Synthesis and biological evaluation of bergenin analogues as mushroom tyrosinase inhibitors

Article abstract of DOI:10.1007/s12272-012-0903-3

In this manuscript, we synthesized a series of bergenin analogues, analyzed their structural importance for two biologic activities (anitioxidant activity (ORAC) and mushroom tyrosinase inhibitory activity). Among them, compound 5 which contains catechol moiety exhibited the most antioxidant activity (3.75 μmol of Trolox equiv. per μmol of 5). Furthermore, compound 5 was found to be the most potent (IC₅₀ value = 17.5 ± 0.04 μM) when compared with the standard tyrosinase inhibitors of arbutin (IC₅₀ value = 221.8 ± 1.9 μM) and kojic acid (IC₅₀ value = 46.6 ± 3.8 μM). The bergenin moiety, the ester linkage, and benzoic acid moiety of bergenin derivatives affected two biologic activities. Tyrosinase inhibitory activity was affected by substituents of benzoic acid moiety. This manuscript provides a good foundation for the design and development of new tyrosinase inhibitors.

Full text of DOI:10.1007/s12272-012-0903-3

Arch Pharm Res Vol 35, No 9, 1533-1541, 2012
DOI 10.1007/s12272-012-0903-3
Synthesis and Biological Evaluation of Bergenin Analogues as
Mushroom Tyrosinase Inhibitors
Yusei Kashima and Mitsuo Miyazawa
Department of Applied Chemistry, Faculty of Science and Engineering, Kinki University, Osaka 577-8502, Japan
(Received March 2, 2012/Revised April 26, 2012/Accepted April 29, 2012)
In this manuscript, we synthesized a series of bergenin analogues, analyzed their structural
importance for two biologic activities (anitioxidant activity (ORAC) and mushroom tyrosinase
inhibitory activity). Among them, compound
5
which contains catechol moiety exhibited the
mol of ). Furthermore, compound 5
most antioxidant activity (3.75 mol of Trolox equiv. per
µ
µ
µ
5
was found to be the most potent (IC50 value = 17.5 ± 0.04
tyrosinase inhibitors of arbutin (IC50 value = 221.8 ± 1.9
M) when compared with the standard
M) and kojic acid (IC50 value = 46.6 ±
µ
3.8
µ
M). The bergenin moiety, the ester linkage, and benzoic acid moiety of bergenin derivatives
affected two biologic activities. Tyrosinase inhibitory activity was affected by substituents of
benzoic acid moiety. This manuscript provides a good foundation for the design and development
of new tyrosinase inhibitors.
Key words: Bergenin, Bergenin derivatives, Tyrosinase inhibitor, Antioxidant activity, Structure-
activity relationship
faster degradation and shorter shelf life (Takahashi
and Miyazawa, 2011; Lin et al., 2012). Recently, an in-
INTRODUCTION
Tyrosinase (monophenol or o-diphenol, oxygen oxido- vestigation demonstrated that various dermatological
reductase, EC 1.14.18.1), also known as polyphenol oxi- disorders, such as age spots and freckles, were caused
dase (PPO), is a copper-containing monooxygenase that by the accumulation of an excessive level of epidermal
is widely distributed in microorganisms, animals, and pigmentation (Thanigaimalai et al., 2010). Tyrosinase
plants (Song et al., 2006). Tyrosinase could catalyze has also been linked to Parkinson’s and other neuro-
two distinct reactions involving molecular oxygen in degenerative diseases (Xu et al., 1997). Therefore, tyro-
the hydroxylation of monophenols to
o-diphenols (mono- sinase inhibitors have become increasingly important
phenolase) and in the oxidation of -diphenols to
o
o-qui- in agriculture, the cosmetic industry, and medication,
nones (diphenolase) (Chen et al., 2003). Due to their high which makes the development and screening of potent
reactivity, quinones could polymerize spontaneously inhibitors of tyrosinase extremely important. An impor-
to form a higher molecular weight brown pigments tant group of browning inhibitors is constituted by com-
(melanins) or react with amino acids and proteins to pounds structurally analogous to phenolic substrate.
enhance the brown color of the pigment produced (Nihei Recently, alkoxybenzoic acids have been targeted for
and Kubo, 2003; Matsuura et al., 2006). Previous reports the inhibition of the enzyme (Khan et al., 2010; Zhu et
confirmed that tyrosinase not only was involved in al., 2011).
melanising in animals, but also was one of the main
However, antioxidant therapies have been increas-
causes of quality loss of most fruits and vegetables dur- ingly recognized as a potential strategy for disorder
ing post harvest handling and processing, leading to development prevention, including but not limited to
cancer, cardiovascular disease, inflammation, neurode-
generative diseases, and the aging processes (Laguerre
et al., 2007; Bicas et al., 2011). In addition, applications
Correspondence to: Mitsuo Miyazawa, Department of Applied
Chemistry, Faculty of Science and Engineering, Kinki Univer-
of antioxidants such as preservatives in the food indus-
try and skin-protective ingredients in cosmetics are
also receiving increasing attention and interests. Re-
sity, 3-4-1, Kowakae, Higashiosaka-shi, Osaka 577-8502, Japan
Tel: 81-6-6721-2332, Fax: 81-6-6727-2024
E-mail: miyazawa@kindai.ac.jp
1533

1534
Y. Kashima and M. Miyazawa
cently, the oxygen radical absorbance capacity (ORAC) of bergenin derivatives in detail. In this study, there-
assay has gained much attention because it deals with fore, bergenin and its derivative ( 10) were prepared
1
-
peroxyl radicals, the most abundant radicals in bio- from naturally occurring bergenin (Fig. 1) and the struc-
logical systems. The ORAC assay was considered to be tural importance of two biologic activities was analyzed.
both the inhibition time and the inhibition degree as This is first report of an antioxidant activity evaluation
the reaction is completed and was evaluated directly using the ORAC assay and the structure-activity re-
as the chain-breaking antioxidant activity (Ou et al., lationships of tyrosinase inhibitory activity exhibited
2002).
by bergenin and its derivatives.
Bergenin (
1) is a C-glucoside of 4-O-methylgallic acid
that occurs naturally in several plant genera. Different
studies have shown the various pharmacological effects
MATERIALS AND METHODS
of
1, such as anti-inflammatory (Swarnalakshmi et al.,
Plant
1984), hypolipidemic (Jahromi et al., 1992), anti-HIV
Roots of Bergenia ligulata were provided by Koei
(Piacente et al., 1996), antiarrhythmic (Pu et al., 2002), Kogyo Co., Ltd. of Tokyo in Japan.
hepatoprotective (Lim et al., 2000), neuroprotective
(Takahashi et al., 2003), gastroprotective, antitussive,
and antinociceptive (DeOliveira et al 2011). Further-
Materials
A thin layer chromatography (TLC) was performed
more, bergenin and its esterified derivatives occur on precoated plates (silica gel 60 F254, 0.25 mm, Merk).
widely in several plants and have been found as ingre- Column chromatography was carried out using 70-230
dients in plant extracts (Yoshida et al., 1982; Saijo et mesh silica gel (Kieselgel 60, Merk). Optical rotations
al., 1990; Jia et al., 1995; Fuji et al., 1996; Lee et al., were measured on a Japan Spectroscopic Co. LTDDIP-
2005). Owing to such a broad spectrum of biological 1000. The melting points (m.p.) were measured on a
activities associated with bergenin, a number of studies Yanaco MP-5000D melting-point apparatus. The in-
have been devoted to either derivatizes of the molecule frared (IR) spectra were recorded with a JASCO FT/
or to synthesize its related compounds to optimize it IR-470 plus Fourier transform infrared spectrometer
as a lead molecule. Bergenin contains five hydroxyl (JASCO Co., Ltd). 1H- and 13C-NMR data were all ob-
groups which are considered to be potentially active. tained on JEOL ECA-400 (400 MHz) spectrometers in
Though bergenin and its esterified derivatives (
1
-
10
)
DMSO-d6 or CDCl3 with TMS as the internal standard
are a new class of potent tyrosinase inhibitors and anti- (chemical shift in
δ
, ppm). values are reported in Hertz.
J
oxidants, to our knowledge, there is no report on the cor- EI-MS spectra were obtained on a JEOL JMS-700
relation between tyrosinase inhibition and structure Tandem MS station (Japan Electron Optics Laboratory
Fig. 1. Structures of bergenin (1) and bergenin derivatives (2-10).

Bergenin Derivatives as Tyrosinase Inhibitor
1535
Co., Ltd.). The absorbance was measured with a MTP- and Ph3P (2 equiv) were dissolved. DIAD (1.7 equiv) was
800 Lab microplate reader. Mushroom tyrosinase (EC added in a drop wise manner when the temperature
1.14.18.1), Tween® 20 Sigma ultra, fluorescein sodium dropped below 0 degrees Celsius. The mixture was
salt, trolox, and 2,2’-azobis (2-methylpropionamidine) stirred for 2 h with nitrogen. It was concentrated in
dihydrochloride (AAPH) were purchased from Sigma- vacuo. The residue was chromatographed on silica gel
Aldrich,
acid, -anisic acid, protocatechuic acid, 3,4-dimethoxy white powder.
benzoic acid, vanillic acid, isovanillic acid, 3,5-dimethoxy Secondly, the allyl protected esters and Pd (PPh3)4
L-tyrosinse, benzoic acid, p-hydroxy benzoic to afford benzoic acid ester of 1a (yield 79-92%) as a
p
benzoic acid, syringic acid, diisopropyl azodicarboxyl- (1 mol%, freshly prepared) was dissolved in degassed
ate (DIAD), tetrakis (triphenyl phosphine) palladium anhydr. THF (2 mL/mmol) and morpholine (10 equiv
(Pd(PPh3)4), triphenylphosphine (Ph3P), sodium iodide per allylgroup to be cleaved) was added dropwise. The
(NaI), morpholine, allylbromide, and dibutylhydroxy- mixture was stirred at room temperature and concen-
toluene (BHT) were purchased from Wako Pure Chemi- trated in vacuo. The residue was taken up in EtOAc.
stry, arbutin and kojic acid were purchased from Tokyo The organic layer was washed several times with small
Kasei Kogyo, all solvents were purchased from Kanto amounts of 1 N HCl, dried and concentrated. The crude
Chemical.
material was purified by silica gel column chromatog-
raphy. The purified material was bergenin derivatives
(2-4, 6-10).
Extraction and isolation
The bark (5.0 kg) of B. ligilata was extracted once
with MeOH at room temperature. The solution was
evaporated until dry in vacuo to create methanol ex-
Synthesis of 11-
First, the phenolic hydroxy groups in protocatechu-
O-protocatechuoylbergenin (5)
tract (480 g). The residue was re-extracted successively aldehyde were selectively allylated. Secondly, 3,4-diall-
with -hexane, CH2Cl2, EtOAc, -BuOH and water. yloxy-protocatechuic acid was accomplished according
Each fraction was concentrated to dryness in vacuo to to previous study (Pearl, 1963). Then, the same procedure
give -hexane extract (12.5 g), CH2Cl2 extract (40.5 g), for compound 10 was used starting from 1a
EtOAc extract (88.2 g), -BuOH extract (98.1 g) and
water fraction (240.7 g), respectively.
The -BuOH extract (98.1 g) was fractionated to
n
n
n
2-4,
6-
.
n
Physical and spectroscopic data of the synthe-
sized compounds
n
fraction 1-3 by silica gel column chromatography with 8,10-diallyloxy-bergenin (1a)
CH2Cl2-MeOH (9:1, 4:1, 7:3, v/v) as eluents. Fraction 2 White powder; yield: 84%; 1H-NMR (CDCl3, 400 MHz)
was recrystallized from CH2Cl2-MeOH (12:1, v/v), there-
by bergenin ( ) (3.7 g) was isolated. The structure of
was identified by the comparison of its physical and 10.8, 1.8 Hz, 2 × =CHa), 4.72 (1H, d,
spectral data with those described in the literature 10b), 4.55-4.41 (4H, m, 2 × O-CH2), 4.01-3.83 (1H, m,
δ
7.34 (1H, s H-7), 6.12-5.97 (2H, m, 2 × –CH=), 5.40
(2H, dd, = 17.2, 1.8 Hz, 2 × =CHb), 5.26 (2H, dd,
= 10.4 Hz, H-
1
1
J
J =
J
(Jia et al., 1995; Wang et al., 2005).
H-4), 3.98 (1H, d, J = 9.2 Hz, H-11b), 3.91 (3H, s,
OMe), 3.73-3.68 (1H, m, H-11a), 3.59-3.54 (1H, m, H-
2, H-3).
Synthesis of bergenin derivatives 1a, 2-10
Synthesis of 8,10-diallyloxy-bergenin (1a)
First, the phenolic hydroxy groups in bergenin (
1)
11-O-benzoylbergenin (2)
25
(700 mg, 2.13 mmol) was selectively allylated. Allyl Amorphous powder; [
α
]
+66.8o (
c = 0.15, MeOH);
D
bromide and NaI were added to the solution bergenin overall yield: 45%; m.p. 248.8-249.1oC; IR (KBr) νmax
and K2CO3 in anhydr. DMF (5 mL). After the mixture cm−1: 3474 (OH), 1722 (ester), 1610 (arom. C=C); EI-
was stirred 2 h at 55oC under nitrogen, it was concen- MS:
m/z
432 [M]+; 1H-NMR (DMSO-d6, 400 MHz)
δ
8.01
trated in vacuo. The residue was treated with water (2H, m, H-2’, 6’), 7.67 (1H, m, H-4’), 7.54 (2H, m, H-3’,
(15 mL) and extracted with EtOAc (3 × 10 mL). The 5’), 7.00 (1H, s, H-7), 5.03 (1H, d, = 10.8 Hz, H-10b),
organic extract was washed successively with brine (3 4.78 (1H, dd, = 12.4, 2.0 Hz, H-11b), 4.34 (1H, dd,
× 15 mL). The extract was dried over Na2SO4 and con- = 12.4, 7.2 Hz, H-11a), 4.03 (1H, t, = 10.0 Hz, H-4a),
centrated in vacuo. The residue was chromatographed 3.93 (1H, ddd, = 9.2, 7.2, 2.0 Hz, H-2), 3.74 (3H, s, 9-
J
J
J
J
J
on silica gel to afford 1a (554 mg, 84%) as a white powder. OCH3), 3.69 (1H, m, H-4), 3.41 (1H, dd,
J = 9.2, 8.8
Hz, H-3); 13C-NMR: Table I.
Synthesis of bergenin derivatives (2-4, 6-10)
First, to a solution of 1a in anhydrous tetrahydro- 11-
furan (anhydr. THF) (2 mL/mmol), various benzoic acids Amorphous powder; [α D
O-p-hydroxybenzoylbergenin (3)
25
]
+18.3o (
c = 0.15, MeOH);

1536
Y. Kashima and M. Miyazawa
Table I. 13C-NMR spectral data for compounds
1-10 (δ in ppm)
Carbon number
1
2
3
4
5
6
7
8
9
10
2
3
81.7
70.8
78.5
70.2
78.6
70.2
78.5
70.2
78.6
70.1
78.5
70.5
78.5
70.5
78.6
70.1
78.3
70.5
78.5
70.8
4
79.8
73.5
73.5
73.5
73.5
73.5
73.5
73.5
73.5
73.5
4a
79.6
79.6
79.6
79.6
79.6
79.5
79.6
79.6
79.5
79.5
6
6a
7
8
9
10
10a
10b
163.2
117.9
109.5
150.8
140.6
148.0
115.8
72.1
163.4
118.1
109.5
150.9
140.6
148.0
115.8
72.2
163.4
118.1
109.5
150.9
140.6
148.0
115.8
72.2
163.4
118.1
109.5
150.9
140.6
148.0
115.8
72.2
163.4
118.1
109.5
150.9
140.6
148.0
115.8
72.2
163.3
118.1
109.6
151.0
140.6
148.0
115.7
72.1
163.3
118.1
109.6
151.0
140.6
148.0
115.8
72.1
163.3
118.1
109.5
151
140.6
148
163.3
118.1
109.6
151.0
140.6
148.0
115.7
72.1
163.2
118.1
109.6
151.1
140.6
148.0
115.7
72.1
115.8
72.1
11
63.8
64.0
63.5
63.7
63.4
64.0
63.8
63.8
64.1
64.0
9-OCH3
59.7
59.7
59.7
59.7
59.8
59.7
59.7
59.7
59.7
59.7
1'
2'
3'
4'
5'
6'
129.5
129.2
128.8
133.5
128.8
129.2
165.6
120
121.7
131.4
114.1
163.3
114.1
131.4
165.4
120.2
116.4
145.1
150.7
115.3
122.0
165.6
123.4
111.2
148.4
153.1
111.7
121.6
165.3
55.5
120.2
112.6
147.4
151.8
115.2
123.7
165.4
55.6
121.6
114.4
146.3
152.1
111.5
122
131.5
106.9
160.5
105.5
160.5
106.9
165.2
55.5
119.0
107.0
147.6
141.1
147.7
107.0
165.3
56.1
131.6
115.4
162.2
115.4
131.6
165.5
COO
3'-OCH3
4'-OCH3
5'-OCH3
165.5
55.6
55.7
55.7
55.5
56.1
13C-NMR spectral data were recorded at 100 MHz in DMSO-d6 using tetramethylsilane (TMS) as internal standard.
25
overall yield: 48%; m.p. 214.6-215.5oC; IR (KBr) νmax Amorphous powder; [α D
]
+25.2o (
c
= 0.25, MeOH);
cm−1 : 3380 (OH), 1712 (ester), 1608 (arom. C=C), 1238 overall yield: 26%; m.p. 167.2-173.5oC; IR (KBr) νmax
1
(C-O); EI-MS:
MHz) 7.85 (2H, d,
H-7), 6.86 (2H, d, = 8.8 Hz, H-3’, 5’), 5.02 (1H, d,
10.0 Hz, H-10b), 4.71 (1H, dd, = 12.0, 1.6 Hz, H- 8.4 Hz, H-6’), 7.00 (1H, s, H-7), 6.81 (1H, d,
11b), 4.26 (1H, dd, = 12.0, 6.4 Hz, H-11a), 4.03 (1H, H-5’), 5.05 (1H, d, = 10.4 Hz, H-10b), 4.67 (1H, brd,
t, = 9.8 Hz, H-4a), 3.87 (1H, ddd, = 9.2, 6.4, 1.6 Hz, = 11.6 Hz, H-11b), 4.26 (1H, dd, = 11.6, 6.0 Hz, H-
H-2), 3.73 (3H, s, 9-OCH3), 3.69 (1H, t, = 9.2 Hz, H- 11a), 4.01 (1H, t, = 10.0 Hz, H-4a), 3.86 (1H, m, H-
4), 3.40 (1H, m, H-3); 13C-NMR: Table I.
2), 3.74 (3H, s, 9-OCH3), 3.69 (1H, t, = 9.2 Hz, H-4),
3.39 (1H, t,
= 9.2 Hz, H-3); 13C-NMR: Table I.
m/z
448 [M]+; H-NMR (DMSO-d6, 400 cm−1 : 3364 (OH), 1709 (ester), 1610 (arom. C=C), 1227
1
δ
J
= 8.8 Hz, H-2’, 6’), 7.00 (1H, s, (C-O); EI-MS,
m
/
z
464 [M]+; H-NMR (DMSO-d6, 400
J
J
=
MHz) 7.38 (1H, d,
δ
J
= 1.6 Hz, H-2’), 7.35 (1H, d,
J =
J
J
= 8.4 Hz,
J
J
J
J
J
J
J
J
J
J
11-
Amorphous powder; [α D
overall yield: 44%; m.p. 258.9-260.4oC; IR (KBr) νmax Amorphous powder; [
O-p-methoxybenzoylbergenin (4)
25
]
+44.4o (
c
= 0.15, MeOH); 11-
O
-(3’,4’-dimethoxybenzoyl)-bergenin (6)
α]
D
25 +32.6o (
c
= 0.025, MeOH);
cm−1 : 3373 (OH), 2959 (C-H), 1720 (ester), 1609 (arom. overall yield: 48%; m.p. 260.6-262.7oC; IR (KBr) νmax
C=C), 1237 (C-O); EI-MS:
d6, 400 MHz) 7.95 (2H, d,
(2H, d, = 8.8 Hz, H-3’, 5’), 7.00 (1H, s, H-7), 5.03 (1H, MHz)
d, = 10.0 Hz, H-10b), 4.74 (1H, dd, = 12.0, 2.1 Hz, d, = 1.6 Hz, H-2’), 7.09 (1H, d,
H-11b), 4.29 (1H, dd, = 12.0, 6.8 Hz, H-11a), 4.04 (1H, (1H, s, H-7), 5.05 (1H, d, = 10.0 Hz, H-10b), 4.83
t, = 10.0 Hz, H-4a), 3.89 (1H, m, H-2), 3.83 (3H, s, 4’- (1H, brd, = 12.0 Hz, H-11b), 4.21 (1H, dd, = 12.0,
OCH3), 3.74 (3H, s, 9-OCH3), 3.69 (1H, t, = 8.8 Hz, H- 7.2 Hz, H-11a), 4.04 (1H, dd, = 10.0, 9.6 Hz, H-4a),
4), 3.40 (1H, t, 3.91 (1H, m, H-2), 3.84 (3H, s, 4’-OCH3), 3.83 (3H, s,
= 8.8 Hz, H-3); 13C-NMR: Table I.
m
/
z
462 [M]+; 1H-NMR (DMSO- cm−1 : 3395 (OH), 1716 (ester), 1598 (arom. C=C), 1225
1
δ
J
= 8.8 Hz, H-2’, 6’), 7.05 (C-O); EI-MS:
m/z
492 [M]+; H-NMR (DMSO-d6, 400
J
δ
7.63 (1H, dd,
J
= 8.4, 1.6 Hz, H-6’), 7.48 (1H,
= 8.4 Hz, H-5’), 7.00
J
J
J
J
J
J
J
J
J
J
J
J
3’-OCH3), 3.73 (3H, s, 9-OCH3), 3.70 (1H, t,
J = 9.6 Hz,
11-O-protocatechuoylbergenin (5)
H-4), 3.38 (1H, m, H-3); 13C-NMR: Table I.

Bergenin Derivatives as Tyrosinase Inhibitor
1537
11-
O
-vanilloylbergenin (7)
9.6 Hz, H-4a), 3.91 (1H, ddd,
(c = 0.1, MeOH); 3.81 (6H, s, 3’, 5’-OCH3), 3.73 (3H, s, 9-OCH3), 3.69 (1H,
J
= 9.2, 7.6, 1.6 Hz, H-2),
+56.4o
25
Amorphous powder; [
α]
D
overall yield: 40%; m.p. 268.5-270.5oC; IR (KBr) νmax m, H-4), 3.38 (1H, m, H-3); 13C-NMR: Table I.
cm−1 : 3365 (OH), 1717 (ester), 1599 (arom. C=C), 1222
1
(C-O); EI-MS:
MHz) 7.51 (1H, dd,
d,
8.4 Hz, H-5’), 5.05 (1H, d,
(1H, dd,
11.6, 8.0 Hz, H-11a), 4.04 (1H, dd,
4a), 3.89 (1H, ddd, = 8.0, 7.6, 2.0 Hz, H-2), 3.85 (3H, d,
s, 3’-OCH3), 3.73 (3H, s, 9-OCH3), 3.69 (1H, dd, H-11b), 4.12 (1H, dd,
9.6, 8.8 Hz, H-4), 3.39 (1H, m, H-3); 13C-NMR: Table I. t,
= 10.0 Hz, H-4a), 3.92 (1H, ddd,
m/z
478 [M]+; H-NMR (DMSO-d6, 400 11-
O
-syringylbergenin (10)
J = 8.4, 2.0 Hz, H-6’), 7.48 (1H, Amorphous powder; [α]D (c = 0.1, MeOH);
+24.5o
25
δ
J
= 2.0 Hz, H-2’), 7.00 (1H, s, H-7), 6.88 (1H, d,
J
=
overall yield: 40%; m.p. 240.7-241.3oC; IR (KBr) νmax
J
= 10.8 Hz, H-10b), 4.81 cm−1 : 3371 (OH), 1718 (ester), 1610 (arom. C=C), 1235
1
J
= 11.6, 2.0 Hz, H-11b), 4.19 (1H, dd,
= 10.4, 9.6 Hz, H- MHz)
= 10.8 Hz, H-10b), 4.89 (1H, dd,
= 11.6, 8.8 Hz, H-11a), 4.06 (1H,
= 10.0, 8.1, 2.4
J
=
(C-O); EI-MS:
m/z
508 [M]+; H-NMR (DMSO-d6, 400
J
δ
7.26 (2H, s, H-2’, 6’), 7.00 (1H, s, H-7), 5.07 (1H,
= 11.6, 2.4 Hz,
J
J
J
J
=
J
J
J
Hz, H-2), 3.85 (6H, s, 3’, 5’-OCH3), 3.72 (3H, s 9-OCH3),
11-O-isovanilloylbergenin (8)
3.69 (1H, m, H-4), 3.38 (1H, m, H-3); 13C-NMR: Table I.
25
Amorphous powder; [
α
]
+42.4o
(c = 0.1, MeOH);
D
overall yield: 36%; m.p. 226.6-227.1oC; IR (KBr) νmax
Oxygen radical absorbance (ORAC) assay
The ORAC method used with fluorescein (FL) as the
cm−1 : 3376 (OH), 1714 (ester), 1608 (arom. C=C), 1237
1
(C-O); EI-MS:
MHz) 7.47 (1H, dd,
d, = 2.0 Hz, H-2’), 7.02 (1H, d,
(1H, s, H-7), 5.03 (1H, d, = 10.4 Hz, H-10b), 4.70 tric Co., Ltd.) with fluorescence filters for an excitation
(1H, dd, = 12.0, 2.0 Hz, H-11b), 4.28 (1H, dd, wavelength of 480 nm and emission wavelength of 540
12.0, 6.4 Hz, H-11a), 4.02 (1H, dd, = 10.4, 9.2 Hz, H- nm. The measurements were made in a plate with 96
4a), 3.89 (1H, ddd, = 7.6, 6.4, 2.0 Hz, H-2), 3.82 (3H, black flat-bottom wells (Greiner Bio-One). The reaction
m/z
478 [M]+; H-NMR (DMSO-d6, 400 “fluorescent probe” was described in the literature
δ
J
= 8.4, 2.0 Hz, H-6’), 7.40 (1H, method (Zulueta et al., 2009). The automated ORAC
= 8.4 Hz, H-5’), 7.00 assay was carried out on a MTP-800AFC (Corona Elec-
J
J
J
J
J =
J
J
s, 4’-OCH3), 3.73 (3H, s, 9-OCH3), 3.69 (1H, dd,
J
=
was performed at 37oC as the reaction was started by
9.2, 8.8 Hz, H-4), 3.39 (1H, m, H-3); 13C NMR: Table I. thermal decomposition of AAPH in 75 mM phosphate
buffer (pH 7.4) because of the sensitivity of FL to pH.
11-
O
-(3’,5’-dimethoxybenzoyl)-bergenin (9)
Fig. 2 is a diagram of the reaction of the fluorescein
Amorphous powder; [
α] c = 0.025, MeOH); during the ORAC assay. The fluorescein stock solution
25 +157.9o (
D
overall yield: 46%; m.p. 185.9-194.2oC; IR (KBr) νmax was made in 75 mM phosphate buffer (pH 7.4) and
cm−1 : 3355 (OH), 1717 (ester), 1602 (arom. C=C), 1237 stored under dark conditions at 4oC. AAPH and trolox
1
(C-O); EI-MS:
MHz) 7.19 (2H, d,
7), 6.79 (1H, t, = 2.4 Hz, H-4’), 5.05 (1H, d,
Hz, H-10b), 4.90 (1H, dd, = 12.0, 1.6 Hz, H-11b), 4.23 perature while being continuously mixed. The reaction
(1H, dd, = 12.0, 7.6 Hz, H-11a), 4.04 (1H, dd, = 10.4, was performed in 75 mM phosphate buffer (pH 7.4) and
m/z
492 [M]+; H-NMR (DMSO-d6, 400 solutions in 75 mM phosphate buffer (pH 7.4) were pre-
δ
J
= 2.4 Hz, H-2’, 6’), 7.00 (1H, s, H- pared daily. The samples were dissolved in 50% acet-
= 10.4 one. The mixture was incubated for 1 h at room tem-
J
J
J
J
J
Fig. 2. The proposed Fluorescein pathway in the presence of AAPH.

1538
Y. Kashima and M. Miyazawa
the final assay mixture (200
µ
L) contains fluorescein Then 16
µ
L of mushroom tyrosinase (500 units/mL,
(160
substrate, AAPH (20
as an oxygen radical generator, and trolox (20
20 M final concentration) or samples (20 L, 5-20 µM
µ
L, 63 nM final concentration) as an oxidizable 0.1 M phosphate buffer at pH 7.0) was added, and the
L, 12.8 mM final concentration) assay mixture was incubated at 37oC for 20 min. Before
L, 5- and after incubation, the amount of dopachrome pro-
duced in the reaction mixture was measured at 492
µ
µ
µ
µ
final concentration). The reaction was performed at nm in a microplate reader (Corona Electric Co., Ltd).
37oC, and fluorescence was recorded every minute for Arbutin and kojic acid were used as a positive control.
120 min. A blank (control) using phosphate buffer in- The extent of tyrosinase inhibition by the different com-
stead of the antioxidant was carried out in each experi- pounds added was calculated and expressed as the
ment. All reaction mixtures were prepared in duplicate, percentage necessary for 50% inhibition concentration
and at least three independent runs were performed (IC50).
for each sample. BHT was used as a reference stand-
The percentage of tyrosinase activity was calculated
ard for antioxidant. Fluorescent measurements were as follows:
normalized to the curve of the blank (no antioxidant).
Tyrosinase activity (%) = [(
where is the absorbance at 492 nm without test
sample, is the absorbance at 492 nm without test
sample and substrate, is the absorbance at 492 nm
with test sample, is the absorbance at 492 nm with
M) of trolox, CSample test sample, but without substrate. All data are the
C − D)/(A − B)] × 100,
The ORAC values were expressed as Trolox equivalents
(mean ± S.D.) by applying the following formula:
A
B
CTrolox×(AUCSample–AUCBlank
)
--------------------------------------------------------------------------------------
ORAC value =
C
CSample×(AUCTrolox–AUCBlank
)
D
where CTrolox is the concentration (
µ
is the concentration of the sample, and AUC is the mean of three experiments.
area below the fluorescence decay curve of the sample,
blank and trolox, respectively, calculated by applying
the following formula:
RESULTS AND DISCUSSION
Antioxidant activities of bergenin (1) and its
derivatives (2-10)
Bergenin (1) and all bergenin derivatives (2-10) were
AUC = 1+f1⁄f0+f2⁄f0+f3⁄f0…+fn⁄f0
where f0 is the initial fluorescence and fn is the
fluorescence at time
n
.
tested that showed antioxidative activity Trolox equiva-
lents (Table II). The antioxidative activities against
peroxyl radicals were in the order of bergenin derivatives
Tyrosinase inhibitory assay
The tyrosinase assay was performed by the method > Trolox > BHT, indicating that bergenin and all
of Baek et al. (2008) with slight modifications, using bergenin derivatives ( 10) were more active than the
tyrosine as the substrate. 100 L of 0.1 M phosphate reference antioxidants. 11- -(3’,4’-Dimethoxybenzoyl)-
buffer (pH 7.0), 36 L of 1.5 mM -tyrosine, and 10 bergenin ( ) showed the lowest activity in the assay
of sample solution containing DMSO + 0.05% Tween® (1.27 ± 0.07
mol of Trolox equiv. per mol of ). 11-
L-
2-
µ
O
µ
L
µL
6
µ
µ
6
O-
20 to dissolve the sample were added to each well of a Protocatechuoylbergenin (5), on the other hand, pres-
96-well plate and then incubated at 37oC for 10 min. ented the highest antioxidant activity (3.75 ± 0.20
Table II. Antioxidant activity of the bergenin derivatives by ORAC assay
Compound
Trolox-equivs.
Compound
benzoic acid (2a
-hydroxy benzoic acid (3a
-anisc acid (4a
protocatechuic acid (5a
Trolox-equivs.
bergenin (
-benzoylbergenin (
-hydroxybenzoylbergenin (
-methoxybenzoylbergenin (
-protocatechuoylbergenin (
-(3',4'-dimethoxybenzoyl)-bergenin (
11- -vanilloylbergenin (
11- -isovanilloylbergenin (
-(3',5'-dimethoxybenzoyl)-bergenin (
11-
1)
2)
3)
4)
5)
6)
7)
8)
9)
)
1.79 ± 0.11
1.56 ± 0.12
2.49 ± 0.39
1.86 ± 0.10
3.75 ± 0.20
1.27 ± 0.07
2.94 ± 0.08
1.69 ± 0.33
1.50 ± 0.05
1.54 ± 0.06
)
)
)
)
)
)
)
)
)
0.05 ± 0.02
1.84 ± 0.88
0.02 ± 0.002
2.26 ± 0.33
0.02 ± 0.001
1.17 ± 0.09
1.16 ± 0.08
0.10 ± 0.01
1.25 ± 0.14
0.12 ± 0.06
1.00
11-O
p
11-
11-
11-
O
-
p
O
p
p
O-
3,4-dimethoxy benzoic acid (6a
vanillic acid (7a
isovanillic acid (8a
3,5-dimethoxy benzoic acid (9a
syringic acid (10a
11-
11-
O
O
O
O
O-syringylbergenin (10
BHT
trolox
Results are presented as the mean ± S.D., n=3.

Bergenin Derivatives as Tyrosinase Inhibitor
1539
Fig. 3. Fluorescence intensity decay curve of fluorescein in
the presence of compound
3, 5, 7 and trolox. Concentrations
for curves were 20 M, respectively.
µ
Fig. 4. The dose-dependent inhibitory effects of bergenin (
and bergenin derivatives ( 10) on tyrosinase activity.
Tyrosinase activity was measured using -tyrosine as the
substrate. Values are means ± S.E. of n = 3 determination.
Error bars show S.E. of triplicates.
1)
2-
µ
mol of Trolox equiv. per
compounds, compound and
dant activities (2.49-2.94 mol of Trolox equiv. per
mol of and ). As presented in Fig. 3, the protective
effect of an antioxidant was measured by assessing
the fluorescence decay curve (AUC) of the sample com- exhibited potent tyrosinase inhibitory effect with IC50
pared to a blank, in which no antioxidant was presented. value of 17.5 ± 0.04 M as compared to positive control,
Furthermore, compound and 10 showed arbutin (IC50 value = 217 ± 2.0 M) and kojic acid (IC50
moderate antioxidant activities. Whereas, all benzoic value = 46.6 ± 3.8 M).
acid derivatives (2a 10a) had lower activities than Among the bergenin derivatives
bergenin derivatives. These results indicate that the of substituents, the addition of the OH or OMe func-
bergenin moiety affects the antioxidant activity, and tional group at the 4’-position makes compound (IC50
the presence of the hydroxy group at 4’-position on value = 216.9 ± 3.5 M) and (IC50 value = 259.6 ± 2.8
bergenin derivative was a potent factor of antioxidant M) about five times more active than compound
activity. When the antioxidant activities of compound which has no substituents of benzoic acid moiety (12.5
and were compared, it was found that the potency ± 1.3% at 300 M), which were much the same effect
increased in 5 which was greater than 6. Therefore, as a potent inhibition of arbutin (IC50 value = 217.0 ±
we considered that the bergenin moiety and catechol 2.0 M). When an OH group was located at the 3’ posi-
tion, and the para-hydrogen was substituted by an OH
group or OMe group (compound and ), inhibitory
activities were enhanced, (IC50 value = 17.5 ± 0.04 and
79.8 ± 3.3 M). In the case of compound , the introduc-
µ
mol of
5). The other active
L
3
7, also exhibited antioxi-
µ
µ
3
7
µ
2,
6,
8,
9
µ
µ
-
2-10, introduction
3
µ
4
µ
2
5
6
µ
µ
moiety have a great influence on antioxidant activity.
5
8
Tyrosinase inhibitory activities of bergenin (1)
and its derivatives (2-10)
µ
5
We investigated the tyrosinase inhibitory activity of tion of another OH group in an ortho-position to an OH
bergenin derivatives. Bergenin and its derivatives were group led to a dramatic increase in inhibitory activity
used as the effectors. Tyrosinase is a copper-containing relative to compound
enzyme that catalyzes two reactions: the hydroxylation On the other hand, when an OMe group was located
of -tyrosine to 3,4-dihydroxyphenylalanine (DOPA), at the 4’ position, and the meta-hydrogen was substi-
and the conversion of DOPA to DOPA quinones (Hearing tuted by an OMe group or OH group (compound and
3 and 7.
L
6
and Jimenz, 1987). The tested compounds were assayed
for tyrosinase inhibitory activity, which demands
8
), the inhibitory activity of compound 8 was much more
L-
potent than that of compound 6. Other compounds (9
tyrosine as substrate. All of them have inhibitory effects and 10) had shown a weak inhibition of mushroom
on enzyme activity with dependence on the concentra- tyrosinase.
tions as shown in Fig. 4. Four compounds (
showed tyrosinase inhibition with IC50 varying from of benzoic acid derivatives and bergenin (
17.5 to 259.6 M. Among the compounds, compound whether the inhibitory activities of compounds 3, 4, 5,
3,
4,
5,
8)
Furthermore, we evaluated the inhibitory activities
1) to clarify
µ
5

1540
Y. Kashima and M. Miyazawa
Table III. Inhibition effects of bergenin (
8a) against tyrosinase
1), bergenin derivatives (2-10), and benzoic acid derivatives (3a, 4a, 5a, and
Concentration
M)
Inhibition
IC50
µ
Concentration
M)
Inhibition
(%)
IC50
M)a
Compound
Compound
(
µ
(%)
(
M)a
(
µ
(
µ
1
75.0
0.0 ± 0.5
0.0 ± 0.7
19.1 ± 3.7
0.0 ± 2.6
8
50.0
75.0
112.5
75.0
150.0
300.0
75.0
150.0
300.0
75.0
150.0
300.0
75.0
150.0
300.0
75.0
150.0
300.0
75.0
150.0
300.0
41.3 ± 1.5
48.1 ± 3.1
60.2 ± 1.0
10.1 ± 1.7
14.0 ± 1.9
17.9 ± 3.1
11.6 ± 1.7
13.9 ± 0.9
17.7 ± 2.9
6.1 ± 1.5
9.2 ± 1.5
18.6 ± 3.5
0.0 ± 3.5
4.49 ± 4.0
13.0 ± 3.5
4.6 ± 0.24
9.5 ± 2.6
12.0 ± 1.7
0.0 ± 0.3
0.7 ± 0.3
1.1 ± 1.7
150.0
300.0
75.0
> 300
> 300
79.8 ± 3.3
> 300
2
9
150.0
300.0
150.0
112.5
300.0
150.0
225.0
300.0
12.5
25.0
50.0
75.0
150.0
300.0
75.0
150.0
300.0
0.0 ± 4.2
12.5 ± 1.3
42.3 ± 1.1
50.7 ± 0.3
58.4 ± 0.4
42.3 ± 1.7
48.2 ± 3.7
51.7 ± 2.4
31.4 ± 1.4
52.9 ± 1.1
78.5 ± 0.7
3.94 ± 0.3
6.4 ± 2.4
3
10
3a
4a
5a
8a
216.9 ± 3.50
259.6 ± 2.80
017.5 ± 0.04
> 300
> 300
4
> 300
5
> 300
6
7
15.7 ± 1.7
2.4 ± 2.1
5.4 ± 0.49
18.9 ± 3.7
> 300
> 300
> 300
arbutinb
217.0 ± 2.00 kojic acidb
46.6 ± 3.8
aThe results are the means ± S.E. of three experiments; bStandard inhibitors of the enzyme tyrosinase
and
8 are due to benzoic acid moiety or bergenin moiety. data indicated that the two activities are not directly
p-Hydroxybenzoic acid (3a) (inhibition: 18.6 ± 3.5% at correlated. However, the substituent groups of bergenin
300
300
1.7% at 300
1.7% at 300
µ
M), anisic acid (4a) (inhibition: 13.0 ± 3.5% at derivatives affected both oxygen radical absorbance
µ
M), protocatechuic acid (5a) (inhibition: 12.0 ± capacity and tyrosinase inhibitory activity. Furthermore,
µ
M), isovanillic acid (8a) (inhibition: 1.1 ± in the biologic activities (antioxidant activity and tyro-
M) and bergenin ( ) (inhibition: 19.1 ± sinase inhibitory activity), the linkage of “bergenin
M) exhibited low inhibitory activity. moiety” and “benzoic acid moiety” affected them. In this
µ
µ
1
3.7% at 300
Each of the benzoic acid derivatives exhibited a rela- study, we reported the correlation between tyrosinase
tively poor inhibitory activity as compared with these inhibition and structures of bergenin derivatives for
bergenin derivatives. The results are summarized in the first time. Some bergenin derivatives (2, 6, 7, 9
Table III. Due to the inhibitory activities of these ben- and 10) might not be drug candidates themselves since
zoic acid derivatives, we assumed that the linkage of their inhibitory activity on tyrosinase is not strong.
“benzoic acid” and “bergenin” as well as benzoic acid However, these results suggest that bergenin derivatives
moiety and bergenin moiety are important factors of may serve as the designed development of novel tyro-
tyrosinase inhibitory activity. In addition to these sinase inhibitors.
reports, the observation of the inhibitory activities of
compound
5 and 8 indicated that 3’ position OMe
group is also a potent factor of tyrosinase inhibitory
activity.
ACKNOWLEDGEMENTS
We are grateful to Professor Horibe I. for number of
In conclusion, we have undertaken synthesis a series valuable suggestions and discussions on this study. The
of bergenin analogues. Compound and showed the author is grateful to assistant Professor Minemastsu T.
most potent inhibition of mushroom tyrosinase (IC50 (School of Pharmaceutical Sciences, Kinki University)
5
8
value = 17.5
of compound
µ
M and 79.8
µ
M), but antioxidant activity for the NMR spectrometry experiments on this study.
8
is less than compound
7. The inhibition

Bergenin Derivatives as Tyrosinase Inhibitor
1541
Med. Chem. Lett., 13, 2409-2412 (2003).
REFERENCES
Ou, B., Huang, D., Hampsch-Woodill, M., Flangan, J. A., and
Deemer, E. K., Analysis of antioxidant activities of com-
mon vegetables employing oxygen radical absorbance cap-
acity (ORAC) and ferric reducing antioxidant power (FRAP)
assays: a comparative study. J. Agric. Food Chem., 50,
3122-3128 (2002).
Pearl, I. A., Vanillic acid. Org. Synth., 4, 972 (1963).
Piacente, S., Pizza, C., DeTommasi, N., and Mahmood, N.,
Constituents of Ardisia japonica and their in vitro anti-
HIV activity. J. Nat. Prod., 59, 565-569 (1996).
Baek, S., Kim, J., Kim, D., Lee, C., Kim, J., Chung, D. K., and
Lee, C., Inhibitory effect of dalbergioidin isolated from the
trunk of Lespedeza cyrtobotrya on melanin biosynthesis.
J. Microbiol. Biotechnol., 18, 874-879 (2008).
Bicas, J. L., Neri-Nmuma, I. A., Ruiz, A. L. T. G., DeCarvalho,
J. E., and Pastore G. M., Evaluation of the antioxidant
and antiproliferative potential of bioflavors. Food Chem.
Toxicol., 49, 1610-1615 (2011).
Chen, Q. X., Liu, X. D., and Huang, H., Inactivation kinetics
of mushroom tyrosinase in the dimethyl sulfoxide solution.
Pu, H. L., Huang, X., Zhao, J. H., and Hing, A., Bergenin is the
Biochemistry (Moscow), 68, 644-649 (2003).
antiarrhythmic principle of Fluggea virosa. Planta Med.,
DeOliveira, C. M., Nonato, F. R., DeLima, F. O., Couto, R. D.,
David, J. P., David, J. M., Soares, M. B. P., and Villarreal,
68 372-374 (2002).
Saijo, R., Nonaka, G., and Nishioka, I., Gallic acid esters of
bergenin and norbergenin from Mallotus japonicus Phyto-
chemistry, 29, 267-270 (1990).
C. F., Antinociceptive properties of bergenin. J. Nat. Prod.
,
.
74, 2062-2068 (2011).
Fuji, M., Miyaichi, Y., and Tomimori, T., Studies on nepalese
crude drugs. XXII on the phenolic constituents of the rhi-
zome of Bergenia ciliata (Haw.) STERNB. Nat. Med., 50,
404-407 (1996).
Hearing, V. J. and Jiménez, M., Mammalian tyrosinase –
the critical regulatory control point in melanocyte pig-
mentation. Int. J. Biochem., 19, 1141-1147 (1987).
Jahromi, M. A. F., Chansouria, J. P. N., and Ray, A. B., Hypoli-
pidemic activity in rats of bergenin, the major constituent
Song, K. K., Huang, H., Han, P., Zang, C. L., Shi, Y., and Chen,
Q. X., Inhibitory effects of cis- and trans-isomers of 3,5-
dihydroxystilbene on the activity of mushroom tyrosinase.
Biochem. Biophys. Res. Commun., 342, 1147-1151 (2006).
Swarnalakshmi, T., Sethuraman, M. G., Sulochana, N., and
Arivudainambi, R., A note on the anti-inflammatory activity
of bergenin. Curr. Sci., 53, 917 (1984).
Takahashi, H., Kosaka, M., Watanabe, Y., Nakade, K., and
Fukuyama, Y., Synthesis and neuroprotective activity of
bergenin derivatives with antioxidant activity. Bioorg.
Med. Chem., 11, 1781-1788 (2003).
of Flueggea microcarpa. Phytother. Res., 6, 180-183 (1992).
Jia, Z., Mitsunaga, K., Koike, K., and Ohmoto, T., New bergenin
derivatives from Ardisia crenata. Nat. Med., 49, 187-189
(1995).
Takahashi, T. and Miyazawa, M., Synthesis and structure-
activity relationships of phenylpropanoid amides of serotonin
on tyrosinase inhibition. Bioorg. Med. Chem. Lett., 21,
1983-1986 (2011).
Thanigaimalai, P., Hoang, T. A. L., Lee, K. C., Bang, S. C.,
Sharma, V. K., Yun, C. Y., Roh, E., Hwang, B. Y., Kim, Y., and
Jung, S. H., Structural requirement(s) of N-phenylthiureas
and benzaldehyde thiosemicarbazones as inhibitors of
melanogenesis in melanoma B16 cells. Bioorg. Med.
Chem. Lett., 20, 2991-2993 (2010).
Khan, S. B., Khan, M. T. H., Jang, E. S., Akhtar, K., Seo, J.,
and Han, H., Tyrosinase inhibitory effect of benzoic
derivatives and their structure-activity relationships. J.
Enzyme Inhib. Med. Chem., 25, 812-817 (2010).
Laguerre, M., Lecomte, J., and Villeneuve, P., Evaluation of
the ability of antioxidants to counteract lipid oxidation:
Existing methods, new trends and challenges. Prog. Lipid
Res., 46, 244-282 (2007).
Lee, Y. Y., Jang, D. S., Jin, J. L., and Yun-Choi, H. S., Anti-
Wang, D., Zhu, H. T., Zhang, Y. J., and Yang, C. R., A carbon-
carbon-coupled dimeric bergenin derivative biotransformed
platelet aggregating and anti-oxidative activities of 11-
O-
(4’- -methylgalloyl)-bergenin, a new compound isolated
O
by Pleurotus ostreatus. Bioorg. Med. Chem. Lett., 15,
from Crassula cv. ‘Himaturi’. Planta. Med., 71, 776-777
(2005).
Lim, H. K., Kim, H. S., Choi, H. S., Oh, S., and Choi, J.,
Hepatoprotective effects of bergenin, a major constituent
of Mallotus japonicus, on carbon tetrachloride-intoxicated
rats. J. Ethnopharmacol., 72, 469-474 (2000).
4073-4075 (2005).
Xu, Y., Stokes, A. H., Freeman, W. M., Kumer, S. C., Vogt, B.
A., and Vrana, K. E., Tyrosinase mRNA is expressed in
human substantia nigra. Mol. Brain Res., 45, 159-162 (1997).
Yoshida, T., Seno, K., Takama, Y., and Okuda, T., Bergenin
derivatives from Mallotus japonicus. Phytochemistry, 21,
Lin, Y. S., Chen, S. H., Huang, W. J., Chen, C. H., Chien, M. Y.,
Lin, S. Y., and Hou, W. C., Effects of nicotinic acid derivatives
on tyrosinase inhibitory and antioxidant activities. Food
Chem., 132, 2074-2080 (2012).
Matsuura, R., Ukeda, H., and Sawamura, M., Tryrosinase
inhibitory activity of citrus essential oils. J. Agric. Food
Chem., 54, 2309-2313 (2006).
1180-1182 (1982).
Zhu, Y. J., Zhou, H. T., Hu, Y. H., Tang, J. Y., Su, M. X., Guo,
Y. J., Chen, Q. X., and Liu, B., Antityrosinase and anti-
microbial activities of 2-phenylehtanol, 2-phenylacetal-
dehyde and 2-phenylacetic acid. Food Chem., 124, 298-
302 (2011).
Zulueta, A., Esteve, M. J., and Frígola, A., ORAC and TEAC
assays comparison to measure the antioxidant capacity of
food products. Food Chem., 114, 310-316 (2009).
Nihei, K. and Kubo, I., Identification of oxidation product of
arbutin in mushroom tyrosinase assay system. Bioorg.