Tyrosinase inhibition by p‐coumaric acid ethyl ester identified from camellia pollen

Abstract A tyrosinase inhibitor was separated from camellia pollen with the aid of solvent fraction, macroporous adsorptive resin chromatography, and high‐speed countercurrent chromatography. The inhibitor was identified to be p‐coumaric acid ethyl ester (p‐CAEE) by nuclear magnetic resonance and mass spectrum. Its inhibitory activity (IC50 = 4.89 μg/ml) was about 10‐fold stronger than arbutin (IC50 = 51.54 μg/ml). The p‐CAEE inhibited tyrosinase in a noncompetitive model with the K I and K m of 1.83 μg/ml and 0.52 mM, respectively. Fluorescence spectroscopy analysis showed the p‐CAEE quenched an intrinsic fluorescence tyrosinase. UV‐Vis spectroscopy analysis showed the p‐CAEE did not interact with copper ions of the enzyme. Docking simulation implied the p‐CAEE induced a conformational change in the catalytic region and thus changed binding forces of L‐tyrosine. Our findings suggest that p‐CAEE plays an important role in inhibiting tyrosinase and provides a reference for developing pharmaceutical, cosmetic, and fruit preservation products using pollen.

dichroism (CD), UV-Visible spectral, and fluorescence techniques (Guo et al., 2018). Furthermore, X-ray and computational techniques have been successfully applied to analyze the binding sites and molecular interactions between inhibitors and enzymes (Fujieda et al., 2020;Song et al., 2020).
Ellagic acid has poor bioavailability (Arulmozhi et al., 2013) and potential adverse interactions with taxane chemotherapy (Eskra et al., 2019). So it is of interest to identify and characterize novel tyrosinase inhibitors from plants.
Pollen that plays an essential role in the sexual propagation of plants carries a variety of nutrients and bioactive compounds necessary for survival and fusion with a female gamete (Edlund et al., 2004). It contains an abundance of proteins, lipids, vitamins, minerals (Ares et al., 2018), and bioactive compounds such as phenolic compounds, flavonoids, and tocopherols (Almeida-Muradian et al., 2005;Li, Wang et al., 2018;Li, Yuan, et al., 2018). Moreover, studies have demonstrated that pollen has the ability to inhibit tyrosinase (Zhang et al., 2015) and regulate melanogenesis of B16 cells in the cAMP/MITF/TYR pathway (Sun et al., 2017). Some tyrosinase inhibitors, including kaempferol, levulinic acid, and 5-hydroxymethyl furfural, have been characterized from pollens (Yang et al., 2016(Yang et al., , 2019. Thus far, tyrosinase inhibitors in pollen have not been sufficiently elucidated. In the present study, a new tyrosinase inhibitor has been separated and characterized from camellia pollen. The specific contents included the following: (a) separate and purify the tyrosinase inhibitors by macroporous adsorption resin chromatography and HSCCC; (b) identify the structure of the tyrosinase inhibitor; (c) characterize the inhibition by kinetic, CD, fluorescence, and UV-Visible spectral analyses; and (d) elucidate the interaction between the inhibitor and the enzyme using computational simulation. This study could help people to understand the tyrosinase inhibition activity of pollens and provide a reference for developing effective tyrosinase inhibitors for whitening agents and other beneficial products.

| Chemicals and reagents
Analytical-grade ethanol, petroleum ether, ethyl acetate, and n-butanol were purchased from Sinopharm Chemical Reagent Corporation. Chromatographic grade acetonitrile was purchased from Tedia Company Inc. Diaion HP-20 macroporous adsorption resin was purchased from Beijing Green Herbs Co., Ltd. Mushroom tyrosinase (E.C.1.14.18.1), L-tyrosine, and standards of arbutin and p-coumaric acid ethyl ester were purchased from Merck Life Science Co., Ltd. Camellia pollen was purchased from Beijing Tong Ren Tang Co., Ltd.

| Purification tyrosinase inhibitor from pollen
Dry camellia pollen (500 g) was powdered by a kitchen blender (JP-500C, Jiu Pin Dian Qi Co., Ltd). Then, the pollen powder was extracted with 5 L 95% (v/v) ethanol for three times (each for 8 hr) at 50°C. The extracts were pooled and vacuum concentrated to yield a slurry (200 g) that was then further extracted by 1 L 99.5% (v/v) ethyl acetate for four times. The ethyl acetate extract was vacuum concentrated at 50°C. The concentrate (10 g) was dissolved in 1 L water and separated by HP-20 macroporous adsorptive resin using batch elution by 30% (v/v) ethanol. The eluents were vacuum concentrated at 50°C using a rotary evaporator. Fraction 3 of the eluent was hydrolyzed using a previously reported method with minor modifications (Nuutila et al., 2002). Briefly, 0.1 g fraction 3 eluent was dissolved in 50 ml 95% (v/v) ethanol and then mixed with 50 ml 4 M hydrochloric acid, sealed, and incubated in a water bath at 90°C for 90 min. After cooling down to room temperature, the hydrolyzate was evaporated in a rotary evaporator at 50°C under vacuum.
The dried hydrolyzate was submitted to high-speed countercurrent chromatography separation (TBE-300C, Shanghai Tauto Biotech Co., Ltd.) using the solvent mixture of n-hexane, ethyl acetate, methanol, and water at the ratio of 4:6:4:6 (v/v/v/v). The flow rate was 3 ml/ min, and the column temperature was kept at 25°C. The effluent from the outlet of the column was monitored with a UV detector at 280 nm. Fractions were collected according to the chromatograms.

| High-performance liquid chromatography analysis
Agilent 1260 HPLC system (Agilent Technologies Co.) and reversephase Symmetry C18 column (150 mm × 4.6 mm i.d., 3.5 µm, Waters) were used for HPLC analysis. The mobile phase was composed of ultrapure water (mobile phase A) and acetonitrile (mobile phase B).

| Structure identification of the tyrosinase inhibitor
The nuclear magnetic resonance spectroscopy (Bruker AVANCE III 400, Bruker BioSpin Corporation) and deuterated methanol solution (CD 3 OD) were used for analysis. Mass spectrometry data were obtained in positive ionization mode from an Agilent 6460 triple quadrupole tandem mass spectrometer (Agilent Technologies Co.) with an ESI interface. The gas temperature was 325°C, gas flow was 12 L/min, nebulizer was 45 psi, sheath gas temperature was 300°C, sheath gas flow was 12 L/min, capillary voltage was 4,000 V, and nozzle voltage was 450 V.

| Tyrosinase inhibitory activity assay
The inhibition of tyrosinase in vitro was measured by a previously reported method with minor modifications (Rezaei et al., 2018). Different concentrations of collected samples (40 µl) were transferred to a 96well plate and then mixed with 80 µl 20 mM phosphate buffer (pH 6.8) and 40 µl 250 U/ml tyrosinase solution. After incubation at 25°C for 10 min in the darkness, 40 µl 0.85 mM substrate L-tyrosine was added to the mixture. The absorbance of each well was measured using a microplate reader (FLUO star OPTIMA) at wavelength 492 nm. Arbutin was severed as a positive control. Each experiment was done in triplicate, and the tyrosinase inhibitory activity (%) was calculated as flow: where A1 is the absorbance of solution with active tyrosinase and sample; A2 is the absorbance of solution with inactivated tyrosinase and sample; A3 is the absorbance of solution with active tyrosinase and methanol; A4 is the absorbance of solution with inactivated tyrosinase and methanol.
The IC 50 value is defined as the inhibitor concentration required to reach a 50% inhibition of tyrosinase activity and measured by the linear fitting.

| Kinetic analysis
The enzyme activity was measured at different concentrations of peak IV (0, 1.5, 2.0, 2.5, and 3.0 μg/ml) and different concentrations of tyrosinase (0, 62.5, 125, 250, and 500 U/ml). A linear regression was created in a double reciprocal plot of the reaction rate and the concentration of the substrate. In addition, the Lineweaver-Burk plot was used to determine the type of tyrosinase inhibition according to a previously described method with modifications (Espin et al., 2000). The reaction rate was measured at different concentration of the substrate L-tyrosine (0.1, 0.2, 0.3, 0.4, and 0.5 mM), and different concentrations of peak IV (0, 1.5, 2.0, 2.5, and 3.0 μg/ml). For the noncompetitive type inhibition, Michaelis constant K m was calculated by the equation below: where K m is the Michaelis constant; V max is the maximum velocity; V is the reaction rate, and [S] is the substrate concentration.

| Circular dichroism spectroscopy assay
The CD spectroscopy (Jasco-810 spectrophotometer, JASCO) analysis was done according to a previously reported method (Biswas et al., 2017). Briefly, 285 μL of 1.86 mg/ml tyrosinase solution was mixed with 20 μL different concentrations of peak IV solution to reach a final molar ratio of 0:1, 1:1, or 4:1. Then, each tyrosinase inhibitor mixture was injected into a 1-cm-path length quartz cuvette and the CD spectrum. The detection conditions were as follows: The scanning wavelength was from 190 to 250 nm, the scanning rate was 100 nm/min, the resolution time was 0.5 s, the response time was 1 s, and the bandwidth was 2 nm. After subtracting the blank (0.5% methanol solution) signal, structure parameters (α-helices, β-turns, β-sheets, random Coils) were analyzed by the CDNN program v.2.1 from Applied Photophysics, Ltd.

| Fluorescence emission spectroscopy assay
A mixture of 50 μL tyrosinase (250 U/ml) and 925 μL 20 mM phosphate buffer (pH 6.8) was incubated with 25 μL peak IV at different concentrations of 0, 100, 300, and 500 μg/ml. The emission spectra were recorded at 20°C by using a 1-cm-path length quartz cuvette in a Varian Cary Eclipse fluorescence spectrometer (Varian, Inc.). The excitation wavelength was 280 nm, and the emission spectra were recorded from 300 to 480 nm. The excitation and emission slits were 5 nm, the scanning rate was 1,000 nm/min, and the resolution was 1.0 nm.

| UV-Vis spectra analyses
Five samples were prepared, sample 1 contained 1 mg/ml of p-coumaric acid ethyl ester only; sample 2 contained 0.125 mM of copper (II) sulfate solution only; sample 3 contained 500 U/ml of tyrosinase only; sample 4 was composed of 1 mg/ml of p-coumaric acid ethyl ester and 0.125 mM of copper (II) sulfate solution; and sample 5 was composed of 1 mg/ml of p-coumaric acid ethyl ester and 500 U/ml of tyrosinase. The change in the absorbance and wavelength on the interaction was recorded using UV-Vis spectrophotometer with a wave scan range from 200 to 600 nm.

| Computational simulation analysis
The crystal structure of tyrosinase (PDB code: 2Y9X) (Chai et al., 2018) was applied as model of molecular docking.
ChemBioDraw Ultra 12.0 was used to prepare the three-dimensional

| Statistical analysis
All samples were analyzed and evaluated in triplicate. The results were expressed as a mean ± SD, and the data were analyzed by using one-way ANOVA followed by a t test to determine any significant differences. The values of p < .05 were considered as statistically significant.   Previous study reports that p-coumaric acid ethyl ester has the same molecular weight (Mussatto et al., 2006). Figure 2c shows the nu-    (Kubo et al., 2003). These results indicate that the inhibition of the p-CAEE on tyrosinase was reversible by noncovalent bonds. The enzyme can be reactivated by physical removal of the bonds or inhibitors (Chen & Kubo, 2002). Figure 4 shows that the x-intercept (−1/K m ) remained the same value but the y-intercept (1/V max ) increased with increasing concentrations of p-CAEE. K m remained constant about 0.52 mM, but V max decreased after the addition of p-CAEE. Furthermore, the parameter K i was determined to be 1.83 μg/ml. The fixed K m value indicates that no competition existed between the substrate and p-CAEE.

| Tyrosinase inhibitory kinetics
Early study reveals that the inhibitor binds reversibly to the free enzyme or enzyme-substrate complex equally in case of the noncompetitive mechanism (Lopes et al., 2018). Noncompetitive inhibition means that the inhibition by inhibitors cannot be overcome by increasing substrate concentrations (Chen et al., 2005). Thus, our results indicate that p-CAEE inhibited tyrosinase in a noncompetitive mechanism. Figure 5a shows that the spectral datum of tyrosinase exhibits 2 negative bands at 208 and 220 nm. This character is the typical feature of α-helix resulting from the transition of amide groups in tyrosinase (VanGelder et al., 1997). Furthermore, the secondary conformation of tyrosinase was modified between 190 and 260 nm. The inserted table of Figure 5a shows that the content of α-helix, β-turn, and random coil decreased from 20. 9, 26.1, and 35.1% to 20.5, 24.7, and 33.3% and the content of β-sheet increased from 22.0% to 24.3% after the addition of p-CAEE.

| Circular dichroism, fluorescence, and UV-Vis spectra analyses of the conformation changes of the tyrosinase
This result suggests that p-CAEE changed the conformation of tyrosinase. Figure 5b shows the fluorescence spectrum of tyrosinase presented a strong emission with maximum wavelength at 340 nm.
With the addition of p-CAEE to the tyrosinase solution, the fluorescence intensity reduced progressively. According to a previous study, the microenvironment of tryptophan influences the fluorescence intensity at fluorescence emission maximum wavelength (340 nm) (Yousefi et al., 2013). Therefore, our result indicates that p-CAEE reduced the fluorescence intensity without changing the microenvironment of tyrosinase.  (Rolff et al., 2011). Some inhibitors inhibit tyrosinase by chelating the copper ion in the active site (Hridya et al., 2015) such as kojic acid (Chang, 2009) and oxalic acid (Son et al., 2000). Our results indicate that p-CAEE did not directly chelate copper ions at the active site of tyrosinase. His 259, His 263, and His 294 of the set B). Early study reports that each set is oriented by coordinately binding with copper ion (Guo et al., 2018). These copper ions are directly involved in enzymatic oxidation (Rolff et al., 2011). Figure 6b shows that p-CAEE is bound with tyrosinase at a site outside the active center and far away from the copper ions. p-CAEE did not directly act on the copper ions or compete with the ligand.

| Computational simulation of the molecular interaction between p-coumaric acid ethyl ester and tyrosinase
But p-CAEE changed the location of the copper ions and the conformation of a loop adjacent to the active center. Figure 6c shows that p-CAEE did not interact with copper ions directly, but three histidine residues combining with copper ions were dispersed. The result indicates that p-CAEE changed the location of copper ions in the active center.

CO N FLI C T S O F I NTE R E S T
The authors declare that there is no conflict of interest regarding the publication of this paper.

DATA AVA I L A B I L I T Y S TAT E M E N T
Data sharing is not applicable to this article as no new data were created or analyzed in this study.