Anti-allergic activity of glycyrrhizic acid on IgE-mediated allergic reaction by regulation of allergy-related immune cells

Glycyrrhizic acid (GA), the major bioactive triterpene glycoside of glycyrrhiza, has been shown to possess a wide range of pharmacological properties, including anti-inflammatory and anti-viral properties. However, few studies have examined the anti-allergic activity and exact mechanism of action of GA. In the present work, the anti-allergic activity and possible mechanisms of action of GA on an immunoglobulin (Ig) E-mediated allergic reaction has been studied using three models of allergic reaction in vivo and in vitro. Active systemic allergic reaction in Balb/c mice showed that GA can suppress the increased level of IL-4 to restore the immune balance of TH1/TH2 cells in a dose-dependent manner. Additionally, GA attenuated significantly the B cells producing allergen-specific IgE and IgG1 partly because of the low levels of TH2 cytokines. Both passive cutaneous anaphylaxis in vivo and an RBL-2H3 cell-based immunological assay in vitro indicated that GA acted as a “mast cell stabilizer”, as it inhibited mast cell degranulation and decreased vascular permeability by inhibiting the expression of Orai1, STIM1 and TRPC1, which blocked extracellular Ca2+ influxes. The current study suggests that GA may serve as an effective anti-allergic agent derived from food for the prevention and treatment of IgE-mediated allergic reaction.

levels and also decrease interleukin (IL)-4, IL-5, eosinophilia and OVA-specific IgE 16 . In addition, GA (10 mg/ kg•bw) can attenuate the development of carrageenan-induced acute inflammation by preventing the activation of NF-κB and STAT-3 17 . Based upon these observations, we hypothesized that GA might be a contributing factor in the medicinal or nutritional uses of glycyrrhiza for relieving allergic reaction. However, few reports are available on the anti-allergic activity of GA. The present study was designed to investigate the anti-allergic effect of GA and to explore its possible underlying mechanism using active systemic allergic reaction and passive cutaneous anaphylaxis in vivo and an RBL-2H3 cell-based immunological assay in vitro.

GA reduces OVA-induced systemic allergic reaction in Balb/c mice through the regulation of T-helper (Th) cell differentiation.
To assess the anti-allergic effect of GA on the IgE-mediated allergic reaction, we examined an active systemic allergic reaction in Balb/c mice. OVA-induced food allergy symptoms were evaluated and scored for allergic symptoms and rectal temperature after a challenge for 40 min. Several allergic symptoms of OVA-induced food allergy were observed in the sensitization group, including strongly reduced activity, scratching, bristled fur and sometimes laboured respiration (1.80 ± 0.84 points). In contrast, the 100 mg/kg•bw GA-treated group showed significant suppression of the allergic symptoms (0.60 ± 0.55 points, Fig. 1B). In addition, the rectal temperature in the sensitization group decreased by −1.60 ± 0.1 °C compared to the Alum control group, whereas in the 1 mg/kg•bw GA-treated group, the rectal temperature was only reduced by −0.90 ± 0.1 °C (Fig. 1C). The suppressive effect of 100 mg/kg•bw of GA was similar to that of hydrocortisone, a common drug for the treatment of anaphylaxis, which was used as a positive control.
We then investigated the cytokine patterns in mouse spleen cells. T H 2-related cytokine (IL-4) was increased by the allergy induction and inhibited by GA, especially at the dose of 100 mg/kg•bw of GA ( Fig. 2A). T H 1-related cytokine (IFN-γ) decreased in the sensitization group, the GA group (1 or 10 mg/kg•bw) and the hydrocortisone group compared to the non-sensitized group. However, it was significantly increased by 100 mg/kg•bw of GA compared to the levels observed in the sensitization group (P < 0.05, Fig. 2B). Alum, which can activate T H 2-type immune cells 18 , also decreased the IFN-γ level (Fig. 2B). The result for IFN-γ/IL-4 was similar to that of IFN-γ (Fig. 2C), and 100 mg/kg•bw of GA ultimately results in a T H 1-type immune response. These results demonstrated that an oral dose of 1-100 mg/kg•bw of GA may affect T H cells by modulating the T H 1/T H 2 immune balance, thus attenuating the allergic reaction. A high concentration of GA can also affect the immune balance.
GA inhibits OVA-specific IgE and IgG 1 production by affecting OVA-specific antibody-producing B cells. We next investigated the effect of GA on the production of IgE and IgG 1 , the T H 2-type antibodies, against the OVA. The production of OVA-specific IgE and IgG 1 was significantly increased in the sensitization group compared to the Alum control group (P < 0.05) and inhibited by GA ( Fig. 3A and B). Only 100 mg/kg•bw can significant decrease the OVA-specific IgE production (P < 0.05, Fig. 3A). The significant inhibitory activity  of GA against the OVA-specific IgE and IgG 1 production was similar to that of hydrocortisone. These results demonstrated that GA also influenced OVA-specific antibody-producing B cells.
GA can also act as a "mast cell stabilizer" to relieve allergic symptoms by suppressing mast cell-mediator release. Mast cells are responsible for IgE-induced anaphylaxis 19 through the secretion of various inflammatory cytokines and mediators that can strengthen allergic symptoms. We then tested whether GA also regulates mast cell activation using passive cutaneous anaphylaxis (PCA) and an RBL-2H3 cell-based immunologic assay. GA significantly attenuated the mast cell-dependent PCA reaction in a dose-dependent manner, exhibiting 49.1%, 47.1% and 26.9% suppression at 1, 10 and 100 mg/kg•bw of GA based upon Evans blue extravasation, respectively (P < 0.05, Fig. 4B and C). Both the quantitative and qualitative PCA results indicated that GA can inhibit the decreased vascular permeability to reduce the albumin leakage; this effect is similar to the sodium cromoglycate.
Similar results with GA treatment were also obtained using the RBL-2H3 cell assay. We first examined the cytotoxic effect of GA on RBL-2H3 cells using the WST-8 assay and found that GA did not affect cell viability at 100~1000 μg/mL (Fig. 5A). Therefore, concentrations of GA < 1000 μg/mL were used for subsequent experiments. To investigate the effect of GA on degranulation, we measured the release of β-hexosaminidase in the presence or absence of GA. GA strongly suppressed β-hexosaminidase release from 87.46% ± 7.52% to 45.23% ± 8.64% as the dose of GA increased from 100 to 1000 μg/mL (P < 0.05, Fig. 5B). GA stabilizes mast cells by reducing the expression of the calcium channel proteins. As reported, the degranulation of RBL-2H3 cells depends on Ca 2+ release from the endoplasmic reticulum (ER) and  calcium release-activated calcium (CRAC)-mediated Ca 2+ influxes 20 . We further investigated the effect of GA on Ca 2+ influx. Fluo-3AM, a fluorescent Ca 2+ indicator, was used to determine the intracellular Ca 2+ concentration. A significant increase of [Ca 2+ ] i (nM, intracellular Ca 2+ concentration) was observed after DNP-HSA challenge at 30 s, and 1000 μg/mL of GA completely inhibited IgE/Ag-stimulated Ca 2+ influx (Fig. 6A).
We further investigated the expression of Ca 2+ influx-related proteins (calcium release-activated calcium channel protein 1 (Orai1), stromal interaction molecule 1 (STIM1), transient receptor potential channel 1 (TRPC1) and inositol 1, 4,5-trisphosphate receptor (IP3R)). Based upon RT-PCR and Western blotting results, the expression levels of Orai1, STIM1 and TRPC1 were significantly decreased by GA (P < 0.05, Fig. 6B and C, Fig. S1). IP3R, a receptor expressed on the ER membrane, was unaffected by GA at the mRNA level (P > 0.05, Fig. 6B). We then confirmed that GA had no effect on the depletion of ER calcium store, but can stabilize mast cells by inhibiting the Ca 2+ influx due to the lower expression of calcium channel proteins (Orai1, STIM1 and TRPC1).

Discussion
Natural triterpenoid compounds, such as glycyrrhizic acid (GA), ursolic acid, oleanolic acid and nomilin, exert similar effects on the immune system of Balb/c mice 12 , which may be related to the similarities in their chemical structures. In addition to the anti-inflammatory, anti-viral, antineoplastic and immune regulatory pharmacological effects, GA was found to possess anti-allergic activity in our study. The three main mechanisms of anti-allergic   (3) acts as a "mast cell stabilizer" to reduce mediator release through the inhibitory effect of Ca 2+ influx due to the lower expression of calcium channel proteins.
Many reports have identified that GA can affect the secretion of cytokines to modulate the immune microenvironment. In a cell-mediated immune response, GA (313 pg/ml) triggered a reduction of the highly elevated level of IL-6 compared with the control animals. In contrast, the level of IL-2 was enhanced with 37.9 pg/ml of GA in metastatic tumour-bearing C57BL/6 mice 21 . Moreover, GA still enhanced IFN-γ levels and reduced IL-4 levels in an allergic rhinitis mouse model 22 , which points to many biological roles including suppressing the stimulation of activated B-cell and T-cell proliferation and the differentiation of CD4 + T cells into T H 2 cells 23 . In a murine model of asthma, GA exerted a therapeutic effect on OVA-induced experimental asthma partly by regulating the T H 1/ T H 2 balance through suppressing OX40-OX40 L signalling and p38 MAPK activity 24 . The results of above reports are consistent with our study, which identified the modulatory effect of GA on T H cells. Furthermore, we found that a high concentration of GA (100 mg/kg•bw) triggers a T H 1-type immune response.
As previously reported, except for its regulatory effect on the T H 1/T H 2 immune balance, GA can reduce total IgE and OVA-specific IgE levels in serum 24 . OVA-specific IgE was decreased significantly in a dose-dependent manner after GA treatment in an allergic rhinitis mouse model, which may be induced by inhibiting T H 2 cell differentiation and maturation, and IL-4 production subsequently prevented allergic rhinitis development 22 . That GA can suppress the production of T H 2 antibodies (IgE and IgG 1 ) from OVA-specific antibody producing B cells is probably because of the effect of GA on the T H cell differentiation. GA produced a more significant suppressive effect on IgG 1 , which may subsequently inhibit the IgG 1 -mediated basophil activation 25 .
Previous studies have found that GA can inhibit histamine synthesis and release in mast cells co-cultured with Swiss 3T3 fibroblasts 26 . In our study, passive cutaneous anaphylaxis, which mainly depends on mast cells in vivo, showed that GA significantly reduced vascular permeability in a way similar to sodium cromoglycate. Similarly, GA can inhibit the release of β-hexosaminidase, a biomarker of degranulation, in RBL-2H3 cells. Our findings confirmed that GA acts as a "mast cell stabilizer" by inhibiting mast cell-mediator release.
Ca 2+ is an essential cofactor for the degranulation of RBL-2H3 cells 27 , which can regulate the granule-plasma membrane fusion and the release of mediators 28 . Based upon [Ca 2+ ] i measurement, the increased level of intracellular Ca 2+ concentration was significantly reduced after GA treatment, which suggested that GA also plays a stabilizing role on mast cells by inhibiting the extracellular Ca 2+ influx process. No difference in the mRNA expression of IP3R in the presence or absence of GA was found, which indicated that GA has no effect on the depletion of ER Ca 2+ store. The decreased expression of Orai1, STIM1 and TRPC1 both at the mRNA and protein levels indicate that GA might suppress the Ca 2+ -dependent degranulation due to the lower expression of these calcium channel proteins.
Based on the combined in vitro and in vivo analysis of GA treatment, we can conclude that GA exerts an anti-allergic effect by influencing T H helper cells, OVA-specific antibody-producing B cells and mast cells (or basophils) (Fig. 7). After the allergen is captured by dendritic cells through the disrupted epithelium, allergen-activated dendritic cells mature and migrate to regional lymph nodes where they present processed allergen epitopes to cognate T cells. Such T cells differentiate and become activated T H 2 cells, but GA can suppress this process to restore the T H 1/T H 2 immune balance. IL-4, which may be derived from T H 2 cells, mast cells, and basophils, also activates immunoglobulin heavy chain gene CSR for allergen-specific IgE production 29 . However, GA inhibits the synthesis and production of OVA-specific IgE and IgG 1 from the antibody producing B cells. Allergen-specific IgE can bind to FcεRI to stimulate mast cell degranulation 30 and to FcγRIII to activate PAF release from basophils 19 ; these processes recruit and activate T H 2 cells 31 to begin a positive feedback loop. However, GA, as a "stabilizer", reduces the release of allergic mediators by blocking extracellular Ca 2+ influxes due to the lower expression of calcium channel proteins (Orai1, STIM1 and TRPC1). In conclusion, as confirmed by active systemic allergic reaction, passive cutaneous anaphylaxis and RBL-2H3 cell-based immunology assay, GA exerts anti-allergic activity and can be used as a potential anti-allergic nutrient in the future.

Materials and Methods
Drugs and chemicals. The

Establishment of active systemic allergic reaction in Balb/c mice.
Thirty female Balb/c mice (4-weeks old, weighing 18-22 g) were divided into six groups (n = 5) with an initial body weight difference of ±20% after 3 day acclimation. The mice in the Alum immune adjuvant control group, were injected intraperitoneally with 200 μL of Imject TM Alum Adjuvant (100 mg in 0.9% NaCl) on day 0, 7 and 14. Mice in the sensitization group, GA treatment group and hydrocortisone treatment control group were sensitized by intraperitoneal injection of 200 μL of OVA solution (50 μg of OVA and 100 mg of Imject TM Alum Adjuvant in 0.9% NaCl) on day 0. The second and third sensitizing doses of OVA were increased to 100 μg on day 7 and 14. Finally, the sensitized mice were challenged intragastrically with a high dose of OVA (5 mg of OVA in 0.9% NaCl) on day 28. Before the challenge, GA was administered orally at a concentration of 1 mg/kg•bw, 10 mg/kg•bw or 100 mg/kg•bw daily between day 16 to day 27. A dose of 5 mg/kg•bw of hydrocortisone, a common drug for treating allergy diseases, was used as a treatment control. The experimental treatment design is shown in Fig. 1A.
Clinical allergic symptoms score system and rectal temperature. We first determined the anti-allergic effect of GA based upon the clinical allergic symptom score system and rectal temperature. The clinical allergic symptoms were scored 30 min post-challenge as previously described 32 : 0, no signs; 1, mice are scratching between 4 and 10 times over 15 min; 2, mice are scratching more than 10 times over 15 min, or display reduced activity or bristled fur; 3, mice have a strongly reduced activity, watery diarrhoea, difficulty walking normally, bristled fur and sometimes laboured respiration; 4, similar to degree 3 but stronger with cyanosis around the mouth and tail; and 5, death. Rectal temperature was measured before and 40 min after the challenge using a WI88375 probe (Beijing Science and Technology, Beijing, China).
ELISA assays for serum OVA-specific antibodies and spleen cell cytokines. To measure the serum OVA-specific IgE and IgG 1 levels by ELISA as previously described 33 , we collected serum samples from the orbital sinus after challenge with 5 mg OVA on day 28. Meanwhile, spleens were isolated from mice under sterile conditions after sacrifice, and spleen cells were seeded at 2 × 10 5 cells/well on a 96-well cell culture plate and incubated in RPMI1640 medium containing 200 μg/mL of OVA for 72 h at 37 °C in a 5% CO 2 incubator. Cytokines were quantified using a commercial mouse ELISA kit (eBioscience, Inc., San Diego, CA).

Establishment of passive cutaneous anaphylaxis in Balb/c mice.
Twenty-five female Balb/c mice (4-weeks old, weighing 18-22 g) were divided into five groups (n = 5). All tested mice received an intradermal injection of 0.5 μg of anti-DNP IgE in 30 μL of saline in the right ear and 30 μL of saline only in the left ear. On day 2, 3 and 4, Balb/c mice were administered orally 1 mg/kg•bw, 10 mg/kg•bw or 100 mg/kg•bw of GA. At the same time, 50 mg/kg•bw of disodium cromoglycate was administered orally as the treatment control. On day 5, each mouse was injected intraperitoneally with 200 μL of DNP-HSA and Evans blue solution (100 μg DNP-HSA and 2% Evans blue in 0.9% NaCl). The experimental treatment design is summarized in Fig. 4A.
Evans blue extravasation assay in Balb/c mice ears. After challenge, Evans blue extravasation in the right ears was recorded by a Canon EOS camera to qualitatively analysis the vascular permeability. After mice were sacrificed at 50 min, ears were collected and incubated with formamide at 64 °C for 12 hours. The concentrations were determined at 620 nm using the Thermo Scientific Varioskan Flash (Thermo, USA) to quantitatively evaluate the vascular permeability.

RBL-2H3 cells culture.
The rat basophilic leukemia cell line (RBL-2H3) obtained from National platform of experimental cell resources (Beijing, China) was cultured in MEM medium supplemented with 15% fetal bovine serum (FBS) and 1 × 10 5 U/L penicillin/streptomycin at 37 °C in a humidified 5% CO 2 incubator.
Water-soluble tetrazolium-8 (WST-8) cell viability assay. RBL-2H3 cells were preincubated with or without GA at a final concentration of 100 μg/mL, 500 μg/mL, 1000 μg/mL or 2000 μg/mL for 24 h. After washing the cells 3 times with PBS, 10 μL of WST-8 was added for incubating another 1 h at 37 °C. Finally, supernatants were transferred into another 96-well plate for measurement at 450 nm with the Thermo Scientific Varioskan Flash (Thermo, USA).  where F min is the background fluorescence with 5 mM EGTA and F max is the maximum fluorescence with 0.1% Triton X-100 instead of DNP-HSA. K d , the effective dissociation constant, of Fluo-3 and Ca 2+ is 400 nM.
Western Blotting. The cell samples were homogenized in a lysis buffer with protease inhibitors. The protein concentration of the supernatant was determined using a BCA Protein Assay Kit. Protein samples (40 μg) from different experimental groups were separated by SDS-PAGE (10%), transferred to nitrocellulose membranes, blocked in TBST solution containing 5% BSA for 1 h at room temperature, and incubated overnight at 4 °C with antibodies against STIM1, Orai1, TRPC1 or β-actin. After washing 6 times with TBST, the membranes were next incubated with HRP-conjugated secondary antibody for 1 h at room temperature. The signal was visualized by enhanced chemiluminescence and exposure to an X-ray film (Sage creation Mnin Chemi II, China). Statistical analyze. Statistical significance was determined by one-way analysis of variance (ANOVA) using GraphPad Prism 5.01 (GraphPad Software, Inc., USA). All data are presented as the mean values ± standard deviation (SD) with three times biological replicates and statistical significance was set at P-value < 0.05.