Discovery of Indole–Thiourea Derivatives as Tyrosinase Inhibitors: Synthesis, Biological Evaluation, Kinetic Studies, and In Silico Analysis

Tyrosinase, a key enzyme in melanin synthesis, represents a crucial therapeutic target for hyperpigmentation disorders due to excessive melanin production. This study aimed to design and evaluate a series of indole–thiourea derivatives by conjugating thiosemicarbazones with strong tyrosinase inhibitory activity to indole. Among these derivatives, compound 4b demonstrated tyrosinase inhibitory activity with an IC50 of 5.9 ± 2.47 μM, outperforming kojic acid (IC50 = 16.4 ± 3.53 μM). Kinetic studies using Lineweaver–Burk plots confirmed competitive inhibition by compound 4b. Its favorable ADMET and drug-likeness properties make compound 4b a promising therapeutic candidate with a reduced risk of toxicity. Molecular docking revealed that the compounds bind strongly to mushroom tyrosinase (mTYR) and human tyrosinase-related protein 1 (TYRP1), with compound 4b showing superior binding energies of −7.0 kcal/mol (mTYR) and −6.5 kcal/mol (TYRP1), surpassing both kojic acid and tropolone. Molecular dynamics simulations demonstrated the stability of the mTYR−4b complex with low RMSD and RMSF and consistent Rg and SASA values. Persistent strong hydrogen bonds with mTYR, along with favorable Gibbs free energy and MM/PBSA calculations (−19.37 kcal/mol), further support stable protein–ligand interactions. Overall, compound 4b demonstrated strong tyrosinase inhibition and favorable pharmacokinetics, highlighting its potential for treating pigmentary disorders.


Introduction
In skin-brightening research, tyrosinase inhibition is essential due to its pivotal role in melanin biosynthesis [1].Tyrosinase catalyzes the hydroxylation of monophenols (e.g., L-tyrosine) into o-diphenols (e.g., L-DOPA), which are then oxidized to o-quinones (e.g., o-dopaquinone).These reactive o-quinones either polymerize to form dark pigments or conjugate with other molecules, enhancing pigment intensity [2].Increased pigmentation can contribute to various skin disorders, such as hyperpigmentation, melasma, and age spots, potentially leading to aesthetic concerns and psychological effects [3,4].
Tyrosinase inhibitors are categorized by their binding mechanisms-competitive (binding at the active site, competing with substrates), noncompetitive (binding elsewhere, altering enzyme conformation or activity), and uncompetitive (affecting enzyme activity or synthesis) [5,6].So far, an extensive range of tyrosinase inhibitors have been identified, including polyphenolic compounds (flavonoids, coumarins) [7], benzaldehyde and benzoate derivatives [8], long-chain lipids and steroids [9], various natural and synthetic inhibitors, and irreversible inactivators [6].These inhibitors critically modulate tyrosinase activity and substrate reactions, proving valuable not only in skin brightening but also in exploring new avenues for treating various skin disorders [10].
The synthesis of derivatives based on an indole core presents a promising strategy for developing tyrosinase inhibitors (Figure 1).Natural products such as indole-3carbaldehyde [22] and indole-3-ethanol (tryptophol) [23] have exhibited significant tyrosinase inhibition activity, while indole-3-carboxylic acid [24] and indole-3-acetic acid [25] were shown to have inhibitory effects on B16F10 melanoma cells in previous studies.Reported syntheses of indole derivatives with tyrosinase inhibitory activity typically involve derivatization at positions 1, 2, 3, and 5 of the indole core.Substitution at position 1 often involves benzyl groups [20], while position 2 is linked to phenol via an amide bond or other active groups [26].Position 3 serves as a crucial branch point for introducing nitrogen-containing heterocycles known for their pharmacological effects, whereas electrondonating or -withdrawing substituents are commonly introduced at position 5 [27][28][29].Additionally, the thiosemicarbazone group represents a critical moiety for tyrosinase inhibition.Structure-activity relationship (SAR) analyses indicate that the carbon-sulfur double bond is pivotal for activity [30].An unsubstituted amino group at the outermost end of the thiourea moiety enhances inhibition, while further introduction of amino groups diminishes inhibition [31,32].The introduction of methyl substituents on unsaturated carbons in imine structures enhances activity compared with that of unsubstituted forms [33], while prolonged alkyl chains negatively impact inhibition under in vitro conditions; however, in B16 mouse melanoma cells, the trends of the inhibition of melanin synthesis are contradictory [34].
Int. J. Mol.Sci.2024, 25, x FOR PEER REVIEW 3 of 20 thiosemicarbazone moieties at the 3-position.These compounds were screened for tyrosinase inhibition activity, followed by kinetic analysis to elucidate the inhibition mechanism.Molecular docking studies were employed to further investigate the interaction between active compounds and tyrosinase, while molecular dynamics (MD) simulations revealed the stability and strength of these interactions, which accounted for protein flexibility.
Figure 1.Some candidates that were identified for representative tyrosinase inhibitors.

Chemistry
Indole-thiourea derivatives 4a-4k were synthesized using a convergent approach, as outlined in Scheme 1.The structural elucidation of all synthesized compounds was confirmed through comprehensive spectroscopic analyses, including 1 H NMR, HRMS, and HPLC.Initially, intermediates 2f and 2g were prepared via nucleophilic substitution of halogenated alkanes with indole-3-carbaldehyde (1a).The 1 H NMR spectra of these intermediates revealed specific chemical shifts that were indicative of their structures, intermediate 2f displayed a quartet at δ 4.30 (q, J = 7.3 Hz, 2H) and a triplet at δ 1.42 (t, J = 7.2 Hz, 3H), which corresponded to the ethyl group, while intermediate 2g showed a singlet Melanin biosynthesis requires three essential enzymes: tyrosinase (TYR), tyrosinaserelated protein 1 (TYRP1), and tyrosinase-related protein 2 (TYRP2) [35].Mushroom tyrosinase (mTYR) is commonly used to screen tyrosinase inhibitors due to its availability and ease of use, despite its limited sequence homology with human tyrosinase (only 12% sequence identity).In contrast, human tyrosinase (hTYR) shares a higher sequence homology with human-derived proteins (40% sequence identity) [36][37][38].Due to the complexity, flexibility, and high glycosylation of human tyrosinase, crystallization is extremely challenging, leading to contradictory findings regarding its physiological functions [39].Homology modeling of tyrosinase-related protein 1 (PDB ID: 5M8M) revealed significant structural differences from mushroom tyrosinase (PDB ID: 2Y9W), with an RMSD of 11.477 and binding site shifts [40].Structural differences between human and mushroom tyrosinase imply distinct molecular characteristics for effective inhibitors [41,42].Previous molecular docking studies on mushroom tyrosinase identified residues N81, M280, and N260 as important for substrate binding, with H263 forming a π interaction [43][44][45].In contrast, studies on tyrosinase-related protein 1 have revealed key residues involved in ligand binding.The crystal structures of TYRP1 with various bound ligands have shown that Y362, R374, H381, S394, and T391 are crucial for ligand binding [46].These findings provide a valuable foundation for drug design targeting human tyrosinase.
In this study, the tyrosinase inhibitor indole-3-carbaldehyde was utilized as a starting material to synthesize a series of indole-thiourea derivatives containing thiosemicarbazone moieties at the 3-position.These compounds were screened for tyrosinase inhibition activity, followed by kinetic analysis to elucidate the inhibition mechanism.Molecular docking studies were employed to further investigate the interaction between active compounds and tyrosinase, while molecular dynamics (MD) simulations revealed the stability and strength of these interactions, which accounted for protein flexibility.
stored at room temperature, showed better stability after 2 and 4 days.After two months, degradation was observed in all compounds, but only compound 4i exhibited significant degradation (main peak area at 65%), while others remained above 90% (Tables S1 and  S2).Most compounds containing thiosemicarbazone groups in aqueous solution showed some stability, likely due to the increased stability of conjugated imines formed between the imine and indole double bonds, as conjugated imine structures are commonly observed in marine alkaloids [49].This enhanced stability may also be due to the π-electron delocalization effect or the inductive effect of X2 = N or O [50], which contribute to the greater stability of hydrazones compared to imines.Scheme 1. Synthetic route of indole-thiourea derivatives.

Anti-Tyrosinase Activity and Structure-Activity Relationship (SAR)
The synthesized compounds (4a-4k) were evaluated for their tyrosinase inhibitory activity using IC50 values, with kojic acid as a positive control.Table 1 summarizes the results.All compounds displayed varying degrees of inhibitory activity, with IC50 values ranging from 5.9 to 163.3 μM.In comparison, kojic acid had an IC50 value of 16.4 ± 3.53 μM.The overall structure of a molecule and its specific substituents influence inhibitory activity, revealing structure-activity relationships (SARs).Variations in the indole group substituents significantly affected the inhibitory potential of these compounds.The stability of indole-thiourea derivatives with imine structures in aqueous solution is crucial for drug design, as their hydrolysis and reactivity influence efficacy, safety, and pharmacokinetics [48].The synthesized compounds, dissolved in aqueous solution and stored at room temperature, showed better stability after 2 and 4 days.After two months, degradation was observed in all compounds, but only compound 4i exhibited significant degradation (main peak area at 65%), while others remained above 90% (Tables S1 and S2).Most compounds containing thiosemicarbazone groups in aqueous solution showed some stability, likely due to the increased stability of conjugated imines formed between the imine and indole double bonds, as conjugated imine structures are commonly observed in marine alkaloids [49].This enhanced stability may also be due to the π-electron delocalization effect or the inductive effect of X 2 = N or O [50], which contribute to the greater stability of hydrazones compared to imines.

Biological Evaluations 2.2.1. Anti-Tyrosinase Activity and Structure-Activity Relationship (SAR)
The synthesized compounds (4a-4k) were evaluated for their tyrosinase inhibitory activity using IC 50 values, with kojic acid as a positive control.Table 1 summarizes the results.All compounds displayed varying degrees of inhibitory activity, with IC 50 values ranging from 5.9 to 163.3 µM.In comparison, kojic acid had an IC 50 value of 16.4 ± 3.53 µM.The overall structure of a molecule and its specific substituents influence inhibitory activity, revealing structure-activity relationships (SARs).Variations in the indole group substituents significantly affected the inhibitory potential of these compounds.
Third, the effect of alkyl chain length on the indole amino group: We further investigated the effect of the length of the alkyl chain connected to the indole amino group.As the carbon chain's length increased, the molecular hydrophobicity and inhibitory activity also increased.Compound 4f (R1 = CH2CH3) outperformed 4g (R1 = CH3).Previous MD simulations and free energy calculations indicated that longer carbon chains enhance protein-ligand binding free energy by increasing hydrophobic contact and reducing the desolvation energy [52].Additionally, longer carbon chains improve spatial filling, induced fit, and molecular flexibility [53,54].These findings underscore the importance of specific substituents and structural modifications in optimizing the inhibitory activity of the compounds under study.

No.
Compound Second, the effect of the indole amino group: Compounds 4g (R 2 = H), 4h (R 2 = OCH 3 ), and 4i (R 2 = Br) with a methyl group at the amino position exhibited significantly lower inhibitory activity than non-substituted compounds 4a (R 2 = H), 4e (R 2 = OCH 3 ), and 4d (R 2 = Br).The unsubstituted amino group acts as a hydrogen bond donor, facilitating tight binding to the protein, which enhances the inhibitory activity [51].
Third, the effect of alkyl chain length on the indole amino group: We further investigated the effect of the length of the alkyl chain connected to the indole amino group.As the carbon chain's length increased, the molecular hydrophobicity and inhibitory activity also increased.Compound 4f (R 1 = CH 2 CH 3 ) outperformed 4g (R 1 = CH 3 ).Previous MD simulations and free energy calculations indicated that longer carbon chains enhance protein-ligand binding free energy by increasing hydrophobic contact and reducing the desolvation energy [52].Additionally, longer carbon chains improve spatial filling, induced fit, and molecular flexibility [53,54].These findings underscore the importance of specific substituents and structural modifications in optimizing the inhibitory activity of the compounds under study.

Determination of the Inhibitory Mechanism through Enzyme Kinetics
Due to the potent inhibition of mushroom tyrosinase by compound 4b, kinetic studies were conducted using Lineweaver-Burk double reciprocal plots to elucidate its inhibitory mechanisms.Three concentrations (0 µM, 3 µM, 6 µM, and 12 µM) of the indole derivative 4b in the presence of varying L-tyrosine concentrations (0.1, 0.2, 0.4, 0.6, 0.8, 1 mM) were employed for the kinetic analyses.The results are summarized in Figure 2. The Lineweaver-Burk plots generated four distinct lines with different slopes corresponding to each concentration of the derivative.All lines converged at a single point on the y-axis, indicating that the Vmax value of compound 4b (0.0135 mM/min) was consistent across the concentrations.Additionally, the data revealed that the K M values of 4b increased in a concentration-dependent manner, implying the competitive inhibition of mTYR alongside L-tyrosine binding.Specifically, the K M values were 1.515, 7.846, 15.740, and 23.586 mM at 0, 3, 6, and 12 µM of 4b, respectively.Furthermore, the kinetic studies yielded K i values of 1.68 × 10 −7 , 5.68 × 10 −8 , and 2.06 × 10 −8 M at 3, 6, and 12 µM of 4b, indicating the potency of inhibition at these concentrations.

Determination of the Inhibitory Mechanism through Enzyme Kinetics
Due to the potent inhibition of mushroom tyrosinase by compound 4b, kinetic studies were conducted using Lineweaver-Burk double reciprocal plots to elucidate its inhibitory mechanisms.Three concentrations (0 μM, 3 μM, 6 μM, and 12 μM) of the indole derivative 4b in the presence of varying L-tyrosine concentrations (0.1, 0.2, 0.4, 0.6, 0.8, 1 mM) were employed for the kinetic analyses.The results are summarized in Figure 2. The Lineweaver-Burk plots generated four distinct lines with different slopes corresponding to each concentration of the derivative.All lines converged at a single point on the y-axis, indicating that the Vmax value of compound 4b (0.0135 mM/min) was consistent across the concentrations.Additionally, the data revealed that the KM values of 4b increased in a concentration-dependent manner, implying the competitive inhibition of mTYR alongside L-tyrosine binding.Specifically, the KM values were 1.515, 7.846, 15.740, and 23.586 mM at 0, 3, 6, and 12 μM of 4b, respectively.Furthermore, the kinetic studies yielded Ki values of 1.68 × 10 −7 , 5.68 × 10 −8 , and 2.06 × 10 −8 M at 3, 6, and 12 μM of 4b, indicating the potency of inhibition at these concentrations.

Free Radical-Scavenging Activity
Radical-scavenging assays evaluate a compound's capacity to neutralize free radicals, providing insight into its potential to reduce oxidative damage caused by tyrosinase activity [55].The DPPH radical-scavenging activity of the eleven synthesized compounds (4a-4k) and a reference compound, ascorbic acid, was evaluated.The SC50 values, which represent the concentration required to scavenge 50% of the DPPH radicals, were determined and are summarized in Table 2 and Figure S1.The SC50 values indicated that compound 4k exhibited the most potent DPPH radical-scavenging activity with an SC50 of 24.6 ± 3.7 μM, followed closely by ascorbic acid with an SC50 of 37.9 ± 0.3 μM.This suggested that compound 4k had a comparable or slightly superior antioxidant capacity relative to that of ascorbic acid.Compounds 4j and 4h exhibited moderate radical scavenging activity, with SC50 values of 275.7 ± 2.7 μM and 340.2 ± 3.1 μM, respectively.On the other hand, compounds 4a, 4b, 4c, 4e, and 4g showed relatively weaker activity, with SC50 values ranging from 482.7 ± 1.3 μM to 659.2 ± 0.5 μM.Compounds 4d and 4f, with SC50 values of 769.0 ± 2.6 μM and >1000 μM, respectively, exhibited the least effective antioxidant activity among the tested compounds.
The data indicate that structural differences significantly influenced the antioxidant activities of the synthesized compounds.Hydroxyl groups in phenolic compounds act as

Free Radical-Scavenging Activity
Radical-scavenging assays evaluate a compound's capacity to neutralize free radicals, providing insight into its potential to reduce oxidative damage caused by tyrosinase activity [55].The DPPH radical-scavenging activity of the eleven synthesized compounds (4a-4k) and a reference compound, ascorbic acid, was evaluated.The SC 50 values, which represent the concentration required to scavenge 50% of the DPPH radicals, were determined and are summarized in Table 2 and Figure S1.The SC 50 values indicated that compound 4k exhibited the most potent DPPH radical-scavenging activity with an SC 50 of 24.6 ± 3.7 µM, followed closely by ascorbic acid with an SC 50 of 37.9 ± 0.3 µM.This suggested that compound 4k had a comparable or slightly superior antioxidant capacity relative to that of ascorbic acid.Compounds 4j and 4h exhibited moderate radical scavenging activity, with SC 50 values of 275.7 ± 2.7 µM and 340.2 ± 3.1 µM, respectively.On the other hand, compounds 4a, 4b, 4c, 4e, and 4g showed relatively weaker activity, with SC 50 values ranging from 482.7 ± 1.3 µM to 659.2 ± 0.5 µM.Compounds 4d and 4f, with SC 50 values of 769.0 ± 2.6 µM and >1000 µM, respectively, exhibited the least effective antioxidant activity among the tested compounds.
The data indicate that structural differences significantly influenced the antioxidant activities of the synthesized compounds.Hydroxyl groups in phenolic compounds act as antioxidants by donating hydrogen atoms, thereby neutralizing free radicals and stabilizing them through resonance and electron delocalization [56,57].This prevents oxidative damage and maintains cellular stability.Compound 4k's high activity was likely due to the hydroxyl group at the 5-position of the indole-benzene ring.Ascorbic acid, with its unsaturated enediol structure, also showed strong antioxidant capacity.Both compound 4k and ascorbic acid share crucial hydroxyl groups, suggesting a common feature for enhancing radical scavenging efficiency.The SC 50 values of antioxidant-functionalized niosomes were determined using a DPPH-scavenging assay.The results are presented as the mean ± standard deviation (SD) of three independent experiments, with each performed in triplicate.

In Silico Experiments 2.3.1. Molecular Properties and Drug Likeness
The results of ADMET (absorption, distribution, metabolism, excretion, and toxicity) analysis showed that compound 4b exhibited several pharmacokinetic and toxicological advantages over kojic acid and tropolone (Table S3).It had a higher LD 50 (3.1502mol/kg), indicating lower acute toxicity than kojic acid (LD 50 = 2.0673 mol/kg) and tropolone (LD 50 = 2.4899 mol/kg).With a 97.0%plasma protein-binding rate, compound 4b ensures prolonged drug action and reduced side effects.It selectively inhibits only CYP1A2, minimizing drug-drug interactions.Importantly, compound 4b does not cross the bloodbrain barrier, thereby reducing side effects in the central nervous system (CNS) [58].It also shows no hepatotoxicity, Ames toxicity, skin sensitization, or hERG inhibition.Despite its moderate permeability, compound 4b achieves high human intestinal absorption (89.027%), ensuring effective bioavailability.In conclusion, compound 4b offers a more favorable safety profile with higher acute toxicity tolerance, extensive plasma protein binding, minimal CYP450 inhibition, and no CNS penetration in comparison with kojic acid and tropolone.These characteristics make compound 4b a promising candidate with potentially lower risks of systemic toxicity and drug-drug interactions, resulting in a safer pharmacokinetic profile for therapeutic applications.
In drug-likeness analysis (Table S4), all compounds showed superior drug likeness to kojic acid and tropolone, adhering to all key drug-likeness rules: RO5 [59], the Ghose Filter [60], the Veber Rule [61], and the Egan Rule [62].This comprehensive compliance indicates that compound 4b possesses excellent pharmacokinetic properties and a high potential for successful development as an orally bioavailable drug.Conversely, while kojic acid and tropolone are promising, their violations of the Ghose Filter criteria (three violations: MW < 160, MR < 40, atoms < 20) may pose challenges in drug development.

Molecular Docking Simulation
The docked complexes of the synthesized ligands (4a-4k) with mushroom tyrosinase and human tyrosinase-related protein 1 were analyzed based on the minimum energy values and ligand interaction patterns.Tropolone and kojic acid served as co-crystallized inhibitors for polyphenol oxidase (PPO) family PPO3 protein (PDB ID: 2Y9X) and human tyrosinase-related protein 1 (PDB ID: 5M8M), respectively.The re-docking results of these co-crystallized ligands demonstrated that tropolone had a binding energy of −6.3 kcal/mol with mTYR, with an RMSD value of 1.565 Å between the pre-and post-docking conformations (Figure 3a).Kojic acid exhibited a binding energy of −5.9 kcal/mol with TYRP1, with an RMSD value of 1.935 Å (Figure 3b).
tyrosinase-related protein 1 (PDB ID: 5M8M), respectively.The re-docking results of these co-crystallized ligands demonstrated that tropolone had a binding energy of −6.3 kcal/mol with mTYR, with an RMSD value of 1.565 Å between the pre-and post-docking conformations (Figure 3a).Kojic acid exhibited a binding energy of −5.9 kcal/mol with TYRP1, with an RMSD value of 1.935 Å (Figure 3b).
Both co-crystallized ligands bound to the same site on their respective proteins with minimal conformational differences, achieving RMSD values that were lower than the widely accepted validation standard of 2 Å [63,64], thereby validating the docking methodology.
(a) (b) In this study, we conducted molecular docking analyses to investigate the binding affinities of 13 compounds (4a-4k, tropolone, and kojic acid) with mTYR and TYRP1.The results indicated that all synthesized ligands displayed favorable binding energy values and effectively interacted within the active site of the target proteins, as detailed in Table 3.A comparative analysis of the docking scores for both mTYR and TYRP1 revealed that compounds 4a and 4b exhibited strong binding to mTYR (−7.0 kcal/mol) but only moderate binding to TYRP1 (−5.8 and −6.5 kcal/mol).Conversely, compounds 4b, 4d, and 4f showed similar high binding affinities for both mTYR and TYRP1.Compounds 4h and 4k exhibited nearly identical binding energies for mTYR and TYRP1, suggesting they may have similar binding modes and affinities for both proteins.Compound 4i displayed a notable difference in binding energy between mTYR (−6.0 kcal/mol) and TYRP1 (−4.5 kcal/mol), indicating significant variability in the binding interactions.
In conclusion, the molecular docking results highlight the importance of specific structural features in determining the binding affinities of tyrosinase inhibitors.Both co-crystallized ligands bound to the same site on their respective proteins with minimal conformational differences, achieving RMSD values that were lower than the widely accepted validation standard of 2 Å [63,64], thereby validating the docking methodology.
In this study, we conducted molecular docking analyses to investigate the binding affinities of 13 compounds (4a-4k, tropolone, and kojic acid) with mTYR and TYRP1.The results indicated that all synthesized ligands displayed favorable binding energy values and effectively interacted within the active site of the target proteins, as detailed in Table 3.A comparative analysis of the docking scores for both mTYR and TYRP1 revealed that compounds 4a and 4b exhibited strong binding to mTYR (−7.0 kcal/mol) but only moderate binding to TYRP1 (−5.8 and −6.5 kcal/mol).Conversely, compounds 4b, 4d, and 4f showed similar high binding affinities for both mTYR and TYRP1.Compounds 4h and 4k exhibited nearly identical binding energies for mTYR and TYRP1, suggesting they may have similar binding modes and affinities for both proteins.Compound 4i displayed a notable difference in binding energy between mTYR (−6.0 kcal/mol) and TYRP1 (−4.5 kcal/mol), indicating significant variability in the binding interactions.
In conclusion, the molecular docking results highlight the importance of specific structural features in determining the binding affinities of tyrosinase inhibitors.Compounds with higher binding energies, such as 4a, 4b, 4f, and 4k, could be promising candidates for further development as tyrosinase inhibitors.
Compound 4b demonstrated the highest tyrosinase inhibitory activity and strong binding affinity in docking simulations with both mTYR and TYRP1, as shown in to its high electronegativity, attracts partial positive charges from amino acid side chains, forming weak hydrogen-like interactions with HIS-85 and HIS-259 (blue dashed line), which help stabilize the ligand-receptor binding [65][66][67].The docking energy of compound 4b with mTYR was −7.0 kcal/mol, which was better than that of the co-crystallized ligand tropolone (−6.3 kcal/mol) and the positive control kojic acid (−5.3 kcal/mol).

Molecular Dynamics (MD) Simulation
As previously mentioned [23], we conducted 100 ns MD simulations on the complexes of mTYR with the co-crystallized ligand (tropolone) and compounds to further assess binding stability and affinity.This was necessary because the semi-flexible docking method does not fully account for protein flexibility.This study analyzed the MD simulation trajectories of the complexes, focusing on the root mean square deviation (RMSD), root mean square fluctuation (RMSF), radius of gyration (Rg), number of hydrogen bonds (H-bonds) between the protein and compound, solvent-accessible surface area (SASA), Gibbs free energy landscape (FEL-energy), energy contribution of binding residues (Res- In TYRP1, compound 4b forms a hydrogen bond with THR-391 (3.0 Å, green dashed line) and a carbon-hydrogen bond with TYR-362 via its imine group (Schiff base).The amino groups on both sides of the thiourea moiety form two hydrogen bonds with GLU-216 (2.4 Å, 2.5 Å), while the outer amino group of the thiourea forms a hydrogen bond with ASP-212 (2.4 Å).The fluorine on the indole benzene ring forms a hydrogen bond with ARG-374 (3.0 Å), indicating strong hydrogen bond stability.The pyrrole ring of the indole interacts with HIS-381 via Pi-Pi T-shaped and carbon-hydrogen bonds, and the indole benzene ring interacts with LEU-382 via Pi-alkyl interactions.The docking energy of compound 4b with TYRP1 was −6.5 kcal/mol, which was superior to that of the co-crystallized ligand kojic acid (−5.9 kcal/mol).
The detailed interactions of other compounds with mTYR and TYRP1 are provided in Table S5.The results show that compound 4f's ethyl groups interact with a greater number of amino acid residues than methyl groups do, forming more interactions (e.g., alkyl, Pi-alkyl, Pi-sigma).These interactions allow ethyl groups to occupy more active sites than methyl groups do, thereby enhancing their inhibitory activity [68][69][70].
The reproducibility of the docking results was confirmed with ten repeated docking simulations for compound 4b with both mTYR and TYRP1, showing binding energies within ±0.1 kcal/mol (Table S6).These findings indicate that compound 4b exhibits high binding affinity to both mTYR and TYRP1, suggesting a potential pharmacological effect that is similar to that of the co-crystallized ligands, likely impacting the structure, function, and bioactivity of tyrosinase proteins.

Molecular Dynamics (MD) Simulation
As previously mentioned [23], we conducted 100 ns MD simulations on the complexes of mTYR with the co-crystallized ligand (tropolone) and compounds to further assess binding stability and affinity.This was necessary because the semi-flexible docking method does not fully account for protein flexibility.This study analyzed the MD simulation trajectories of the complexes, focusing on the root mean square deviation (RMSD), root mean square fluctuation (RMSF), radius of gyration (Rg), number of hydrogen bonds (Hbonds) between the protein and compound, solvent-accessible surface area (SASA), Gibbs free energy landscape (FEL-energy), energy contribution of binding residues (Residueenergy), average binding free energy (Total-energy), and conformational structures of the complexes at 0 ns, 25 ns, 50 ns, 75 ns, and 100 ns of the MD simulation.
An RMSD curve was used to assess the protein-ligand complex stability; lower values indicate minimal structural changes and greater stability.The PPO3 protein complexed with compound 4b showed consistently low RMSD fluctuations, closely mirroring the co-crystallized ligand's curve, suggesting that the PPO3-4b complex is stable (Figure 5a).The Rg assesses structural compactness and stability, with larger values indicating greater expansion in dynamic simulations and smaller values indicating stability.In the PPO3-4b complex (Figure 6a), the Rg curve stabilizes around 2.1 nm, demonstrating overall stability with minimal fluctuations.Compared to the co-crystallized ligand, PPO3-4b exhibits smaller Rg fluctuations, indicating the formation of a tight and stable complex with compound 4b.Importantly, the inclusion of compound 4b did not significantly alter the protein's overall structure.
To investigate hydrogen bonding at the binding site of the complex, we computed the number of key hydrogen bonds stabilizing ligand-protein interactions.During 100 ns of dynamic simulation, stable hydrogen bonding interactions (1-3 bonds) formed between the PPO3 protein and compound 4b (Figure 6b), indicating robust binding stability.In contrast, the co-crystallized ligand group maintained a stable hydrogen bond count of one throughout, suggesting slightly superior stability of the PPO3-4b complex compared with that of the co-crystallized ligand group.The RMSF curve illustrates residue fluctuation in proteins during dynamic simulations, where higher values denote greater flexibility and lower values indicate stability.In the PPO3-4b complex (Figure 5b), the RMSF remains within 1 nm, with slight fluctuations (~0.4 nm) observed at residues 70-80 near the protein periphery.The RMSF curve of compound 4b closely overlaps with that of the co-crystallized ligand group, suggesting a minimal impact on residue stability in PPO3 and affirming the complex's overall stability.
The Rg assesses structural compactness and stability, with larger values indicating greater expansion in dynamic simulations and smaller values indicating stability.In the PPO3-4b complex (Figure 6a), the Rg curve stabilizes around 2.1 nm, demonstrating overall stability with minimal fluctuations.Compared to the co-crystallized ligand, PPO3-4b exhibits smaller Rg fluctuations, indicating the formation of a tight and stable complex with compound 4b.Importantly, the inclusion of compound 4b did not significantly alter the protein's overall structure.
the number of key hydrogen bonds stabilizing ligand-protein interactions.During 100 ns of dynamic simulation, stable hydrogen bonding interactions (1-3 bonds) formed between the PPO3 protein and compound 4b (Figure 6b), indicating robust binding stability.In contrast, the co-crystallized ligand group maintained a stable hydrogen bond count of one throughout, suggesting slightly superior stability of the PPO3-4b complex compared with that of the co-crystallized ligand group.The SASA is a factor in studying protein structures' folding and stability.Proteins with stable structures typically exhibit consistent SASA curves.The SASA curve of the PPO3-4b complex shown in Figure 7 shows stable fluctuations throughout, without significant changes.Moreover, the SASA curve of the co-crystallized ligand group closely overlaps with that of the PPO3-4b group, indicating the high stability of the complex formed by the PPO3-4b group.The Gibbs FEL was computed using the Gromacs scripts g_sham and xpm2txt.py,and the RMSD and Rg values were utilized to calculate the Gibbs free energy.In the Gibbs FEL plots, the RMSD, Rg, and Gibbs free energy are depicted on the X, Y, and Z axes, illustrating favored conformations during dynamic simulations.Weak protein-ligand interactions are characterized by rough, multi-clustered FEL surfaces, while strong interactions exhibit smooth, single-clustered surfaces.Dark purple/blue spots in the FEL denote To investigate hydrogen bonding at the binding site of the complex, we computed the number of key hydrogen bonds stabilizing ligand-protein interactions.During 100 ns of dynamic simulation, stable hydrogen bonding interactions (1-3 bonds) formed between the PPO3 protein and compound 4b (Figure 6b), indicating robust binding stability.In contrast, the co-crystallized ligand group maintained a stable hydrogen bond count of one throughout, suggesting slightly superior stability of the PPO3-4b complex compared with that of the co-crystallized ligand group.
The SASA is a factor in studying protein structures' folding and stability.Proteins with stable structures typically exhibit consistent SASA curves.The SASA curve of the PPO3-4b complex shown in Figure 7 shows stable fluctuations throughout, without significant changes.Moreover, the SASA curve of the co-crystallized ligand group closely overlaps with that of the PPO3-4b group, indicating the high stability of the complex formed by the PPO3-4b group.
greater expansion in dynamic simulations and smaller values indicating stability.In the PPO3-4b complex (Figure 6a), the Rg curve stabilizes around 2.1 nm, demonstrating overall stability with minimal fluctuations.Compared to the co-crystallized ligand, PPO3-4b exhibits smaller Rg fluctuations, indicating the formation of a tight and stable complex with compound 4b.Importantly, the inclusion of compound 4b did not significantly alter the protein's overall structure.
To investigate hydrogen bonding at the binding site of the complex, we computed the number of key hydrogen bonds stabilizing ligand-protein interactions.During 100 ns of dynamic simulation, stable hydrogen bonding interactions (1-3 bonds) formed between the PPO3 protein and compound 4b (Figure 6b), indicating robust binding stability.In contrast, the co-crystallized ligand group maintained a stable hydrogen bond count of one throughout, suggesting slightly superior stability of the PPO3-4b complex compared with that of the co-crystallized ligand group.The SASA is a factor in studying protein structures' folding and stability.Proteins with stable structures typically exhibit consistent SASA curves.The SASA curve of the PPO3-4b complex shown in Figure 7 shows stable fluctuations throughout, without significant changes.Moreover, the SASA curve of the co-crystallized ligand group closely overlaps with that of the PPO3-4b group, indicating the high stability of the complex formed by the PPO3-4b group.The Gibbs FEL was computed using the Gromacs scripts g_sham and xpm2txt.py,and the RMSD and Rg values were utilized to calculate the Gibbs free energy.In the Gibbs FEL plots, the RMSD, Rg, and Gibbs free energy are depicted on the X, Y, and Z axes, illustrating favored conformations during dynamic simulations.Weak protein-ligand interactions are characterized by rough, multi-clustered FEL surfaces, while strong interactions exhibit smooth, single-clustered surfaces.Dark purple/blue spots in the FEL denote The Gibbs FEL was computed using the Gromacs scripts g_sham and xpm2txt.py,and the RMSD and Rg values were utilized to calculate the Gibbs free energy.In the Gibbs FEL plots, the RMSD, Rg, and Gibbs free energy are depicted on the X, Y, and Z axes, illustrating favored conformations during dynamic simulations.Weak proteinligand interactions are characterized by rough, multi-clustered FEL surfaces, while strong interactions exhibit smooth, single-clustered surfaces.Dark purple/blue spots in the FEL denote stable conformations, whereas red/yellow indicates instability.PPO3 complexes with both compound 4b (Figure 8a) and the co-crystallized ligand (Figure 8b) exhibit single concentrated energy clusters, indicating stable interactions between PPO3 and compound 4b.Upon achieving stability in the complex systems, we computed the MM/PBSA binding free energies for the complexes of the PPO3 protein with compound 4b (Figure 9a) and the co-crystallized ligand (Figure 9b).The average binding free energies of the PPO3 Upon achieving stability in the complex systems, we computed the MM/PBSA binding free energies for the complexes of the PPO3 protein with compound 4b (Figure 9a) and the co-crystallized ligand (Figure 9b).The average binding free energies of the PPO3 protein with compound 4b and the co-crystallized ligand were −19.37 and −12.47 kcal/mol, respectively.This indicated that the PPO3 protein bound more strongly with compound 4b than with the co-crystallized ligand.Upon achieving stability in the complex systems, we computed the MM/PBSA binding free energies for the complexes of the PPO3 protein with compound 4b (Figure 9a) and the co-crystallized ligand (Figure 9b).The average binding free energies of the PPO3 protein with compound 4b and the co-crystallized ligand were −19.37 and −12.47 kcal/mol, respectively.This indicated that the PPO3 protein bound more strongly with compound 4b than with the co-crystallized ligand.In the residue energy analysis, compound 4b showed optimal binding with the amino acid residues VAL-283 and HIS-263 in the PPO3 protein, with binding energies of -2.35 and −1.69 kcal/mol, respectively (Figure 10a).This highlights the significant roles of VAL-283 and HIS-263 in the interaction between compound 4b and the PPO3 protein.
Similarly, the co-crystallized ligand exhibited strong binding with the amino acid residues VAL-283, ASN-260, and HIS-263 in the PPO3 protein (Figure 10b), indicating their prominent involvement in the interaction.The residue energy results corroborate the findings from the docking simulations for compound 4b and the co-crystallized ligand.In the residue energy analysis, compound 4b showed optimal binding with the amino acid residues VAL-283 and HIS-263 in the PPO3 protein, with binding energies of -2.35 and −1.69 kcal/mol, respectively (Figure 10a).This highlights the significant roles of VAL-283 and HIS-263 in the interaction between compound 4b and the PPO3 protein.Similarly, the co-crystallized ligand exhibited strong binding with the amino acid residues VAL-283, ASN-260, and HIS-263 in the PPO3 protein (Figure 10b), indicating their prominent involvement in the interaction.The residue energy results corroborate the findings from the docking simulations for compound 4b and the co-crystallized ligand.To explore the binding dynamics of the complex in pre-and post-molecular dynamics simulations, we aligned the structures of compound 4b (Figure 11a) and the co-crystallized ligand (Figure 11b) with the PPO3 protein at 0, 25, 50, 75, and 100 ns from the simulation trajectories.The binding positions for both compounds showed minimal changes, underscoring the robust stability of these complexes.To explore the binding dynamics of the complex in pre-and post-molecular dynamics simulations, we aligned the structures of compound 4b (Figure 11a) and the co-crystallized ligand (Figure 11b) with the PPO3 protein at 0, 25, 50, 75, and 100 ns from the simula-tion trajectories.The binding positions for both compounds showed minimal changes, underscoring the robust stability of these complexes.To explore the binding dynamics of the complex in pre-and post-molecular dynamics simulations, we aligned the structures of compound 4b (Figure 11a) and the co-crystallized ligand (Figure 11b) with the PPO3 protein at 0, 25, 50, 75, and 100 ns from the simulation trajectories.The binding positions for both compounds showed minimal changes, underscoring the robust stability of these complexes.

Chemicals and Instruments
A melting-point apparatus (Shanghai YICE Equipment Co., Ltd., WRS-2, Shanghai, China) was used to determine the melting points (uncorrected) of the synthesized compounds.The synthesized compounds were analyzed using a Waters 2695 system (Waters Corp., Milford, MA, USA) equipped with a 250 × 4.6 mm Kromasil C18 column (Eka Nobel, Bohus, Sweden, EU).The 1 H NMR spectra of the compounds were recorded on a JNM-ECX 400 (FT-NMR system, 400 MHz, JEOL Co., Akishima, Japan) with 400 MHz spectrophotometers DMSO-d6 as a solvent.HRMS was performed on a SYNAPT G2 mass spectrometer (Waters Corp., Manchester, UK) with positive ion mode, m/z 50-1200 range, and time-of-flight (TOF) analysis for accurate molecular weight determination.All commercially available chemicals and reagents were utilized without further purification.The reaction progress was monitored using thin-layer chromatography (TLC).Mushroom tyrosinase, L-tyrosine, and kojic acid were purchased from Sigma Chemicals (St. Louis, Missouri, USA).

Chemicals and Instruments
A melting-point apparatus (Shanghai YICE Equipment Co., Ltd., WRS-2, Shanghai, China) was used to determine the melting points (uncorrected) of the synthesized compounds.The synthesized compounds were analyzed using a Waters 2695 system (Waters Corp., Milford, MA, USA) equipped with a 250 × 4.6 mm Kromasil C18 column (Eka Nobel, Bohus, Sweden, EU).The 1 H NMR spectra of the compounds were recorded on a JNM-ECX 400 (FT-NMR system, 400 MHz, JEOL Co., Akishima, Japan) with 400 MHz spectrophotometers using DMSO-d6 as a solvent.HRMS was performed on a SYNAPT G2 mass spectrometer (Waters Corp., Manchester, UK) with positive ion mode, m/z 50-1200 range, and time-of-flight (TOF) analysis for accurate molecular weight determination.All commercially available chemicals and reagents were utilized without further purification.The reaction progress was monitored using thin-layer chromatography (TLC).Mushroom tyrosinase, L-tyrosine, and kojic acid were purchased from Sigma Chemicals (St. Louis, MO, USA).

Tyrosinase Inhibition Assay
In the enzymatic assay, a reaction mixture was prepared containing 0.1 M potassium phosphate buffer (pH 6.8), 2 mM L-tyrosine, and 2500 units of tyrosinase.Samples at various concentrations were dispensed as 20 µL aliquots into a 96-well plate.Each well received 130 µL of substrate solution (buffer and L-tyrosine), followed by 5 µL of tyrosinase and 45 µL of buffer solution.The reaction mixture was then incubated at 37 • C for 10 min, after which absorbance at 490 nm was measured.

Kinetic Mechanism Analysis
The kinetic experiment was intended to elucidate the inhibitory behavior of compound 4b against tyrosinase.Four different doses of 4b (0, 3, 6, and 12 µM) were investigated to characterize its inhibition pattern with respect to tyrosinase activity.Various concentrations of L-tyrosine substrate ranging from 0.1 to 1 mM were employed across all experiments.The initial incubation time and reading period mirrored those outlined in the anti-tyrosinase inhibition method.The maximum initial velocity (V 0 ) was determined from the linear phase of absorbance measurements taken over 5 min following the addition of the enzyme solution.The inhibition mode was further delineated using a Lineweaver-Burk plot, where the reciprocal of velocities (1/V) was plotted against the reciprocal of substrate concentrations utilized.

Free Radical-Scavenging Assay
The DPPH radical-scavenging activity was evaluated using Blois' method.A 0.2 mM solution of DPPH in ethanol (EtOH) was prepared as the working reagent.Each sample (20 µL) was introduced into a 96-well plate, followed by the addition of 180 µL of the DPPH solution.The reaction mixture was incubated at room temperature for 15 min.After the reaction, the absorbance was recorded at 515 nm using a microplate reader.The SC 50 value, representing the concentration required to scavenge 50% of the DPPH radicals, was determined for each sample.Each measurement was performed in triplicate, and the mean value was reported.Ascorbic acid served as the positive control throughout the experiment.

Molecular Docking Simulation
The receptor protein PPO3 mushroom tyrosinase (PDB ID: 2Y9X) and human tyrosinaserelated protein 1 (PDB ID: 5M8M) were obtained from the PDB database, and their 3D structure files were downloaded.The protein structures were inspected using the PyMOL 3.0.3software in preparation for molecular docking.All synthesized compounds were drawn in 3D using ACD/ChemSketch Freeware, and their structures were optimized using the MMFF94 force field in OpenBabel 2.0.2 software to obtain the most energetically favorable conformations.AutoDock Tools 1.5.6 was employed to prepare the proteins by adding hydrogen atoms and preprocess the ligand by hydrogenating and determining rotatable bonds.The grid parameters for molecular docking were set based on the crystallographic position of the ligand in the protein structures.The mTYR protein was set as follows: Center (X, Y, Z) = (−10.2,−30.3, −44.4),Size (X × Y × Z) = (15.0× 15.0 × 15.0).The TYRP1 protein was set as follows: Center (X, Y, Z) = (−25.8,−26.1, 22.8), Size (X × Y × Z) = (15.0× 15.0 × 15.0).Semi-flexible docking was performed with an exhaustiveness of 25 using the Lamarckian genetic algorithm in AutoDock Vina 1.2.0.The docking simulations yielded binding free energies and docking result files.Re-docking experiments were conducted using the co-crystallized ligand from the protein structures, and the root mean square deviation (RMSD) was calculated between the initial and re-docked poses.An RMSD value less than 2 Å indicated successful methodological validation.Additionally, to validate the reliability of molecular docking, the proteins were docked with the ligand under identical conditions in 10 independent runs, and the resulting data were compared to identify any differences.

Molecular Dynamics (MD) Simulations
MD simulations of the PPO3 protein complexes with the co-crystallized ligand and compound 4b were conducted using Gromacs 2022.The Charmm36 force field was applied for the protein, while Gaff2 was utilized for the ligand.The TIP3P water model was em-ployed to solvate the protein-ligand systems within a periodic boundary box of 1.2 nm.To mimic realistic experimental conditions, sodium and chloride ions were added to neutralize the system's charge.The simulation protocol consisted of three main stages of preparation.Initially, energy minimization was carried out using the steepest descent algorithm for 50,000 steps, terminating when the maximum force reached below 1000 kJ/mol.Subsequently, NVT pre-equilibration was performed for 50,000 steps at a constant temperature of 310 K, with a time step of 2 fs.This was followed by NPT pre-equilibration for 50,000 steps at a constant temperature of 310 K and pressure of 1 atm, also using a time step of 2 fs.After completing energy minimization and equilibration steps, a 100 ns MD simulation was conducted without constraints while employing a time step of 2 fs and saving coordinates every 10 ps.

Conclusions
In this study, a series of synthesized compounds (4a-4k) were evaluated for tyrosinase inhibition.Compound 4b showed the highest inhibitory activity with an IC 50 of 5.9 ± 2.47 µM, outperforming kojic acid (16.4 ± 3.53 µM).SAR analysis highlighted the importance of the fluorine substituent and unsubstituted amino group on the indole core.Compound 4k demonstrated significant antioxidant activity due to its hydroxyl.Compound 4b exhibited favorable ADMET and drug-likeness properties, positioning it as a more promising drug candidate compared to kojic acid and tropolone.
Molecular docking of compounds 4a-4k targeting tyrosinase revealed strong binding affinities for compounds 4a, 4b, 4f, and 4k to mTYR and TYRP1.Compound 4b exhibited superior binding energies (−7.0 kcal/mol with mTYR and −6.5 kcal/mol with TYRP1) to those of tropolone and kojic acid, emphasizing its efficacy.In MD simulations, compound 4b displayed stable interactions with PPO3 over 100 ns, which were characterized by low RMSD and RMSF values, indicating structural stability and minimal residue fluctuations.Rg analysis confirmed the consistent compactness of the PPO3-4b complex.The study demonstrated that compound 4b forms a highly stable complex with PPO3, as evidenced by the consistent SASA curves, single-clustered Gibbs FEL surfaces, and stronger MM/PBSA binding free energy (−19.37 kcal/mol) than those of the co-crystallized ligand.Hydrogen bonding analysis showed persistent interactions with active site residues of tyrosinase, highlighting its reliable binding profile.Residue analysis confirmed robust interactions, particularly with VAL-283 and HIS-263, corroborating the docking simulation results.These findings provide valuable insights for the rational design and optimization of novel tyrosinase inhibitors with enhanced therapeutic potential.However, further in vivo studies are necessary to verify the efficacy and safety.

Figure 1 .
Figure 1.Some candidates that were identified for representative tyrosinase inhibitors.

Figure 5 .
Figure 5. Plots of the MD simulation results for the complex: (a) RMSD curves; (b) RMSF curves.

Figure 5 .
Figure 5. Plots of the MD simulation results for the complex: (a) RMSD curves; (b) RMSF curves.

Figure 6 .
Figure 6.Plots of the MD simulation results for the complex: (a) Rg curves; (b) H-bond plot.

Figure 7 .
Figure 7. Plot of the SASA results for the complex in MD simulations.

Figure 6 .
Figure 6.Plots of the MD simulation results for the complex: (a) Rg curves; (b) H-bond plot.

Figure 6 .
Figure 6.Plots of the MD simulation results for the complex: (a) Rg curves; (b) H-bond plot.

Figure 7 .
Figure 7. Plot of the SASA results for the complex in MD simulations.

Figure 7 .
Figure 7. Plot of the SASA results for the complex in MD simulations.

Table 1 .
Tyrosinase inhibitory activity of the synthesized compounds.

Table 1 .
Tyrosinase inhibitory activity of the synthesized compounds.

R 1 R 2 Tyrosinase Activity IC 50 ± SEM (µM)
SEM, standard error of the mean; values are expressed as the mean ± SEM.

Table 2 .
Antioxidation activity of synthetic compounds in a DPPH assay.

Table 3 .
Docking energy values of synthesized compounds with tyrosinase.