Kojic acid–natural product conjugates as mushroom tyrosinase inhibitors

Article abstract of DOI:10.1016/j.ejmech.2020.112480

As part of our effort to develop potential tyrosinase inhibitors, we have conjugated the well-known tyrosinase inhibitor kojic acid (KA) with several phenolic natural products such as umbelliferone, sesamol, thymol, carvacrol, eugenol, isoeugenol, vanilli

Full text of DOI:10.1016/j.ejmech.2020.112480

Journal Pre-proof
Kojic acid–natural product conjugates as mushroom tyrosinase inhibitors
Morteza Ashooriha, Mehdi Khoshneviszadeh, Mahsima Khoshneviszadeh, Alireza
Rafiei, Mostafa Kardan, Rezvan Yazdian-Robati, Saeed Emami
PII:
S0223-5234(20)30451-7
DOI:
https://doi.org/10.1016/j.ejmech.2020.112480
EJMECH 112480
Reference:
To appear in: European Journal of Medicinal Chemistry
Received Date: 20 March 2020
Revised Date: 15 May 2020
Accepted Date: 16 May 2020
Please cite this article as: M. Ashooriha, M. Khoshneviszadeh, M. Khoshneviszadeh, A. Rafiei,
M. Kardan, R. Yazdian-Robati, S. Emami, Kojic acid–natural product conjugates as mushroom
tyrosinase inhibitors, European Journal of Medicinal Chemistry (2020), doi: https://doi.org/10.1016/
j.ejmech.2020.112480.
This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition
of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of
record. This version will undergo additional copyediting, typesetting and review before it is published
in its final form, but we are providing this version to give early visibility of the article. Please note that,
during the production process, errors may be discovered which could affect the content, and all legal
disclaimers that apply to the journal pertain.
© 2020 Published by Elsevier Masson SAS.

Graphical Abstract

Kojic acid–natural product conjugates as mushroom tyrosinase inhibitors
Morteza Ashooriha a, Mehdi Khoshneviszadeh b,c, Mahsima Khoshneviszadeh c, Alireza
Rafiei d, Mostafa Kardan d, Rezvan Yazdian-Robati e, Saeed Emami*,f
aDepartment of Medicinal Chemistry, Student Research Committee, Faculty of Pharmacy,
Mazandaran University of Medical Sciences, Sari, Iran
b Department of Medicinal Chemistry, Faculty of Pharmacy, Shiraz University of Medical Sciences,
Shiraz, Iran
c Medicinal and Natural Products Chemistry Research Center, Shiraz University of Medical Sciences,
Shiraz, Iran
d Department of Immunology and Molecular and Cell Biology Research Center, Faculty of Medicine,
Mazandaran University of Medical Sciences, Sari, Iran
e Molecular and Cell Biology Research Center, Faculty of Medicine, Mazandaran University of
Medical Sciences, Sari, Iran
f Department of Medicinal Chemistry and Pharmaceutical Sciences Research Center, Faculty of
Pharmacy, Mazandaran University of Medical Sciences, Sari, Iran
Abstract: As part of our effort to develop potential tyrosinase inhibitors, we have
conjugated the well-known tyrosinase inhibitor kojic acid (KA) with several phenolic natural
products such as umbelliferone, sesamol, thymol, carvacrol, eugenol, isoeugenol, vanillin,
isovanillin, and apocynin that some reports have shown their activity on tyrosinase enzyme.
The designed compounds were synthesized using click reaction and 1,2,3-triazole formation.
All compound showed potent anti-tyrosinase activity significantly higher than KA. The best
activities were observed with apocynin and 4-coumarinol analogs (10c and 16c) displaying
IC50 values of 0.03 and 0.02 µM, respectively. The potency of 16c was >460-times more
than that of KA. Cell-based assays against B16F10 and HFF cells revealed that the
representative compounds can efficiently suppress the melanogenesis without
significant toxicity on cells.
Keywords: Natural products; kojic acid; hyperpigmentation; tyrosinase inhibitor
*Correspondence: Saeed Emami, Department of Medicinal Chemistry, Faculty of Pharmacy,
Mazandaran University of Medical Sciences, Sari, Iran
E-mail: sdemami12@gmail.com; semami@mazums.ac.ir
1

1. Introduction
Tyrosinase (also known as monophenol and polyphenol oxidase) is a type III copper-
containing enzyme that is responsible for synthesis of natural pigments (e.g., melanins) in
wide range of organisms and microorganisms, such as, bacteria, fungi, insects, plants and
animals [1]. Despite the structural differences of the enzyme in various types of organisms,
the structure and function of the active site are similarly preserved. The process of melanin
biosynthesis begins with oxidation of tyrosine amino acid by the tyrosinase enzyme in two
steps, producing L-DOPA and dopaquinone, respectively [2].
Generally, various functions are attributed to melanin in different species; such as UV
absorption, free radical scavenging activity [3], skin, eyes and hair coloration, fruits and
vegetables pigmentation [4], as well as defense function in insects and fungus [5]. However,
the increase of tyrosinase activity and expression in the human melanocytes lead to disorders
associated with hyperpigmentation, including senile lentigines, melasma, and freckles.
Furthermore, fruits and vegetables browning is related to the tyrosinase activity [6].
Accordingly, the discovery and development of tyrosinase inhibitors have great values in
cosmetic and food industries.
Several natural and synthetic compounds, such as hydroquinone, arbutin, azelaic acid and
kojic acid (KA, 1a), have been widely used as whitening agents in pharmaceuticals and
cosmetics products [7]. Certainly, KA is one of the most important compounds used in the
cosmetic products [8]. Nevertheless, this drug has profound drawbacks like skin irritation,
mutagenic effect on mammalian cells, poor efficacy and low formulation stability. Thus,
many efforts have been recently made to improve potency and safety profile of tyrosinase
inhibitors by modifying kojic acid structure and designing kojic acid-derived analogs or
hybrids [9, 10].
On the other hand, a survey on literatures revealed that numerous natural compounds have in
vitro tyrosinase inhibitory activity. For example, umbelliferone 2a [11, 12], sesamol 3a [13-
15], thymol 4a [16-19], carvacrol 5a [18, 19], eugenol 6a [19, 20], isoeugenol 7a [19, 21],
vanillin 8a [20, 22, 23], and isovanillin 9a [23] are reported compounds as tyrosinase
inhibitors or anti-melanogenic agents (Fig. 1). Despite the benefits of the natural compounds,
they have not yet found their true position in the pharmaceutical products due to their
inadequate potency as tyrosinase inhibitors. As presented in Fig. 1, with the exception of
2

sesamol, these natural compounds are found to be less potent than KA against tyrosinase
enzyme.
Hybridization of different bioactive molecule into a single entity is an attractive approach in
the drug design and discovery, especially amongst the natural compounds [24]. An important
type of hybridization is design of non-cleavable hybrids, in which two distinct bioactive
compounds are conjugated by a metabolically-stable linker [25].
As part of our effort to develop potential tyrosinase inhibitors, we have designed novel
conjugated analogs of KA (1a), and above mentioned natural compounds (2a-9a) via click
reaction. Previous studied indicated that the 3-hydroxypyran-4-one part of KA mimics the
native substrates of tyrosinase enzyme (tyrosine and L-DOPA) and can enter in the active site
and coordinate with the copper ion [9]. Thus for designing new KA analogs, we used
alcoholic tail of KA for modification and conjugation with other natural molecules.
Promisingly, the rational combination of KA (1a) with a component of natural origin may
lead to desired hybrid molecules with improved pharmacokinetics and pharmacodynamic
profiles, which synergistically augment potency of each other and reduce the dosage.
Figure 1. Structures of kojic acid and some natural compounds with tyrosinase inhibitory activity.
3

Thus we report here, the synthesis and biological evaluation of KA conjugates (2c-16c)
bearing diverse scaffolds of natural products and related compounds as potent tyrosinase
inhibitors.
2. Results and discussion
2.1. Chemistry
The conjugated analogs of KA (1a) and phenolic natural products (2a-11a) were synthesized
by click reaction and formation of triazole ring (Scheme 1). To carry out this type of reaction,
kojic azide (1b) and appropriate terminal alkyne 2b-11b were needed. The kojic azide (1b)
was synthesized from KA (1a) in two steps as reported previously [10]. On the other hand,
the natural compounds 2a-11a were propargylated with propargyl bromide in the presence of
K2CO3 as a base, in acetone or DMF as solvent (Scheme 1). In the final step, the click
reaction of O-propargylated phenols 2b-11b with key intermediate 1b was performed in a
mixture of solvents containing tetrahydrofuran, water and tert-butyl alcohol in a 10:3:1 ratio,
to afford compounds 2c-11c. The required Cu(I) for the click reaction was provided in situ by
the reaction between sodium ascorbate and copper sulfate pentahydrate (Scheme 1).
4

O
O
O
a
HO
HO
OH
N3
b
O
1b
Kojic acid (1a)
N
c
O
O
O
e
N
O
N
HO
O
O
O
O
O
2c
HO
Umbelliferone (2a)
2b
O
O
O
N
O
HO
O
O
O
N
e
O
d
N
3c
4c
O
O
HO
Sesamol (3a)
3b
O
N
O
N
O
HO
O
e
d
N
HO
O
4b
Thymol (4a)
N
O
HO
O
5c
N
d
e
O
N
HO
O
5b
O
O
Carvacrol (5a)
R
N
N
HO
R
O
O
O
d
N
e
R
O
HO
O
6c-11c
6b-11b
Eugenol (6a): R= 4-allyl
Isoeugenol (7a): R= 4-(prop-1-en-1-yl)
Vanillin (8a): R= p-formyl
Isovanillin (9a): R= m-formyl
Apocynin (10a): R= p-acetyl
Syringaldehyde (11a): R= o-methoxy-p-formyl
Reagents and conditions: (a) thionyl chloride, rt, 1 h; (b) NaN3, DMF, rt, 24 h; (c) propargyl bromide, K2CO3,
DMF, rt, 12 h; (d) propargyl bromide, K2CO3, acetone, reflux, 8 h; (e) compound 1b, THF, t-BuOH, H2O,
CuSO4 .5H2O, sodium ascorbate, rt, 1 h.
Scheme 1. The synthesis of compounds 2c–11c.
5

For further structure-activity relationship study, we have also synthesized a series of analogs
12c–16c as illustrated in Scheme 2. The synthetic strategy was the same as method explained
for compounds 2c–11c. It should be noted that the propargyl ether of para-hydroxy benzoic
acid (compound 15b) was prepared from hydrolysis of related methyl ester 14b. Direct
reaction of para-hydroxy benzoic acid with propargyl bromide could produce a mixture of
products due to the O-propagylation of phenolic and carboxylic acid groups.
O
O
O
R
b
a
N
R
O
R
N
O
N
O
HO
12b:
13b:
14b: R= OMe
=
R H
12c:
13c:
=
R H
12a:
13a:
14a: R= OMe
=
R H
HO
=
R Me
O
=
R Me
=
R Me
14c: R= OMe
15c: R= OH
R= OMe
O
c
b
OH
O
15b
O
OH
O
N
O
O
N
b
d
O
N
O
O
O
O
HO
O
16c
16b
16a
Reagents and conditions: (a) propargyl bromide, K2CO3, DMF, rt, 12 h; (b) compound 1b, THF, t-BuOH,
H2O, CuSO4 .5H2O, sodium ascorbate, rt, 1 h; (c) MeOH, KOH, reflux, 6 h; (d) propargyl bromide, K2CO3,
acetone, reflux, 8 h.
Scheme 2. The synthesis of compounds 12c–16c.
2.2. Anti-tyrosinase activity
The anti-tyrosinase activity of synthesized compounds 2c-16c was evaluated against
mushroom tyrosinase using L-DOPA as substrate. The obtained IC50 values of the test
compounds are listed in Table 1 and compared to that of KA as reference drug.
6

Interestingly, all synthesized compounds exhibited potent tyrosinase inhibitory activity with
IC50 values ≤3.75 µM, being more potent than KA (IC50= 9.28 µM). Certainly, sesamol,
carvacol, apocynin, and 4-hydroxycoumarin (4-coumarinol) derivatives (3c, 5c, 10c and 16c)
showed very potent activity in the range of 0.02-0.07 µM. These compounds were 133-464
times more potent than KA. The 4-coumarinol analog 16c was the most active compound.
Structurally, the compounds 4c and 5c, 6c and 7c, 8c and 9c, 2c and 16c are the regioisomers
of each other. Compounds 5c and 16c were 10-fold more potent than their corresponding
regioisomers 4c and 2c, respectively.
The difference between eugenol and isoeugenol derivatives (6c and 7c) is the location of
double bond in the side chain attached to the aromatic ring. The isoeugenol derivative 7c was
2-fold more active than eugenol analog 6c. However, isomers 8c and 9c derived from vanillin
and isovanillin had similar potency against tyrosinase. The comparison of compounds 8c and
10c revealed that the replacement of formyl with acetyl on the methoxyphenoxy moiety
dramatically increased the anti-tyrosinase activity. Indeed, acetyl analog 10c was 66-times
more active than formyl counterpart 8c. Notably, among the methoxylated compounds 6c-
11c, the apocynin derivative 10c showed the highest activity, indicating the preference of
acetyl group compared to the other type of substituents. While the removal of o-methoxy
group in compound 10c resulted in the compound 13c with diminished activity, the
elimination of o-methoxy from 8c could produce simpler compound (12c) with slightly better
activity. In contrast, the introduction of the second MeO group on the ortho position in
compound 8c had negative effect on activity (compare 11c with 8c).
As described above, to study the structural features of prototype compound 10c, we
synthesized simpler phenoxy compounds 12c-15c with different substituents on the para-
position. Among these derivatives, the methyl ester analog 14c had superior activity and the
carboxylic acid derivative 15c was the less potent compound. Nevertheless, these
modifications could not result in more potent compound respect to 10c. Another modification
was isomerization of coumarin derivative 2c. As seen from data of the coumarin derivatives
(2c and 16c), the displacement of connection position on coumarin core from 7 to 4
significantly increased the inhibitory activity.
7

Table 1. The IC50 values of compounds 2c-16c against tyrosinase enzyme.
Compound
2c
R
IC50 (µM) a
0.23 ± 0.03
0.06 ± 0.02
0.69 ± 0.08
3c
4c
0.07 ± 0.03
5c
1.07 ± 0.16
0.52 ± 0.24
1.99 ± 0.08
6c
7c
8c
1.95 ± 0.45
0.03 ± 0.11
3.75 ± 1.03
9c
10c
11c
1.05 ± 0.68
1.21 ± 0.84
0.46 ± 0.06
3.13 ± 1.40
0.02 ± 0.001
12c
13c
14c
15c
16c
9.28 ± 2.04
Kojic acid
a Each experiment was conducted in triplicate and the IC50 values were presented as mean ± SE.
8

The mechanism of inhibition of the most potent compounds 5c, 10c and 16c were
investigated by determining the kinetic behavior of tyrosinase for the oxidation of L-DOPA.
Accordingly, the inhibition constants and inhibition types on mushroom tyrosinase activity
were determined for compounds 5c, 10c and 16c (Figs 2-4 and Table 2).
The kinetic studies of the enzyme by the Lineweaver–Burk plotting of 1/V versus 1/[S] in the
presence of different inhibitors concentrations gave a series of straight lines as shown in
Figures 2a-4a. The slope of lines was significantly increased by the increment of inhibitor
concentration. The results of these experiments showed that when the concentrations of
compounds 5c, 10c, and 16c were increased, the corresponding Vmax values decreased while
the related Km values increased.
These behaviors of tyrosinase enzyme exposed to kojic acid derivatives 5c, 10c, and 16c,
indicated that all compounds inhibit tyrosinase by two different mechanism; forming
enzyme–inhibitor (EI) complex competitively, and interrupting enzyme–substrate–inhibitor
(ESI) complex in non-competitive manner.
The secondary plots of slopes versus concentration of compounds 5c, 10c, and 16c was used
for determination of EI dissociation constants KI (Figures 2b-4b). Also, the ESI dissociation
constants KIS were determined by secondary plots of intercept versus concentration of
compounds 5c, 10c, and 16c (Figures 2c-4c). The obtained kinetic parameters were also listed
in Table 2. As seen from data, all compounds showed mixed-type inhibition. However, a
lower value of KI in comparison with KIS indicated that there was stronger binding between
enzyme and compounds 5c, 10c, and 16c which suggested preferred competitive over
noncompetitive manner.
9

a
c
Figure 2. The kinetic behaviors of tyrosinase for the oxidation of L-DOPA in the presence of compound 5c: (a)
Lineweaver–Burk plot for the inhibition of tyrosinase by 5c at the concentrations of 0, 0.025, 0.05 and 0.1 µM.
The concentrations of substrate L-DOPA were 0.1, 0.25, 0.5, 0.75 and 1.5 mM; (b) The plot of slopes versus
inhibitor (5c) concentrations to determine KI; (c) The plot of vertical intercepts versus inhibitor (5c)
concentrations to determine KIS.
10

a
b
Figure 3. The kinetic behaviors of tyrosinase for the oxidation of L-DOPA in the presence of compound 10c.
(a) Lineweaver–Burk plot for the inhibition of tyrosinase by 10c at the concentrations of 0, 0.025, 0.05 and 0.1
µM. The concentrations of substrate L-DOPA were 0.1, 0.25, 0.5, 0.75 and 1.5 mM; (b) The plot of slopes
versus inhibitor (10c) concentrations to determine KI; (c) The plot of vertical intercepts versus inhibitor (10c)
concentrations to determine KIS.
11

a
b
c
Figure 4. The kinetic behaviors of tyrosinase for the oxidation of L-DOPA in the presence of compound 16c.
(a) Lineweaver–Burk plot for the inhibition of tyrosinase by 16c at the concentrations of 0, 0.025, 0.05 and 0.1
µM. The concentrations of substrate L-DOPA were 0.1, 0.25, 0.5, 0.75 and 1.5 mM; (b) The plot of slopes
versus inhibitor (16c) concentrations to determine KI; (c) The plot of vertical intercepts versus inhibitor (16c)
concentrations to determine KIS.
12

Table 2. Kinetic parameters of the mushroom tyrosinase for the oxidation of L-DOPA in the presence of
different concentration of selected inhibitors 5c, 10c and 16c.
Conc. (μM) Vmax (ΔOD475 nm/min) Km (mM)
Inhibition type
KI (μM)
KIS (μM)
Compound
5c
0
0.267
0.241
0.219
0.192
0.257
0.232
0.208
0.183
0.259
0.210
0.176
0.122
1.413
1.517
1.831
2.016
1.403
1.582
1.916
2.113
1.406
1.712
2.018
2.246
0.025
0.05
0.1
Mixed-inhibition
0.09
0.21
0
10c
16c
0.025
0.05
0.1
Mixed-inhibition
Mixed-inhibition
0.07
0.04
0.19
0
0.025
0.05
0.1
0. 12
2.3. Effect of copper (II) on the tyrosinase inhibition of compound 16c
The most active compounds 16c was selected for Cu-binding study to investigate the ability
of the inhibitor for the specific binding to the binuclear copper center in the active site of
tyrosinase enzyme [26]. In this test, the enzyme was fully inhibited by the highest
concentration of 16c, followed by treating with a range of CuSO4 concentrations (0.005–1.2
mM). The catalytic activity of tyrosinase was observed before and after addition of copper
ions at 475 nm. The results indicated that the activity of the pre-inhibited tyrosinase was
recovered by increasing concentration of copper ions. It should be noted that the alone effect
of Cu2+ on tyrosinase was also measured by treating the enzyme with different concentrations
of Cu2+. Indeed, compound 16c at the concentration of 0.5 μM, maximally inhibited the
tyrosinase by showing a ΔAbs475 nm of 0.021/min (Fig. 5), when compared to 0.09/min for
control (no inhibitor). The maximum concentration of Cu2+ (1.2 mM) resulted in the
inversion of the tyrosinase activity by about 85%. Based on the obtained result, it could
13

propose that the complexation of compound 16c with the excess of copper ions can prohibit
the binding of inhibitor to the copper center of the active site.
90
(a)
80
70
60
50
40
30
20
10
0
0
0.2
0.4
0.6
0.8
1
1.2
CuSO4 mM
0.1
0.09
0.08
0.07
0.06
0.05
0.04
0.03
0.02
0.01
0
(b)
0
0.2
0.4
0.6
0.8
1
1.2
1.4
CuSO4 mM
Figure 5. Effect of copper ion addition on the inhibition of compound 16c against tyrosinase enzyme; a)
Inhibition of tyrosinase activity by CuSO4 without drug; b) Recovery of tyrosinase activity pre-inhibited by 16c,
after addition of CuSO4.
2.4. Study of antioxidant activity
Since tyrosinase catalyzes the oxidation reactions, antioxidant compounds also can exhibit
tyrosinase inhibitory activity without interaction with enzyme [27]. Hence, assessment of
antioxidant activity can improve our understanding about the mechanism action of the title
14

compounds. The antioxidant potential of compounds 2c-16c was evaluated by testing their
radical scavenging activity using DPPH method. The target compounds at 10 µM
concentration showed no significant radical scavenging activity against DPPH. Therefore,
tyrosinase inhibitory activity of these derivatives cannot be as a result of antioxidant activity
and this property is due to direct inhibition of enzyme.
2.5. Melanin content assay
The effect of the selected compounds 5c and 16c on the melanin production was evaluated in
invasive melanoma B16F10 cell line. In this experiment, B16F10 cells were treated with 5c
and 16c (5 µM), and KA (100 µM) and compared to the control group (untreated). The
percentages of melanin content for 5c and 16c and KA are presented as mean ± SD of
triplicate samples relative to control. As shown in Fig. 6, compounds 5c and 16c at 5 µM
concentration significantly reduced the melanin content in skin melanoma cells. In particular,
the efficacy of 5c at the concentration of 5 µM was equal to that of KA at 100 µM.
Figure 6. Effect of compounds 5c and 16c (5 µM) on melanin content in B16F10 murine melanoma cells as
compared to positive KA (kojic acid, 100 µM) and negative control. ****Statistically significant difference with
control (P<0.001)
2.6. Evaluation of cytotoxic effect on B16F10 and HFF cells
Compounds 5c, 10c and 16c with excellent inhibition potential on tyrosinase were selected to
study their cytotoxicity on melanoma (B16F10) and human foreskin fibroblast (HFF) cells by
15

MTT assay [28]. As shown in Table 3, the IC50 values of tested compounds were compared to
those of KA. The IC50 value of 16c against B16F10 cell was the same as that of KA. The
cytotoxicity of 5c and 10c was even less than that of KA. In addition, all tested compounds
had very weak cytotoxicity over normal cells HFF (IC50s ≥193 µM). As described above,
compounds 5c, 10c and 16c display anti-tyrosinase activity at very low concentrations, thus
at those level have no significant toxicity against cells, indicating better toxicity profile of
them respect to KA.
Table 3. The IC50 values of tyrosinase ihhibitors 5c, 10c, and 16c in comparison with KA against B16F10 and
HFF cells.
Compound
B16F10
408.0
415.9
143.2
144.4
HFF
193.0
212.6
194.1
>150
5c
10c
16c
KA
2.7. Docking study
In order to rationalize the obtained anti-tyrosinase activity data, the most potent hybrid
compound 16c along with two parent compounds (KA and 4-coumarinol) were subjected to
docking into the active site of tyrosinase. The 3D-structure of mushroom tyrosinase (2Y9X)
was obtained from the Protein Data Bank. Firstly, to check validity of the docking protocol,
the co-crystallized ligand was re-docked into the active site of tyrosinase. The root mean
square deviation (RMSD) of re-docked ligand and native ligand was 1.02 Å.
The results of docking for compound 16c were illustrated as 2D and 3D presentations in Fig.
7. Since the linker between pyranone moiety and coumarin ring has enough flexibility, the
molecule is folded in the active site. The carbonyl group of pyran-4-one is coordinated with
Cu ions in active site, playing essential role in the binding of inhibitor to the target enzyme.
Additionally, this group forms hydrogen bond with His 85. The middle of molecule
containing 1,2,3-triazole ring can make hydrogen bond with the His 244 with 2.82 Å of
distance. The coumarin part of 16c locates into the hydrophobic pocket, forming hydrophobic
16

interactions with Glu322, Cys83, Asn81, and Thr324 residues. The calculated binding free
energy of compound 16c to the tyrosinase was −7.40 kcal/mol.
Figure 7. 2D and 3D presentations for the binding mode of compound 16c in the active site of tyrosinase
(2Y9X).
On the other hand, the best docking pose of KA (1a) and 4-coumarinol (16a) in the active site
of tyrosinase is illustrated in Fig. 8. In the case of KA, one challenge was the number of
poses that can be generated, due to the small size of molecule that can bind in multiple
different fashions. However, the selected pose in Fig. 8 in accordance with experimental data
17

due to the fact that the coordination with copper center in the active site of tyrosinase enzyme
is essential. Furthermore, the carbonyl group forms hydrogen bond with His85. As depicted
in Fig. 8, other parent compound 4-coumarinol (16a) can accommodate in the proximity of
copper center in the active site, engaging the oxygen of carbonyl with Cu ions. In addition the
oxygen of 4-hydroxy group is involved in H-bonding with Asn260.
Figure 8. 2D and 3D presentations for the binding modes of KA (1a) (left) and 4-coumarinol (16a) (right) in the
active site of tyrosinase (2Y9X).
The comparison of docking modes of hybrid compound 16c and its parent compounds
revealed that the 3-hydroxypyran-4-one part of 16c mimics KA and other parts of the
inhibitor including triazole linker and coumarin ring are involved in the interactions with
18

active site and help for the strong inhibition of tyrosinase enzyme. As mentioned above, the
free energy of binding in the active site of tyrosinase for hybrid compound 16c was estimated
to be −7.40 kcal/mol, which is significantly higher than those of KA (−4.09 kcal/mol) and 4-
coumarinol (−5.39 kcal/mol). Therefore, the higher potency of compound 16c may be
attributed to the additional interactions of the triazole linker and coumarin ring with the target
site.
Since the syringaldehyde–kojic acid conjugate (compound 11c) was the less active one
among the tested compounds, thus we also investigated its docking mode and compared to
that of the most potent compound 16c (Fig. 9).
a
b
Figure 9. (a) 3D presentations for the binding mode of compound 11c in the tyrosinase active site; (b)
superimposition of the docking poses of compound 11c (as less potent compound) and 16c (as the most potent
compound) in the active site of tyrosinase (2Y9X).
19

The binding energy of compound 11c with tyrosinase was found to be −5.74 kcal/mol which
was significantly lower than that of compound 16c (−7.40 kcal/mol). However, the less
potent compound (11c) showed better interactions and more favorable binding energy when
compared to the parent compound KA.
The superimposed docking-poses of compound 11c and 16c in Fig. 9b indicated that while
the hydroxypyran-4-one (kojic residue) of both compounds has same orientation and position
in the active site close to the copper center, the conformation of the triazolic linker and the
interactions of pendent aryloxy are different for 11c and 16c. These findings further
demonstrated the importance of the triazolic linker and the aryloxy part in the more favorable
interactions with tyrosinase and high potency of the prototype compound 16c.
3. Conclusion
We have coupled a series of natural products with reported in vitro activity on tyrosinase to
the well-known tyrosinase inhibitor kojic acid (KA) via click reaction. The phenolic natural
products umbelliferone, sesamol, thymol, carvacrol, eugenol, isoeugenol, vanillin, isovanillin,
and apocynin and some related phenolic and enolic compounds including syringaldehyde, 4-
hydroxybenzaldehyde, 4-hydroxyacetophenone, methylparaben, and 4-hydroxycoumarin (4-
coumarinol) have been conjugated to the side chain of KA. All analogs showed potent anti-
tyrosinase activity (IC50 values = 0.02-3.75 µM), significantly greater than KA (IC50 value =
9.28 µM). Certainly, the conjugates of umbelliferone, sesamol, thymol, carvacrol, isoeugenol,
apocynin, methylparaben, and 4-coumarinol showed sub-miromolar activities (IC50s < 1 µM).
The apocynin and 4-coumarinol analogs (10c and 16c) with IC50 values of 0.03 and 0.02 µM,
were the most potent compounds. The potency of 16c was >460-times more than that of
KA.
The SAR study by comparing the highly potent compound 10c with its analogs such as 8c,
9c, 11c, and 12c-15c revealed the importance of ortho-methoxy and para-acetyl groups on
aryl residue of the target compound. Furthermore, the 4-oxycoumarin derivative 16c was
>10-times more active than its regioisomer 2c. DPPH assay showed that the target
compounds had no significant radical scavenging activity. Thus, the potent inhibition of
tyrosinase by these derivatives cannot be as a result of antioxidant activity. Addition of Cu(II)
ions resulted in the activity reversal of the inhibited enzyme, indicating the binding of
20

inhibitor (16c) to the copper center of the active site. Cell-based assays against B16 and HFF
cells revealed that the representative compounds can efficiently suppress the
melanogenesis without significant toxicity on cells. Results of this study indicate that the
design of twin drug approach is good strategy for discovery and development of potent
tyrosinase inhibitors especially derived from KA and natural compounds.
4. Experimental
All reagents and solvents were obtained from local commercial sources and used without any
further purification. The starting materials 1a-14a and 16a were purchased from Merck
Company. Azidokojic acid (1b) was prepared from KA according to the literature method
[10]. Thin-layer chromatography (TLC, pre-coated with silica gel 60 F254 on aluminum
sheet) was used to monitor completion of reactions. TLC visualization and detection of spots
were carried out by UV lamp (254 nm). The instruments used for characterizations were
mentioned in Supplementary Material.
4.1. General procedure for the preparation of propargylated compounds 3b-14b
A mixture of appropriate phenolic compound (3a-14a, 1.0 eq.) and K2CO3 (1.2 eq.) in
acetone was stirred, Then, propargyl bromide (1.2 eq.) was added and the reaction was
refluxed with stirring for 8 h. After filtration and separation of suspended salts, the solvent of
filtrate was evaporated to give the O-propargyated product (3b-14b). These intermediates
were used for click reaction without further purification.
4.2. General procedure for the preparation of propargyloxy coumarins 2b and 16b
7-Hydroxycoumarin or 4-hydroxycoumarin (2a or 16a, 1.0 eq.) were added to a stirring
mixture of K2CO3 (1.2 eq.) in DMF, followed by adding propargyl bromide (1.2 eq.). The
reaction mixture was stirred at room temperature for 12 h. After completion of the reaction,
water (30 mL) was added and the resulting mixture was extracted using ethyl acetate/n-
hexane (4:1). The organic layers were washed with saturated brine and the volatile solvents
were evaporated under reduced pressure. The obtained product (2b and 16b) was used for the
next step without further purification.
21

4.3. Preparation of 4-(prop-2-ynyloxy) benzoic acid (15b)
The methyl ester 14a (613 mg, 3.0 mmol) was dissolved in THF (20 mL) at room
temperature and then treated with KOH solution (840 mg, 15 mmol in 15 mL methanol).
After the reaction mixture was stirred at room temperature overnight, solvent was evaporated
under vacuum. Then, the concentrated sample was acidified with aqueous HCl (6.0 M) and
extracted with ethyl acetate (3×15 mL). The combined organic layers were washed with brine
and dried over Na2SO4. After evaporation of the solvent, the product was obtained as a white
crystalline solid with 78% yield.
4.4. Typical procedure for the preparation of 1,2,3-triazole derivatives 2c-16c
Synthesis of 7-((1-((5-Hydroxy-4-oxo-4H-pyran-2-yl)methyl)-1H-1,2,3-triazol-4-yl)methoxy)-
2H-chromen-2-one (2c). Firstly, the required catalyst copper (I) was prepared by adding
copper (II) sulfate pentahydrate (CuSO4.5H2O, 30 mg, 10 mol%) to the solution of sodium
ascorbate (95.1 mg, 40 mol%) in deoxygenated water (3 mL). After dissolving of
CuSO4.5H2O and alteration the color of solution to green, it was ready for next step. The
propargyl derivative 7-(prop-2-yn-1-yloxy)-2H-chromen-2-one (2b, 1.0 mmol) and azide 1b
(1.0 mmol) were dissolved in a mixed solvents THF (10 mL) and tert-butyl alcohol (1 mL).
Then, the pre-prepared solution of Cu (I) was added dropwise to the clickable reactants
solution and the mixture was stirred at r.t. for 1h. After confirming the completion of the
reaction by TLC, the solvents were removed by evaporation under reduced pressure. The
obtained residue was mixed with water (30 mL, containing 10% EDTA disodium salt) and
extracted with ethyl acetate (3×15 mL). The collected organic phases were washed with brine
and dried (Na2SO4). After solvent evaporation under reduced pressure, the product was
triturated with a small amount of cold acetone and recrystallized from methanol to give pure
1
final compound 2c. White solid; yield 83%; m.p. 205-207 °C; H NMR (500 MHz, DMSO-
d6) δ: 9.29 (br s, 1H, OH), 8.39 (s, 1H, H-6 pyran), 8.05 (s, 1H, triazole), 7.99 (d, 1H, J = 9.5
Hz, H-4 coumarin), 7.64 (d, 1H, J = 8.5 Hz, H-5 coumarin), 7.16 (d, 1H, J = 1.5 Hz, H-8
coumarin), 7.02 (dd, 1H, J = 8.5 and 1.5 Hz, H-6 coumarin), 6.41 (s, 1H, H-3 pyran), 6.30 (d,
13
1H, J = 9.5 Hz, H-3 coumarin), 5.63 (s, 2H, OCH2), 5.29 (s, 2H, NCH2). C NMR (125
MHz, DMSO-d6) δ: 174.07, 161.52, 160.97, 160.70, 155.76, 146.51, 144.72, 142.92, 140.49,
129.99, 126.20, 113.57, 113.36, 113.19, 113.11, 102.06, 62.00, 50.48. MS (m/z, %): 367 (M+,
6), 178 (20), 162 (18), 134 (62), 105 (41), 89 (25), 67 (63), 41 (100). Anal. Calcd for
C18H13N3O6: C, 58.86; H, 3.57; N, 11.44. Found: C, 58.82; H, 3.56; N, 11.53.
22

The physicochemical and spectroscopic data of other compounds 3c-16c could be found in
Supplementary material.
4.5. Tyrosinase inhibition assay
Mushroom tyrosinase (EC 1.14.18.1) was obtained from Sigma Chemicals as lyophilized
powder (≥1000 unit/mg solid). The sample solutions (40 mM) of compounds 2c-16c as well
as KA were provided in DMSO and subsequently diluted with PBS (pH= 6.8) to obtain
different concentrations. The DMSO concentration in the final test solutions was up to 2.0%.
L-DOPA was used as a substrate for enzymatic inhibition assay and dopachrome formation
was detected at 475 nm by an ELISA reader (BioTek). The protocol used for tyrosinase
inhibition assay was according to reported method [10, 28]. Each assay was conducted as
triplicate and the concentration required for 50% inhibition was determined as IC50 value.
4.6. Kinetic analysis of tyrosinase inhibition
A series of experiments were performed to determine the inhibition kinetics of derivatives 5c,
10c and 16c by using the method reported previously [10]. The inhibitor concentrations were
0, 0.025, 0.05 and 0.1 μM for all compounds. The concentration of L-DOPA as a substrate
was between 0.1 and 1.5 mM in all kinetic studies. Pre-incubation and measurement time was
the same as described in mushroom tyrosinase inhibition assay protocol. Maximal initial
velocity was determined from initial linear portion of absorbance up to 10 minutes after
addition of enzyme at a 30 s interval. The inhibition type was assayed by Lineweaver–Burk
plots of inverse of velocities (1/V) versus inverse of substrate concentration (1/[S]). The EI
dissociation constant KI was determined by secondary plot of slope versus inhibitors
concentrations while ESI dissociation constant KIS was determined by plotting of intercept
versus inhibitors concentrations.
4.7. Antioxidant assay
The radical scavenging activity of the target compounds 2c-16c was evaluated by using
diphenylpicrylhydrazine (DPPH) method as reported previously [10, 29].
23

4.8. MTT cell viability assay
The cytotoxicity of representative compounds 5c, 10c and 16c was studied against B16F10
melanoma and Human Foreskin Fibroblast (HFF) cells, obtained from Pasture Institute, Iran.
The test compounds and KA were dissolved in DMSO and then diluted with the desired
medium (RPMI-1640 or DMEM) to the required concentrations. The viability of cells after
exposure with compounds was determined by MTT (3-(4,5-dimethylthiazol-2-yl)-2,5-
diphenyltetrazolium bromide) as described previously [28]. The absorbance of produced
formazan dyes was recorded by an ELISA reader (BioTek) at 570 nm. All experiments were
conducted in triplicate and the IC50 values were calculated using Prism 6, GraphPad
Software.
4.9. Determination of melanin content
The effect of compounds 5c and 16c on melanin content was studied in Murine B16F10
melanoma cells [30]. The obtained B16F10 cells from Pasteur Institute (Iran, Tehran) were
seeded in a 12-well plate overnight in RPMI media supplemented with 10% fetal bovine
serum (FBS; Gibco, USA), penicillin (100 units/mL), and streptomycin (0.1 mg/mL). Then
compounds 5c and 16c were added to the cell culture to reach the concentration of 5 µM.
After 24 h, the cells were harvested by trypsin and washed with PBS. The obtained cell
pellets were treated with aqueous NaOH (2 M, 100 μL) and incubated at 100 °C for 0.5 h.
The absorption of each well was recorded at 405 nm using a Synergy H4 Hybrid Multi-Mode
Microplate Reader (BioTek). Three wells containing all components without drug were used
as control.
4.10. Computational study
Docking study was performed using AutoDock 4.2 software according to the previously
described method [10]. Discovery Studio and LIGPLOT plus were used for 2D and 3D
visualizations.
24

Acknowledgments
This work was supported by a grant (No. 4924) from the Research Council of Mazandaran
University of Medical Sciences, Sari, Iran.
References
[1] R. Burke, J. Cairney, Mycorrhiza, 12 (2002) 105-116.
[2] M. Goldfeder, M. Kanteev, S. Isaschar-Ovdat, N. Adir, A. Fishman, Nature
Communications, 5 (2014) 1-5.
[3] G. Goodman, D. Bercovich, Medical Hypotheses, 71 (2008) 190-202.
[4] F. Solano, S. Briganti, M. Picardo, G. Ghanem, Pigment Cell Research, 19 (2006) 550-
571.
[5] S. Seyedmousavi, M.G. Netea, J.W. Mouton, W.J. Melchers, P.E. Verweij, G.S. de Hoog,
Clinical Microbiology Reviews, 27 (2014) 527-542.
[6] E. Roh, J.-E. Kim, J.Y. Kwon, J.S. Park, A.M. Bode, Z. Dong, K.W. Lee, Critical
Reviews in Food Science and Nutrition, 57 (2017) 1631-1637.
[7] T. Pillaiyar, M. Manickam, V. Namasivayam, Journal of Enzyme Inhibition and
Medicinal Chemistry, 32 (2017) 403-425.
[8] M. Saeedi, M. Eslamifar, K. Khezri, Biomedicine & Pharmacotherapy, 110 (2019) 582-
593.
[9] S.M. Hashemi, S. Emami, Pharmaceutical and Biomedical Research, 1 (2015), 1-17.
[10] M. Ashooriha, M. Khoshneviszadeh, M. Khoshneviszadeh, S.E. Moradi, A. Rafiei, M.
Kardan, S. Emami, Bioorganic Chemistry, 82 (2019) 414-422.
[11] S. Asthana, P. Zucca, A.V. Vargiu, E. Sanjust, P. Ruggerone, A. Rescigno, Journal of
Agricultural and Food Chemistry, 63 (2015) 7236-7244.
[12] Z. Ashraf, M. Rafiq, S.-Y. Seo, M.M. Babar, N.-u.-S.S. Zaidi, Journal of Enzyme
Inhibition and Medicinal Chemistry, 30 (2015) 874-883.
[13] S.h. Baek, S.H. Lee, Experimental Dermatology, 24 (2015) 761-766.
25

[14] P.-Y. Wu, Y.-J. You, Y.-J. Liu, C.-W. Hou, C.-S. Wu, K.-C. Wen, C.-Y. Lin, H.-M.
Chiang, International Journal of Molecular Sciences, 19 (2018) 1108.
[15] M. Srisayam, N. Weerapreeyakul, S. Barusrux, K. Kanokmedhakul, J Cosmet Sci, 65
(2014) 69-79.
[16] H. Satooka, I. Kubo, Journal of Agricultural and Food Chemistry, 59 (2011) 8908-8914.
[17] Z. Ashraf, M. Rafiq, S.-Y. Seo, K.S. Kwon, M.M. Babar, European Journal of Medicinal
Chemistry, 98 (2015) 203-211.
[18] N. Brotzman, Y. Xu, A. Graybill, A. Cocolas, A. Ressler, N.P. Seeram, H. Ma, G.E.
Henry, Bioorganic & Medicinal Chemistry Letters, 29 (2019) 56-58.
[19] M.d.M. Garcia-Molina, J.L. Muñoz-Muñoz, F. Garcia-Molina, P.A. García-Ruiz, F.
Garcia-Canovas, Journal of Agricultural and Food Chemistry, 60 (2012) 6447-6453.
[20] S. Shirota, K. Miyazaki, R. Aiyama, M. Ichioka, T. Yokokura, Tyrosinase inhibitors
from crude drugs. Biol Pharm Bull. 17 (1994) 266-269.
[21] Zuo AR, Dong HH, Yu YY, Shu QL, Zheng LX, Yu XY, Cao SW. The antityrosinase
and antioxidant activities of flavonoids dominated by the number and location of phenolic
hydroxyl groups.Chin Med. 2018 Oct 19;13:51. doi: 10.1186/s13020-018-0206-9.
eCollection 2018.
[22] Z. Ashraf, M. Rafiq, S.-Y. Seo, M.M. Babar, Bioorganic & Medicinal Chemistry, 23
(2015) 5870-5880.
[23] A. Rescigno, G.M. Casañola‐Martin, E. Sanjust, P. Zucca, Y. Marrero‐Ponce, Drug
Testing and Analysis, 3 (2011) 176-181.
[24] S. Choudhary, P. K. Singh, H. Verma, H. Singh, O. Silakari, European Journal of
Medicinal Chemistry, 151 (2018) 62-97
[25] R. Morphy, Z. Rankovic, Journal of Medicinal Chemistry, 48 (2005) 6523-6543.
[26] U. Ghani, N. Ullah, Bioorganic & Medicinal Chemistry, 18 (2010) 4042-4048.
[27] S.-Y. Seo, V.K. Sharma, N. Sharma, Journal of Agricultural and Food Chemistry, 51
(2003) 2837-2853.
[28] Y.S. Lee, J.H. Park, M.H. Kim, S.H. Seo, H.J. Kim, Archiv der Pharmazie, 339 (2006)
111-114.
26

[29] K.-H. Lee, F.H.A. Aziz, A. Syahida, F. Abas, K. Shaari, D.A. Israf, N.H. Lajis,
European Journal of Medicinal Chemistry, 44 (2009) 3195-3200.
[30] S.A. Emami, R. Yazdian-Robati, M. Sadeghi, J. Baharara, E. Amini, F. Salek, Z.
Tayarani-Najaran, Research in Pharmaceutical Sciences, 12 (2017) 160-167.
27

Research Highlights
• A series of kojic acid (KA)-natural product conjugates were prepared via click reaction.
• All compounds 2c-16c were highly potent tyrosinase inhibitors (IC50s ≤3.75 µM).
• Apocynin and 4-coumarinol derivatives (10c and 16c) were the most potent compounds.
• The 4-coumarinol analog (16c) was >460-fold more potent than KA.
• Certainly, 16c can suppress melanogenesis without significant toxicity on B16F10 cells.