Rational design of an N-terminal cysteine-containing tetrapeptide that inhibits tyrosinase and evaluation of its mechanism of action

Abstract

There has been a resurgence of interest in bioactive peptides as therapeutic agents. This is particularly interesting for tyrosinase, which can be inhibited by thiol-containing peptides. This work demonstrates that an N-terminal cysteine-containing tetrapeptide can be rationally designed to inhibit tyrosinase activity in vitro and in cells. The tetrapeptide cysteine (C), arginine (R), asparagine (N) and leucine (L) or CRNL is a potent inhibitor of tyrosinase activity with an IC₅₀ value of 39.62 ± 6.21 μM, which is comparable to currently used tyrosinase inhibitors. Through structure-activity studies and computational modeling, we demonstrate the peptide interacts with the enzyme via electrostatic (R with E322), hydrogen bonding (N with N260) and hydrophobic (L with V248) intermolecular interactions and that a combination of these is required for potent activity. Moreover, copper chelating activity might be one of the mechanisms of tyrosinase inhibition by CRNL. Kinetic studies show that tetrapeptide is a competitive inhibitor with two-step irreversible inhibition. In addition, CRNL had no toxicity and could reduce melanin levels in the murine melanoma cell line (B16F1). Overall, CRNL is a very promising candidate for hyperpigmentation treatment.

Graphical abstract

Highlights

• •All designed N-terminal cysteine-containing tetrapeptides exhibited potent tyrosinase inhibitory activity.
• •Electrostatic, hydrogen bonding, and hydrophobic interactions were essential for CRNL tetrapeptide's tyrosinase inhibition.
• •Kinetic studies showed that the CRNL tetrapeptide was a competitive inhibitor of tyrosinase.
• •The CRNL tetrapeptide was not toxic and could reduce melanin levels in B16F1 cells.

1. Introduction

Melanin is a pigment produced in plants, microorganisms, and animals (Alghamdi et al., 2023). In skin, its role is to protect the skin from ultraviolet (UV) radiation damage (Shen et al., 2019; Nazir et al., 2022). However, the overproduction and abnormal accumulation of melanin can result in dark skin, melasma, freckles, lentigo and Riehl melanosis (Nazir et al., 2022). Melanin synthesis is also associated with enzymic browning, which turns food brown and can affect food quality (Moon et al., 2020). Hence, the food and cosmetic industries have tried to control melanin synthesis to prevent food discoloration and skin pigmentation (Shin et al., 2023).

Tyrosinase is a key rate-limiting enzyme in melanogenesis (Joompang et al., 2022; Nazir et al., 2022), the process that creates melanin. As a consequence, inhibitors of this enzyme are key for hyperpigmentation treatment (Lee et al., 2015; Joompang et al., 2020, 2022; Nazir et al., 2022). Known tyrosinase inhibitors such as kojic acid and α-arbutin are widely used in cosmetics, however they can cause skin allergies, inflammation, and other adverse side effects (Curto et al., 1999; DeCaprio, 1999; Hermanns et al., 2000; Parvez et al., 2006; Lee et al., 2015). For example, kojic acid has shown tumorigenic potential in the liver of CBA mice (Takizawa et al., 2003), while arbutin can be metabolized to yield hydroquinone which is potentially carcinogenic (Blaut et al., 2006). Moreover, genetic variation in the human population means the potency of these inhibitors varies from individual to individual (Bachtiar and Lee, 2013). Therefore, the identification of new tyrosinase inhibitors is crucial to improve hyperpigmentation therapy.

Peptides have attracted interest due to their potency and more importantly their minimal adverse side effects (Ubeid et al., 2009, 2012). As a result, tyrosinase inhibitor peptides from both natural peptides (Ateş et al., 2001; Kubglomsong et al., 2018) and artificial peptide mimetics (Ubeid et al., 2009; Hsiao et al., 2014; Ochiai et al., 2016) have been reported and characterized mechanistically. They bind to the catalytic copper active site and interact with nearby residues (Ubeid et al., 2012; Lee et al., 2015; Ochiai et al., 2016). While some information is known on how the amino acid sequence and composition affects tyrosinase inhibition, the studies are not comprehensive and more information is vital to improve peptide inhibitors of tyrosinase.

Herein, we rationally design a tyrosinase inhibiting tetrapeptide (CRNL) that binds to the tyrosinase active site. The potency and mechanism of action are evaluated by biochemical, computational and cellular studies. Our study shows that the peptide is a strong inhibitor of tyrosinase through electrostatic and hydrophobic interactions, and has comparable activity to kojic acid. Most importantly, the peptide is non-toxic in cells and can reduce melanin levels suggesting it could be used in hyperpigmentation treatments. Our studies also provide a rational approach to designing other peptides that strongly inhibit enzymes.

2. Materials and methods

2.1. Materials and reagents

3,4-Dihydroxy-L-phenylalanine, dimethyl sulfoxide (DMSO), mushroom tyrosinase, and pyrocatechol violet were purchased from Sigma-Aldrich (St. Louis, MO, USA). Kojic acid was purchased from TCI (Tokyo, Japan). 3-[4,5-Dimethylthiazol-2-yl]-2,5-diphenyltetrazolium bromide (MTT) was purchased from Sigma-Aldrich (Eugene, OR, USA). Dulbecco's modified Eagle's medium (DMEM), Fetal bovine serum (FBS), penicillin/streptomycin and L-glutamine were purchased from Lonza (Walkersville, MD, USA). Trypsin-EDTA was purchased from Corning Inc. (Manassas, VA, USA).

2.2. Tetrapeptide design

Our designed peptide is expected to interact with amino acids/catalytic copper atoms in the active site of tyrosinase, which has four key regions in close proximity (Fig. 1): R1 (a dinuclear copper complex that interacts with six histidine residues); R2 (contains negatively charged residues); R3 (contains polar residues) and R4 (contains hydrophobic residues). As a consequence, we decided to design a peptide with four amino acids. We refer to each position of the tetrapeptide from N-terminus to C-terminus as P1, P2, P3 and P4 and assigned amino acids to these positions that form complementary interactions with the four positions of the enzyme active site – cysteine (C, P1, metal co-ordination interactions with R1), arginine (R, P2, electrostatic interactions with R2), asparagine (N, P3, polar interactions with R3) and leucine (L, P4, hydrophobic interactions with R4). To study structure-activity relationships, amino acids at positions P2, P3 and P4 were replaced with glutamic acid (E), leucine (L) or asparagine (N) respectively. Various combinations of these substitutions were tested (7 peptides in total, Fig. 1A, Table 1). All peptides were synthesized by GL Biochem Ltd. (Shanghai, China) using Fmoc solid-phase synthesis.

Fig. 1.

Tyrosinase active site and peptide design. (A) Schematic of rationally designed tetrapeptides with a single, double and triple position modifications. (B) Mushroom tyrosinase (PDB: 2Y9X) active site with catalytic amino acids shown. In this research the tyrosinase actvity site is divided into four regions: R1, R2, R3 and R4. The figure was constructed from Discovery Studio 2017 R2 Client.

Table 1.

The ability of CRNL peptide and its variants to inhibit tyrosinase processing of L-DOPA.

Peptide sequence	Relative tyrosinase activity (%)						IC50 (μM)
	0 μM	12.50 μM	25 μM	50 μM	100 μM	200 μM
CRNL (504.6110 Da)	100.00 ± 4.86a	78.38 ± 4.32b	71.89 ± 5.62b	46.85 ± 8.10c	6.74 ± 3.70d	3.69 ± 1.99d	39.62 ± 6.21b
CENL	100.00 ± 0.77a	90.49 ± 3.37ab	84.15 ± 7.35b	81.23 ± 6.68b	65.81 ± 7 .20c	37.96 ± 6.10d	150.67 ± 7.37d
CRLL	100.00 ± 0.75a	85.71 ± 2.01b	81.95 ± 0.25c	84.21 ± 4.01bc	70.43 ± 0.75d	31.91 ± 1.26e	143.50 ± 4.82d
CRNN	100.00 ± 2.62a	85.33 ± 4.71b	84.69 ± 2.48b	72.89 ± 3.50c	51.60 ± 0.87d	12.90 ± 0.64e	87.23 ± 1.64c
CELL	100.00 ± 3.21a	83.60 ± 3.22b	80.39 ± 4.01b	76.74 ± 2.94b	59.54 ± 6.33c	40.64 ± 1.93d	148.60 ± 16.13d
CRLN	100.00 ± 1.39a	88.64 ± 0.55b	86.70 ± 2.49b	86.15 ± 0.83b	79.50 ± 0.83c	49.31 ± 1.11d	234.27 ± 8.98e
CENN	100.00 ± 0.65a	86.36 ± 2.60b	80.84 ± 0.97b	79.55 ± 1.62b	63.30 ± 2.20d	37.03 ± 3.03e	144.20 ± 8.02d
CELN	100.00 ± 0.26a	90.32 ± 5.91b	77.63 ± 4.63cd	81.49 ± 0.26c	72.75 ± 0.77d	47.81 ± 0.51e	232.70 ± 4.95e
Kojic acid (KA)	100.00 ± 5.45a	54.74 ± 1.32b	49.73 ± 3.27c	38.77 ± 0.88d	25.85 ± 1.30e	13.80 ± 1.62f	20.80 ± 2.10a

Values are the mean ± the standard deviation. The different superscript letters are used to indicate the significant difference at p < 0.05.

2.3. Tyrosinase inhibition assay and inhibition mode

The inhibition assay was performed as described by Joompang et al. (2020). Briefly, L-DOPA (3.60 mM, 50 μL in 100 mM phosphate buffer pH 6.8) was mixed with the desired peptide (0 – 400 μM, 47.50 μL) at room temperature before the addition of tyrosinase (400 U/mL, 2.50 μL) to initiate the enzymatic reaction. The absorbance at 475 nm was monitored every 30 s using a SpectraMax M5 plate reader (Molecular Devices, Sunnyvale, CA, USA). The initial rate (slope) of tyrosinase activity was obtained by plotting the absorbance at 475 nm vs time. Kojic acid was used as a positive control. The tyrosinase inhibitory activity and tyrosinase activity were calculated using the following equation:

$$\text{Tyrosinase inhibitory activity}\left(\%\right)=\left(\frac{\Delta \mathrm{A}{475}_{\text{control}}-\Delta \mathrm{A}{475}_{\text{test}}}{\Delta \mathrm{A}{475}_{\text{control}}}\right)\times 100$$ (1)

$$\text{Tyrosinase activity}\left(\%\right)=\left(\frac{\Delta \mathrm{A}{475}_{\text{test}}}{\Delta \mathrm{A}{475}_{\text{control}}}\right)\times 100$$ (2)

where ΔA475_control and ΔA475_test represent the initial rate (slope) of the reaction without the inhibitor and the reaction with the inhibitor, respectively. GraphPad Prism7 (GraphPad Software Inc., California, USA) was used to calculate IC₅₀ values using a non-linear regression model.

To determine the mode of inhibition, the tyrosinase activity assay was performed with the different concentrations of L-DOPA (0 – 0.80 mM) and peptide (0 – 50 μM). The changes of K_m and V_max from the Lineweaver-Burk plot were used to infer the type of inhibition.

The reversible or irreversible inhibition of the peptide was examined from plotting tyrosinase concentration (0 – 10 μg/mL) and tyrosinase activity using different concentrations of the peptides (0 – 50 μM).

2.4. Kinetics of irreversible inhibition

The kinetics of irreversible inhibition was evaluated by pre-incubation of peptide (0 – 50 μM, 47.50 μL) with tyrosinase (400 U/mL, 2.50 μL) for 0, 5, 10, 15, 20, 25, 30, 45 and 60 min at room temperature. Then L-DOPA (3.6 mM, 50 μL in 100 mM phosphate buffer, pH 6.8) was added to initiate the reaction. Tyrosinase activity was detected according to section 2.3. The plot of Ln (remaining tyrosinase activity (%)) vs peptide concentration was used to determine the observed inactivation rate (k_obs). The reciprocal plot of 1/k_obs vs 1/peptide concentration was used to estimate K_I and K_inact.

2.5. Molecular docking

Molecular docking was performed using GOLD Suite 5.5 (Cambridge Crystallographic Data Center, Cambridge, UK) (Jones et al., 1997). Three-dimensional (3D) structures of peptides were constructed using Discovery Studio 2017 R2 Client (Dassault Systèmes BIOVIA, BIOVIA Workbook, Release, 2017; BIOVIA Pipeline Pilot, Release, 2017; San Diego: Dassault Systèmes). The crystal structure of mushroom tyrosinase (PDB: 2Y9X) was selected as the target receptor, and water was removed from the structure before docking. The automated molecular docking was performed with the addition of hydrogen atoms. Then, the tropolone molecule was removed, and the binding site of peptides was set as the tropolone binding site with the radius of 10 Å. Goldscore scoring function was used with 1000 GA running. The search efficiency was flexible and set at 200%. The most feasible complex with the highest gold score was selected to represent the interaction between the peptide and tyrosinase. Discovery Studio 2017 R2 Client was used to detect the modes of interactions.

2.6. Peptide copper chelation assay

Peptide copper chelation activity was determined using a modification of the protocol reported by Fu et al. (2014). Briefly, phosphate buffer (500 mM, 10 μL, pH 6.8) was mixed with peptides (0 – 200 μM, 50 μL) and CuSO₄ (aq., 0.69 mM, 20 μL). Subsequently, pyrocatechol violet (PV, 0.69 mM, 20 μL) was added to the mixture and incubated for 20 min. Then the absorbance at 632 nm was recorded. The reaction without the peptide and CuSO₄ was performed as a “blank”. The copper chelating activity was calculated using the equation below:

$$\text{Relative}\text{Copper}(\%)=\left(\frac{{\mathrm{A}632}_{\text{test}}-{\mathrm{A}632}_{\text{blank}}}{{\mathrm{A}632}_{\text{control}}-{\mathrm{A}632}_{\text{blank}}}\right)\times 100$$ (3)

where A632_test and A632_control represent the absorbance of the reaction with and without the peptide respectively. A632_blank is the absorbance of the reaction without the peptide and CuSO₄. GraphPad Prism7 (GraphPad Software Inc., California, USA) was used to calculate IC₅₀ values using a non-linear regression model.

2.7. Cell viability assay

Cell viability was determined by a modification of protocol reported by Phosri et al. (2014). B16F1 cells (1.5 x 10⁵ cells/well) were seeded and grown overnight on a 12-well plate, followed by the addition of different concentrations of the CRNL peptide (0 – 200 μM) and incubation at 37 °C, 5% CO₂ with 95% relative humidity for 48 h. Next, the culture medium was removed and MTT (0.5 mg/mL) was added to the cells, followed by the incubation at 37 °C for 2 h. The absorbance was then detected at 570 nm using a microplate reader. The percentage of cell viability was calculated using the equation below:

$$\text{Cell}\text{Viability}(\%)=\left(\frac{{\mathrm{A}570}_{\text{test}}}{{\mathrm{A}570}_{\text{control}}}\right)\times 100$$ (4)

where A570_control is the absorbance of untreated cells and A570_test is the absorbance of cells treated with the peptide.

2.8. Total melanin content investigation

B16F1 cells (1.5 x 10⁵ cells/well) were seeded and grown overnight on a 12-well plate, followed by the addition of different concentrations of CRNL (0 – 200 μM) at 37 °C, 5% CO₂ with 90% relative humidity. After 48 h, the culture medium was removed, and the cells were washed twice with PBS buffer (pH 7.4). Then a DMSO/NaOH (1 M) mixture (1:9, 200 μL) was added and the reaction was incubated at 80 °C for 1 h. The absorbance at 490 nm of 100 μL of supernatant was detected using a microplate reader. The relative melanin content was calculated using the equation below:

$$\text{Melanin}\text{Content}(\%)=\left(\frac{{\mathrm{M}}_{\text{test}}}{{\mathrm{M}}_{\text{control}}}\right)\times 100$$ (5)

where M_test is the absorbance of the cells treated with the peptide and M_control is the absorbance of the untreated cells.

2.9. Statistical analysis

All experiments were conducted in triplicate (n = 3). The data are presented as the mean ± the standard deviation. The ANOVA test according to Duncan (p < 0.05) was carried out for comparison.

3. Results and discussion

3.1. Peptide inhibition of tyrosinase activity

The tyrosinase active site can be divided into four regions: R1 (a dinuclear copper complex that interacts with six histidine residues; copper atom CuA interacts with H85, H94 and H61; copper atom CuB interacts with H259, H263 and H296); R2 (negatively charged residues such as E256 and E322); R3 (N260 and R268) and R4 (F264, V283, P284, V248, H244 and A246) (Fig. 1B). To target the active site, a tetrapeptide was rationally designed. Previously, cysteine-containing peptides were reported to serve as potent tyrosinase inhibitors (Lee et al., 2015; Joompang et al., 2020). As a consequence, at position 1 (P1), a cysteine (C) was used in order to favor coordinate bond formation between the sulfhydryl group and the catalytic copper atom in R1. At position 2 (P2), arginine (R) was selected to electrostatically interact with the guanidinium and E322 in R2. At position 3 (P3), the very polar asparagine (N) amino acid was selected in order to interact with the polar amino acids of the R3 site. At position 4 (P4), leucine (L) was chosen to facilitate hydrophobic interactions with R4 (Fig. 1A and B). The selection of leucine ensures the rationally designed peptide contains one of the two amino acids that have been reported to give good tyrosinase inhibitory peptides (the other being arginine) (Schurink et al., 2007). To validate the rationale behind the design, the amino acids at the position 2, 3 and 4 were substituted by those with opposite properties (E, L and N respectively), while the C terminal amino acid at P1 was kept constant. Different combinations of these amino acid modifications yield seven versions of the CRNL parent peptide (Fig. 1A and Table 1).

To evaluate their potency as an inhibitor, L-DOPA and tyrosinase were incubated with varying concentrations of the peptides (0 – 200 μM) and the IC₅₀ value was determined with kojic acid as a positive control (IC₅₀ = 20.80 ± 2.10 μM). Promisingly, the rationally designed peptide CRNL gave an IC₅₀ value of 39.62 ± 6.21 μM, which is close to that of the control (Table 1). On the other hand, the other peptide variants had higher IC₅₀ values ranging from 143.50 to 232.70 μM. CRNN was notably better with an IC₅₀ value of 87.23 μM. Overall, these results demonstrate our rational design approach was reasonable.

3.2. Molecular docking

To further support our structure-activity inhibition assay, molecular docking simulations were performed with the CRNL and its variants. All peptides were able to interact with the active site of tyrosinase (Fig. 2A–H and Tables S1–S8). In all cases, the sulfhydryl group of cysteine was predicted to be orientated towards the catalytic copper atoms enabling coordination bonds between the sulfur and copper atoms. This is consistent with previous literature reports (Hsiao et al., 2014; Lee et al., 2015). CRNL formed five hydrogen bonds, one electrostatic interaction (R at P2–E322), and two hydrophobic interactions (R at P2–H244 and L at P4–V248) with the active site. These interactions are consistent with our hypothesis.

Fig. 2.

Tyrosinase complexed with (A) CRNL, (B) CENL, (C) CRLL, (D) CRNN, (E) CELL, (F) CRLN, (G) CENN and (H) CELN. The conventional hydrogen bond and carbon hydrogen bond are represented by the green and cyan dash lines, respectively. The electrostatic interaction, Pi-cation, and Pi-hydrogen bond donor are represented by red, brown, and orange dash lines, respectively. The hydrophobic interactions of alkyl-alkyl and Pi-alkyl are represented by the blue and violet dash lines, respectively. The figure was constructed from Discovery Studio 2017 R2 Client.

From our tyrosinase inhibition assays, all CRNL variants (except CRNN) had similar IC₅₀ values in the range of 143.50 to 232.70 μM. The corresponding molecular docking studies demonstrate that these variants either disrupted electrostatic interactions (CENL and CRLL; IC₅₀ = 150.67 and 143.50 μM) or had more drastic effects on peptide–enzyme interactions (CRLN and CELN; IC₅₀ = 234.27 and 232.70 μM). The CRNL ⟶ CENL variant caused the loss of the electrostatic interaction between the peptide and E322; the CRNL ⟶ CRLL variant disrupted electrostatic interactions (R at P2–E322) and two hydrogen bonds (N at P3 with N260), but did promote hydrophobic interactions. This change does appear to decrease the inhibition of tyrosinase. Finally, CRNL ⟶ CRLN and CRNL ⟶ CELN modifications dramatically altered tyrosinase interactions losing both electrostatic interactions between R at P2 and E322 and hydrophobic interactions between L at P4–V248, which naturally results in the highest IC₅₀ and least potency. This finding suggests the disrupting such both electrostatic and hydrophobic interactions is severely detrimental to tyrosinase inhibition.

Interestingly, for the CRNL ⟶ CRNN variant, docking confirmed that hydrophobic interaction (L at P4–V248) was disrupted. However, given the potency of the CRNN variant was still good (IC₅₀ = 87.23 ± 1.64 μM cf. CRNL IC₅₀ of 39.62 ± 6.21 μM), it appears this hydrophobic interaction is less important than electrostatic interactions.

In combination, these in silico molecular docking studies corroborated our hypothesis and experimental results. They provide interesting extra mechanistic information on how peptides interact with the active site of tyrosinase.

3.3. Peptide copper chelation activity

Compounds with copper chelating functional groups, especially thiol group (-SH) are known to decrease the tyrosinase activity by sequestering the copper atoms of tyrosinase (Pillaiyar et al., 2017; Kubglomsong et al., 2018). The tetrapeptides tested herein all contain an N-terminal cysteine, which is an excellent copper chelator. To evaluate their respective affinity for copper, the peptides were mixed with copper before the addition of pyrocatechol violet. Stronger binders prevent pyrocatechol violet from interacting with copper and therefore changing the absorbance of the dye. The difference in signal from a control (no peptide) and a test sample (with a peptide) gave an indicator of copper binding of the peptide. These results showed that none of the peptides strongly bind to copper unless the concentration of the peptide is higher than 100 μM. At 200 μM, CRNN, CENN and CRNL showed some affinity for copper (41.12, 53.72 and 69.63%; lower values are better chelators; Table 2). However, this does not correlate with their tyrosinase inhibitory activity (IC₅₀ = 87.23, 144.20 and 39.62 μM respectively, Table 1). While this suggests copper chelation is not the primary mechanism by which these peptides inhibit tyrosinase, it must be noted that this assay is binding of the peptide to free copper. When the peptide binds to copper in tyrosinase, other intermolecular interactions may favor copper interactions (see section 3.5 and irreversible inhibition studies). Considering the relation of copper chelating activity and tyrosinase inhibitory activity, it was found that the inhibitory activity of most peptides tends to intensify regarding the increase of the copper chelating activity. However, such relation was not applied to some peptides. These might be due to the different peptide structures and sequences, resulting in the different modes of inhibition towards tyrosinase, consequently affecting the inhibitory activity.

Table 2.

Free copper chelating activity of CRNL and its variants using a pyrocatechol violet assay.

Peptide	Relative copper (%)						IC50 (μM)
	0 μM	12.5 0 μM	25 μM	50 μM	100 μM	200 μM
CRNL	100.00 ± 8.85a	107.70 ± 1.25ab	102.13 ± 6.92ab	103.89 ± 2.54ab	95.01 ± 2.11b	69.63 ± 3.00c	>200
CENL	100.00 ± 8.85a	102.13 ± 6.74a	100.95 ± 8.05a	102.13 ± 7.57a	95.16 ± 4.06a	73.15 ± 1.43b	>200
CRLL	100.00 ± 10.46a	95.08 ± 5.99ab	93.10 ± 4.47abc	91.24 ± 2.38abc	87.82 ± 2.56bc	83.31 ± 0.45d	>200
CRNN	100.00 ± 10.46a	96.28 ± 3.61a	92.92 ± 2.56a	91.18 ± 1.00a	77.61 ± 4.79b	41.12 ± 2.18c	171.97 ± 7.70
CELL	100.00 ± 8.85a	106.60 ± 3.21a	106.68 ± 1.47a	104.11 ± 3.42a	99.34 ± 4.33a	84.15 ± 0.46b	>200
CRLN	100.00 ± 8.85a	109.17 ± 4.08ab	103.23 ± 3.24ab	101.69 ± 3.24ab	97.29 ± 5.86b	87.45 ± 3.00c	>200
CENN	100.00 ± 10.46a	94.42 ± 1.88ab	94.96 ± 2.98ab	90.28 ± 2.52b	79.11 ± 0.95c	53.72 ± 2.55d	>200
CELN	100.00 ± 10.46a	97.42 ± 4.15ab	93.16 ± 3.72ab	93.94 ± 0.28ab	88.90 ± 2.45c	72.81 ± 3.26d	>200

Values are the mean ± the standard deviation. The different superscript letters are used to indicate the significant difference at p < 0.05.

3.4. Type of inhibition

In order to determine the type of inhibition that CRNL causes Lineweaver-Burk plots were made to determine K_m and V_max values. The result indicated that the increase in the concentration of peptides caused an increase of K_m, while V_max did not change. This suggests that CRNL is a competitive inhibitor with K_i of 11.49 ± 1.65 μM (Fig. 3A and B and Table 3). This is consistent with previous reports that a cysteine-containing peptide inhibits tyrosinase a competitive inhibitor (Hsiao et al., 2014; Lee et al., 2015; Tseng et al., 2015).

Fig. 3.

Kinetic studies. (A) Lineweaver-Burk plot of tyrosinase kinetic when L-DOPA was used as a substrate and in the presence of the CRNL peptide from 0 to 50 μM. (B) Secondary plot of Lineweaver-Burk plot in (A). (C) An irreversible inhibition of tyrosinase by the CRNL peptide. (D) The plot of Ln (% remaining tyrosinase activity) vs pre-incubation time at different CRNL concentrations. (E) The plot of k_obs (min^-1) vs CRNL concentration. (F) Kitz–Wilson re-plot. CRNL at concentration of 0 – 50 μM was used.

Table 3.

Kinetic constants of tyrosinase.

Peptide name	Concentration (μM)	Km or K'm (μM)	Vmax or V'max (△A475/min)	Mode of inhibition
CRNL	0	256.71 ± 21.18a	0.049 ± 0.0022a	competitive
	12.50	495.35 ± 30.97b	0.047 ± 0.0013a
	20.00	715.67 ± 53.43c	0.049 ± 0.0038a
	50.00	1279.65 ± 143.95d	0.048 ± 0.0040a

Values are the mean ± the standard deviation. The different superscript letters are used to indicate the significant difference at p < 0.05.

The thiol group of cysteine could coordinate to the catalytic copper atoms resulting in the irreversible inhibition of tyrosinase. While CRNL peptide does not seem to bind free copper, it may do so once the other intermolecular interactions it forms with the enzyme orientate it for interaction with the copper atoms. As a result, this was evaluated by looking at the linear relationship between each concentration of tyrosinase and its activity in the presence and absence of CRNL (Fig. 3C). The parallel straight lines denoted the occurrence of the irreversible inactivation of these peptides. Based on the irreversible inhibition, the kinetics of peptides then studied for understanding the inactivation mechanism. The plot of natural logarithm (Ln) of the remaining tyrosinase activity vs the pre-incubation times was shown in Fig. 3D. This plot was used to estimate the observed rate constant (k_obs). Increasing peptide concentration was found to increase the observed rate constant as follow hyperbolic characteristic (k_obs, Fig. 3E), suggesting a two-step irreversible inhibition mechanism, which is consistent with previously reported cysteine-containing peptide, TILI-2 (Joompang et al., 2020). After CRNL reversibly binds to tyrosinase (E•I), tyrosinase is inactivated through coordination bond formation (E-I). The following equation was used to express the two-step inactivation:

k_obs = k_inact[I]/(K_I+I)

where k_inact is the maximum rate of the enzyme inactivation (min^-1), and K_I is an inhibition constant (μM) of the reversible binding. In order to obtain the estimated k_inact and K_I, Kitz–Wilson re-plot (1/k_obs = K_I/k_inact[I] + 1/kinact) was performed (Fig. 3F). The results indicated that CRNL exhibited K_I and k_inact of 7.87 ± 0.28 μM and 0.027 ± 0.0007 min^-1, respectively. In addition, CRNL was observed to have the constant rate of covalent bond formation (k_inact/K_I) of 0.0035 ± 0.00004 μM^-1 min^-1.

3.5. Cell viability and melanin content

The cytotoxicity and melanin inhibition of CRNL (50 – 200 μM) was investigated in B16F1 cells. Promisingly, no toxicity was observed at the tested concentration (Fig. 4A), while melanin decreased in a dose-dependent manner. At the highest concentration used (200 μM), the melanin content was reduced by 32.13 ± 1.25% (n = 6, Fig. 4B). These results indicated that CRNL could decrease melanin without any toxicity towards B16F1, suggesting the possibility it could be used in hyperpigmentation treatments.

Fig. 4.

The effect of tyrosinase inhibitors on (A) cell viability and (B) melanin content in B16F1 cells. Values are the mean ± the standard deviation. The different superscript letters are used to indicate the significant difference at p < 0.05.

4. Conclusions

A potent tyrosinase inhibitor peptide, CRNL, was developed. Evaluation of structural analogues along with molecular docking suggested that the electrostatic interactions between the peptide's P2 R and the tyrosinase's E322 amino acids, and hydrophobic interactions between the peptide's P4 L and the tyrosinase's V248 amino acids were important for CRNL's strong inhibition of tyrosinase. In addition, all modified tetrapeptides were copper chelators with CRNN being the most potent chelator. Interestingly, there was a good correlation between peptide's ability to chelate copper and its ability to inhibit tyrosinase. Kinetic studies indicated that the CRNL peptide was a competitive inhibitor of tyrosinase through a two-step inactivation mechanism. Moreover, CRNL was not cytotoxic and could also decrease melanin levels in B16F1 cells. Collectively, these results suggest that CRNL is a tyrosinase inhibitor that could be used for hyperpigmentation treatment or as a copper chelator.

CRediT authorship contribution statement

Anupong Joompang: Conceptualization, Investigation, Data curation, Methodology, Formal analysis, Writing – original draft. Preeyanan Anwised: Investigation, Formal analysis. Sompong Klaynongsruang: Conceptualization, Resources, Methodology, Supervision. Lapatrada Taemaitree: Data curation, Visualization, Writing – review & editing. Anuwat Wanthong: Data curation, Visualization. Kiattawee Choowongkomon: Methodology, Supervision. Sakda Daduang: Resources, Methodology, Supervision. Somporn Katekaew: Supervision, Visualization, Writing – review & editing. Nisachon Jangpromma: Conceptualization, Resources, Methodology, Supervision, Data curation, Visualization, Writing – review & editing.

Declaration of competing interest

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 paper.

Handling Editor: Professor Aiqian Ye

Abbreviations

R1

the tyrosinase active site region 1

R2

the tyrosinase active site region 2

R3

the tyrosinase active site region 3

R4

the tyrosinase active site region 4

P1

position 1 from the N-terminus to the C-terminus of the designed tetrapeptide

P2

position 2 from the N-terminus to the C-terminus of the designed tetrapeptide

P3

position 3 from the N-terminus to the C-terminus of the designed tetrapeptide

P4

position 4 from the N-terminus to the C-terminus of the designed tetrapeptide

C

cysteine

R

arginine

N

asparagine

L

leucine

E

glutamic acid

V

valine

F

phenylalanine

P

proline

H

histidine

A

alanine

Appendix A. Supplementary data

The following is the Supplementary data to this article:

Data availability

The data that has been used is confidential.