Synthesis and evaluation of coumarin derivatives on antioxidative, tyrosinase inhibitory activities, melanogenesis, and in silico investigations

New coumarin derivatives were designed using a 2-(2-oxo-2H-chromen-4-yl)acetic acid scaffold conjugated with amino acid esters or tyramine. The anti-tyrosinase and anti-lipid peroxidation activities of the synthesized compounds were investigated. Coumarin derivatives 7,9, 11–13, 15–18 showed strong anti-lipid peroxidation activity. Compound 13 exhibited uncompetitive tyrosinase inhibitory activity with an IC50 value of 68.86 µM. Compound 14 (% activity = 123.41) showed stronger tyrosinase activating activity than 8-methoxypsolaren (8-MOP, % activity = 109.46). In silico studies revealed different poses between the inhibitors and activators near the tyrosinase catalytic site. Compounds 13 (25–50 μM) and 14 (25–100 μM) did not show cytotoxicity against B16F10 cells. In contrast to the tyrosinase inhibition assay, compound 13 (50 μM) suppressed melanogenesis in B16F10 cells with two times higher potency than KA (100 μM). Compound 14 at 100 μM showed melanogenesis enhancement in B16F10 cells in a dose-dependent manner, however, inferior to the 8-MOP. Based on the findings, compound 13 and 14 offer potential for development as skin-lightening agents and vitiligo therapy agents, respectively.


Tyrosinase inhibition assay
The synthesized compounds were screened for their effect on mushroom tyrosinase using KA as a reference tyrosinase inhibitor and 8-methoxy-psoralen (8-MOP) as a reference tyrosinase activator.Table 1 displays the % activity of mushroom tyrosinase following incubation with the studied compounds; when the % activity value is less than 100, the molecule is considered an inhibitor, and when it is greater than 100, the compound is considered an activator 20 .

Enzyme kinetic study
Double-reciprocal plots were used to investigate the mechanisms of mushroom tyrosinase inhibition by compound 13 (Fig. 2).Plotting 1/V vs. 1/[S] yielded parallel straight lines with equal slopes, indicating uncompetitive inhibition.

Molecular docking
Tyrosinase enzyme is made up of two identical H and two identical L subunits.The two copper ions (Cu 2+ ), each of them paired with three histidine residues, are located at the active site of the H subunit.The first Cu 2+ was Table 1.Lipid peroxidation and mushroom tyrosinase inhibitory activity of the coumarin derivatives.a %lipid peroxidation inhibition at 1 mg/ml of tested compound.b %mushroom tyrosinase activity at 100 µM of tested compound.BHT butylated hydroxytoluene, KA kojic acid, 8-MOP 8-methoxypsoralen, n.d.not determined, N.A. not applicable.The docking behavior of the synthesized compounds (7-18) and the reference compounds (KA, 8-MOP, and tropolone) with the mushroom tyrosinase protein (PDB ID: 2Y9X) was evaluated using the docking score (kcal/mol) from AutoDock4 22 as shown in Fig. 3.In Fig. 3, the synthesized compounds (7-18) were separated into two groups based on the coumarin substituent(s).Tyrosinase inhibitors included phenylalanine ethyl ester derivatives (7, 11, 15) and tryptophan ethyl ester derivatives (9, 13, 17), while tyrosine ethyl ester derivatives (8, 12, 16) and tyramine derivatives (10, 14, 18) acted as tyrosinase activators.Compounds 7, 9, 13, and 17 were bound at the active site with different conformational poses and binding energy values, except compounds 11 and 15, which were located far from the protein's pocket site (see Fig. 4A,B).The coumarin ring could bind with both copper ions in this group, the   www.nature.com/scientificreports/same as the KA and tropolone (Fig. 4C).Except for compound 10, all activators (8, 10, 12, 14, 16, 18) were well accommodated in the binding site, and their phenolic group of tyrosine or tyramine moiety interacted with the copper ions in the same manner as 8-MOP (Fig. 4D-F).
Compounds 13 and 14 were found to be tyrosinase inhibitors and activators in vitro, respectively.As a result, these docking complexes were used for conformational analysis.The docking score of compound 13 (inhibitor, ΔG = − 8.09 kcal/mol) was lower than that of KA (ΔG = − 4.95 kcal/mol).Compound 13's carbonyl oxygen atom formed a hydrogen bond with His244.The coumarin ring interacted with His259 and His263 through π-π stacking.Carbonyl oxygen on the coumarin ring also forms hydrogen bonds with His61 and His85.In the case of the reference KA, the molecule has a π-π interaction with His263 and forms hydrogen bonds with Met280 and Ser282 (Fig. 5).According to the tyrosinase inhibitor's docking results, replacing the hydroxyl group at the C7-position with the methoxy group may reduce the predictive binding score, for example, in compounds 7 and 11, as well as 9 and 13.The lower binding energy caused by methoxy-substituted coumarin (compound 13 with respect to the compound 9) may affect tyrosinase inhibitory efficacy because the coumarin ring remained close to the Cu 2+ in both compounds 9 and 13, and the coumarin ring can participate in the Cu 2+ -associated electron transfer process.In this case, it is possible that if the binding scores were similar, indicating that the compounds have the same preference for the site, it would not contribute significantly to inhibitory efficacy as long as the coumarin ring remained close to the Cu 2+ ions (Fig. 5B).However, if the hydroxyl group was added to the C6 position along with the methoxy at the C7-position, as in compounds 13 and 17, compound 17's docking score decreased, and the coumarin ring remained farther away from the Cu 2+ than compound 13.These suggested that the hydroxyl group at the C6-position could reduce inhibitory efficacy by inducing the binding pose, resulting in a greater distance between the coumarin ring and the Cu 2+ .
For the tyrosinase activator, the compound 14 (ΔG = − 7.98 kcal/mol) had a lower binding energy than 8-MOP (ΔG = − 6.98 kcal/mol) in the case of the activator.The carbonyl oxygen atom on the coumarin ring interacted with Arg268 via a hydrogen bond.Furthermore, the benzyl group on the tyramine moiety interacted with His259 and His263, whereas 8-MOP only interacted with His263 via π-π stacking and His296 via hydrogen bonding (Fig. 6).Likewise, the methoxy substitution at the C7-position in compounds 10 and 14 reduced the docking score, but compound 14 converted the phenolic group to Cu 2+ , whereas compound 10 used the coumarin ring (Fig. 4E).Despite the fact that compound 18 converted the phenolic group to Cu 2+ similar to the compound 14, the hydroxyl group, inserted at the C6-position in the coumarin, reduced the docking score, which could be explained by the fact that compound 18 had a lower binding preference for this site than compound 14.
To summarize, the docking results show that the inhibitor and activator efficacy can be weighed by two factors: binding preference and binding characteristics.The inhibitory efficacy could be attributed to the distance between the coumarin ring and Cu 2+ , whereas the activator efficacy could be attributed to the phenolic group's proximity to Cu 2+ .However, the compound needed to have a strong binding preference for the tyrosinase copper site.

Cell viability assay
From the synthetic series, we selected the most potent tyrosinase inhibitor (compound 13) and activator (compound 14) and evaluated their toxicity to the skin cells by MTT assay in the B16F10 melanoma cells from Mus musculus skin.B16F10 cells were incubated with compounds 13 (25-100 µM) and 14 (25-150 µM) for 48 h.KA (50-200 µM) and 8-MOP (50-150 µM) were used as reference compounds for the tyrosinase inhibitor and the activator, respectively.A percentage of cell viability was calculated and illustrated in Fig. 7A-D.Compounds 13 did not show cytotoxicity to B16F10 cells at 25-50 µM (%viability > 80%).Cytotoxicity increased when the concentration of the compounds increased to 75 µM.(%viability decreased to 50%).Compound 13 had a narrow cytotoxicity profile in comparison to KA. Compound 14 was less cytotoxic than compound 13 and showed a cytotoxicity profile similar to 8-MOP.

Melanin content assay
Compound 13 and compound 14 were also determined for their melanogenesis activities in the B16F10 cells.Intracellular melanin content was measured.In contrast to the result from the tyrosinase inhibition assay, compound 13 at 50 μM suppressed melanogenesis with approximately two times higher potency than KA at 100 μM.When compared to the untreated cells, compound 13 significantly suppressed melanin synthesis at 50 μM whereas 25 μM inhibited melanogenesis equally to KA at 100 μM (Fig. 8A).As an activator, compound 14 showed less stimulation at 25 and 50 μM and effectively enhanced melanogenesis in B16F10 cells in a dose-dependent manner.However, the increase in melanin content in the cells was lower than cells treated with 8-MOP (Fig. 8B).

Discussion
Twelve coumarin derivatives (compound 7-18) were developed, their anti-oxidative activity and effect on the tyrosinase enzyme were evaluated.The effect of melanogenesis in B16F10 cells was studied on the most potent tyrosinase inhibitor (compound 13) and activator (compound 14).Compounds 7, 9, 11-13, and 15-18 demonstrated excellent anti-lipid peroxidation activity (Table 1).The anti-lipid peroxidation activity tended to depend on R 1 and R 2 substituents on the coumarin moiety of the designed compounds.In general, the 2-( 7 other series bearing a phenolic group on the coumarin ring (compounds 7-10 and 15-18).In each series, the compounds derived from phenylalanine ethyl ester (compounds 7, 11, and 15) showed higher potency than the tyrosine (compounds 8, 12, and 16), tryptophan (compounds 9, 13, and 17), and tyramine (compounds 10, 14, and 18) derivatives.The tyramine derivatives showed the weakest anti-lipid peroxidation activity than other amino acid ester derived-compounds in each series.This might be due to lower lipophilicity of tyramine comparing to the amino acid esters.The anti-lipid peroxidation results suggested additional benefits of the coumarin derivatives for cosmeceutical applications.
In the tyrosinase inhibition activity evaluation, the synthesized compounds showed different impacts on the enzyme (Table 1).The phenylalanine ethyl ester (compounds 7, 11, 15) and tryptophan ethyl ester derivatives (compounds 9, 13, 17) inhibited the function of tyrosinase, while, surprisingly, the tyrosine ethyl ester (compounds 8, 12, 16) and tyramine derivatives (compounds 10, 14, 18) tended to activate the enzyme.The most potent tyrosinase activator and inhibitor were compounds 13 and 14, respectively.www.nature.com/scientificreports/An enzyme activator is a molecule that binds to an enzyme and thus increases its activity.Enzyme activators often bind at the allosteric site and modify the rate of enzyme function.In some cases, when a substrate binds to one catalytic subunit of an enzyme, this can trigger an increase in substrate affinity as well as catalytic activity in the enzyme's other subunits.Many chalcones, phenolic compounds 20,23,24 , flavonoids 25 , and coumarins 26,27 were reported as tyrosinase activators, but the mechanism of the activity is still unknown.Similar to our findings, previous studies have shown that substituent changes can cause a switch from an activator to an inhibitor 20,28 .For example, You, A. et al. 28 showed that adding a thiosemicarbazide group to amino-acetophenones made them inhibitors instead of activators.However, the underlying mechanism was still unknown.
An enzyme kinetic study revealed that compound 13, the most potent inhibitor, uncompetitively inhibited the tyrosinase enzyme.In uncompetitive inhibition, the inhibitor binds at a different site from the substrate and combines with the enzyme-substrate complex (ES), but not with the free enzyme (E).From molecular docking, different binding poses between the inhibitors and the activators at the active site of the mushroom tyrosinase enzyme were observed (Fig. 4).These findings might explain their different effects on the tyrosinase enzyme.Compounds 13 and 14 had lower binding energies than KA and 8-MOP, respectively (Fig. 3).The enzyme inhibition assay suggested that compound 13 (IC 50 68.86µM) was less potent than KA (IC 50 4.75 µM).Compound 14 (%activity 123.41) was determined to be a stronger tyrosinase activator than the reference compound 8-MOP (%activity 101.05).Compounds 13 exhibited a minimally different profile in cytotoxicity against B16F10 melanoma cells in comparison to compound 14.In the melanin content assay, compound 13 exhibited higher melanin suppression at a lower concentration (50 μM) than KA (100 μM) (Fig. 8A); however, this contrasted with the results obtained from the enzyme inhibition assay.SwissADME prediction (accessed from http:// www.swiss adme.ch) 29 showed that compound 13 (cLogP 3.27) exhibited higher lipophilicity than KA (cLogP − 0.16).This suggested that compound 13 is a lipophilic molecule, allowing its ease of penetrating through the cell membrane, thus demonstrating approximately two folds higher melanin suppression.KA, currently used as an inhibitor for melanin inhibition, also showed activity against murine tyrosinase in a broad range (> 100 μM) 30 .Mushroom tyrosinase was used in the enzyme inhibition assay; however, B16F10 cells from murine were used in the melanin content assay.Certain amino acid sequence differences in the primary structure of tyrosinase from different species contribute to variances in binding affinity, resulting in irrelevant results 31 .In addition, mechanisms involved in melanosome activity by other proteins involved in melanogenesis may explain the differences between results from melanoma cells and mushroom tyrosinase 32 .Compound 14 stimulated melanin production in B16F10 cells in the same manner as the mushroom enzyme inhibition assay.

Conclusion
The effects of 12 synthetic coumarin derivatives on tyrosinase activity and melanogenesis have been studied.Compound 13 was discovered to possess strong anti-lipid peroxidation activity and a moderately potent, uncompetitive tyrosinase inhibitor.Compound 14 demonstrated potent enzyme-activating potency.Molecular docking showed that the tyrosinase activators and inhibitors bound to different areas near the catalytic site.Compound 13 suppressed melanogenesis with a higher potency than KA in B16F10 cells.Compound 14 stimulated melanin production in a dose-dependent manner.Compounds 13 and 14 are potential candidates for the treatment of skin pigmentation disorders and as cosmeceuticals.

Methods
All chemicals were purchased from Sigma-Aldrich or Merck AG.The progress of the reactions and the purities of the compounds were checked by thin layer chromatography (TLC) on silica gel 60 F 254 aluminum sheets (Merck AG).

General method for preparation of 2-(2-oxo-2H-chromen-4-yl)acetic acids (4-6)
At 0 °C, resorcinol derivatives (27.25 mmol) were dissolved in 70% sulfuric acid (20 mL), and then 1,3-acetonedicarboxylic acid (27.25 mmol) was added.The reaction mixture was then stirred for 24 h at room temperature.Iced water was poured into the reaction mixture to form precipitates.The solid was collected by vacuum filtration and washed with cold water.The desired product was purified by silica gel column chromatography using a mixture of dichloromethane:methanol (97:3) as the mobile phase.

Lipid peroxidation inhibitory activity
This method was modified from our previous work 18 .Briefly, the test compound (2.0 mg) was dissolved in 0.5 mL of 2.5% linoleic acid in ethanol, 8.0 mL of 10 mM phosphate buffer (pH 7.0), and 2.0 mL of purified water in a screw-capped vial.The control was prepared in the same way but without the test sample.Butylated hydroxytoluene (BHT) was used as a positive control.The experiment was performed in duplicate.Each vial was incubated at 45 °C in the dark for 7 days.The reaction mixture (2 mL) was mixed with 20% trichloroacetic acid (1.0 mL) and 1% thiobarbituric acid in 0.05 N sodium hydroxide (1.0 mL) and heated in a water bath for 20 min.The mixture was left to cool down to room temperature, centrifuged at 4000 rpm for 20 min, and the supernatant was drawn to measure absorbance at 532 nm.Percent inhibition was calculated as follows: %Inhibition = [(A control -A sample )/A control ] × 100, when A control is the absorbance of TBARS solution without sample, A sample is the absorbance of TBARS solution with sample solution.

Mushroom tyrosinase activity assay 20
Samples were dissolved in a mixture of DMSO and the phosphate buffer solution (40%DMSO, 20 µL) and were added into a 96-well microplate filled with phosphate buffer (20 mM, pH 6.8, 140 µL), mushroom tyrosinase (30 U/mL, 20 µL).After 10 min of pre-incubation at room temperature (25 °C), 20 µL of a 0.85 mM L-DOPA (3,4-dihydroxyphenylalanine) solution in phosphate buffer was added to the mixture.(Final concentration of the tested compound equaled 100 µM and final DMSO content equaled 4%).After 20 min of additional incubation, the absorbance at 475 nm was measured using a microplate reader (Power Wave X, California, USA).KA and 8-MOP were used as reference compounds for tyrosinase inhibitors and activators, respectively.Instead of an inhibitor solution, phosphate buffer was added as the negative control.The percentage of tyrosinase activity was calculated as follows: %Activity = (Sample) × 100/Blank, where Blank = (OD of Blank solution at 10 min)-(OD of Blank solution at 0 min) and Sample = (OD of Sample solution at 10 min)-(OD of Sample solution at 0 min).Each sample was analyzed in triplicate (n = 3).The 50 percent inhibition of tyrosinase activity (IC 50 ) studies were carried out using six concentrations of compound 13 (ranging from 0.390625 to 80 µM) and KA (ranging from 0.00305 to 400 µM), with the data analyzed using Sigmoidal dose-response model built in the GraphPad Prism 8.0.2 software.Each sample was examined in triplicate.

Enzyme kinetic study
Utilizing the same protocol as the tyrosinase activity assay, an enzyme kinetics study was conducted with compound 13.A Michaelis-Menten plot was generated by varying the substrate concentration (25-10,000 µM) and reaction rate under conditions with and without inhibitors.A Lineweaver-Burk plot was generated, and the inhibition mode of the inhibitor was determined based on the variation of K m and V max as the inhibitor concentration changed.

Molecular docking
The structure of coumarin was obtained from the PubChem database (CID: 323).The PCCDB project (www.pccdb.org) created the reference compounds, and Molview software (www.molvi ew.org) created the other ligands.The Online SMILES Translator and Structure File Generator (http:// cactus.nci.nih.gov/ trans late/) was used to convert a ligand structure into a Protein Data Bank (PDB) format file.The Protein Data Bank (PDBID: 2Y9X; resolution: 2.78) was used to acquire the crystal structure of mushroom tyrosinase (Agaricus bisporus).The cocrystallized ligand (tropolone) and water molecules were removed.AutoDock Tools (ADT) version 1.5.6 was used to add the hydrogen atoms.AutoDock4 was used to evaluate the molecular docking studies between synthesis compounds (7-18) and mushroom tyrosinase.The grid points in x-y-z were set to 90-90-90 with a spacing of 0.375 Å and the center coordinate (− 11.495, − 23.224, − 40.241).Other parameters in the program were set to default values.As in previous studies 34 , the sampling was carried out using a Lamarckian genetic algorithm (GA) with 50 iterations and a population size of 200.BIOVIA Discovery Studio version 2021 and Visual Molecular Dynamics (VMD) package 35 were used to visualize all molecular interactions and the complex structure.

Cell viability assay
B16F10 cells (ATCC-CRL-6475, USA) were cultured in Dulbecco's modified eagle medium (DMEM) containing 10% fetal bovine serum (FBS), 100 U/mL penicillin, and 1 μg/mL streptomycin.Cell viability was evaluated using the MTT assay 36 .B16F10 cells (5 × 10 3 cells/well) were cultured in a 96-well plate at 37 °C in a 5% CO 2 incubator for 24 h.Tested compounds dissolved in DMSO/H 2 O (40% v/v) at different concentrations were added to sample wells, and the cells were further incubated at 37 °C in a 5% CO 2 incubator for 48 h.Then, 50 µL of methylthiazolyldiphenyl-tetrazolium bromide (MTT) solution (0.5 mg/ml in 10 mM phosphate buffer, pH 7.4) was added to each well and left under a CO 2 incubator for 2 h.The formazan crystals formed were dissolved in DMSO, and the absorbance was measured at 570 nm.Percent viability was calculated as the percent of surviving cells in the well compared to the control cells.

Melanin content assay
The method for the melanin content assay was modified from Lee et al. 37 .B16F10 (8 × 10 4 cell/mL) cells were seeded in a 6-well plate and incubated at 37 °C in a 5% CO 2 incubator for 24 h.The tested and reference

Figure 6 .
Figure 6.Activators (compounds 14 and 8-MOP) bound with the tyrosinase protein.(A) Molecular docking model of compounds 14 (violet) and 8-MOP (blue) on the tyrosinase active site.Ligand interaction diagrams of compounds 14 (B,C) and 8-MOP (D,E), involving the hydrogen bond and π-π interaction, as respectively illustrated in red and blue.

Figure 7 .
Figure 7. %Cell viability of B16F10 cells after incubation with the synthetic coumarin derivatives for 48 h.