Tyrosinase Inhibitory Activity of Extracts from Prunus persica

Abstract

The demand for skin-whitening agents is high across the world, including Asian countries. An extensive screening using a tyrosinase inhibition assay was performed in order to discover novel plant materials. In our research program investigating a safe and effective agent, 50% ethanolic extracts prepared from discarded parts of Prunus persica were screened for in vitro tyrosinase inhibitory activity. Among the extracts tested, twig extract showed the most potent inhibitory activity: 38% inhibition at 500 µg/mL. The investigation of active compounds in twig extract found four flavanones that acted as moderate inhibitors, including (−)-prunin, persiconin, (+)-dihydrokaempferol, and (−)-naringenin. These compounds were only observed in the twig extract following preliminary quantification by HPLC, with the following concentrations: (−)-prunin, 1.8 mg/g sample; persiconin, 0.8 mg/g sample; (+)-dihydrokaempferol, 0.8 mg/g sample; (−)-naringenin, 1.7 mg/g sample. These results suggest that twig extracts can be more useful for skin-whitening compared with other parts of the plant. In addition, a new constituent of twig extract was identified, namely isoquercitrin, which suggests that twig extract can be a potent source of flavones and flavanones. Further studies on the identification of novel compounds from twig extract are now underway in our laboratory.

1. Introduction

Demand for skin-whitening is relatively high in Asian countries, especially in China, Korea, and Japan [1]. Particularly in Japan, safe and potent skin-whitening agents are highly desired. Many cosmetic suppliers prefer natural sources of skin-whitening agents. In our research program, we focused on plants as sources of novel agents.

The fruit of Prunus persica is known as the “peach”, which is a popular fruit consumed all over the world. In ancient China, peaches were consumed as a remedy for curing a variety of symptoms. However, the fruits were very small and hard at that time. Peaches underwent repeated selective breeding that made the fruits bigger and juicier. Although peaches were recognized as a remedy in ancient times, the medicinal or useful functions of peaches in modern times are uncertain.

We have previously screened the extracts from Prunus plants and non-fruit parts of P. persica for anti-tyrosinase activity. The flowers of P. persica showed potent tyrosinase inhibitory activity, and afzelin and naringenin were identified as active compounds for skin-whitening, possessing a suppressive activity in melanin biosynthesis [2]. Thus, P. persica was recognized as a functional material for skin-whitening. This result prompted us to screen various parts of the plant for skin-whitening agents using an in vitro anti-tyrosinase assay. This comprehensive screening has not yet been reported, and the results could provide information on the usefulness of the modern peach as a supplemental and/or functional food material.

In the cultivation of peaches, it is necessary to trim twigs, leaves, and immature fruits annually. The resultant waste products are disposed of; therefore, we focused on the discarded materials as resources. In this report, a screening for the anti-tyrosinase activity of extracts from unwanted parts of the plant, along with the fruits of P. persica, was performed. In addition, active compounds of tyrosinase inhibitory activity were isolated from the extract and identified. Moreover, the concentrations of active compounds in the extracts were quantified. During purification, an unexpected constituent was detected, and the chemical structure of the compound was identified.

2. Materials and Methods

2.1. Materials

Mushroom tyrosinase was purchased from Sigma-Aldrich, Inc. (St. Louis, MO, USA). All the other reagents used in this study were analytical grade and purchased from FUJIFILM Wako Pure Chemical Corporation (Osaka, Japan) or NACALAI TESQUE, INC. (Kyoto, Japan). P. persica samples (twigs, leaves, pericarps, seeds, immature fruits, and mature fruits) were obtained from a farm located in Wakayama, Japan (34°14′56″ N and 135°21′0″ E) from 2017 to 2020.

2.2. Preparation of Extracts for Screening

Dried samples were pulverized, and 10 g of each sample was extracted using 50% ethanol (EtOH) under reflux conditions for 2 h. EtOH in the filtrate was evaporated under reduced pressure and lyophilized to obtain crude 50% EtOH extracts. Their yields were as follows: twigs, 12%; leaves, 24%; pericarps, 41%; seeds, 4%; immature fruits, 43%; and mature fruits, 66%.

2.3. Anti-Tyrosinase Assay

Anti-tyrosinase activity was evaluated by the method reported previously [3], with optimization for the utility of a 96-well plate. A 50 μL sample solution of 5% dimethylsulfoxide (DMSO) in phosphate-buffered saline (PBS) was mixed with 0.03% L-DOPA solution (50 μL) in PBS and incubated for 10 min at 25 °C. After incubation, 50 μL tyrosinase solution was added to PBS and incubated for 5 min at 25 °C. After incubation, the optical density at 475 nm of the reaction mixture was measured with a microplate reader (Sunrise Rainbow Thermo, Tecan, Männedorf, Switzerland). The percentage of inhibition was determined by comparing the difference in the absorption of the sample with the control groups that exhibit minimum inhibition. Kojic acid was used as a reference sample.

2.4. Purificaiton of Active Compounds

The twig sample (500 g) was pulverized and soaked in 5 L of 50% EtOH. The suspension was exposed to reflux conditions for 5 h. The filtrate was evaporated under reduced pressure and lyophilized to obtain a crude extract with a 12% (61.8 g) yield. The crude extract (1 g) was subjected to partitions between ethyl acetate (EtOAc, 500 mL × 2)-water (500 mL) and 1-butanol (500 mL × 2)-water (500 mL). The yield of each fraction was obtained as follows: EtOAc fraction, 24% (236 mg); 1-butanol fraction, 25% (253 mg); and water fraction, 48% (485 mg) (Figure 1).

The EtOAc fraction (200 mg) was subjected to further purification using reversed-phase flash column chromatography (Smart Flash, EPCLC-AI-580S, Yamazen, Osaka, Japan) with the following conditions: column, Universal 2L ODS (Yamazen, 3.0 i.d. × 20.0 cm); mobile phase, 20% to 95% acetonitrile (MeCN) for 34 min (linear gradient); flow rate, 20 mL/min; column temperature, ambient; and detection, UV 254 nm. Fractions 2 to 9 (0 to 135 mL eluent) were combined (113 mg, 57% yield). This combined fraction was further purified with the same conditions except for the mobile phase, 5% to 50% MeCN for 34 min. Fractions 2 to 15 (0 to 225 mL eluent) were combined (104 mg, 92% yield). The combined fraction was further purified with reversed-phase HPLC under the following conditions: column, SunFire C18 (Waters Corp., Milford, MA, USA, 5 µm, 10 i.d. × 250 mm); mobile phase, 20% MeCN (0.1% formic acid) to 80% MeCN (0.1% formic acid) for 30 min; flow rate, 5 mL/min; column temperature, ambient; detection, UV 280 nm. Five dominant peaks at retention times of 9.3, 10.2, 11.6, and 16.2 min were observed, and the combination fractionated yielded CP-1 (3.1 mg), CP-2 (1.7 mg), CP-3 (2.6 mg), and CP-4 (3.1 mg), respectively.

2.5. Spectral Analyses

The chemical structures of isolated compounds were determined by spectral analysis using combinations of NMR, MS, and optical rotation. NMR spectra were obtained at 400 MHz (JNM-ECS400, JEOL, Tokyo, Japan) or 600 MHz (Avance III HD, Bruker Biospin, Ettlingen, Germany) in CD₃CN or CD₃OD. MS was obtained with ESI mode using 50% MeCN as the solvent (QDa, Waters CORP.). Optical rotation was measured with a polarimeter using methanol (MeOH) as the solvent (P-2200, Jasco, Tokyo, Japan).

2.6. Quantification of Active Compounds in Twigs, Leaves, Pericarps, Seeds, Immature Fruits, and Mature Fruits of P. persica

Preliminary quantification of active compound concentrations in P. persica extracts were performed using the isolated compounds as standards. CP-1, 2, 3, and 4 were dissolved in MeOH to obtain 0.5 mg/mL solutions, and a series of dilutions were prepared using MeOH to make 0.25, 0.13, 0.063, 0.031, and 0.016 mg/mL solutions. Plant samples (10 g) were pulverized and suspended in MeOH (100 mL). After stirring at room temperature for 2 h, the suspensions were filtered, and the filtrates were filled up to 100 mL in a volumetric flask to make sample solutions. Standard and sample solutions were analyzed by HPLC under the following conditions: column, Imtakt Cadenza CD-C18 (3 µm, 4.6 i.d. × 150 mm, Kyoto, Japan); mobile phase, 5% MeCN (0.1% formic acid) to 95% MeCN (0.1% formic acid) for 30 min (linear gradient); flow rate, 1 mL/min; column temperature, 40 °C; detection, UV 284 nm; and injection volume, 10 µL. The concentrations of each compound were calculated from the calibration curves. HPLC analyses were carried out in triplicates, and the quantification was performed once.

2.7. Purification of an Unknown Compound (UK-1) in the Twig Extract

Detection of UK-1 was performed by analytical HPLC using UK-1 obtained from a preliminary isolation study as a standard under the following conditions: column, Imtakt Cadenza CD-C18 (3 µm, 4.6 i.d. × 150 mm); mobile phase, 5% MeCN (0.1% formic acid) to 95% MeCN (0.1% formic acid) for 30 min (linear gradient); flow rate, 0.8 mL/min; column temperature, ambient; and detection, UV 254 and 280 nm.

The twigs of P. persica (1 kg) were extracted with MeOH at room temperature 3 times. The filtrates were combined, and MeOH was evaporated under reduced pressure. The resulting residue was partitioned between hexane (1.5 L × 2) and 80% MeOH (1.5 L). From the 80% MeOH fraction, MeOH was evaporated under reduced pressure to yield a water suspension. The water suspension was partitioned between EtOAc (1.5 L × 2) and water (1.5 L). The EtOAc fraction (25.2 g) was further purified with Diaion HP-20 (Mitsubishi Chemical Corp., Tokyo, Japan). Diaion HP-20 (1.3 L) was added to the water suspension of EtOAc fraction and filtered to obtain the charge fraction. The resin was first washed with water (3 L), then 40% MeOH (3 L), and then 80% MeOH (3 L). The 40% MeOH fraction was subjected to flash column chromatography under the following conditions: column, Universal 2L ODS (Yamazen, 3.0 i.d. × 20.0 cm); mobile phase, 10% to 40% MeCN (0.1% formic acid) for 17 min (linear gradient) and 40% MeCN (0.1% formic acid) for 9 min; flow rate, 20 mL/min; column temperature, ambient; and detection, UV 254 nm. The fractions containing UK-1 were combined and subjected to preparative HPLC purification under the following conditions: column, SunFire C18 (5 µm, 10 i.d. × 250 mm); mobile phase, 5% MeCN (0.1% formic acid) to 30% MeCN (0.1% formic acid) for 30 min; flow rate, 5 mL/min; column temperature, ambient; and detection, UV 254 nm. UK-1 (2.4 mg) was isolated as a yellow amorphous powder.

2.8. Statistical Analysis

Biological activity data were statistically analyzed with Statcel4 (OMS Publishing, Saitama, Japan). The data were analyzed by one-way ANOVA followed by a multiple comparison test (Bonferroni/Dunn algorithm) to detect significant differences, p < 0.05 and p < 0.01.

3. Results

3.1. Screening of Extracts from Various Parts of P. persica for Tyrosinase Inhibition

The results are shown in

Table 1. Inhibitory activity of extracts from twigs, leaves, pericarps, seeds, immature fruits, and mature fruits from P. persica.

Samples	Concentration	Optical density at 475 nm 1	Inhibition (%)
Control	-	528 ± 6	-
Twigs	250	388 ± 4 **	27
	500	328 ± 6 **	38
Leaves	250 µg/mL	480 ± 10 **	9
	500	454 ± 15 **	14
Pericarps	250	497 ± 2 **	6
	500	485 ± 4 **	8
Kojic acid	5 µM	385 ± 10 **	27
	10	293 ± 13 **	45
	50	95 ± 4 **	82
Control	-	512 ± 10	-
Seeds	250	455 ± 1 **	11
	500	446 ± 0 **	13
Immature fruits	250 µg/mL	473 ± 1 **	8
	500	450 ± 2 **	12
Fruits	250	497 ± 10	3
	500	502 ± 17	2
Kojic acid	5 µM	375 ± 18 **	27
	10	298 ± 16 **	42
	50	101 ± 3 **	80

¹ Each value represents the mean ± S.D. of triplicates. Significantly different from control group, **: p < 0.01.

. Among the extracts tested, the twig extract showed the most potent inhibitory activity: 27 and 38% at 250 and 500 µg/mL, respectively. This result prompted us to investigate the active compounds in the twig extract.

3.2. Characterization of Active Compounds in Twig Extract

CP-1, 2, 3 and 4 were identified as (−)-prunin (naringenin 7-O-β-D-glucopyranoside) [4], persiconin [5], (+)-dihydrokaempferol [6], and (−)-naringenin [7] from their spectral data, which included 1D-NMR, 2D-NMR, MS, and optical rotations (Figure 2). The data were consistent with the reported data. However, the stereochemistry of persiconin at C2 was not identified because no data on optical rotation or stereochemistry were available from previous reports. Similarly, the derivatization of isolated persiconin was not possible due to its limited availability.

Tyrosinase inhibitory activity by (−)-prunin, persiconin, (+)-dihydrokaempferol, and (−)-naringenin were 55, 46, 49 and 57% at 500 µM, respectively. This is the first report to declare the tyrosinase inhibitory activity of (−)-prunin and persiconin. The inhibitory activity of crude extracts was attributed to the cooperative effects of the moderate inhibitors.

3.3. Quantification of Active Compounds in Twig, Leaves, Pericarp, Seeds, Immature Fruit, and Mature Fruit Extracts from P. persica

The twigs of P. persica may be used as a skin-whitening material since they have a relatively high tyrosinase inhibition activity. The usefulness of twig extract was supported by the quantification of four active compounds contained in the extract.

All calibration curves showed a good linearity from 0.016 to 0.5 mg/mL (Figure 3). Among the tested extracts, only twig extract showed high concentrations of the four compounds: (−)-prunin, 1.8 mg/g sample; persiconin, 0.8 mg/g sample; (+)-dihydrokaempferol, 0.8 mg/g sample; and (−)-naringenin, 1.7 mg/g sample. The concentrations of these four compounds in the other extracts were below the quantification limit of 0.16 mg/mL. However, trace amounts of (−)-prunin and (−)-naringenin were detected in seed extract, and trace amounts of (−)-prunin were detected in pericarp extract.

3.4. Characterization of UK-1

A peak was detected in the preparative HPLC chromatogram during the purification of active compounds in the twig extract and named UK-1. The investigation of the chemical constituents in twig extract from P. persica had not previously been performed, and this prompted us to clarify the chemical structure of UK-1.

NMR, MS, and optical rotation analyses of UK-1 identified the compound as isoquercitrin (Figure 4) [8,9]. This compound did not show a tyrosinase inhibitory activity at 500 µM.

4. Discussion

We examined twig extract from P. persica as a tyrosinase inhibitor. Although the acute toxicity of twig extract should be tested, its potency as a skin-whitening agent is attractive due to its natural origin. We also revealed four active compounds of tyrosinase inhibitory activity in the twig extract: (−)-prunin, persiconin, (+)-dihydrokaempferol, and (−)-naringenin. These compounds possessed moderate tyrosinase inhibitory activity. The inhibitory activity of crude extract from twigs can be attributed to the cooperative activity of these four compounds. This was the first report to characterize these compounds, as well as a flavone and isoquercitrin, in the twig extract of P. persica. In addition, the concentrations of the four active compounds in extracts from twigs, leaves, pericarps, seeds, immature fruits, and mature fruits revealed that twig extract was the most potent in flavonoids and flavonoid glycosides.

From previous studies, the constituents of bark extract from P. persic a were identified as flavanones, including persicogenin (5,3′-dihydroxy-7,4′-dimethoxyflavanone), persiconin, naringenin, aromadendrin (dihydrokaempferol), persicogenin 3′-O-β-D-glucopyranoside, eriodictyol, and hesperetin 5-O-β-D-glucopyranoside [5,10]. Some of these flavanones were also found in twig extract, as demonstrated in this study, which suggests that the constituents of bark and twig extract are similar. Further studies to examine the chemical constituents of twig extract, which may reveal other novel compounds and biological activities for improving quality of life, are ongoing in our laboratory.

5. Patents

This work has applied for Japan Patent No. 2021-160171.