GCA and CA dried leaves were cultivated under identical conditions for the same duration as previously described (25) and preserved at -40°C under regulated conditions. These dried samples were chopped into tiny pieces that were pulverized into a fine powder with an RT-34 mixer from Rong Tsong Precision Technology. The powder was then filtered through a ~20-mesh sieve. To perform the extraction, 300 mg of the powdered GCA and CA samples were immersed in 30 ml of distilled water at 60°C for 30 min. Following extraction, the solution was centrifuged at 3,000 × g for 10 min at room temperature, filtered through Whatman™ filter paper No. 4 (Cytiva), and then freeze dried.

Dulbecco's modified eagle medium was purchased from Hyclone; Cytiva. A solution of penicillin and streptomycin was purchased from CellGro by Mediatech (http://www.cellgro.com). Fetal bovine serum was purchased from Seradigm (https://us.vwr.com/cms/avantor_seradigm). The antibody against tyrosinase (cat. no. AB170905) was obtained from Abcam. Antibodies against MITF (cat. no. AB59201), phospho-PKA (p-PKA; cat. no. 4781S), PKA (cat. no. 4782S), p-CREB (cat. no. 9198S) and CREB (cat. no. 9197S) were purchased from Cell Signaling Technology, Inc. The anti-body against β-actin (cat. no. SC-8432) was obtained from Santa Cruz Biotechnology, Inc. A protein assay reagent kit was purchased from Bio-Rad Laboratories, Inc. Arbutin was purchased from MilliporeSigma and dissolved in dimethyl sulfoxide.

Murine melanoma B16F10 cells were obtained from the Korean Cell Line Bank (Seoul, Republic of Korea). B16F10 cells were seeded at a density of 2×10⁵ cells per 60 mm² dish. After overnight incubation at 37°C, the cells were pretreated with GCA and CA (25-100 µg/ml) for 1 h. Then, 200 nM of α-MSH was added, and the cells were incubated for 72 h. The conditioned media containing extracellular melanin was collected and centrifuged at 13,572 × g for 10 min. Then, the supernatant was transferred to a 96-well plate, and an Epoch microplate reader from BioTek Instruments was used to measure the absorbance at 490 nm (1).

The forced-floating method was used to establish a three-dimensional melanoma cell-culture system. In the present study, 1×10⁴ B16F10 cells were cultured in an ultra-low attachment (ULA) 96-well round plate (SPL Life Sciences) at 37°C. The next day, the cells were co-treated with 200 nM α-MSH, GCA (25-100 µg/ml), or arbutin (100 µg/ml). The ULA plate was incubated for 3 days at 37°C. The melanin content in the 3D cell cultures was analyzed by measuring the absorbance at 490 nm (1).

To measure the intracellular cAMP concentration, the cAMP assay was performed with cell lysate using a cAMP assay kit (R&D Systems, Inc.) in accordance with the manufacturer's protocol. B16F10 cells (2×10⁵ cells per 60 mm² dish) were incubated with the indicated concentration of GCA (25-100 µg/ml) and α-MSH for 15 min. Lysates were centrifuged 13,572 × g at 4°C, and then lysate was used directly. The concentration of cAMP was observed by measuring the absorbance at 450 nm in a plate reader.

To determine the tyrosinase inhibitory effect, 3,4-dihydroxy-L-phenylalanine (L-DOPA) was used as a substrate. To assay the tyrosinase inhibition of GCA, 80 µl of distilled water or 80 µl of L-DOPA, 80 µl of mushroom tyrosinase (27.8 U/ml) and various concentrations of GCA (25-100 µg/ml) were added to each well of a 24-well plate. Arbutin (100 µg/ml) was used as a positive control. The sample was mixed with tyrosinase and a 1 mM L-DOPA substrate to react at 37°C for 30 min. Tyrosinase activity was measured at 475 nm (2).

Cells were lysed in a RIPA lysis solution that included 10 mM Tris (Ph 7.5), 150 mM NaCl, 5 mM EDTA, 0.1% Triton X-100, 1 mM DTT, 0.1 mM PMSF, 10% glycerol, and a protease inhibitor cocktail tablet from GenDEPOT, LLC. BCA protein assay kits were used to quantify protein concentrations in the lysates following the manufacturer's protocol. For protein separation, 10 µg of protein samples underwent electrophoresis in a 10% sodium dodecyl sulfate-polyacrylamide gel electrophoresis system and were then transferred to PVDF membranes supplied by MilliporeSigma. Blocking of the membranes was achieved with 5% skim milk for 1 h at room temperature, followed by overnight incubation at 4°C with the designated primary antibody (1:1,000 dilution). After attaching the horseradish peroxidase-conjugated with goat anti-mouse IgG secondary antibody (cat. no. 1721019) or goat anti-rabbit IgG (cat. no. 1721019) secondary antibody (both from Bio-Rad Laboratories, Inc.) at a 1:5,000 dilution for 1 h at room temperature, an E-9150 Ez-Capture II device by Atto Corporation was used to detect protein bands (26). The ImageJ software version 1.53k (National Institutes of Health) was used to measure the relative density.

GCA and CA extracts were applied in a Thermo Vanquish HPLC system (Thermo Fisher Scientific Inc.) to identify the phytochemicals. An HPLC Cortecs C18 column (2.1×50 mm, 1.6 µm) from Waters Corporation was used to perform the chromatographic separations of the metabolites. The flow rate was set to 0.3 ml/min. A 1-µl aliquot of a 1,000-ppm GCA and CA extract was injected, and the column oven was set to 45°C. The mobile phases were 0.1% formic acid in HPLC-grade water (Solvent A) and 0.1% formic acid in HPLC-grade acetonitrile (Solvent B). Gradient elution was achieved by running the following gradient program: 0-0.5 min, 5% B; 0.5-3.5 min, 5-100% B; 3.5-4 min, 100% B; 4-4.1 min, 5% B; and a 2-min hold time followed by a 3-min re-equilibration to the starting conditions. High-resolution mass spectrometry (MS) data were obtained on a Thermo TSQ Altis high-resolution mass spectrometer equipped with a hybrid quadrupole-Orbitrap mass analyzer. Electrospray ionization (H-ESI) in positive-ion mode was used to acquire all MS data. The optimized MS parameters were as follows: Spray voltage maintained statically at 3,500 V, sheath gas flow rate set at 50 arbitrary units, auxiliary gas at 10 arbitrary units, and sweep gas at 1 arbitrary unit. The ion-transfer tube and vaporizer temperatures were maintained at 325°C and 350°C, respectively. Selected Reaction Monitoring was conducted in positive-ion mode with a cycle time of 0.5 sec. The resolutions of quadrupole 1 (Q1) and quadrupole 3 (Q3) were set to 0.7 and 1.2 full-width at half height, respectively, without the use of a calibrated RF lens. Collision-induced dissociation was performed with a gas pressure of 1.5 mTorr.

Molecular docking simulations were conducted to investigate the binding affinity and potential interactions between the melanocortin-1 receptor (MC1R) and selected ligands. The 3D structure of MC1R (PDB ID: 7F4D, α-MSH-bound melanocortin-1 receptor) was obtained from the Protein Data Bank (https://www.rcsb.org/), and the 3D structures of the ligands were sourced from the Drug Bank. Autodock Vina in the AMdock platform (27) was used to conduct docking simulations between MC1R and four selected ligands. A previous study identified α-MSH binding sites at GLU94 and LEU106 on MC1R (28); therefore, a search space of 60 cubic Angstroms was centered on these sites. AMdock generated simulations for the 10 most probable binding configurations from which the results with the lowest energy were selected. The two compounds were further analyzed using the MolDock software (version 7.0.0, http://molexus.io/) to calculate their MolDock scores. The same grid box parameters were applied, and the MolDock optimizer algorithm was used to generate the scores (29).

B16F10 cellular supernatant (500 µg) was incubated at 4°C with asiaticoside-Sepharose 4B beads, madecassoside-Sepharose 4B beads (or Sepharose 4B alone as a control) (100 µl, 50% slurry) in reaction buffer [50 mM Tris (pH 7.5), 5 mM EDTA, 150 mM NaCl, 1 mM DTT, 0.01% Nonidet P-40, 2 µg/ml bovine serum albumin, 0.02 mM PMSF and 1X protease inhibitor mixture] for 24 h at 4°C. For competition assays, α-MSH (0.2, 2, 20, or 200 µM) was added to the reaction mixture to a final volume of 500 µl and incubated at 4°C for an additional 24 h. After incubation, the beads were washed five times with buffer [50 mM Tris (pH 7.5), 5 mM EDTA, 150 mM NaCl, 1 mM DTT, 0.01% Nonidet P-40 and 0.02 mM PMSF]. The proteins bound to the beads were then analyzed by immunoblotting.

The 3D human skin-equivalent model, Neoderm^®-ME from Tego Science, Inc., consists of human epidermal keratinocytes and human melanocytes. Neoderm^®-ME was transferred to a 12-well plate in a maintenance medium (Tego Science, Inc.) containing GCA or arbutin and incubated at 37°C in 5% CO₂ for 7 days. The medium was changed once every 2 days. On day 7, skin pigmentation was observed by viewing with an Olympus AX70 light microscope (Olympus Corporation). Lysate was collected and subsequently centrifuged at 4°C for 10 min at 13,572 × g. BCA assay reagent kits were used to measure the protein concentration of the lysate following the manufacturer's protocol to analyze tyrosinase expression.

3D human skin-equivalent samples were fixed overnight in 4% formaldehyde at room temperature, embedded in paraffin, and then cut into 3-µm-thick sections with a microtome. The slides were treated with an ammoniacal silver solution at 56°C for 30 min, followed by rinsing in distilled water. The slides were then incubated in a 0.2% gold chloride solution at room temperature for 30 sec and then in a 5% sodium thiosulfate solution for 2 min. The sections were incubated in nuclear fast red solution for 5 min and dehydrated three times by use of fresh absolute alcohol. An AX70 light microscope (Olympus Corporation) was used to examine the staining results.

Statistical analyses were performed using the SPSS-WIN 12.0K program (SPSS, Inc.). All data are presented as the mean ± standard deviation or standard error of the mean. Unpaired Student's t-test was used for single statistical comparisons and one-way ANOVA followed by Tukey's Honest Significant Difference (HSD) test for multiple comparisons. Differences were considered statistically significant for values of P<0.05.

Research on B16F10 cells has shown that α-MSH increases melanin expression through various pathways (30,31). The concentration of α-MSH and treatment time used in the present study were selected on the basis of previous research results (32,33). It was found that α-MSH effectively induced melanin expression up to 2.4-fold relative to the control. Arbutin was used as a positive control (34,35). Melanin secretion was inhibited more effectively by the GCA extract than by the CA extract (Fig. 1A). Especially at a concentration of 100 µg/ml, the GCA extract exhibited superior efficacy in skin-whitening improvement, with a 77% reduction relative to that of α-MSH induction, whereas the CA extract showed a 45% reduction. To further confirm the anti-melanogenic effect of GCA, a 3D cell-culture model was used, which more closely mimics the in vivo skin environment (1). The 3D model visually demonstrated the skin-whitening effect of the GCA extract (Fig. 1B). Quantitative analysis of the melanin content in the 3D model revealed that the GCA extract reduced melanin production dose-dependently, with the 100 µg/ml concentration demonstrated efficacy comparable to that of arbutin (Fig. 1C). These results consistently demonstrated the superior melanin-inhibiting effects of the GCA extract in both 2D and 3D cell-culture systems.

These results suggested that GCA extract effectively inhibits the PKA-CREB-MITF signaling cascade, which is critical for melanin synthesis. In B16F10 cells, α-MSH significantly elevates cAMP levels, triggering enhanced melanin synthesis (36). This increase in cAMP activates PKA, phosphorylating critical proteins in the melanogenesis pathway (9). The present results showed that α-MSH treatment significantly increased intracellular cAMP levels, which in turn activated PKA. In the presence of GCA extract, a dose-dependent reduction in cAMP levels was observed, suggesting that GCA modulates this signaling cascade by inhibiting the α-MSH-induced cAMP increase (Fig. 2A). To further investigate the effects observed in cells, additional experiments with L-DOPA were performed to determine how the GCA extract affects tyrosinase activity. The experiment targeted tyrosinase extracted from mushrooms and evaluated the inhibitory effect of an GCA extract at various concentrations. By using L-DOPA as a substrate to measure the activity of tyrosinase, it was possible to gain a deeper understanding of how the GCA extract regulates its enzyme's activity, which is crucial in the melanin production process (2,37). Particularly at concentrations of 50 µg/ml and 100 µg/ml (Fig. 2B), the GCA extract exhibited pronounced effects in inhibiting tyrosinase activity. The inhibitory effect of the GCA extract at 100 µg/ml was comparable to that of arbutin, a known tyrosinase inhibitor, indicating that the GCA extract effectively modulated the activity of enzymes critical for melanin production. Tyrosinase is crucial for melanin production, and its activity is regulated by various factors and signaling pathways, including the MC1R-cAMP-PKA-CREB-MITF axis (28,38). The current investigation focused on assessing GCA extract's potential to modulate this pathway and its downstream effects on melanogenesis. Treatment of B16F10 melanoma cells with GCA extract resulted in a dose-dependent decrease in the phosphorylation of PKA and CREB, as well as reduced expression levels of MITF and tyrosinase (Fig. 2C).

To identify the key components contributing to the melanin inhibition effect of GCA extract, HPLC-MS/MS analysis of the major triterpenoids was performed in both the GCA and CA extracts. The molecular structures of the four main triterpenoids (madecassoside, asiaticoside, madecassic acid and asiatic acid) are presented in Fig. 3A. Quantitative analysis revealed significant differences in the triterpenoid composition between the GCA and CA extracts (Fig. 3B). Notably, the GCA extract contained substantially higher levels of madecassoside (10.43±0.05 mg/g) than those in the CA extract (2.53±0.01 mg/g). Similarly, asiaticoside content was higher in the GCA extract (1.65±0.01 mg/g) than in the CA extract (0.95±0.01 mg/g). These compounds have been previously reported to possess anti-melanogenic properties (39,40). Interestingly, the levels of madecassic acid and asiatic acid were lower in the GCA extract than in the CA extract. This differential composition suggests that the enhanced melanogenesis inhibitory effect of the GCA extract is primarily attributed to its higher content of madecassoside and asiaticoside. These findings provide insight into the potential active components responsible for the superior skin-whitening effects of the GCA extract observed in the aforementioned experiments.

Treatment with asiaticoside and madecassoside significantly reduced melanin production (Fig. 4A). Asiaticoside showed a more potent inhibitory effect than that of madecassoside. Conversely, madecassic acid and asiatic acid had minimal effects on inhibiting melanin production, which may explain GCA's superior efficacy in suppressing melanin production considering the higher content of asiaticoside and madecassoside in GCA than in CA (Fig. 3B). Pull-down assay results confirmed that asiaticoside and madecassoside directly bound to MC1R (Fig. 4B and C). Interestingly, increasing concentrations of α-MSH led to decreased binding of asiaticoside and madecassoside to MC1R. This finding suggests that asiaticoside and madecassoside compete with α-MSH for binding to MC1R.

To further investigate the binding of asiaticoside or madecassoside to MC1R, molecular docking studies were performed using the human MC1R structure. Previous studies have suggested that the GLU94 of MC1R greatly influences ligand binding and receptor function (41). The present docking model results revealed that both asiaticoside and madecassoside bound to GLU94 of MC1R, with docking energies of -9.9 and -8.6 kcal/mol, respectively (Fig. 5A and B). By contrast, the docking energy of α-MSH was higher at -6.9 kcal/mol. These results suggested that asiaticoside and madecassoside may bind more strongly to MC1R than α-MSH, which is consistent with the findings from the pull-down assay (Fig. 4B and C). While the aforementioned Autodock results indicated that asiaticoside has higher potency than madecassoside, further analysis using MolDock confirmed this observation. The MolDock binding scores for asiaticoside were -224.697, which were higher than those for madecassoside, which scored -209.118. These results provide stronger evidence of the higher binding affinity of asiaticoside (Fig. 5B). However, due to the large molecular weight of α-MSH, it was not possible to obtain α-MSH MolDock results.

To further validate the anti-melanogenic effects of the GCA extract in a more physiologically relevant context, a 3D human skin-equivalent model (Neoderm^®-ME) was used. This model incorporates human melanocytes and exhibits natural melanogenesis over time, closely mimicking human skin (1). Neoderm^®-ME was treated with 25, 50 and 100 µg/ml GCA extract, 100 µg/ml CA extract, and arbutin as a positive control. After 7 days of treatment, the GCA extract at 100 µg/ml significantly reduced melanin content by 64% relative to that in the control, outperforming both the CA extract (30% reduction) and arbutin (Fig. 6A). The expression of tyrosinase, a critical enzyme in melanin synthesis, was also markedly decreased in the presence of the GCA extract, particularly at 100 µg/ml (Fig. 6B). This reduction in tyrosinase expression is associated with the observed decrease in melanin production (Fig. 6A). Furthermore, Fontana-Masson staining, which enables visualization of melanin in tissue sections, revealed a notable reduction in staining intensity in samples treated with the GCA extract (Fig. 6C), corroborating the quantitative data and tyrosinase expression patterns. Collectively, these findings demonstrated that the GCA extract effectively inhibits melanogenesis in a 3D human skin-equivalent model, suppressing both melanin production and tyrosinase expression. These results, in conjunction with the molecular docking studies and binding assays, provide compelling evidence that the anti-melanogenic effects of the GCA extract were superior to those of the CA extract, highlighting its potential as a natural skin-whitening agent.