Fluorescence was read at 485-nm excitation and 516-nm emission filters. DNP-HSA (0.1 µg/mL) was added at 100 s, and calcium responses were recorded simultaneously.
Cell protein samples were separated on a NuPAGE Novex 4-12% Bis-Tris gel under non-reduced condition. The proteins were then transferred to a nitrocellulose membrane at 20 V for 1 h. The membrane was blocked with 4% BSA-TBST for 1 h and probed with appropriate antibodies overnight at 4 • C. The membrane was washed three times with TBST and incubated with HRP-conjugated goat anti-rabbit IgG or HRP-conjugated goat anti-mouse IgG diluted with 4% BSA-TBST for 1 h at room temperature. Finally, the membrane was developed with a chemiluminescent peroxidase substrate for 1 min and exposed to chemiluminescence film.

Immunoprecipitation
IgE-sensitized BMMCs were washed, resuspended in HEPES buffer at 5 × 10 7 cells/mL, treated with or without SA (25 µmol/L) for 30 min, and stimulated with DNP-HSA (0.1 µg/mL) for 10 min at 37 • C and 5% CO 2 . Or non-stimulated BMMCs (5 × 10 7 cells/mL) were treated with SA (5, 10, and 25 µmol/L) for 30 min. The reactions were terminated by adding the lysis buffer. Samples were put on ice for 30 min, span at 12,000 Frontiers in Immunology | www.frontiersin.org rpm, 15 min to get the soluble lysates. The soluble lysates were immunoprecipitated with the appropriate antibodies for 2 h at 4 • C. Protein A/G Sepharose beads (Santa Cruz Biotechnology) were then added, and the samples were incubated overnight at 4 • C. After washing three times, the samples were boiled at 95 • C for 5 min in 1 × NuPAGE sample buffer. Then, the samples were resolved on SDS-PAGE gels (10%) under non-reduced condition and blotted as described.

Passive Cutaneous Anaphylaxis
Passive cutaneous anaphylaxis (PCA) was performed as described previously (9). In brief, 1 µg of anti-DNP-IgE was intradermally injected into the right ear of 7-week-old male mice. The next day, the mice received oral administration of 25, 50, or 100 mg/kg SA or 50 mg/kg ketotifen (Sigma-Aldrich). After 30 min, the mice were challenged for 30 min by intravenous injection of 100 µg DNP-HSA in 300 µL saline containing 0.5% Evans blue. Finally, Evans blue was extracted after 24 h at room temperature with 300 µL of formamide and measured by absorbance at 630 nm.

Passive Systemic Anaphylaxis
Seven-week-old male mice were sensitized by intravenous injection of 2 µg of anti-DNP-IgE for 24 h. The next day, the mice received oral administration of 25, 50 mg/kg SA or 50 mg/kg ketotifen. After 30 min, the mice were challenged by intravenous injection of 1 mg of DNP-HSA. After antigen challenge, rectal temperature was measured and recorded every 5 min for 50 min with a digital thermometer. Mice were sacrificed 1 h after antigen challenge. Blood serum was collected. Ear tissues were fixed with 4% paraformaldehyde and embedded in paraffin. Sections were stained with H&E, toluidine blue. For immunofluorescence (Lyn and FcεRIβ), mice were sacrificed 10 min after antigen challenge. Briefly, sections were incubated overnight with anti-Lyn and anti-FcεRIβ. Staining was detected using the appropriated secondary antibody. Nuclei were stained with DAPI. The stained sections were scanned using panoramic slide scanner (3D HISTECH, Hungary). The images were calculated using Image-Pro plus 6.0.

Statistical Analysis
Experiments were conducted in triplicate, and data are presented as mean (n = 3) ± standard error of the mean (SEM). P values were determined using one-way ANOVA using GraphPad Prism 6.0. Statistical significance was set at P of ≤ 0.05, ≤ 0.01, or ≤ 0.001.

Preparation of SA
In order to enrich enough SA to meet the needs of pharmacological experiments, we obtained a total of 200 mg of SA through extraction, separation and chemical transformation ( Figure 1A).

SA Inhibits Degranulation and Lipid Mediators in IgE/Ag-Stimulated BMMCs
Before investigating the effect of SA (Figure 1A) on mast cell activation, we examined the cytotoxicity of SA on BMMCs by using XTT assay. BMMCs were treatmed with various concentrations (1, 10, 25, and 50 µmol/L) of SA for 24 h ( Figure 1B); SA did not exhibit any cytotoxicity, even at 50 µmol/L. Therefore, we used SA at concentrations of up to 25 µmol/L for all in vitro experiments. To determine the effect of SA on BMMC degranulation, we first sensitized BMMCs with anti-DNP-IgE for 20 h and then incubated cells with various concentrations (2.5, 5, 10, and 25 µmol/L) of SA. After 30 min, BMMCs were stimulated with DNP-HSA for 30 min, and then mast cells released β-hex and histamine. SA treatment significantly decreased β-hex ( Figure 1C) and histamine (Figure 1D) release in a dose-dependent manner. Furthermore, the inhibitory effect of SA on histamine was better than that obtained by PP2, a tyrosine kinase inhibitor.
FcεRI-mediated mast cells also release lipid mediators after activation-prostaglandin D 2 (PGD 2 ) and LTC 4 . To further clarify the effect of SA, we sensitized BMMCs with anti-DNP-IgE, followed by incubation with or without SA for 30 min. The cells were then activated with DNP-HSA for 6 h. When BMMCs were activated for 6 h, cells released a large amount of LTC 4 (Figure 1E), but SA treatment significantly decreased LTC 4 release in a dose-dependent manner. SA also considerably attenuated PGD 2 generation after cells were stimulated with DNP-HSA for 6 h (Figure 1F).

SA Attenuated FcεRI-Mediated Cytokine Production in BMMCs
Engagement of the FcεRI receptor leads to the production of multiple cytokines and chemokines essential for mast cell function and allergic response. To determine the effect of SA on FcεRI-mediated cytokines and chemokines, BMMCs were sensitized with anti-DNP-IgE overnight, and pretreatment with SA for 30 min, then stimulated with DNP-HSA for 1 h. We found that SA significantly inhibited the expression of cytokine mRNAs in FcεRI-mediated BMMCs. These include Th2-related cytokines and inflammation-related cytokines. However, the expression of MIP-1α which belongs to chemokines remained unaltered (Figure 2A). Subsequently, we used ELISA to confirm the effect of SA on FcεRI-mediated secretion of TNF-α, IL-6, and IL-13. As shown in Figure 2B, treatment with SA significantly reduced TNF-α, IL-6, and IL-13 secretion in a dose-dependent manner. In particular, SA at 25 µmol/L considerably inhibited IL-6 release, and the inhibition rate was up to 79.9%. These data indicate that SA plays a role in stabilizing mast cells, inhibiting their activation.

Effect of SA on RBL-2H3 Cells
We next evaluated the effect of SA on IgE-activated RBL-2H3 cells. IgE-sensitized RBL-2H3 cells were pretreatment with SA for 30 min, then stimulated with DNP-HSA for 30 min or 1 h. After 30 min of DNP-HSA stimulation, the supernatant was assayed for β-hex release (Figure 3A), SA significantly inhibited β-hex release in a concentration-dependent manner. And SA also attenuated the mRNA expressions of TNF-α (Figure 3B), CCL2 (Figure 3C), IL-13 (Figure 3D), and IL-4 ( Figure 3E) after RBL-2H3 cells were stimulated with DNP-HSA for 1 h. This further demonstrated that SA suppressed the release of allergic mediators from mast cells.

SA Suppresses Anaphylactic Responses in Mice
Given that SA downregulated the release of allergic mediators in vitro, we used PCA in the ear and passive systemic anaphylaxis (PSA) to evaluate the action of SA in anaphylactic responses in vivo. PCA was elicited by subcutaneous injection of anti-DNP-IgE into the ear. After 24 h, mice treated with various concentrations (25,50, and 100 mg/kg) of SA for 30 min. Then, mice received intravenous DNP-HSA solution containing 0.5% Evans blue dye. Mice challenged with DNP-HSA showed a clear PCA response in the ear. SA significantly reduced the PCA response on visual inspection ( Figure 4A) and the Evans blue dye extracted from the reaction site of the ear (Figure 4B), but no difference was found between SA at 50 and 100 mg/kg. Therefore, we used SA at 25 and 50 mg/kg in the PSA assay.
PSA is an IgE-mediated type I immediate hypersensitivity reaction. Mice were sensitized with anti-DNP-IgE overnight and treated with or without SA for 30 min; then, they were intravenously injected with DNP-HSA to elicit an anaphylactic response. Rectal temperature was recorded every 5 min to monitor the magnitude of the PSA response. Mice challenged with DNP-HSA showed an evident rectal temperature drop by 10-25 min after injection. By contrast, mice treated with SA or ketotifen showed significantly reduced rectal temperature drop ( Figure 4C). Histological analysis of ear sections at 1 h after injection of DNP-HSA also confirmed the antiallergic effect of SA (Figures 4D,E). In addition, SA significantly inhibited serum levels of PGD 2 ( Figure 4F) and histamine ( Figure 4G). In particular, at the same concentration, SA was comparable to ketotifen, a H1 blocker and mast cell stabilizer.
Because SA inhibited mast cell activation in vitro, we hypothesized that SA achieves the inhibition effect of anaphylactic responses by stabilizing mast cells. Thus, we examined whether mast cell degranulation occurs or not in PSA Frontiers in Immunology | www.frontiersin.org with SA. Toluidine blue staining of ear sections showed nearly activated mast cells after allergen exposure (red arrow), whereas SA significantly reduced mast cell activation, particularly at 50 mg/kg, most mast cells were at rest status ( Figure 4H). In addition, as shown as Figure 4I, SA suppressed the absolute numbers of activated mast cell. These data indicate that SA inhibits allergic response in vivo by suppressing mast cell activation.

SA Inhibits Ca 2+ Mobilization in IgE-Stimulated BMMCs
Intracellular calcium is an important second messenger in cells, and a transient increase in intracellular calcium is crucial for the activation of mast cells (6,20). Therefore, we explored the effect of SA on Ca 2+ influx in BMMCs. BMMCs had high levels of calcium flux after allergen exposure and reached a peak after BMMCs challenged DNP-HSA for 50 s ( Figure 5A). As expected, SA dramatically decreased intracellular calcium levels (Figure 5A), and effectively inhibited the amount of calcium levels which was accumulated from BMMCs challenged DNP-HSA 0 s to 300 s ( Figure 5B).

SA Altered FcεRI Signaling Events in BMMCs
The aforementioned data indicated that SA inhibits FcεRImediated mast cell activation both in vitro and in vivo. We next examined whether SA affects the activation of intracellular signaling pathways by FcεRI in BMMCs. IgEsensitized BMMCs were pretreated with SA for 30 min, followed by DNP-HSA crosslinking for 5 min or 15 min. After 5 min of DNP-HSA stimulation, SA dose-dependently inhibited the phosphorylation of Lyn (Figures 6A,E), which   is a FcεRI-proximal tyrosine kinase. We next detected the downstream signaling molecules of Lyn: FcεRIβ, FcεRIγ, Syk, LAT, and PLCγ1; as expected, SA significantly suppressed their phosphorylation (Figures 6B,C,F-J). However, Fyn, another FcεRI-proximal tyrosine kinase, was not significantly changed after treatment with SA (Figures 6A,D). Consistently, SA could not change the phosphorylation of Akt (Figures 7A,F), a downstream Fyn-dependent protein (21).
MAPK signaling is distal signaling during mast cell activation and is vital for the FcεRI-mediated production of cytokines and chemokines in mast cells (6,22,23). As shown in Figures 7A,C-E, after 15 min of DNP-HSA stimulation, SA considerably inhibited the phosphorylation of Erk1/2 and JNK, but p-P38 remained unaltered; in addition, SA significantly inhibited p-cPLA 2 (Figures 7A,G), which is consistent with the inhibition of LTC 4 release (24). Moreover, SA significantly promoted AMP-activated protein kinase (AMPK) phosphorylation at Thr172 after 15 min of DNP-HSA stimulation (Figures 7A,B), while it did not alter p-AMPK (Ser485/491). In addition, SA pretreatment alone did not affect AMPK phosphorylation in non-stimulated mast cells (Figures 7H,I).
Previous report showed that Lyn-specific siRNA increase human mast cell activation (25). To further clarify the mechanism of SA in allergic response, we assessed the effects of SA on Lyn activity by using an in vitro enzyme system. However, SA did not inhibit Lyn kinase activation (data not shown). We then examined the effect of SA on AMPK/Lyn interaction before any FcεRI aggregation. As shown as Figures 7J,K, the interaction between AMPK and Lyn was not altered with SA. In the Lyn-related pathway, the Lyn-FcεRIβ interaction is indispensable for FcεRI-mediated mast cells (16). We examined whether SA influenced the Lyn-FcεRIβ interaction. As shown as Figures 8A,B, there was a weak binding between Lyn and FcεRIβ when mast cells were at rest status, and SA did not affect the binding between Lyn and FcεRIβ. However, allergen exposure for 10 min considerably enhanced the binding between Lyn and FcεRIβ, and SA blocked this binding, reduced it to that at mast cell rest status. On investigating whether SA blocked the binding between Lyn and FcεRIβ in vivo in the PSA model, we found that SA significantly suppressed the binding between Lyn and FcεRIβ after exposure to the allergen for 10 min, whereas ketotifen did not (Figures 8C,D).

DISCUSSION
Abietane-type diterpenes are abundant in nature and have various biological activities and pharmacological properties, such as antitumor (26), antiviral (27), antibacterial (28,29), and antiinflammatory (30). However, their antiallergic effect has not been reported. SA is an abietane-type diterpene from S. dentata Royle ex Benth. In this study, we first report that SA inhibits mast cell activation and mast cell-mediated allergic reactions.
Mast cells play a key role in allergic response in conditions such as asthma, rhinitis, and atopic dermatitis. IgE/Agactivated mast cells activate a complex intracellular signaling pathway. Lyn and Fyn are essential for initiating FcεRImediated mast cell activation: Lyn mediates a pivotal signaling pathway following aggregation of FcεRIs, whereas Fyn is key in a complementary pathway for mast cell activation (31). Several studies have reported compounds that suppress IgEmediated allergic response through the inhibition of Lyn and Fyn. For example, AZD7762 suppressed IgE-mediated mast cells and allergic responses by inhibiting the activity of Lyn and Fyn kinases (32). Elaeocarpus, which was isolated from Elaeocarpus sylvestris L., inhibited Fyn and Lyn phosphorylation, thereby preventing mast cell degranulation and expression of proinflammatory cytokines (33).
Unlike the aforementioned compounds, SA inhibited the phosphorylation of only Lyn and not Fyn (Figure 6A), and had no inhibitory effect on Akt (Figure 7A), a protein that belongs to Fyn-regulated compensatory signaling pathways. In experiments with RBL-2H3 cells, where Fyn gene and protein expressions are hardly detected (34), SA inhibited the degranulation and expression of proinflammatory cytokines in a dose-dependent manner (Figure 3). Therefore, speculating that SA only inhibits the Lyn-dependent pathway in IgE/Ag-induced mast cells, we investigated the effect of SA on phosphorylation of FcεRIβ, FcεRIγ, Syk, LAT, PLCγ, and MAPKs. SA significantly suppressed the phosphorylation of Syk, LAT, PLCγ, Erk1/2, and JNK, but did not affect the phosphorylation of P38.
AMPK, a regulator of energy metabolism, negatively regulates mast cell activation and anaphylaxis. AMPK activation suppressed IgE/Ag-induced mast cell activation through the inhibition of Erk and JNK phosphorylation, but did not change P38 phosphorylation (35). Therefore, we investigated the involvement of SA in IgE/Ag-induced AMPK activation. As expected, SA significantly increased AMPK phosphorylation at Thr172 after 15 min of DNP-HSA stimulation (Figure 7A), while it did not alter p-AMPK (Ser485/491). However, the AMPK phosphorylation at Thr 172 is not affected with Lyn knockdown, whereas Fyn counter-regulates the AMPK signaling pathway in mast cells (35). Thus, we suggest that the target of SA in mast cells might be simultaneously related to Lyn and AMPK.
FcεRIβ amplifies the FcεRI-mediated activation signal. The Lyn-FcεRIβ interaction is indispensable for FcεRIβ-mediated human mast cell activation (16). Lyn can pre-associate with the FcεRIβ chain before receptor crosslinking, and receptor stimulation can further increase their binding (11). In addition, AMPK binds to Lyn when mast cells are at rest, and after mast cell activation, AMPK binds to FcεRIβ (36). We hypothesized that the AMPK-FcεRIβ binding might be caused by the binding of Lyn to FcεRIβ, and that the Lyn-FcεRIβ interaction might simultaneously affect Lyn and AMPK phosphorylation. Although previous study showed that AMPK activation blocked the formation of Lyn-FcεRIβ complexes, their results are based on the AMPK phosphorylation at Ser485/491 (36). In our study, we did not observe significant effect of SA on p-AMPK (Ser 485/491) after FcεRI activation; instead, we observed that SA significantly increased AMPK phosphorylation at Thr172 after 15 min of DNP-HSA stimulation ( Figure 7A). In addition, we observed that SA did not alter the interaction between AMPK and Lyn in non-stimulated mast cell ( Figure 7J). And SA blocked the binding Lyn to FcεRIβ in IgE/Aginduced mast cell both in vitro (Figures 8A,B) and in vivo (Figures 8C,D). Above all, we suggest that SA suppressed mast cell-mediated allergic response by blocking the Lyn-FcεRIβ interaction in vitro and in vivo, which is a key step in mast cell activation. Thus, SA may be a promising therapeutic agent for allergic diseases.

ETHICS STATEMENT
All animal experiments were performed according to the Health Guidelines of the Shanghai University of Traditional Chinese Medicine, China, and protocols were approved by the Institutional Animal Ethics Committee of Shanghai University of Traditional Chinese Medicine (No. SZY201710088).

AUTHOR CONTRIBUTIONS
YL and JX designed the study. FQ, LZ, SL, and GM preformed the experiments. FQ and YL analyzed the data and wrote the manuscript. FG and PL provided scientific discussion.