Epigallocatechin-3-Gallate and Glabridin Synergistically Inhibited Melanogenesis by Molecular Interaction and the MicroRNA–Microphthalmia-Related Transcription Factor Pathway

Abstract

Epigallocatechin-3-gallate (EGCG) and glabridin (GLA) are two phytochemicals derived from homologous plants used for medicine and food that have demonstrated efficacy in inhibiting excessive production of melanin and promoting skin whitening properties. However, it is unknown whether EGCG and GLA can synergistically inhibit melanogenesis. The aim of this study was to explore the effects of the combination of EGCG and GLA on the expression and activity of microphthalmia-related transcription factor (MITF) and to reveal the mechanism of the synergistic treatment with EGCG and GLA on excessive melanin deposition. This study used B16F10 cells to screen for the synergistic effect on melanin deposition and determined the best ratio of EGCG and GLA. Dual-luciferase reported gene assays were used to explore whether GLA and EGCG could regulate the mRNA level of MITF. Molecular docking and molecular dynamics simulation were used to explain the synergistic interaction of EGCG and GLA with MITF, and then Masson–Fontana and hematoxylin–eosin were used to observe the skin changes of mice. The results showed that the combination of EGCG and GLA could synergistically regulate MITF and thus inhibit melanogenesis by modulating microRNA expression. EGCG and GLA were also able to synergistically interact with MITF to inhibit excessive melanogenesis.

INTRODUCTION

There is a global consensus that tea is a healthy beverage. Tea contains large amounts of tea polyphenols, of which epigallocatechin-3-gallate (EGCG) is one of the main active components.^1,2 Many studies have shown that EGCG can play a role in skin whitening by reducing melanin production and is a potential natural plant-derived cosmetic ingredient.^3,4 Glabridin (GLA) is an isoflavone compound derived from the leguminous plant Glycyrrhiza glabra L., which has anti-oxidative, anti-inflammatory, and anti-melanogenesis effects. It has attracted much attention in the field of skin health care and has been widely used in various types of cosmetics.⁵ However, the synergistic inhibition of melanin production by EGCG and GLA has not been elucidated and needs to be further studied.

Melanin is synthesized by melanocytes in the skin’s epidermis. Melanogenesis is a complex synthetic process that is affected by melanin-related enzymes, proteins, and genes, such as microphthalmia-related transcription factor (MITF), tyrosinase (TYR), DHICA oxidase (TRP-1), and DOPA pigment isomerase (TRP-2). MITF regulates transcription by binding to the promoter regions of TYR, TRP-1, and TRP-2 to regulate melanogenesis.⁶ However, certain studies have shown that the MITF expression levels were associated with the regulation of some microRNAs (miRNAs).⁷ MicroRNAs are small, single-stranded RNA molecules of 21–25 nt length that do not code proteins.⁸ They regulate gene transcription by binding to the untranslated region (UTR) of mRNAs, either by increasing the transcription of mRNAs or by decreasing the level of mRNAs.⁹ Therefore, miRNAs promote or inhibit melanogenesis by modulating the expression of relevant target genes (MITF).

Melanin is the main cause of skin pigmentation. Normal melanin deposition can protect the body from light damage, whereas excessive production of melanin can lead to hyperpigmentation in the skin.¹⁰ Ultraviolet radiation is one of the primary external factors that causes excessive pigmentation in skin.¹¹ Ultraviolet radiation can cause many pathological effects in the skin, including wrinkles, roughness, and epidermal thickening.¹² Ultraviolet radiation B (UVB) is the main band that affects the synthesis and transport of melanin in melanocytes.¹³

Melanin is an important factor in the skin, while excessive deposition is detrimental to skin homeostasis. We were surprised to find that EGCG and GLA have obvious synergistic inhibition on melanin production in B16F10 cells, and a significant synergistic interaction of EGCG and GLA with MITF was observed by molecular dynamics (MD) simulation and molecular docking. Interestingly, EGCG and GLA synergistically regulated the miRNA levels, which further modulated the mRNA transcription of MITF by dual luciferase reporter gene assay. Therefore, the aim of this study was to reveal the synergistic inhibition mechanism on MITF and the consequent reduction in melanin deposition by EGCG and GLA.

MATERIALS AND METHODS

Materials

EGCG and GLA (purity >98%) were obtained from Shanghai Yuanye Biotechnology Co., Ltd. Arbutin (Ab) was used as a positive control and purchased from Thermo Fisher Co., Ltd. Polyvinyl alcohol 205 (PVA-205) and polyvinylpyrrolidone K29/32 (PVP-K29/32) were used to prepare the micro-needles and purchased from Shanghai Yuanye Biotechnology Co., Ltd.

Cell culture

B16F10 cells were purchased from the Institute of Biochemistry and Cell Biology, Chinese Academy of Sciences and inoculated in DMEM high-glucose medium (Gibco) containing a sterile 10% fetal bovine serum (Gibco) and 1% penicillin–streptomycin mixture (Gibco). The culture flask was placed in a CO₂ incubator (37°C, 5% CO₂, relative humidity of 98%). When the cell density reached 80–90% in the flask, the cells were digested with 0.25% trypsin (Gibco) and subcultured.

Cell viability

B16F10 melanoma cells were plated in 96-well plates at an initial density of 1 × 10⁴ cells/mL (100 μL/well) and maintained under standard culture conditions (37°C, 5% CO₂) for 12 h. Post-adhesion, the control group received 100 μL of fresh complete medium, whereas treatment groups (EGCG, GLA, Ab) were administered equivalent volumes of medium containing respective pharmacological agents. In the EGCG group, B16F10 cells were treated with different doses of EGCG (0, 50, 100, 150, 200, 250, and 300 μg/mL) for 48 h. In the GLA group, B16F10 cells were treated with different doses of GLA (0, 25, 50, 75, 100, 125, and 150 μg/mL) for 48 h. In the Ab group, B16F10 cells were treated with different doses of Ab (0, 50, 100, 150, 200, 250, and 300 μg/mL) for 48 h. Post-treatment protocols involved careful removal of supernatants followed by three ice-cold PBS washes. Cellular viability was quantified by adding 10 μL of 10% CCK-8 solution (Gibco) per well, with a subsequent 4-h incubation. Optical density measurements at 450 nm were conducted using a spectrophotometric microplate reader. Experimental validity was ensured through triplicate technical replicates per condition, with data representing mean values from three biologically independent trials.

Relative content of cell melanin

Melanin content in B16F10 melanoma cells was quantified using a sodium hydroxide (NaOH) solubilization assay. Cells were plated in 96-well plates at a density of 1 × 10⁴ cells/mL (100 μL/well) and incubated under standard conditions (37°C, 5% CO₂) for 12 h. Following adhesion, the control group received 100 μL of fresh complete medium, while treatment groups (EGCG, GLA, and Ab) were supplemented with equivalent volumes of medium containing respective pharmacological agents. In the EGCG group, B16F10 cells were treated with different doses of EGCG (0, 25, 50, 75, 100, 125, and 150 μg/mL) for 48 h. In the GLA group, B16F10 cells were treated with different doses of GLA (0, 20, 40, 60, 80, and 100 μg/mL) for 48 h. In the Ab group, B16F10 cells were treated with different doses of Ab (0, 25, 50, 75, 100, 125, and 150 μg/mL) for 48 h. Post-treatment, cells underwent three PBS washes, followed by trypsinization (0.25% trypsin) and centrifugation (1000 rpm, 5 min). Pelleted cells were lysed with 200 μL NaOH (1 mol/L) containing 10% dimethyl sulfoxide and then incubated at 80°C for 1 h. Absorbance was measured spectrophotometrically at 405 nm to calculate melanin content. The half-maximal inhibitory concentration (IC₅₀) for melanogenesis was derived for each treatment. Based on IC₅₀ values of EGCG and GLA, combination regimens were designed and analyzed using CompuSyn software^14

–17 to compute combination indices (CI). Synergistic effects were defined as CI < 1, indicating enhanced suppression of melanin synthesis.

Quantitative real-time PCR

Total RNA isolation and subsequent complementary DNA synthesis were systematically performed with commercial reagent systems (TRIzol® Plus RNA Purification Kit and SuperScript™ III First-Strand Synthesis SuperMix, Thermo Fisher Scientific). Quantitative amplification primers were computationally designed through the integrated application of Primer Premier 6.0 and Beacon Designer 7.8 platforms, with oligonucleotide specifications comprehensively outlined in Table 1. Fluorometric quantification of nucleic acids was achieved using PowerUp™ SYBR™ Green Master Mix (Applied Biosystems) according to established protocols, with all reactions conducted in triplicate. Transcript abundance normalization was implemented through parallel amplification of the constitutively expressed GAPDH gene, where relative expression ratios were computationally derived using the comparative Ct (2^−ΔΔCt) algorithm for quantitative interpretation.

Table 1.

Real-Time PCR Primers

Gene	Genbank accession	Primer sequences(5′ to 3′)	Size (bp)	Annealing (°C)
Mouse GAPDH	GU214026.1	GAAGGTCGGTGTGAACGGATTTG	127	60
Mouse GAPDH	GU214026.1	CATGTAGACCATGTAGTTGAGGTCA	127	60
Mouse MITF	NM_001113198.2	GATCGACCTCTACAGCAACCAG	132	60
Mouse MITF	NM_001113198.2	GCTCTTGCTTCAGACTCTGTGG	132	60
Mouse Trp-1	NM_031202.3	GCCACAGGATGTCACTCAGTG	151	60
Mouse Trp-1	NM_031202.3	CAGGTTGTGAAGGCTTCGAACA	151	60
Mouse Trp-2	NM_010024.3	GCAAGATTGCCTGTCTCTCCAGAA	114	60
Mouse Trp-2	NM_010024.3	GAGTCCAGTGTTCCGTCTGCTT	114	60
Mouse TYR	NM_011661.6	CAGGCTCCCATCTTCAGCAGAT	130	60
Mouse TYR	NM_011661.6	CCCTGTGAGTGGACTGGCAAA	130	60

MITF, microphthalmia-related transcription factor; Trp-1, DHICA oxidase; Trp-2, DOPA pigment isomerase; TYR, tyrosinase.

Western blot

Cellular protein lysates were prepared using a commercial extraction system (Total Protein Extraction Kit, Thermo Fisher Scientific), with protein concentration determined by bicinchoninic acid (BCA) assay (Beyotime Institute of Biotechnology). Following sequential processing steps including SDS-PAGE (12% resolving gel), electrophoretic transfer onto PVDF membranes (0.45 μm pore size), and blocking with 5% non-fat milk/TBST (1 h at 25°C), membranes underwent immunodetection using primary antibodies (Table 2) followed by species-matched HRP-conjugated secondary antibodies(Table 3). Chemiluminescent detection was conducted by incubating membranes with SuperSignal™ West Dura substrate (Thermo Fisher Scientific) for 60 s, followed by X-ray film exposure in a darkroom. Band intensity quantification was performed using ImageJ software (National Institutes of Health) with β-actin serving as a loading control. Statistical analysis of triplicate experiments was executed using Student’s t-test, with data presented as mean ± standard deviation (n = 3 biological replicates).

Table 2.

Primary Antibodies

Antibody	Company	Dilutions	Molecular mass (kDa)
MITF	Abcam ab140606	1: 200	52
TYR	Thermo Fisher MA5-14177	1: 100	60
Trp-1	Abcam ab235447	1: 1000	75
Trp-2	Proteintech 13095–1-AP	1: 500	59
β-actin	Abcam ab8226	1: 10000	42

Table 3.

Secondary Antibodies

Antibody	Company	Dilutions
Goat antimouse IgG (H + L) secondary antibody	Thermo Pierce 31160	1:5000
Goat antirabbit IgG (H + L) secondary antibody	Thermo Pierce 31210	1:5000

Molecular docking and molecular dynamics

The three-dimensional structure of the MITF protein was retrieved from the RCSB Protein Data Bank (https://www.rcsb.org), while the SDF-formatted ligand structures of EGCG and GLA were acquired from the PubChem Compound Database (https://pubchem.ncbi.nlm.nih.gov). Structural format conversion (SDF → PDB) for ligands was performed using OpenBabel 3.1.1. The protein target underwent preprocessing in AutoDock Tools 1.5.6, involving the removal of nonpolar hydrogens and the addition of polar hydrogens. Both ligands and the target protein were subsequently converted into PDBQT format through charge calculation and atom type assignment. Molecular docking simulations were executed using AutoDock Vina 1.2.3.

The investigation of receptor-ligand binding mechanisms employed MD computational approaches, utilizing GROMACS 2019.6 for trajectory calculations.¹⁸ Protein and ligand molecular topologies were constructed using the Amber99sb-ildn and GAFF force fields, respectively. A cubic solvation environment was established with 1.0 nm buffer regions surrounding the biomolecule. The explicit solvent system incorporated SPC216 water molecules, with ionic compensation (Na⁺/Cl⁻) ensuring electrostatic equilibrium.

Structural optimization proceeded through steepest descent energy minimization to resolve atomic steric conflicts. Sequential thermal (NVT, 300 K) and pressure (NPT, 1 bar) equilibration phases were conducted over 100 ps intervals. Following system stabilization, extended MD sampling was performed for 100 ns under periodic boundary constraints, employing V-rescale temperature coupling and Parrinello–Rahman pressure regulation.¹⁹

MD simulations employed the leapfrog Verlet algorithm with a 2-femtosecond integration time step. Long-range electrostatic forces were accounted for through the Particle Mesh Ewald (PME) method, utilizing a 0.16-nm Fourier grid spacing parameter. Bond length constraints were maintained using the LINCS algorithm with fourth-order polynomial interpolation. Trajectory analysis leveraged specialized visualization tools: VMD 1.9.3 and PyMOL 2.4.1 for three-dimensional structural rendering and dynamic process animation. Thermodynamic binding parameters were quantified through the MMPBSA protocol implemented in the gmx_mmpbsa (http://jerkwin.github.io/gmxtool) computational framework.

miRNA levels in B16F10 cells

Using combinatorial algorithms from TargetScan (https://www.targetscan.org/vert) and miRanda (www.microrna.org), MITF was systematically predicted as a potential target of miRNAs in B16F10 melanoma cells. Subsequent refinement via TargetScanMouse identified conserved miRNA candidates specifically interacting with MITF. According to the “relative content of cell melanin” method, B16F10 cells were treated with EGCG (100 μg/mL), GLA (50 μg/mL), Ab (125 μg/mL), and EGCG combined with GLA (EGCG: GLA = 50 μg/mL: 25 μg/mL), respectively. RNA extraction was performed using TRIzol™ reagent (Thermo Fisher Scientific), followed by miRNA isolation via PureLink™ miRNA Isolation Kit. Reverse transcription utilized SuperScript™ III Reverse Transcriptase with stem-loop primers (sequences provided in Table 4). Quantitative real-time PCR amplification (QuantStudio 6 Flex System) evaluated expression levels of miR-182-5p, miR-96-5p, and miR-183-5p using SYBR™ Green chemistry, with primer sequences documented in Table 4.

Table 4.

Reverse Transcription Primer Sequences of miRNAs and Real-Time PCR Primers of miRNAs

Gene	Genbank accession	Primer sequences (5′ to 3′)	Size (bp)
mmu-miR-182-5p-RT	MIMAT0000211	GTCGTATCCAGTGCAGGGTCCGAGGTA	63
mmu-miR-182-5p-RT		TTCGCACTGGATACGACCGGTGT
mmu-miR-182-5p-F		GCGTTTGGCAATGGTAGAACTC
mmu-miR-96-5p-RT	MIMAT0000541	GTCGTATCCAGTGCAGGGTCCGAGGTA	62
mmu-miR-96-5p-RT		TTCGCACTGGATACGACAGCAAA
mmu-miR-96-5p-F		GGCGTTTGGCACTAGCACAT
mmu-miR-183-5p-RT	MIMAT0000212	GTCGTATCCAGTGCAGGGTCCGAGGTA	61
mmu-miR-183-5p-RT		TTCGCACTGGATACGACAGTGAA
mmu-miR-183-5p-F		CGCGTATGGCACTGGTAGAA
U6	M10329.1	GTCGTATCCAGTGCAGGGTCCGAGGTA	107
U6		TTCGCACTGGATACGACGTCATCC
U6-F		CGCTTCGGCAGCACATATACTAA
Universal reverse primer（micro-R）	/	AGTGCAGGGTCCGAGGTATT	/

miRNA, microRNA.

Dual luciferase reporter gene assay

In 96-well plates, 293T cells were seeded at a density of 1 × 10⁵ cells/mL. After 12 h, pmirGLO-MITF WT or MUT was co-transfected with miR-182-5p, miR-183-5p, and miR-96-5p mimic or NC into the above cells. After 48 h, the dual luciferase reporter gene detection system (Promega Co., Ltd.) was used to determine the luciferase activity of fireflies and Renilla according to the operation instructions. Finally, firefly luciferase activity was normalized according to Renilla luciferase activity.

Computational docking analyses revealed specific binding affinities between MITF and either EGCG or GLA. A MITF-responsive luciferase reporter system was engineered through molecular cloning of the pcDNA3.1-luc vector, with constitutive pRL-TK expression serving as internal normalization control; 293T cultures were established in 96-well formats (80% confluency) for parallel pharmacological interventions: EGCG (0, 50, 100, and 150 μg/mL), GLA (0, 25, 50, and 100 μg/mL), and EGCG combined with GLA (0:0, 25:25, 50:50, and 100:100). Lipofectamine™ 3000-mediated co-transfection (Thermo Fisher Scientific) delivered MITF-luc/pRL-TK complexes (10:1 mass ratio). Post-48 h incubation, sequential processing included each well was washed twice with PBS, 250 µL of 1×PLB lysate was added, and cells were then lysed at room temperature. The Firefly and Renilla luciferase activities were then determined with the Dual-Luciferase Reporter Assay System (Promega Co., Ltd.) according to the operating guidance. Then Firefly luciferase activity was normalized according to Renilla luciferase activity.

Preparation and characterization of DMN

Polydimethylsiloxane (PDMS) (Taizhou Microchip Pharmaceutical Technology Co., Ltd.) was used as the casting mold. The main body of the mold was cylindrical (diameter 8 mm, height 150 mm). The mold had a circular array of 80 needles (needle length 1000 μm, tip distance 750 μm, and bottom diameter 420 μm). The concentrations of 60% PV A-205 and 45% PV-k29/32, respectively, were poured into the PDMS molds, centrifuged for 5 min (5000 rpm/min), and dried at 37°C for 24–48 h before demolding. Fresh pig skin was used to test the machine properties of DMN, in vitro skin insertion rate, and to evaluate the transdermal release ability of DMN using the Franz diffusion cell method.^20
–22

Animals

This study was approved by the Animal Experimental Ethics Committee of China Jiliang University (2023-055), which conforms to the principles of animal protection, welfare, and ethics. Healthy 4-week-old C57BL/6 female mice were purchased from Shanghai SLAC Experimental Animal Co., Ltd (SCXK-<Hu>-2022-0004). All mice were placed in an experimental environment of 22 ± 2°C temperature, 50 ± 10% humidity, and 12 h of day and night circulation. During the experiment, all mice were free to drink water and feed.

The design of the animal experiment is shown in Figure 1. After a week of adaptive culture, 70 healthy C57BL/6 female mice were randomly divided into a control group, a model group, and a DMN treatment group. The DMN group was subdivided into the EGCG group (100 μg/mL), GLA group (50 μg/mL), Ab group (150 μg/mL), empty Micro-needle group, EGCG & GLA combination group (EGCG: GLA= 50:25), and 10 mice were used in each group. A 2 × 3 cm area of the back skin of these mice was shaved with a shaver and then smeared with depilatory cream for 10 min, and the remaining hair was scraped with a scraper. Except for the control group, the mice in the other groups were placed in a ventilated box with a 302 nm UVB lamp to receive 180 mJ/cm² irradiation every day for 3 weeks. During the 3 weeks, the control group received normal natural light irradiation, and DMN groups were treated with 302 nm UVB light irradiation (per mouse once per day for 6 h). After 3 weeks, the control group continued to receive normal natural light irradiation, the model group and DMN treatment groups continued to receive UVB irradiation, and DMN treatment groups continued to be treated for a period of 3 weeks. After the end of the treatment, the experimental mice were euthanized by carbon dioxide asphyxia, and the skin tissue on the back of the mice was taken for histopathological observation and tissue biochemical index determination.

FIG. 1.

Animal experiment.

Histopathological observation of the skin

The skin tissues (0.5 cm × 0.5 cm) from the backs of the mice were fixed with 10% formalin for 24 h, dehydrated, and embedded in paraffin, and then sliced by a slicing machine. The slices were divided into two parts with a thickness of 3–4 μm for hematoxylin–eosin (HE) staining and Masson–Fontana staining, respectively. The sections used for HE staining were dewaxed with xylene, rehydrated with a graded series of ethanol, and stained with HE. The thickening of the skin epidermis was observed under a common optical microscope. The sections used for Masson–Fontana staining were subjected to xylene dewaxing and hydration, then immersed in ferrous sulfate for 60 min, washed with distilled water 5–6 times, and then immersed in acidic potassium ferricyanide solution for 30 min. After 4–5 times of distilled water, the sections were counterstained with nuclear fast red staining solution for 5 min, washed with distilled water for 1–2 times, dehydrated, transparent, and sealed with synthetic resin, and then observed by optical microscope. The distribution of black particles in the epidermal basal layer of skin tissue was used to characterize melanin deposition.

Quantitative real-time PCR and Western blot in mice skin

The specific method is shown in “Quantitative Real-Time PCR” and “Western Blot” sections.

Statistical analysis

In this study, one-way analysis of variance was performed using SPSS 24.0 (IBM), and the data were expressed as means ± standard deviations. The statistical significance between the groups was tested by using Tukey tests for post-hoc multiple comparisons, and P < .05 was statistically significant.

RESULTS

EGCG and GLA exhibited no cytotoxicity against B16F10 within the tested concentrations

In this study, the safe concentration of EGCG, GLA, and Ab for B16F10 cells was screened by CCK-8 assay. As shown in Figure 2a, the results showed that the concentrations of EGCG up to 150 μg/mL had no cytotoxic effect on B16F10. As shown in Figure 2b, the results showed that the content of GLA within 100 μg/mL had no effect on B16F10. As shown in Figure 2c, the results showed that the content of Ab up to 150 μg/mL had no cytotoxic effect on B16F10.

FIG. 2.

Cytotoxicity screening of EGCG, GLA, and Ab on B16F10 cells. (a) The effect of EGCG on the activity of cells. (b) The effect of GLA on the activity of cells. (c) The effect of Ab on the activity of cells. The group without drug treatment was used as the control group (cell activity was 100%). The cell viability of each group after drug treatment was calculated as the percentage of the control group. After the ANOVA/Tukey–Kramer multiple comparison test, all values are shown as means ± SEM (n = 10). ***P < .001 vs control group. Ab, arbutin; ANOVA, analysis of variance; EGCG, epigallocatechin-3-gallate; GLA, glabridin; SEM, standard error of mean.

Combination of EGCG and GLA had an obvious synergistic inhibition on melanin production in B16F10

IC₅₀ values were obtained from single compound treatment when the relative contents of melanin reached 50% in B16F10 cells, and then the combination regimen was formulated according to IC₅₀ values. After the proportions of several combinations of EGCG and GLA were tested, CI values were calculated by CompuSyn software. When CI < 1, the combination had a synergistically inhabited melanin production.

As shown in Figure 3a, the results showed that the IC₅₀ of EGCG was 100.04 (μg/mL). As shown in Figure 3b, the results showed that the IC₅₀ of GLA was 50.34 (μg/mL). As shown in Figure 3c, the results showed that the IC₅₀ of Ab was 125.27 (μg/mL). In this study, five combinations of EGCG and GLA were designed according to the IC₅₀ values of a single compound, which were group a (EGCG: GLA = 25:25), group b (EGCG: GLA = 25:50), group c (EGCG: GLA = 50:25), group d (EGCG: GLA = 50:50), and group e (EGCG: GLA = 100:50). The relative content of melanin in each combination was calculated by CompuSyn software. As shown in Table 5, the combination of EGCG and GLA had a synergistic inhibition of melanin production in B16F10 cells. According to CompuSyn software calculation, it was concluded that the inhibition rate of melanin formation in B16F10 cells reached 50% (IC₅₀) when EGCG and GLA were combined at a ratio of 41.74 μg/mL: 20.87 μg/mL, which showed that the CI value was 0.83 (<1). In summary, the ratio of EGCG: GLA = 50 μg/mL: 25 μg/mL was used as the combination group of EGCG and GLA in subsequent experiments.

FIG. 3.

The combination of EGCG and GLA has a synergistic inhibitory effect on melanin production in B16F10 cells. (a) The relative synthesis of melanin in B16F10 cells under different concentrations of EGCG (0, 25, 50, 75, 100, 125, 150 μg/mL). (b) The relative synthesis of melanin in B16F10 cells under different concentrations of GLA (0, 20, 40, 60, 80, 100 μg/mL). (c) The relative synthesis of melanin in B16F10 cells under different concentrations of Ab (0, 25, 50, 75, 100, 125, 150 μg/mL). (d) The relative synthesis of melanin in B16F10 cells was treated with different proportions of EGCG and GLA (25:25, 25:50, 50:25, 50:50, 100:50). (e) Median–effect plot. (f) Dose–effect curve. After the ANOVA/Tukey–Kramer multiple comparison test, all values are shown as means ± SEM (n = 10). ***P < .001 vs control group.

Table 5.

Synergistic Inhibition on Melanogenesis by EGCG and GLA

Groups	Dm (μg/mL)	m	r	CI
EGCG	100.041	0.68039	0.972276	/
GLA	50.3411	0.72369	0.93897	/
Ab	125.265	0.66982	0.97526	/
EGCG and GLA	62.6113 (41.7409 + 20.8704)	2.99431	0.98698	0.83182

Dm is used to indicate the dose that is 50% effective (IC₅₀). The value of m is the slope of the median-effect plot. m = 1, m > 1 and m < 1 indicate hyperbolic, sigmoidal, and flat sigmoidal, respectively. The value of r is the linear correlation coefficient by the median-effect plot.

Ab, arbutin; CI, combination indices; EGCG, epigallocatechin-3-gallate; GLA, glabridin.

Regulation of the expression of melanin-related genes by EGCG and GLA

The study was divided into the following groups: Control group, EGCG group (100 μg/mL), GLA group (50 μg/mL), Ab group (125 μg/mL), and combination group (EGCG:GLA = 50 μg/mL: 25 μg/mL), as shown in Figure 4a–d. Compared with the control group, the four treatment groups (EGCG, GLA, Ab, EGCG and GLA) significantly decreased the transcription levels of MITF, TYR, Trp-1, and Trp-2. The largest decrease was seen in the synergistic treatment group. The protein intensities of MITF, TYR, Trp-1, and Trp-2 were detected by Western blotting in B16F10 cells. In Figure 4e and Figure 4f, the results showed that the combination of EGCG and GLA significantly inhibited MITF expression and suppressed melanin production.

FIG. 4.

Expressions of melanin-related genes and proteins. (a) Relative MITF mRNA expressions in B16F10 cells. (b) Relative TYR mRNA expression in B16F10 cells. (c) Relative Trp-1 expressions in B16F10 cells. (d) Relative Trp-2 mRNA expressions in B16F10 cells. (e) Expressions of melanin-related proteins (MITF, TYR, Trp-1, and Trp-2). (f) Relative intensity of melanin-related protein expression. After the ANOVA/Tukey–Kramer multiple comparison test, all values are shown as means ± SEM (n = 10). **P < .01 vs control group, ^## P < .001 vs control group. MITF, microphthalmia-related transcription factor; Trp-1, DHICA oxidase; Trp-2, DOPA pigment isomerase; TYR, tyrosinase.

Targeting the relationship between miRNA and MITF

As shown in Figure 5a, the interaction sites between miR-182-5p, miR-96-5p, miR-183-5p, and MITF mRNA were hypothesized based on the bioinformatics analysis of TargetScan and miRanda. As shown in Figure 5b, the expression levels of miR-182-5p in each group had better matching degrees with the mRNA of MITI than the levels of miR-183-5p. This study further confirmed the interaction between miR-182-5p and MITF by dual luciferase reporter gene assay. As shown in Figure 5c, the relative fluorescence value of the miR-182-5p + MITF-WT co-transfection group was lower than that of the miR-NC + MITF-WT group (P > .05). The results verified that there was an interaction site between miR-182-5p and MITF mRNA. As shown in Figure 5d, the inhibitory effect of EGCG combined with GLA on MITF was significantly stronger than that of the two components alone.

FIG. 5.

The targeting relationship between miR-182-5p and MITF. (a) Alignment of miR-138-5p and SIRT1 3′-UTR by computational prediction via TargetScan and miRanda. (b) The levels of miR-182-5p in exosomes of B16F10 cells treated with different compounds. (c–d) Dual luciferase reporter assay (n = 10). The same letters indicate no significant difference among groups (P > .05), and different letters indicate significant differences among groups (P < .05). After the ANOVA/Tukey–Kramer multiple comparison test, all values are shown as means ± SEM (n = 10). **P < .01 vs control group. UTR, untranslated region.

Synergistic interaction of EGCG and GLA on MITF as observed by molecular docking and molecular dynamics simulations

To elucidate molecular interactions in protein-ligand complexes, we quantified binding free energies at equilibrium states using the gmx_mmpsa computational package (https://jerkwin.github.io/gmxtool/). The MMPBSA methodology divides thermodynamic binding parameters into four distinct contributions: electrostatic forces, van der Waals interactions, polar solvation energies, and non-polar solvation effects—the latter calculated based on solvent-accessible surface area approximations.

Our computational analysis revealed significant binding affinities for three molecular configurations: MITF-EGCG (−90.997 kJ/mol), MITF-GAL (−82.903 kJ/mol), and MITF-EGCG-GAL (−95.014 kJ/mol). Notably, the ternary EGCG-GAL complex demonstrated enhanced binding stability compared with individual ligand interactions, suggesting synergistic effects from EGCG incorporation. Energy decomposition analysis identified electrostatic and van der Waals forces as primary contributors to molecular recognition, consistent with typical protein–ligand interaction profiles.

These computational findings (presented in Fig. 6 and Table 6) provide quantitative evidence for EGCG’s role as a potential binding enhancer in MITF-GAL molecular complexes. The observed energy differentials of approximately 8–12 kJ/mol between configurations warrant further investigation into cooperative binding mechanisms.

FIG. 6.

The free energy landscapes (left) and 2D interactions (right) of EGCG/GLA and MITF system. (a) 3D constructive map for EGCG binding to MITF and specific binding sites. (b) 3D constructive map for GLA binding to MITF and specific binding sites. (c) 3D tectonic diagrams and specific binding sites for the binding of EGCG and GLA to MITF complexes.

Table 6.

Protein Ligands MMPBSA Analysis

Energy	MITF-EGCG	MITF-GAL	MITF-EGCG-GAL
Van der Waals energy (kJ/mol)	−142.722	−112.499	−131.165
Electrostatic energy (kJ/mol)	−74.168	−34.585	−31.722
Polar solvation energy (kJ/mol)	145.126	79.760	86.038
Nonpolar solvation energy (kJ/mol)	−19.233	−15.578	−18.166
Total binding energy (kJ/mol)	−90.997	−82.903	−95.014

The RMSD curves represented fluctuations in the conformation of the protein-target complexes, and the molecular motions of the trajectories were computationally simulated for 100 ns. The EGCG-GLA-MITF binding was always stable with no fluctuation in the RMSD curve as shown in Figure 7, and the EGCG-MITF complex fluctuated more after 40 ns, while the GLA-MITF complex only fluctuated more after 25 ns. This indicated that EGCG and GLA have a better synergistic effect on inhibiting MITF function.

FIG. 7.

Plot of changes in RSMD of protein-ligand complexes.

Preparation and characterization of DMN

To prepare the DMN that was able to effectively penetrate the stratum corneum of mouse skin, we chose the concentrations of 60% PV-A-205 and 45% PV-k29/32 for the preparation of DMN. The composite DMN (containing 100 μg/mL EGCG, 50 μg/mL GLA, and 100 μg/mL Ab) was prepared using a Franz diffusion cell to determine the drug release rate within 8 h to further verify whether the DMN had a good drug release rate. As shown in Figure 8a–c, the results indicated that DMN prepared with 60% PV-A-205 and 45% PV-k29/32 had the highest release rate of the three drugs over 8 h. In conclusion, the prepared DMN had the advantage of a high drug release rate and was able to be used as an effective drug delivery system for animal experiments.

FIG. 8.

(a–c) The drug release rates of EGCG, GLA, and Ab in the DMNs (60% PVA-205 and 45% PVP-K29/32) at different time points (0 h, 1 h, 2 h, 4 h, 6 h, 8 h). PVA-205, polyvinyl alcohol 205; PVP-K29/32, polyvinylpyrrolidone K29/32.

EGCG and GLA significantly reduced the excessive deposition of skin melanin and epidermal thickening

The back skin state of C57BL/6 female mice is shown in Figure 9a. The results showed that the skin of the model group and the blank micro-needle group had obvious hyperpigmentation, compared with the control group. At the same time, the four treatment groups (EGCG group, GLA group, Ab group, EGCG and GLA group) had slight melanin deposition, and the EGCG and GLA group had the lowest degree of pigmentation.

FIG. 9.

Skin melanin deposition and epidermal thickening. (a) Skin conditions on the backs of mice. (b) Masson–Fontana staining. (c) HE staining. (1)–(7) are the control group, model group, blank DMNs group, EGCG group, GLA group, Ab group, and EGCG and GLA group. (d) Skin epidermal thicknesses of mice in different groups (control group, model group, blank DMNs group, EGCG group, GLA group, Ab group, EGCG and GLA group). After the ANOVA/Tukey–Kramer multiple comparison test, all values are shown as means ± SEM (n = 10). **P < .01 vs model group. ^##P < .001 vs control group. HE, hematoxylin–eosin.

Masson-Fontana staining was performed on the back skin of the C57BL/6 mice. The staining results are shown in Figure 9b. These results showed that there was no melanin particle distribution in the skin of the control group, and there was a large amount of melanin particle deposition in the model group and the blank micro-needle group. There was a small quantity of melanin granules in the single-drug treatment groups (EGCG group, GLA group, Ab group), but there was no obvious melanin granule deposition in the EGCG and GLA group.

HE staining was also performed in C57BL/6 mice. The staining results showed that the epidermis of the model and blank micro-needle group was significantly thickened compared with the control group, whereas the epidermis of the treatment group was slightly thicker. The specific thickness of each group is shown in Figure 9c–d. All the results showed that the epidermal thickening was significantly reduced by UVB irradiation (P < .01) in the four treatment groups. The EGCG and GLA group had the best therapeutic effect on epidermal thickening.

The combination of EGCG and GLA significantly inhibited mice skin melanin production by regulating the MITF signaling pathway

We examined the expression of melanin-related genes and proteins MITF, TYR, TrP-1, and Trp-2 by qRT-PCR and Western blotting to verify the mechanism of melanogenesis inhibition in mouse skin. As shown in Figure 10, the expression levels of MITF, TYR, Trp-1, and Trp-2 were significantly increased in both the model and blank DMN groups compared with the control group. Furthermore, expression levels of MITF, TYR, Trp-1, and Trp-2 were significantly down-regulated in the four treatment groups (EGCG group, GLA group, Ab group, and EGCG and GLA group), with the lowest level in the EGCG and GLA group (EGCG: GLA = 50 g/mL:25 µg/mL). These results showed that EGCG and GLA significantly downregulated the transcription and expression levels of MITF, TYR, TrP-1, and Trp-2, and inhibited melanin production. These results were consistent with the above MD simulations as well as cellular experiments.

FIG. 10.

Expression of melanin-related genes and proteins. (a) Relative MITF mRNA expressions in the skins of C57BL/6 female mice. (b) Relative TYR mRNA expressions in the skins of C57BL/6 female mice. (c) Relative Trp-1 expression in the skin of C57BL/6 female mice. (d) Relative Trp-2 mRNA expressions in the skins of C57BL/6 female mice. (e) Expression of melanin-related proteins (MITF, TYR, Trp-1, and Trp-2). (f) Relative intensity of melanin-related protein expressions. After the ANOVA/Tukey–Kramer multiple comparison test, all values are shown as means ± SEM (n = 10). **P < .01 vs control group. ^## P < .001 vs control group.

DISCUSSION

Melanogenesis is influenced by several transcription factors and signaling pathways. However, MITF is one of the most important regulators of melanogenesis.^23
–25 EGCG and GLA are natural plant-derived components, and some studies have shown that they have potent antioxidant, anti-aging, and anti-dyslipidemia effects.¹⁰ In recent years, some studies have suggested that melanin deposition in skin may be inhibited by EGCG and GLA.²⁶ Zhang et al. reported that EGCG regulated melanogenesis in B16F10 cells through the cAMP/CREB/MITF pathway, which suggests that catechins could be used as anti-melanogenic agents to protect cells from abnormal melanogenesis.¹³ Pan et al. revealed that GLA can effectively inhibit the biological synthesis of melanin both in vitro and in vivo, through the suppression of both PKA/MITF and MAPK/MITF pathways, which suggests that GLA therapy is effective for anti-hyperpigmentation, anti-roughness, and beneficial for skin health.¹ However, the inhibitory mechanism of melanogenesis by EGCG and GLA has rarely been the subject of reports. Especially, the synergistic inhibition mechanism needs further investigation.

In recent years, CompuSyn software has been used to calculate the CI values to infer whether there were synergistic effects among different treatment groups.^16,17 In this study, we evaluated the synergistic inhibition effect on MITF by the CI values from a large number of natural compounds with GLA in B16F10 cells. Interestingly, we found that EGCG and GLA significantly and synergistically inhibited MITF. However, the mechanism of synergistic inhibition of MITF by EGCG and GLA is unclear. Thus, we investigated the synergistic mechanism of inhibition of MITF by molecular docking, MD simulation, and transcriptional regulation.

MD simulation is an indispensable classical technique in CADD. Drug molecules usually exert their pharmacological effects by recognizing and dynamically binding to biomolecular targets such as proteins. It is often difficult to effectively study such dynamic processes by experimental means, so the use of MD to study the dynamic behaviors of target-ligand complex structures is particularly important as a powerful simulation and computational approach. Therefore, we explored whether EGCG and GLA could interact with MITF by MD simulation, and the results showed that the binding within 100 ns in the EGCG-GLA-MITF system was not only stable but also better than that in the EGCG-MITF and GLA-MITF systems. Furthermore, the total binding energy of EGCG-GLA-MITF was the lowest among the above three systems. In conclusion, the combined use of EGCG and GLA can synergistically inhibit MITF, thereby inhibiting melanogenesis.

In this study, the miRDB database search was performed to screen three miRNAs (miR-182-5p, miR-96-5p, and miR-183-5P) for potential effects on MITF. Following this, software simulations were carried out, which indicated that miR-182-5p had the highest binding affinity with MITF. To further validate the simulation reliability, we verified it by the dual luciferase reported assay technique. Interestingly, the experimental results were in good agreement with the theoretical simulations. In summary, EGCG and GLA can regulate the expression of MITF through miR-182-5p, which in turn regulates the synthesis of melanin.

Many studies have shown that DMN can effectively penetrate the skin stratum corneum through a micrometer-sized needle tip and release the whitening ingredients by utilizing the degradable properties of bio-compatible materials.^26

–31 In this study, PV-A-205 and PV-K29/32 composite DMN with good mechanical strength and drug release rate were prepared, and the micro-needle trans-dermal drug delivery loaded with EGCG and GLA (EGCG: GLA = 50 μg/mL: 25 μg/mL) was selected for animal experiments.

CONCLUSIONS

In our study, we found that the melanin production and expression of MITF were significantly inhibited by a combination of EGCG and GLA in B16F10, and there was a synergistic interaction of EGCG and GLA with MITF. The expression of miR-182-5p was obviously modulated by EGCG and GLA in B16F10, and then the mRNA level of MITF was significantly regulated according to the dual-luciferase reported gene assay. The significant interaction of EGCG and GLA with MITF was simulated by MD software GROMACS. To further validate the inhibitory effect, skin experiments were also performed in C57BL/6 mice, which showed that the combination of EGCG and GLA was capable of synergistically inhibiting the expression of MITF and melanin over-deposition.

Footnotes

ACKNOWLEDGMENT

Financial assistance from the National Natural Science Foundation of China and the Major Science and Technology Projects in Zhejiang Province is sincerely appreciated.

AUTHORS’ CONTRIBUTIONS

J.S.: Drafted the article and was involved in the majority of experiments, analysis, and interpretation of data. C.Z.: Drafted the article and was involved in the majority of experiments, analysis, and interpretation of data. T.L.: Performed several experiments and proofread the diagrams in the article. X.W.: Performed several experiments and proofread the diagrams in the article. G.F.: Designed and guided the entire experimental process. G.J.: Designed and guided the entire experimental process.

DATA AVAILABILITY

The data that support the findings of this study are available from the corresponding author upon reasonable request.

ETHICS STATEMENT

The animal experiment protocol was conducted according to the guidelines of the Declaration of Helsinki and approved by the Laboratory Animal Ethics Committee of China Jiliang University (2023-055). All animal procedures were performed in accordance with the institutional and national guidelines.

AUTHOR DISCLOSURE STATEMENT

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this article.

FUNDING INFORMATION

This work was financially supported by the National Natural Science Foundation of China (31100499 and 31672394) and the Key Project of Science and Technology Plan of Zhejiang Provincial Market Supervision and Administration Bureau (ZD2025026).

References

Balasubramanian

, Eckert

. Keratinocyte proliferation, differentiation, and apoptosis—differential mechanisms of regulation by curcumin, EGCG and apigenin. Toxicol Appl Pharmacol, 2007; 224(3):214–219; doi: 10.1016/j.taap.2007.03.020

Messire

, Serreau

, Berteina-Raboin

. Antioxidant effects of catechins (EGCG), andrographolide, and curcuminoids compounds for skin protection, cosmetics, and dermatological. uses an update. Antioxidants, 2023; 12(7):1317–1335; doi: 10.3390/antiox12071317

Kim

, Hwang

, Lee

, et al. Skin protective effect of epigallocatechin gallate. Int J Mol Sci, 2018; 19(1):173–187; doi: 10.3390/ijms19010173

Wang

, Di

, Wang

, et al. EGCG, GCG, TFDG, or TSA inhibiting melanin synthesis by downregulating MC1R expression. Int J Mol Sci, 2023; 24(13):11017–11037; doi: 10.3390/ijms241311017

Zhang

, Li

, et al. Anti-melanogenic effects of epigallocatechin-3-gallate (EGCG), epicatechin-3-gallate (ECG) and gallocatechin-3-gallate (GCG) via down-regulation of cAMP/CREB/MITF signaling pathway in B16F10 melanoma cells. Fitoterapia, 2020; 145:104634–104652; doi: 10.1016/j.fitote.2020.104634

Pan

, Liu

, Zheng

, et al. The mechanisms of melanogenesis inhibition by glabridin: Molecular docking, PKA/MITF and MAPK/MITF pathways. Food Sci Hum Well, 2023; 12(1):212–222; doi: 10.1016/j.fshw.2022.07.011

Huang

, Gao

, Wang

, et al. Hydroxyurea regulates the development and survival of B16 melanoma cells by upregulating MiR-7013-3p. Int J Med Sci, 2021; 18(8):1877–1885; doi: 10.7150/IJMS.52177

Zhao

, Alam

, An

, et al. Antimelanogenic effect of an oroxylum indicum seed extract by suppression of MITF expression through activation of MAPK signaling protein. Int J Mol Sci, 2018; 19(3):760–778; doi: 10.3390/ijms19030760

, Rothenberg

. MicroRNA. J Allergy Clin Immunol, 2018; 141(4):1202–1207; doi: 10.1016/j.jaci.2017.08.034

10.

Meira

, Heinrich

, Cadena

SMSC

, et al. Melanogenesis inhibits respiration in B16-F10 melanoma cells whereas enhances mitochondrial cell content. Exp Cell Res, 2017; 350(1):62–72; doi: 10.1016/j.yexcr.2016.11.006

11.

Huang

, Li

, Hu

, et al. Mimicking melanosomes: Polydopamine nanoparticles as artificial microparasols. ACS Cent Sci, 2017; 3(6):564–569; doi: 10.1021/acscentsci.6b00230

12.

, He

, Chen

, et al. Effects of donkey milk on UVB-induced skin barrier damage and melanin pigmentation: A network pharmacology and experimental validation study. Front Nutr, 2023; 10:1121498–1121510; doi: 10.3389/fnut.2023.1121498

13.

Matsui

, Tanaka

, Higashiguchi

, et al. Protective and therapeutic effects of fucoxanthin against sunburn caused by UV irradiation. J Pharmacol Sci, 2016; 132(1):55–64; doi: 10.1016/j.jphs.2016.08.004

14.

Hespeler

, Kaltenbach

, Pyo

. Glabridin smartPearls–Silica selection, production, amorphous stability and enhanced solubility. Int J Pharm, 2019; 561:228–235; doi: 10.1016/j.ijpharm.2019.02.028

15.

, Yan

, Zhu

, et al. Biological effects of EGCG@ MOF Zn (BTC) 4 system improves wound healing in diabetes. Molecules, 2022; 27(17):5427–5446; doi: 10.3390/molecules27175427

16.

Duarte

, Falcão

, El Mehdi

, et al. Honeybee venom synergistically enhances the cytotoxic effect of CNS drugs in HT-29 colon and MCF-7 breast cancer cell lines. Pharmaceutics, 2022; 14(3):511–536; doi: 10.3390/pharmaceutics14030511

17.

, Xie

, Chen

, et al. Effects of quercetin on the efficacy of various chemotherapeutic drugs in cervical cancer cells. Drug Des Devel Ther, 2021; 15:577–588; doi: 10.2147/DDDT.S291865

18.

, Ge

, Zhu

, et al. Potential toxic mechanisms of neonicotinoid insecticides in rice: Inhibiting auxin-mediated signal transduction. Environ Sci Technol, 2023; 57(12):4852–4862; doi: 10.1021/acs.est.2c09352

19.

Kataria

, Khatkar

. Molecular docking of natural phenolic compounds for the screening of urease inhibitors. Curr Pharm Biotechnol, 2019; 20(5):410–421; doi: 10.2174/1389201020666190409110948

20.

Vora

, Vavia

, Larrañeta

, et al. Novel nanosuspension‐based dissolving microneedle arrays for transdermal delivery of a hydrophobic drug. J Interdiscip Nanomed, 2018; 3(2):89–101; doi: 10.1002/jin2.41

21.

Ramadon

, Sutrisna

LFP

, Harahap

, et al. Enhancing intradermal delivery of lidocaine by dissolving microneedles: Comparison between hyaluronic acid and poly (Vinyl Pyrrolidone) backbone polymers. Pharmaceutics, 2023; 15(1):289–304; doi: 10.3390/pharmaceutics15010289

22.

Khan

, Shah

SNH

, Arshad

, et al. Formulation and optimization of dimenhydrinate emulgels for topical delivery using response surface methodology. Pak J Pharm Sci, 2021; 34(1 Supplementary):245–255; doi: 10.3672/1/PJPS.2021.34.1.SUP.245-255.1

23.

Khames

. Hexyl alginate derivative, an amphiphilic innovative buccal film-forming material of promising mechanical and release characteristics for the improvement of repaglinide bioavailability. Drug Des Devel Ther, 2019; 13:925–940; doi: 10.2147/DDDT.S196425

24.

Ercelik

, Ozden

, Devrim

, et al. Investigation of nafion based composite membranes on the performance of DMFCs. Int J Hydrogen Energ, 2017; 42(4):2658–2668; doi: 10.1016/j.ijhydene.2016.06.215

25.

Liu-Smith

, Meyskens

. Molecular mechanisms of flavonoids in melanin synthesis and the potential for the prevention and treatment of melanoma. Mol Nutr Food Res, 2016; 60(6):1264–1274; doi: 10.1002/mnfr.201500822

26.

Krol

, Loedige

, Filipowicz

. The widespread regulation of microRNA biogenesis, function and decay. Nat Rev Genet, 2010; 11(9):597–610; doi: 10.1038/nrg2843

27.

, Yang

, Zhang

, et al. Dissolving microneedles loaded with etonogestrel microcrystal particles for intradermal sustained delivery. J Pharm Sci, 2018; 107(4):1037–1045; doi: 10.1016/j.xphs.2017.11.013

28.

Lee

, Fakhraei Lahiji

, Jang

, et al. Micro-pillar integrated dissolving microneedles for enhanced transdermal drug delivery. Pharmaceutics, 2019; 11(8):402–414; doi: 10.3390/pharmaceutics11080402

29.

Qiu

, Li

, Zhang

, et al. Systemic delivery of artemether by dissolving micro-needles. Int J Pharm, 2016; 508(1–2):1–9; doi: 10.1016/j.ijpharm.2016.05.006

30.

Rojekar

, Vora

, Tekko

, et al. Etravirine-loaded dissolving microneedle arrays for long-acting delivery. Eur J Pharm Biopharm, 2021; 165:41–51; doi: 10.1016/j.ejpb.2021.04.024

31.

Ren

, Yang

, Wang

, et al. Substrate-free dissolving micro-needles with barbed shape to increase adhesion and drug-delivery efficiency to skin. J Control Release, 2024; 368(24):199–207; doi: 10.1016/j.jconrel.2024.02.009