**2. Results**

#### *2.1. Structural Determination of Viridicatol*

The chemical structure of viridicatol (Figure 1A) was determined primarily by a detailed analysis of its 1D and 2D nuclear magnetic resonance (NMR) spectra (Figures S1–S6).

**Figure 1.** The X-ray crystallography and degranulation efficiency of viridicatol. (**A**) Chemical structure of viridicatol. (**B**) The X-ray crystallography of viridicatol. (**C**) Effects of viridicatol on degranulation. Cells were incubated with 200 ng/mL of dinitrophenyl (DNP)-specific immunoglobulin E (IgE) for 16 h in 48-well plates. The medium was replaced by Tyrode's buffer containing the indicated concentrations of viridicatol (2.5, 5, and 10 μg/mL) followed by stimulation with 500 ng/mL DNP-bovine serum albumin (BSA) for 1 h. The release of β-hexosaminidase was measured. (**D**) Effects of viridicatol on histamine release. The cells were sensitized and treated as described in (C), except for stimulation with DNP-BSA for 15 min, and the level of histamine was measured using an ELISA kit. \*\* *p* < 0.01 compared with the DNP-BSA group. The data represent the mean ± SD of three repeated experiments. Vir: viridicatol.

Viridicatol: 1H-NMR (CD3OD, 400 MHz) δH 7.33 (1H, m, H-5), 7.32 (1H, m, H-8), 7.31 (1H, d, J = 7.6, H-5), 7.23 (1H, d, J = 8.2, H-7), 7.11 (1H, m, H-6), 6.86 (1H, dd, J = 9.2, 2.1, H-4), 6.81 (1H, d, J = 8.1, H-6), 6.80 (1H, s, H-2); 13C-NMR (CD3OD, 100 MHz) δC 160.5 (s, C-2), 127.3 (s, C-3), 134.4 (s, C-4), 128.0 (d, C-5), 123.8 (d, C-6), 126.3 (d, C-7), 116.5 (d, C-8), 136.2 (s, C-1), 117.9 (d, C-2), 158.7 (s, C-3), 116.0 (d, C-4), 136.0 (d, C-5), 122.2 (d, C-6), 123.0 (s, C-4a), 143.1 (s, C-8a). HRESIMS *m*/*z* 252.0744 [M − H]<sup>−</sup>.

#### *2.2. X-ray Crystallography of Viridicatol*

The structure of viridicatol was assigned by the single X-ray (Figure 1B). It was obtained as a colorless needle. The crystal data were recorded with an Xcalibur Eos Gemini single-crystal diffractometer with Cu-Kα radiation (λ = 1.54184 Å). Space group P (−1), a = 6.2014 (2) Å, b = 10.2629

(4) Å, c = 10.7415 (5) Å, α = 117.029 (4), β = 96.135 (3), γ = 100.680 (3), V = 583.98 (5) Å3, Z = 2, Dcalcd = 1.491 g/cm3, μ = 0.888 mm<sup>−</sup>1, F (000) = 274.0. The final *R* indices were [I > 2σ(I)] *R*1 = 0.0524, w *R*2 = 0.1410. Crystallographic data of viridicatol were deposited in the Cambridge Crystallographic Data Center (deposition number: 2008745).

#### *2.3. Viridicatol Decreased the Release of* β*-Hexosaminidase and Histamine in RBL-2H3 Cells*

In the preliminary screening, we found that viridicatol significantly decreased the release of β-hexosaminidase and histamine in RBL-2H3 cells in a dose-dependent manner (Figure 1C,D). Meanwhile, we found that viridicatol had no measurable cytotoxic e ffects on RBL-2H3 cells (data not shown), and the IC50 value of viridicatol was 6.67 ± 0.6 μg/mL (26.3 μM).

#### *2.4. Viridicatol Relieved OVA-Induced Allergic Symptoms in Mice*

An OVA-induced mouse model of food allergy was established to explore the anti-allergic activity of viridicatol in vivo (Figure 2A). As expected, the mice in the OVA group exhibited diarrhea, hypothermia, and a high anaphylactic score after five successive oral challenges [31]. Moreover, after daily treatment with viridicatol, the hypothermia (Figure 2B), loose stool rate (Figure 2C), and anaphylactic score (Figure 2D) were significantly attenuated. Compared with the phosphate-bu ffered saline (PBS) group, the cecum in the OVA group was significantly increased, which was caused by allergic diarrhea. Following treatments with viridicatol, the onset of diarrhea was suppressed (Figure 2E).

**Figure 2.** Food allergy model and effects of viridicatol on the allergic symptoms. (**A**) Food allergy model. Mice were sensitized with 100 μg ovalbumin (OVA) and 2 mg alum in 200 μL of phosphate-buffered saline (PBS) by intraperitoneal injection, orally challenged with 50 mg OVA in 200 μL PBS, and the viridicatol concentrates (5, 10, 20 mg/kg) were orally administered. (**B**) The rectal temperature was measured 1 h after the fifth OVA challenge. (**C**) The rate of loose stools was measured 1 h after each OVA challenge. (**D**) The anaphylactic score was measured 1 h after each OVA challenge. (**E**) The representative large macroscopic view of the intestine in each group of mice. \*\* *p* < 0.01 compared with the OVA group. The data represent the mean ± SD. Vir: viridicatol.

#### *2.5. E*ff*ects of Viridicatol on Immunoglobulins, Anaphylactic Mediators, and Cytokines in the Mouse Serum*

Severe allergic reactions are associated with the increased secretion of allergic mediators. In this study, viridicatol significantly decreased the level of OVA-specific IgE at a concentration of 20 mg/kg (Figure 3A). However, no effects of viridicatol on OVA-specific IgG1 and IgG2a were observed (Figure 3B,C). As shown in Figure 3D–F, the oral administration of viridicatol significantly decreased the levels of serum histamine, mast cell protease-1 (mMCP-1) , and tumor necrosis factor-α (TNF-α) in a dose-dependent manner compared to that of the OVA group. As shown in Figure 3G, following an oral administration of viridicatol, the level of interleukin (IL)-10 was increased in the serum as a slight trend toward a dose-dependent manner.

**Figure 3.** The levels of immunoglobulins, anaphylactic mediators, and cytokines in serum (*n* = 6). (**A**) The level of OVA-specific IgE. (**B**) The level of anti-OVA IgG1. (**C**) The level of anti-OVA IgG2a. (**D**) The level of histamine. (**E**) The level of mMCP-1. (**F**) The level of TNF-<sup>α</sup>. (**G**) The level of IL-10. In (A,B,C,F,G), serum was from retro-orbital blood collection on day 41. In (D,E), the serum from the mice tail veins was collected within 1 h after the last challenge. \* *p* < 0.05, \*\* *p* < 0.01 compared with the OVA group. The data represent the mean ± SD. Vir: viridicatol.

#### *2.6. Viridicatol Alleviated Intestinal Injury and Inflammation of the Jejunum*

While the jejunum villi were regularly arranged without fracture on the surface of the jejunum villi in the PBS group, they were severely damaged in the OVA group. As expected, the damaged jejunum villi were repaired to different degrees following treatments with viridicatol (Figure 4A). In parallel with the clinical assessment, there were obvious signs of inflammation, including serious lymphocytic infiltration using hematoxylin and eosin (H&E). These symptoms were relieved following treatments with viridicatol (Figure 4B).

**Figure 4.** Effects of viridicatol on jejunum tissue injury. (**A**) Injury of the jejunum villi. The scanning electron microscopy (SEM) results shown at 500×, 1000×, and 3000× magnification. (**B**) Representative hematoxylin and eosin (H&E)-stained jejunum sections (magnification: 200×). Vir: viridicatol.

#### *2.7. E*ff*ects of Viridicatol on the Subpopulation of B Cells and Tregs in the Spleen*

As previously reported, regulating the populations of immune cells involved in the allergic responses is one of the most effective methods to treat food allergies [6,32]. To seek the target cells of viridicatol in vivo, we isolated the splenic lymphocytes from all groups of mice on day 41, and the lymphocyte populations were detected by flow cytometry. Compared to the PBS group (54.04%), the population of CD19+ B cells significantly increased to 64.32% in the OVA group. Moreover, the population of CD19+ B cells in the viridicatol group was significantly decreased at a concentration of 20 mg/kg (Figure 5A).

**Figure 5.** Effects of viridicatol on the population of B cells and regulatory T cells (Tregs) in the spleen. (**A**) Histograms of splenic B cells. Splenic lymphocytes were labeled with anti-CD3 and anti-CD4. (**B**) Scatter diagrams of regulatory T cells (Tregs). Splenic lymphocytes were labeled with anti-CD4 and anti-Foxp3. Spleens were isolated from each group of mice 24 h after the last challenge, labeled with various antibodies, and underwent Fluorescence Activating Cell Sorter (FACS) analysis by flow cytometry. \* *p* < 0.05, \*\* *p* < 0.01 compared with the OVA group. The data represent the mean ± SD. Vir: viridicatol.

Considering the effects of viridicatol on the level of serum IL-10 (Figure 3G), the population of Tregs in the spleen was further investigated. Compared to the PBS group (19.60%), the population of Tregs was significantly decreased in the OVA group (6.6%). As expected, the population of Tregs was significantly upregulated following treatment with viridicatol (Figure 5B). In summary, viridicatol displayed a weak and lack of a dose-dependent effect on the population of B cells and Tregs, respectively.

#### *2.8. E*ff*ects of Viridicatol on MC Population and Degranulation*

In light of the significant inhibitory effects on histamine and mMCP-1 in the serum, we determined the various populations of FcεRI<sup>+</sup>c-KIT<sup>+</sup> cells (MCs) by flow cytometry. Compared to the PBS group (0.16%), the population of MCs was significantly increased in the OVA group (1.34%). Following treatments with viridicatol, the population of MCs was significantly decreased compared to the OVA group (Figure 6A). The population of MCs was inhibited after treatments with viridicatol using toluidine blue staining (Figure 6B).

**Figure 6.** Effects of viridicatol on the mast cell (MC) population and degranulation. (**A**) Scatter diagrams of MCs and FACS analysis by flow cytometry. Splenic lymphocytes were labeled with anti-c-kit and anti-FcεR Iα. (**B**) Representative toluidine blue-stained jejunum sections (magnification: 400×). Jejunum tissues were isolated from each group of mice 24 h after the last challenge and fixed in paraformaldehyde. \*\* *p* < 0.01 compared with the OVA group. The data represent the mean ± SD. Vir: viridicatol.

#### *2.9. Viridicatol Decreased the Level of Intracellular Calcium*

The elevation of Ca2+ is a critical process for MC secretory granule translocation [33]. As a calcium ionophore, A23187 is often used to study the status of intracellular Ca2+ influx [34]. Moreover, the elevation of intracellular Ca2+ stimulated by A23187 was found to be much stronger than that observed in IgE antigen-stimulated cells [16]. In the A32187-induced degranulation model, viridicatol significantly inhibited the release of β-hexosaminidase induced by A23187 in RBL-2H3 cells (Figure 7A). To further study the effect of viridicatol on intracellular Ca2+, we used the Ca2+ probe to detect the concentration of intracellular Ca2+. The concentration of intracellular Ca2+ was significantly decreased by viridicatol in a dose-dependent manner compared with the A23187 group, as expected (Figure 7B,C). These results indicated that viridicatol may inhibit MC degranulation by suppressing the influx of intracellular Ca2+.

**Figure 7.** Effects of viridicatol on the degranulation of RBL-2H3 cells induced by A23187 and the concentration of intracellular calcium. (**A**) The release of β-hexosaminidase reduced by the calcium ionophore A23187. RBL-2H3 cells were inoculated in 48-well plates for 12 h and incubated with viridicatol in the presence of 2.5, 5, and 10 μg/mL for 1 h. The cells were then stimulated with 2.5 μM A23187 for 15 min. The release of β-hexosaminidase was measured. (**B**) Effect of Vir on [Ca<sup>2</sup>+]i. (**C**) The accumulation of intracellular calcium (magnification: 200×). \* *p* < 0.05, \*\* *p* < 0.01 compared with the A23187 group. The data represent the mean ± SD of three repeated experiments. Vir: viridicatol.

#### *2.10. Relationship between the Chemical Structure and Inhibitory E*ff*ects on Cell Degranulation*

The activity of compound is closely related to its chemical structure. Two homologues of viridicatol, viridicatin and 3-*O*-methylviridicatin, were also isolated from *Penicillium griseofulvum*. The chemical structures are presented in Figure 8A–C. In light of the effects of viridicatol on OVA-induced food allergy and the RBL-2H3 cell model, we investigated the effects of viridicatol and its homologous on RBL-2H3 cell degranulation to analyze their inhibitory activity. The effect of these compounds on β-hexosaminidase release in antigen and A23187-induced RBL-2H3 cells was measured at a concentration of 10 μg/mL. In the two models, viridicatol displayed an obvious inhibitory effect on β-hexosaminidase release compared to viridicatin (Figure 8D,E), which may be attributed to the missing conjugation of phenolic hydroxyl. Interestingly, 3-*O*-methylviridicatin hardly inhibited RBL-2H3 cell degranulation, which may have been due to the absence of a phenolic hydroxyl, and the hydroxyl on the mother ring was replaced by a methoxy group.

**Figure 8.** Viridicatol and its homologs for the chemical structural, and the inhibitory effect on RBL-2H3 cells. (**A**) The chemical structure of viridicatol. (**B**) The chemical structure of viridicatin. (**C**) The chemical structure of 3-*O*-methylviridicatin. (**D**) Effects of viridicatol and its homologs on IgE-induced degranulation. Same as Figure 1C, except that viridicatol was replaced by its homologs at 10 μg/mL. (**E**) Effects of viridicatol and its homologs on A23187-induced degranulation. Same as Figure 7B, except that viridicatol was replaced by its homologs at 10 μg/mL. \* *p* < 0.05, \*\* *p* < 0.01 compared with the DNP-BSA or A23187 groups. The data represent the mean ± SD of three repeated experiments. Vir: viridicatol.
