A Ganoderma-Derived Compound Exerts Inhibitory Effect Through Formyl Peptide Receptor 2

Formyl peptide receptors (FPRs) are G protein-coupled receptors (GPCRs) widely expressed in neutrophils and other phagocytes. FPRs play important roles in host defense, inflammation, and the pathogenesis of infectious and inflammatory diseases. Because of these functions, FPRs are potential targets for anti-inflammatory therapies. In order to search for potentially novel anti-inflammatory agents, we examined Ganoderma (Lingzhi), a Chinese medicinal herbs known for its anti-inflammatory effects, and found that compound 18 (C18) derived from Ganoderma cochlear could limit the inflammatory response through FPR-related signaling pathways. Further studies showed that C18 could bind to FPR2 and induce conformation change of the receptor that differed from the conformational change induced by the pan-agonist, WKYMVm. C18 inhibited at the receptor level and blocked WKYMVm signaling through FPR2, resulting in reduced superoxide production and compromised cell chemotaxis. These results identified for the first time that a Ganoderma-derived component with inhibitory effects that acts through a G protein-coupled receptor FPR2. Considering its less than optimal IC50 value, further optimization of C18 would be necessary for future applications.

The complex role in various diseases indicates FPRs are potential targets for therapeutic intervention. Therefore, searching for novel agonists and antagnosits of FPRs has drawn significant attention. The FPR subfamily has a variety of structurally diverse ligands, including natural peptides and synthetic non-peptide compounds. Compared with natural peptides, small molecule comounds are more stable to serve as potentially therapeutic agents (He and Ye, 2017). Although numerous small molecule compounds with inhibitory effects have been found through screening of combinatorial compound libraries, very few were fully characterized with high potency (Ki < 10 µM) in the literatures (Schepetkin et al., 2014;He and Ye, 2017). Traditional Chinese medicine (TCM) has been explored in search for novel FPR antagonists, based on its long-term medical practice (Yuan et al., 2016). In this study, Ganoderma (Lingzhi) was chosen because it has been used extensively in Asian countries for more than 2,000 years, due to its various pharmacological effects, including immunomodulation, antibacterial, anticancer, antioxidant, and antiviral activities (Gao et al., 2003;Sliva, 2004;Yuen and Gohel, 2005;Joo et al., 2008;Sanodiya et al., 2009;Ma et al., 2011;Xu et al., 2011). Ganoderma is rich in active compounds, including triterpenoids, fatty acids, polysaccharides, peptides, and other chemicals (Sanodiya et al., 2009;Peng and Qiu, 2018), and that has led to the possibility of identifying FPR agonists and antagnosits.
In this study, 34 Ganoderma-derived compounds that were available in our collection were subjected to initial screening using FPR2-dependent superoxide generation assay and degranulation assay. Among these triterpenoids and meroterpenoids, C18 was identified to have strong inhibitory activities. C18 and 5 other structurally similar compounds (Figure 1), all Ganoderma meroterpenoids (GMs) (Peng and Qiu, 2018), were selected for further studies. (Figure 1). C18, was found to display significant inhibition in several FPR-mediated functional assays, but had no effect on C5a receptor and PKC-mediated signaling pathways. To assess the structure-activity relationship, FLAsH-based fluorescence resonance energy transfer (FRET) detection and molecular docking analysis were performed. The results demonstrated that C18 could inhibit FPR-mediated pro-inflammatory response by targeting FPR2. In short, our work demonstrated the in vitro inhibitory effects of a novel Ganoderma-derived compound through FPR2, further revealing its detailed mechanism with competitive binding assay and FRET detection assay, and finally show its interaction with FPR2 by molecular docking analysis. These results suggest that C18 may be a naturally active component and exert its inhibitory effects through FPR2.

Compounds Preparation
As shown in Supplementary Figure 1. G. cochlear (68 kg) mushrooms were chipped and extracted with 95% ethanol (EtOH, 120 L) under reflux three times at 60°C, each for 3 h. The combined ethanol extracts were evaporated under reduced pressure. The residue was suspended in H 2 O (10 L) and extracted with ethyl acetate (EtOAc, 3 × 10 L) and n-Butanol (3 × 10 L), respectively. The volume of the combined EtOAc extracts was reduced to one-third under reduced pressure. The residue (11.5 kg) was fractionated by macroporous resin (D-101; MeOH/H 2 O, 50:50, 70:30, and 90:10, v/v): fractions I-III.
The molecular formula of testing active compound C18 and the negative control compound C12 were determined by HRESIMS and 13 C-DEPT NMR at Kunming Institute of Botany, China, as previously described (Peng et al., 2016). These compounds were analysis by HPLC and determined to have a purity ≥ 95% (Supplementary Figure 2).

Calcium Mobilization Assay
RBL-FPR2 or RBL-FPR1 cells were cultured in black wall/clear bottom 96-well plate until the confluence reached about 90%. The cells were washed once with DMEM and incubated with FLIPR calcium-sensitive dye (Molecular Devices, Sunnyvale, CA) and different concentration of compounds to be tested or vehicle (0.1% DMSO) in HBSS/BSA for 60 min at 37°C with 5% CO 2 . The agonist (WKYMVm, 10 nM for screening and 10 -12 -10 -7 M for dose-response curve; fMLF, 10 -9 -10 -6 M for dose-response curve) was added and samples were read in a FlexStation III Multi-Mode Microplate Reader (Molecular Devices) with excitation wavelength at 488 nm and emission wavelength at 525 nm according to the manufacturer's protocol (He et al., 2013).

Superoxide Generation Assay
Superoxide production of differentiated HL-60 cells (dHL60-6d, 1×10 5 cells per well) was determined by isoluminol-ECL assay (Dahlgren and Karlsson, 1999), using 96-well, flat-bottom, white tissue culture plates (PerkinElmer Life Sciences, Boston, MA). dHL60 cells were harvested and washed once with 0.5% BSA/ HBSS. Cells were then re-suspended with 0.5% BSA/HBSS buffer and incubated with or without compounds for 30 min, then added with 100 mM isoluminol and 40 U/ml HRP at 37°C for 5 min in the dark. Aliquots (200 Ml) of the cells were added into the 96-well plate, and chemiluminescence (CL) was eventually detected at 37°C with an EnVision Multilabel Plate Reader (PerkinElmer Life Sciences, Boston, MA). The CL counts per second (CPS) was continually recorded, at 16 s intervals, for 20 points before and 200 points after stimulation with 100 ng/ml PMA or 1 mM fMLF or WKYMVm. The relative level of superoxide anion produced was calculated based on the integrated CL during the first 15 min after agonist stimulation.

Cell Degranulation Assay
For release of b-hexosaminidase, RBL-FPR2 cells were cultured in a 24-well plate for 24 h, or differentiated HL-60 cells for 6 days (dHL60-6d, 1×10 5 cells per well), then treated with or without compounds for 1 h. Subsequently, cells were washed briefly and pre-incubated with 10 mM cytochalasin B in HBSS included 20 mM HEPES, pH 7.4, and 0.5% BSA (HBSS-HB) for 15 min on ice followed by 15 min at 37°C, as described in a previous publication (Nanamori et al., 2004). Then, cells were stimulated for 15 min with 1 mM WKYMVm and vehicle at 37°C before chilling on ice to terminate the degranulation reaction. The amount of secreted b-hexosaminidase was quantified by incubating 20 ml of supernatant with 10 ml of 1 mM p-nitrophenyl-N-acetyl-b-D-glucosamide in 0.1 M sodium citrate buffer, pH 4.5, at 37°C for 1 h in a 96-well plate. Then reaction was terminated by adding 200 Ml of 0.1M Na 2 CO 3 and 0.1M NaHCO 3 , pH 10, and absorbance was determined at 405 nm in a Flex Station 3 Multi-Mode Microplate Reader (Molecular Devices). Total cellular b-hexosaminidase was determined with cell lysate in 0.1% Triton X-100.

Cell Chemotaxis Assay
Agonist-induced migration of cells was assessed in a 24-well transwell chamber (Corning costar, Kennebunk, USA), as reported previously (Nanamori et al., 2004). In brief, dHL-60 cells (2×10 5 cells per well) were pre-incubated with different concentrations of the compounds (10 nM-20 mM) for 30 min and then seeded the cells in the upper chamber (100 Ml), WKYMVm (1nM) was placed in the bottom well (600 Ml), which was separated from the lower compartment by a polycarbonate membrane filter with pore size of 5 mm. After incubation at 37°C for 2 h, the upper chamber was removed and the total number of cells in the bottom wells were counted by flow cytometry. Data were presented as chemotaxis index, which was the ratio of cells migrated toward agonists over the cells migrated toward medium. Checkerboard analysis was performed by adding 2×10 5 dHL60 cells/well to the upper chamber, and serial dilutions of C18 were added to the upper chamber as well as the lower chamber. After 2 h, cells that migrated through the polycarbonate membrane were counted in the lower chamber.

FLAsH-Based Fluorescence Resonance Energy Transfer (FRET) Detection
Using the methods described previously (Gaietta et al., 2002;Hoffmann et al., 2005), HEK-293 cells (ATCC ® CRL-1573) were cultured in 24-well plate on coverslips with poly-D-lysine treatment. Plasmid coding for FPR2-ICL3-ECFP (FPR2 protein with ECFP at the C-terminal and FLAsH-binding sequence in the third intracellular loop) was transiently transfected into the cells. Twenty-four hours after transfection, TC-FlAsH™ II In-Cell Tetracysteine Tag Detection Kit (Thermo Fisher Scientific, Waltham, MA USA) was used to label the modified FPR2 proteins according to the manufacturer's instructions. In brief, the cells were washed with HBSS buffer and incubated with FlAsH-EDT2 labeling reagent (500 nM FLAsH/EDT, 12.5 mM EDT, and 5.6 mM glucose in HBSS) for 1 h. The excess and nonspecifically bound FLAsH was removed by incubating the cells with BAL wash buffer (250 mM BAL in HBSS) for 10 min twice followed by another wash with HBSS buffer. The cells were then treated with or without the indicated compounds for 10 min.
The coverslips with the cells were mounted on glass slides and FRET signals were analyzed with a Leica TCS SP8 confocal microscope immediately. The fluorescent FPR2 proteins were excited by a 448-nm laser and two images representing FLAsH emission and ECFP emission, respectively, were taken simultaneously with a dichroic beam splitter D448/D514 under a 40× oil objective. The emission paths are 535 ± 15 nm (FLAsH) and 480 ± 20 nm (ECFP). FRET signals were calculated as ratios of FLAsH intensities to ECFP intensities from the correspondent two images. Five dots with 5 × 5 pixels each at the cell membranes were chosen to obtain the fluorescence intensities for each cell sample.

Cell Morphology Observation
Stably transfected FPR2-RBL cells were seeded in 12-well plate with round coverslips for 24 h. Then the cells were treated with or without C18 (10 mM) for 30 min, and stimulated with WKYMVm (final concentration of 1 mM) for another 15 min at 37°C. Subsequently, remove cell medium and wash the cells once with PBS, then fixed cells with 4% Paraformaldehyde for 15 min at room temperature, remove fixative solution and washed twice with PBS, blocked with 5% BSA+PBS for 1 h, and then stained with rhodamine phalloidin (Cytoskeleton, USA) and Hoechst according to the manufacture's protocol. In brief, stain the cells with 100 nM rhodamine phalloidin, incubate at room temperature in the dark for 30 min wash the cell coverslips three times in PBS, and stain DNA for 5 min with of 2 Mg/ml Hoechst. Rinse the coverslips and invert on a drop of anti-fade mounting media on a glass slide and seal each side with nail polish. Coverslips with the cells were mounted on glass slides and cell images were taken immediately with a 40× oil objective. The images were analyzed by ImageJ 1.49U (Rasband, W.S., ImageJ, U. S. National Institutes of Health, Bethesda, Maryland, USA, https://imagej.nih.gov/ij/, 1997-2018).

Molecular Docking Analysis of the Binding Interaction of C18 With FPR2
Because the crystal structures of FPRs are not currently available, the structure models of FPR2 was obtained from Swiss-model server which was based on C5a receptor (PDB code: 5o9hA) because of its higher similarity (34.5%) and resolution (2.7Å) (Robertson et al., 2018;Waterhouse et al., 2018). The initial conformations of ligands were generated by ChemBio 3D (PerkinElmer). Hydrogens were added by Autodock Tools and molecular docking was performed by Autodock Vina (Trott and Olson, 2010). The search box was set as 46 Å × 34 Å × 60 Å for FPR2. The best conformation was refined with energy minimization and analyzed with PyMOL Molecular Graphics System (Version 2.0 Schrödinger, LLC).

Competitive Binding Assays
RBL-FPR2 or RBL-FPR1 cells (4×10 4 cells per well) were harvest and washed twice with buffer (HBSS supplemented with 20 mM HEPES, pH 7.5, and 0.1% BSA). These methods prepared as described (He et al., 2013). The competitive binding assays is used to measure relative affinity of WKYMVm and compound C18, in which a fixed concentration of WK (FITC) YMVm (50 nM) or fMLFIIK-FITC (100 nM) was added, and then added increasing

Statistical Analysis
Data were shown as mean ± standard deviation (SD) from at least 3 independent experiments. Statistical analyses were performed using GraphPad Prism (Version 6.0, La Jolla, CA), IC 50 values from each assay were calculated from dose response curve that were fitted by non-linear regression analysis. The differences of screening results were analyzed via one-way ANOVA with Dunnett's multiple comparison test. Other samples were analyzed with Student's t-test, and probability values of 0.05 or less were considered statistically significant.

Screening the Active Components of Ganoderma for Inhibitory Properties
To identify components that regulate inflammation, FPR ligandinduced superoxide generation and cell degranulation assays were used for initial screening (Supplementary Figure 3). After exclusion of the cytotoxicity (data not shown), six compounds with similar structure (Figure 1) were selected from a pool of 34 Ganoderma-derived compounds, by using differential HL-60 (dHL60) cells. Since neutrophils have to be freshly isolated from human subjects and their lifespan is short, dHL60 cells were used as a substitution model for neutrophils in this study. HL-60 is a human promyelocytic leukemia cell line that acquires neutrophil-like properties when differentiated with 1.3% DMSO (Hauert et al., 2002). The preliminary data confirmed that both FPR1 and FPR2 were highly expressed in dHL60-6d cells, and the cells could efficiently generate superoxide upon stimulation with fMLF or WKYMVm (Supplementary Figure 4).
The initial screening revealed that among these 6 compounds, cochlearin H (C9) , chizhine D (C10) (Luo et al., 2015), C30 (unpublished), and especially ganomycin F (C18) (Peng et al., 2016) Figure 3A). Meanwhile, C18 also has distinct inhibition on WKYMVm-induced cell degranulation in dHL60 cells (Supplementary Figure 3B). This compound was selected for further analysis because it was the only compound in the small group that also inhibited WKYMVm-induced Ca 2+ mobilization (Figure 2A), although C18 alone could not induce Ca 2+ mobilization even at micromolar concentrations A B C D FIGURE 2 | Screening results of Ganoderma-derived compounds. RBL-FPR2 cells were seeded until the confluence reached nearly 90%, and then incubated with FLIPR calcium-sensitive dye and 10 mM of Ganoderma-derived compounds for 60 min at 37°C. After that, the cells were simulated with 10 nM WKYMVm and relative fluorescence unit (RFU) was recorded. The results show that only C18 could inhibit calcium mobilization in stably transfected RBL-FPR2 cells (A). Differentiated HL-60 cells (dHL60-6d, 1×10 5 cells per well) were pre-incubated with C18 (10 nM-20 mM) for 30 min, 37°C. Chemiluminescence count per second (CPS) was continually recorded after stimulation with 1 mM WKYMVm, as described in Methods. The results show that C18 displayed significant inhibition of superoxide generation, with an IC 50 of 4.0 mM (B). dHL-60 cells (2×10 5 cells per well) were pre-incubated with different concentrations of C18 (10 nM-20 mM) for 30 min and then seeded to transwell plate. WKYMVm (1nM) was placed in the bottom well (600 ml), which was separated from the lower compartment by a polycarbonate membrane filter with pore size of 5 mm. After incubation at 37°C for 2 h, the total number of migrate cells were counted. Data were presented as chemotaxis index, which represents the ratio of cells migrated toward agonists over the cells migrated toward medium. The results show that C18 has an IC 50 of 3.8 mM for cell chemotaxis (C). HEK-293 cells were seeded in 96-well plate for 24 h and then treated with different concentrations of C18 (1 mM-50 mM) for another 24 h. The viability of cells was then measured with CCK8 kit (Dojingdo, Japan) according the manufacturer. The IC 50 value (32.9 mM) for C18 was then calculated (D). Data are shown as Mean ± SD of three independent experiments. *P < 0.05 versus vehicle-treated cells (control group). RBL, Rat basophils leukemia cells; FPR2, formyl peptide receptor 2.
(Supplementary Figure 5). Furthermore, C18 dose-dependently inhibited superoxide generation ( Figure 2B) and cell migration ( Figure 2C) while having low cell toxicity ( Figure 2D). A checker-board analysis was performed and it was confirmed that the inhibition in chemotaxis induced by C18 was due to the concentration difference between the upper chamber and lower chamber (Supplementary Table 1).

Identification of the Pharmacological Target of C18
To further investigate the action mechanism of C18, the inhibitory effects on several GPCR-dependent pathways were compared, including FPR1, FPR2, and the complement component 5a (C5a) receptor. In addition, phorbol-12myristate-13-acetate (PMA), an analogue of diacylglycerol (DAG), can directly stimulates protein kinase C (PKC) for PKC-dependent superoxide generation (Mellor and Parker, 1998) and was therefore included as a GPCR-independent agonist. As shown in Figure 3, C18 was selective for FPRmediated superoxide generation induced by fMLF ( Figure 3A) and WKYMVm ( Figure 3B), as superoxide generation induced by C5a ( Figure 3C) and PMA (Supplementary Figure 6) was not significantly inhibited by 5 µM C18. The results were further confirmed in genetically engineered COS phox cells expressing FPR2 (Nie et al., 2010). Since COS phox cells stably express gp91 phox , p22 phox , p67 phox , and p47 phox and lack the hemopoietic specific proteins such as the FPRs, this cell model could be useful to generate high-level superoxide in a FPR2dependent manner (Price et al., 2002). As shown in Figure 3D, C18 at 5 µM strongly inhibited WKYMVm-induced superoxide generation, similarly to the inhibitory effect seen in the dHL60 cells. Since the C5a receptor signaling mechanism is similar to that of the FPRs, these results indicate that C18 exerts its inhibitory effect at the receptor level of FPR.

Investigation of the Effects of C18 on FPR-Mediated Cellular Functions
As shown in Supplementary Figure 7, besides superoxide generation and chemotaxis assays, formyl peptide receptors (FPRs) mediate other cellular responses such as Ca 2+ mobilization, and degranulation during innate immune response. Based on the previous results that C18 exerts its inhibitory effects at receptor level, a series of functional assays were performed using RBL-FPR2 cells, because C18 only compete with ligand binding to FPR2 but not FPR1 (data not shown), as discussed below. As showed in Figure 4, preincubation with different concentrations of C18 caused a rightshift of the EC 50 values in Ca 2+ mobilization assays, increasing from 42.3 pM with 1 mM of C18 to 255.8 pM with 10 mM of C18 ( Figure 4A), with concomitant reduction in the maximum response ( Figure 4B). Meanwhile, C18 exhibited inhibitory effects in degranulation assays in a dose-dependent manner ( Figure 4C). To investigate the molecular mechanism, competitive binding assays were performed to verify whether C18 could compete for active site of FPR2, and the results show that C18 could not compete effectively with WKYMVm-FITC, but could compete partially at higher concentrations (IC 50 : 3.2 mM). In comparison, WKYMVm could compete with its fluorescein labeled ligand with a higher affinity (IC 50 : 1.7 nM;) ( Figure 4D). As will be discussed below, C18 may have an allosteric effect on FPR2 at higher concentrations.

Observation of the Changes of C18 Targets on FPR-Mediated Cell Morphology
Due to its inhibitory effects on almost entirely FPR-mediated functional assays, cell morphological changes caused by C18 have attracted our attention. As shown in Figure 5, 10 mM; C18 was preincubated with RBL-FPR2 cells for 30 min, then stimulated with 1 mM; WKYMVm for another 15 min and fixed. After staining the actin and nuclei, the results showed that C18 indeed interfered with morphological changes induced by FPR2 agonists. It could limit cell extrusion induced by WKYMVm ( Figure 5), but had no effect on unstimulated RBL-FPR2 cells. Therefore, C18 exerts an effect through FPR2. This may also explain why C18 could inhibit cell chemotaxis towards FPR2 agonist, but the detailed mechanism still need to be further investigated.

Comparison of FPR2 Conformational Changes Induced by C18 and WKYMVm
Considering that C18 could compete for the active binding site of FPR2, FRET detection assay was conducted to investigate molecular dynamics of ligand-induced receptor conformation. FPR2 fluorescent biosensors were generated by placing into one of its intracellular loops a FlAsH binding motif and an enhanced cyan fluorescent protein (ECFP) in its C-terminus ( Figure 6A). Using this FPR2(ICL3) + ECFP construct, WKYMVm induced a decrease in FRET signal, suggesting that the C-terminal ECFP moved away from the ICL3-inserted FlAsH ( Figure 6B). In comparison, C18 but not C12 induced an opposite conformational change, suggesting that the C-terminal ECFP moved closer to the ICL3 upon C18 stimulation. These results confirmed that C18 could act directly on FPR2 and cause conformation changes opposite to those induced by WKYMVm, thereby reducing the stimulation effects. The fact that C12 could not induced conformational changes of FPR2 indicates selectivity of C18 at the receptor level.

Analysis of the Interaction Between C18 and FPR2 With Molecular Docking
Molecular docking analysis was performed to investigate the detailed interaction between C18 and FPR2. As shown in Figure 7, FPR2 has a relatively large binding pocket for its agonists. The pan-agonist WKYMVm has a high calculated affinity for FPR2 and fits well with this C5a-based model by occupying the binding pocket completely. Compared with WKYMVm, C18 could only occupy a small area of the binding pocket with a lower calculated affinity, and the affinity was also confirmed by previous competitive binding assay ( Figure 4C). It was predicted that C18 forms one key hydrogen bond with Ser 84 and several hydrophobic interactions with FPR2. The molecular binding of C18 and FPR2, based on the docking analysis, blocks some of the binding sites of WKYMVm, such as Ser 84 , His 102 , Val 105 , Phe 257 , Asn 285 , Phe 292 , that are vital to ligand binding and hydrogen bond formation for FPR2 (Fujita et al., 2011;Schepetkin et al., 2011;Stepniewski and Filipek, 2015). The docking data provide a structural basis for C18 interaction with FPR2.

DISCUSSION
As FPRs are important regulators in various disease, low-molecularweight compounds that could target FPRs and FPR-related signaling pathways may have great potential in the discovery of drugs for treating inflammatory diseases (Dahlgren et al., 2016). Several FPR-mediated functional assays were adopted for screening of Ganoderma-derived compounds and one compound, C18, was found to clearly limit the FPR2 agonist-induced cellular responses. Further studies investigated the molecular mechanism under these immune-modulating actions, and the results showed that C18 could inhibit FPR-mediated cell extrusion as well as superoxide generation. Meanwhile C18 caused FPR2 conformational changes that are different from the agonist induced conformational changes. Since C18 could compete partially with WKYMVm at higher concentration, it perhaps binds to an allosteric site on FPR2 and then partially limits FPR-mediated cellular responses. The assumption was further confirmed by molecular docking analysis. C18 could form one hydrogen bond and several hydrophobic interactions with FPR2. Based on these results, it was suggested that the Ganderma-derived C18 compound could exert its inhibitory effects through binding to an allosteric site on FPR2. The binding causes a conformational change that limit the activation of the receptor by its agonist, and further inhibit its downstream signaling pathway, resulting in reduced production of superoxide anions and compromised cell chemotaxis, thus relieving the symptoms of inflammation. Based on these results, we believed that Ganoderma-derived C18 may be a potential candidate for antiinflammatory compound that exerts inhibitory effects through FPR2. It is well documented that herbal ingredients, including many that exhibit anti-inflammatory activities, can act on multiple targets. To identify the potential target (s) of actions of C18, we conducted a number of assays including competitive binding and differential activation of the NADPH oxidase through FPRdependent (fMLF and WKYMVm) and FPR-independent (PMA) pathways. In addition, we have included a different chemoattractant receptor, the C5a receptor, that uses similar or even identical signaling pathways for the activation of cellular functions in neutrophils . Our results clearly demonstrate that C18 inhibited the FPR-mediated superoxide generation while having no effects on C5a receptor-mediated and PKC-dependent (PMA-induced) superoxide activation. C18 was not as effective on FPR1-mediated superoxide generation, prompting us to select FPR2 for further investigation on conformational changes induced by WKYMVm, C12 and C18. Using a FLAsH-based single-molecule FRET detection assay, we found that WKYMVm and C18 altered the FRET intensity in opposite directions, suggesting that C18 interacts directly with FPR2 and induces different conformational changes in FPR2. In comparison, C12, a structural analogue of C18, failed to induce FPR2 conformational changes suggesting that C18 is highly selective for FPR2. Combined with the results from the competitive binding assay that showed lower binding affinity, we postulate that C18 serves as a negative modulator of FPR2 in a manner that differs from a typical neutral antagonist.
Due to the absence of crystal structures for FPRs, the detailed recognition between FPRs and ligands remains unclear. To explain ligand-FPR interactions, homology models were adopted and molecular docking analysis based on the previously site-directed mutagenesis studies were used as an alternative method. In this study, the recently acquired crystal structure of C5a receptor was used as a template model due to its higher sequence similarity (34.5%) and resolution (2.7 Å) than the previously used CXCR4 A D E B C FIGURE 7 | Molecular docking analysis of the interaction between FPR2 and its ligands, C18, and WKYMVm. Docking models of FPR2 was obtained from Swissmodel server which was based on C5a receptor (PDB code: 5o9hA) because of its higher sequence similarity (34.5%) and resolution (2.7 Å). The initial conformations of the ligands were generated by ChemBio 3D. Hydrogens were added by Autodock Tools and molecular docking was performed by Autodock Vina.
The search box was set as 46 Å × 34 Å × 60 Å for FPR2. The best conformation was refined with energy minimization and analyzed with PyMOL molecular graphics system. The docking results show that FPR2 has a relatively large binding pocket (A), and C18 could occupy a part of the FPR2 binding pocket (B) compared with WKYMVm (C) (C18 is colored in orange, WKYMVm in green). There was a partial overlap between the C18 binding site and the WKYMVm binding site in FPR2. C18 is predicted to form hydrogen bond with Ser 84 and several hydrophobic interactions with FPR2, such as Leu 81 , Met 85 , Ser 288 , Val 160 , His 102 , Phe 257 , and probably more (D). WKYMVm is predicted to form six hydrogen bonds with His 102 , Thr 177 , Tyr 277 , and Ser 288 and several hydrophobic interactions (E). FPR2, Formyl peptide receptors.
template (He and Ye, 2017). The C5a receptor-based model revealed a relatively larger binding pocket of FPR2, and the natural agonist WKYMVm fits well with this model. More recently, Stepniewski et al. used a dual template approach with CXCR4 and mOR, and not only confirmed the previously identified residues such as His 102 , Phe 257 , and Arg 201 for ligand binding, but also found two novel residues, Ser 84 and Asn 285 , presumably important for the formation of hydrogen bonds (Stepniewski and Filipek, 2015). In the present study, we found that C18 could form hydrogen bond with Ser 84 , while other structural homologs such as C12 could not form hydrogen bond with Ser 84 (data not shown). This amino acid will be of interest in future studies using sitedirected mutagenesis. C18 contains more hydroxyl groups (-OH), that may be acceptors or donors of H-bonding, than its analogues in this group of Ganoderma-derived compounds. This H-bonding donor/acceptor feature indicates that C18 has similar property as those of the reported FPRs antagonists (Schepetkin et al., 2014), that inhibit FPR-mediated pro-inflammatory response.
Our initial experiments were conducted using dHL60 cells that express both FPR1 and FPR2. However, when using stablytransfected FPR1-RBL cells, C18 had no effect on maximal Ca 2+ mobilization induced by fMLF, not did it compete with FITClabelled fMLF (data not shown). These and other results shown in this study support the notion that C18 acts at FPR2. Since FPR1 and FPR2 can form homodimer or heterodimer as reported by Cooray and coworkers (Cooray et al., 2013), each dimer of FPR may yield specific signaling pathways to resolve inflammation (Filep, 2013;Krishnamoorthy et al., 2018). Thus, C18 should be further tested in cells for the possibility of using FPR heterodimer (FPR1-FPR2) or homodimer (FPR2-FPR2) for potential anti-inflammatory activity.
As a new compound of Ganoderma meroterpenoids (GMs), C18 was firstly isolated in 2016, and the molecular formula was determined as C 21 H 30 O 3 by HRESIMS and 13 C-DEPT NMR (Peng et al., 2016). Although the chemical properties have been determined, its biological activities remain unclear. GMs include two parts, a 1,2,4-trisubstituted phenyl and a polyunsaturated terpenoid. Compared the structure of these GMs in Figure 1, it is evident that the hydroxyl group on polyunsaturated terpenoid plays a vital role in bioactivities. These diverse structural skeletons and related bioactivities of GMs, as well as the development of chemical synthesis methods (Peng and Qiu, 2018), have attracted more attention in recently years. Considering that the IC 50 of C18 is still poor, there is a need for further modification of the compound for better activity and reduced cytotoxicity, that will be used for investigation of the anti-inflammatory effects with in vivo studies.

DATA AVAILABILITY STATEMENT
The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation, to any qualified researcher.

AUTHOR CONTRIBUTIONS
HW and XP contributed equally to this work. HW performed experiments, collected and analyzed data, and prepared figures. XP purified compounds and performed initial characterization. YG and SZ prepared fluorescent biosensors and assisted in functional assays. YF performed supplementary chemotaxis assay. ZW and WH performed molecular docking and analysis. MQ and RY designed the study. RY and HW wrote the manuscript. All authors have given approval to the final version of the manuscript.