Kojic acid-amino acid conjugates as tyrosinase inhibitors

Article abstract of DOI:10.1016/j.bmcl.2009.08.041

Kojic acid (KA), a well known tyrosinase inhibitor, has insufficient inhibitory activity and stability. We modified KA with amino acids and screened their tyrosinase inhibitory activity. Among them, kojic acid-phenylalanine amide (KA-F-NH₂) sho

Full text of DOI:10.1016/j.bmcl.2009.08.041

Bioorganic & Medicinal Chemistry Letters 19 (2009) 5586–5589
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Kojic acid–amino acid conjugates as tyrosinase inhibitors
Jin-Mi Noh a, Seon-Yeong Kwak a, Hyo-Suk Seo a, Joo-Hyun Seo b, Byung-Gee Kim b, Yoon-Sik Lee a,
*
a School of Chemical and Biological Engineering, Seoul National University, Seoul 151-744, Republic of Korea
b School of Chemical and Biological Engineering and Institute of Molecular Biology and Genetics, Seoul National University, Seoul 151-744, Republic of Korea
a r t i c l e i n f o
a b s t r a c t
Article history:
Kojic acid (KA), a well known tyrosinase inhibitor, has insufficient inhibitory activity and stability. We
modified KA with amino acids and screened their tyrosinase inhibitory activity. Among them, kojic
acid–phenylalanine amide (KA-F-NH2) showed the strongest inhibitory activity, which was maintained
for over 3 months at 50 °C, and acted as a noncompetitive inhibitor as determined by kinetic analysis.
It also exhibited dopachrome reducing activity. We also propose a new tyrosinase inhibition mechanism
based on the docking simulation data.
Received 13 May 2009
Revised 10 August 2009
Accepted 11 August 2009
Available online 14 August 2009
Keywords:
Kojic acid
Ó 2009 Elsevier Ltd. All rights reserved.
Tyrosinase inhibitor
Stability
Enzyme kinetics
AUTODOCK program
Melanin is a dark pigment produced by skin cells in the inner-
most layer of the epidermis. Melanogenesis is initiated with the
first step of tyrosine oxidation by tyrosinase. When the skin is
exposed to UV radiation, the formation of abnormal melanin pig-
ment occurs, which constitutes a serious esthetic problem that is
particularly prevalent in middle-aged and elderly individuals.1,2
Tyrosinase plays an important role in the pathway of melanin bio-
synthesis from tyrosine. It catalyzes two distinct reactions involv-
improve its properties by converting the C-7 hydroxyl group into
an ester,10 hydroxyphenyl ether,11 glycoside,12 amino acid deriva-
tives, or tripeptide derivatives.13,14,16
Recently, our group introduced various KA-tripeptide deriva-
tives as new tyrosinase inhibitors. In this study, we found that
KA-tripeptide amides, especially KA-FWY-NH2, showed similar
tyrosinase inhibitory activities to those of KA-tripeptide free acids
but exhibited superior storage stability than those of KA and KA-
tripeptide free acids.15,16 During these studies, we noticed the
importance of C-terminal amide form in KA-peptide conjugates.
To find further KA derivatives with higher tyrosinase inhibitory
activity, stability, and synthetic efficiency, we prepared a library
of KA-amino acid amides (KA-AA-NH2) and screened their tyrosi-
nase inhibitory activities.
ing molecular oxygen; the hydroxylation of tyrosine to
L-DOPA as
monophenolase and the oxidation of -DOPA to dopaquinone as
L
diphenolase. Dopaquinone is nonenzymatically converted to dopa-
chrome, and ultimately to dihydroxyindols which cause the pro-
duction of melanin pigments.3 Therefore, the development of
tyrosinase inhibitors is of great concern in the medical, agricul-
tural, and cosmetic fields.4 Among the many kinds of tyrosinase
inhibitors, kojic acid (KA) has been intensively studied. It acts as
a good chelator of transition metal ions such as Cu2+ and Fe3+
and a scavenger of free radicals.5 This fungal metabolite is cur-
rently applied as a cosmetic skin-lightening agent and food addi-
tive to prevent enzymatic browning.6 KA shows a competitive
inhibitory effect on the monophenolase activity and a mixed inhib-
itory effect on the diphenolase activity of mushroom tyrosinase.7,8
However, its use in cosmetics has been limited, because of the skin
irritation caused by its cytotoxicity and its instability during stor-
age. Accordingly, many KA derivatives have been synthesized to
KA was activated with 1,10-carbonyldiimidazole (CDI) and
coupled to the resin-bounded amino acids. After cleavage of the
KA-AA-NH2 from the resin, it was analyzed by HPLC and character-
ized by MALDI-TOF mass spectroscopy.17–19 Twenty kinds of KA-
AA-NH2 were examined for their tyrosinase inhibitory activity on
mushroom tyrosinase (Fig. 1).
As expected, most of them showed better tyrosinase inhibitory
activity than KA itself. Especially, when amino acids which possess
aromatic side chains, such as phenylalanine, tryptophan, tyrosine,
and histidine, were conjugated to KA, its tyrosinase inhibitory activ-
ity was dramatically increased to over 90% at 20
inhibitory activity of KA-F-NH2 toward mushroom tyrosinase was
98.6%, its IC50 value in the inhibition of -DOPA oxidation was
14.7 M and Ki value was 11.0 M. In case of KA, its tyrosinase inhib-
lM. Especially, the
L
* Corresponding author. Tel.: +82 2 880 7073; fax: +82 2 888 1604.
E-mail address: yslee@snu.ac.kr (Y.-S. Lee).
l
l
0960-894X/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2009.08.041

J.-M. Noh et al. / Bioorg. Med. Chem. Lett. 19 (2009) 5586–5589
5587
Figure 1. Tyrosinase inhibitory activities of KA and KA-AA-NH2. The reaction media contained
L-tyrosine (1.67 mM), mushroom tyrosinase (2 units/ll) and tyrosinase
inhibitor (20
lM) in a total volume of 750
l
l of 0.1 M phosphate buffer (pH 6.8). The data were obtained by measuring the UV absorbance at 475 nm.
Figure 2. Lineweaver–Burk plots of mushroom tyrosinase with
L
-DOPA as a substrate in the presence of KA (A), and KA-F-NH2 (B). The data were obtained from triplicate
runs.
Table 1
Effect of KA and KA-AA-NH2 on dopachrome concentration
ABS (475 nm)
Dopachrome production
reducing activityd (%)
No inhibitor (5 min)
No inhibitor (10 min)
KA
KA-F-NH2
KA-W-NH2
0.995a
1.053b
0.973c
0.730c
0.797c
0.910c
0.880c
8
31
24
14
16
KA-Y-NH2
KA-H-NH2
a
Measured after the reaction mixture of tyrosinase and
5 min.
L
-tyrosine reacted for
-tyrosine reacted for
Figure 3. Absorption spectra of reaction solution of: (a) mushroom tyrosinase and
-tyrosine after 10 min of reaction; (b) after 5 min of reaction; (c) after adding KA to
b
Measured after the reaction mixture of tyrosinase and
10 min.
L
L
b; (d) after adding KA-F-NH2 to b.
c
Measured after 5 min of inhibition addition into the reaction mixture of
tyrosinase -tyrosine reacted for 5 min.
L
d
Calculated using the following formula: [(A ꢀ B)/A] ꢁ 100 (A, absorbance of No
itory activity was 18.3%, IC50 value was 722.0
l
M and Ki value was
inhibitor (10 min); B, absorbance of test inhibitor solution).
582.7 lM. Kinetic study of the inhibition of L-DOPA oxidation by
KA and KA-F-NH2 revealed that the Vmax values of mushroom tyros-
inase activity was increased as the concentration of inhibitors was
increased, while the Km values remained unchanged (Fig. 2).20–24
Therefore, KA-F-NH2 is regarded as a non-competitive inhibitor of
mushroom tyrosinase as KA.
intensity of the solution became fainter than before adding. To
evaluate its activity, the reaction mixture was scanned by a spec-
trophotometer (Fig. 3). Two distinct absorption peaks below
350 nm were generated by the reaction of L-tyrosine and L-DOPA
Tyrosinase is an enzyme that catalyzes the hydroxylation of
with tyrosinase. The reaction mixture revealed an absorption peak
at 475 nm which corresponded to the formation of dopachrome.8
L
-tyrosine and the subsequent oxidation of
L-DOPA to dopaqui-
none. When -tyrosine aqueous solution was incubated with tyros-
L
When the reaction solution of tyrosinase and L-tyrosine was incu-
inase at 37 °C, its color turned to red-brown due to dopachrome
formation. When KA was added during the reaction, the color
bated for 10 min, the intensity of the absorption peak at 475 nm
was increased to a greater extent than that of the reaction solution

5588
J.-M. Noh et al. / Bioorg. Med. Chem. Lett. 19 (2009) 5586–5589
tyrosinase inhibitor, the absorption peak at 475 nm, corresponding
to dopachrome formation, was decreased (Table 1). It has been re-
ported that the KA can act as a reducing agent of dopaquinones to
L
-DOPA.8,9 Although it is unclear that KA or KA-F-NH2 have reduced
the dopachrome or conjugated with dopachrome, forming a com-
plex, we measured the dopachrome reducing activity of KA-AA-
NH2 in the same way as we employed in the Mushroom Tyrosinase
Inhibition Assay. From this, we confirmed that KA-F-NH2 was more
effective in reducing the melanin biosynthetic pathway products
than KA itself or other KA-AA-NH2 derivatives. When the reducing
activity of the amount of dopachrome production was calculated in
the same way as that for tyrosinase inhibitory activity, the dopa-
chrome production reducing activity of KA was only 8%. However,
the reducing activities of KA-F-NH2, KA-W-NH2, KA-Y-NH2, and
KA-H-NH2 were calculated to be 31%, 24%, 14%, and 16%, respec-
tively. The dopachrome production reducing activity of KA-F-NH2
was about fourfold higher than that of KA. From this observation,
we concluded that KA-AA-NH2 decreased the amount of dopa-
chrome much better than KA and delayed melanin formation.
When the C-terminus of the peptide was converted to the
amide form, its stability as a function of time, temperature, and
Figure 4. Storage stabilities of KA and KA-F-NH2. (20 lM dissolved in DI water)
were stored at 50 °C for 3 months and their tyrosinase inhibitory activity was
measured by Mushroom Tyrosinase Inhibition Method).
reacted for 5 min. This means that intermediate products in the
melanin biosynthetic pathway were continuously synthesized as
time went by. However, when KA or KA-F-NH2 was added as a
Figure 5. Molecular interactions of KA-AA-NH2 near the tyrosinase binding site after docking simulations (Matoba, Y. et al. J. Biol. Chem. 2006, 281, 8981). (Left up) The
interaction of the phenylalanine group of KA-F-NH2 with His63, His216 and Phe59 near the enzyme binding site is shown with their distances; red = oxygen, blue = nitrogen
(Right up) KA-W-NH2 interaction in tyrosinase active site. (Down) KA-Y-NH2 interaction in tyrosinase active site.

J.-M. Noh et al. / Bioorg. Med. Chem. Lett. 19 (2009) 5586–5589
5589
pH value became higher than that of the free acid form. This was
References and notes
expected from our group’s previous work.16 To compare their stor-
1. Parvez, S.; Kang, M.; Chung, H.-S.; Cho, C.; Hong, M.-C.; Shin, M.-K.; Bae, H.
Phytother. Res. 2006, 20, 921.
2. Parvez, S.; Kang, M.; Chung, H.-S.; Bae, H. Phytother. Res. 2007, 21, 805.
3. Kim, Y. M.; Yun, J.; Lee, C.-K.; Lee, H.; Min, K. R.; Kim, Y. J. Biol. Chem. 2002, 18,
16340.
4. Battaini, G.; Monzani, E.; Casella, L.; Lonardi, E.; Tepper, A. W. J. W.; Canters, G.
W.; Bubacco, L. J. Biol. Chem. 2002, 47, 44606.
5. Kahn, V.; Ben-Shalom, N.; Zakin, V. J. Agric. Food Chem. 1997, 45, 4460.
6. Burdock, G. A.; Soni, M. G.; Carabin, I. G. Regul. Toxicol. Pharmacol. 2001,
33, 80.
7. Schurink, M.; Van Berkel, W. J. H.; Wichers, H. J.; Boeriu, C. G. Peptides 2007, 28,
485.
8. Chen, J. S.; Wei, C.-I.; Marshall, M. R. J. Agric. Food Chem. 1991, 11, 1897.
9. Marguerite, F.; Horowitz, N. H.; Heinemann, S. F. J. Biol. Chem. 1963, 238,
2045.
age stability, KA and KA-F-NH2 were stored at 50 °C as solution
state (20 lM, in DI water), and their inhibition activities were mea-
sured periodically. The tyrosinase inhibitory activity of KA was de-
creased from 18% to 8% in 3 days, however that of KA-F-NH2 was
not changed. We imagine that the aromatic side chain residues
make KA-F-NH2 be piled up and eventually lead to lower its solu-
bility. And the lowered solubility might prevent it from fast oxida-
tion in water. Actually, KA-F-NH2 is less soluble than KA. As turned
out in Figure 4, the tyrosinase inhibitory activity of KA was sharply
decreased, whereas that of KA-F-NH2 was well maintained even
after 3 months.
As mentioned before, when amino acids having an aromatic
ring structure at the side chain were conjugated to KA, its tyrosi-
nase inhibitory activity was enhanced to over 90%. We suggest that
the aromatic ring structure may contribute to the binding of the
inhibitor to the hydrophobic pocket of the enzyme near the binu-
clear copper active site. Recently, the crystallographic structure
of tyrosinase has been revealed. The three-dimensional structure
of tyrosinase enables us to gain a better understanding of the
tyrosinase inhibition mechanism. Although the structure of mush-
room tyrosinase was not determined yet, we can borrow the crys-
tallographic data because there is a high homology for the active
center of most tyrosinase from different origin.25–27 To ascertain
the importance of the aromatic ring structure of KA-AA-NH2, we
docked KA-AA-NH2 into the active site of tyrosinase using the AUTO-
DOCK Tools program. According to the docking calculations, KA-AA-
NH2 having an aromatic ring structure emitted a higher free energy
after docking to the active site of tyrosinase and, in this way, low
energy levels were achieved.28 We also found that quite a few
hydrophobic amino acids were located around the copper active
site of tyrosinase. From these data, we confirmed that there existed
hydrophobic interactions between the aromatic rings of KA-F-NH2
and the hydrophobic side chains in the tyrosinase active site, and
these interactions blocked the accessibility of the substrate to the
active site. For example, the aromatic side chain of KA-F-NH2 and
KA-W-NH2 might interact with the hydrophobic pocket around
His63, His216, and Phe59 surrounding the binuclear copper active
site of tyrosinase. In the case of KA-Y-NH2, the phenol ring might
interact with His216, Phe66, and Trp254 (Fig. 5). Because of these
interactions, we can conclude that KA-AA-NH2, which has an aro-
matic ring structure at the side chain, exhibits much higher tyros-
inase inhibitory activity than KA.
10. Kobayashi, Y.; Kayahara, H.; Tadasa, K.; Tanaka, H. Bioorg. Med. Chem. Lett.
1996, 12, 1303.
11. Kadokawa, J.; Nisikura, T.; Muraoka, R.; Tagaya, H.; Fukuoka, N. Synth. Commun.
2003, 7, 1081.
12. Nishimura, T.; Kometani, T.; Takii, H.; Terada, Y.; Shigetaka, O. J. Ferment.
Bioeng. 1994, 1, 37.
13. O’Brien, G.; Patterson, J. M.; Meadow, J. R. J. Org. Chem. 1962, 27, 1711.
14. Kobayashi, Y.; Kayahara, H.; Tadasa, K.; Nakamura, T.; Tanaka, H. Biosci.,
Biotechnol., Biochem. 1995, 9, 1745.
15. Kim, H.; Choi, J.; Cho, J. K.; Kim, S. Y.; Lee, Y. S. Bioorg. Med. Chem. Lett. 2004, 14,
2843.
16. Noh, J. M.; Kwak, S. Y.; Kim, D. H.; Lee, Y. S. Biopolymers 2007, 2, 300.
17. Kojic acid (5 g, 35 mmol) was dissolved in THF (100 mL). After stirring for 1 h,
with N2 purging, CDI (5.1 g, 0.9 equiv) in THF (50 mL) was added, and the
mixture stirred for 24 h at room temperature. A white solid powder was
filtered and dried. Yield: 75%; 1H NMR (DMSO-d6, 300 MHz, d): 9.34 (1H, s, –
OH), 8.33 (1H, s, N–CH@N), 8.13 (1H, s, –CH–O), 7.67 (1H, s, imidazole), 7.10
(1H, s, imidazole), 6.66 (1H, s, CH–C@O), 5.30 (2H, s, CH2–O). N-Fmoc-amino
acid (2 equiv), in NMP, was quantitatively introduced to the Rink amide AM
SUREÒ resin (0.76 mmol/g) using the general protocol of BOP-mediated solid
phase Fmoc/tBu strategy. After removing Fmoc group, activated kojic acid
(2 equiv) was added and shaken for 6 h. Finally, the resin was treated with 30%
TFA/DMC for 1 h, and filtered. The crude product in the filtrate was
concentrated and precipitated with cold diethyl ether.
KA-F-NH2 (5-hydroxy-4-oxo-4H-pyran-2-yl)methyl-1-amino-1-oxo-3-phenyl-
propan-2-ylcarbamate); White powder, yield: 86%, purity: 95% (HPLC; Waters
lBondapak C18 reverse phase column, 125 Å, 10 lm, 3.9 ꢁ 150 mm; gradient
elution with A: 0.1% TFA/water, B: 0.1% TFA/acetonitrile; from 10% to 90% B
over 50 min, flow: 1 ml/min; detection; UV, 230 nm), 1H NMR (DMSO-d6,
300 MHz) d 8.09 (1H, s, –NH), 7.65 (1H, s, @CH–O), 7.50 (1H, d, benzene), 7.24
(1H, d, benzene), 7.16 (1H, s, benzene), 6.49 (1H, s, O@CH@C), 6.26 (2H, s, NH2),
5.06 (1H, s, NH–CH–CO), 4.77 (2H, d, –CH2–O), 3.14 (2H, d, CH2-benzene), 13C
NMR (DMSO, 300 MHz) d 173.5 (C4), 163.5 (C2), 160.0 (C10), 155.3 (C5), 153.5
(C20), 145.9 (C40), 141.0 (C60), 140.0 (C80), 138.2 (C6), 129.2 (C50), 128.0 (C90),
125.0 (C70), 113.2 (C3), 64.9 (C1), 61.5 (C20), 57.2 (C30), MS (ES): [M+1]+ 333.3.
KA-Y-NH2; Yield: 80%, purity: 93%, KA-W-NH2; yield: 87%, purity: 95%, KA-S-
NH2; yield: 76%, purity: 92%, KA-T-NH2; yield: 73%, purity: 86%, KA-D-NH2;
Yield: 83%, purity: 80%, KA-N-NH2; yield: 79%, purity: 84%, KA-E-NH2; yield:
80%, purity: 92%, KA-Q-NH2; yield: 75%, purity: 93%, KA-K-NH2; yield: 71%,
purity: 89%, KA-R-NH2; yield: 76%, purity: 95%, KA-H-NH2; yield: 80%, purity:
96%, KA-C-NH2; yield: 70%, purity: 70%, KA-M-NH2; yield: 82%, purity: 92%,
KA-P-NH2; yield: 74%, purity: 86%, KA-G-NH2; yield: 76%, purity: 96%, KA-A-
NH2; yield: 77%, purity: 93%, KA-V-NH2; yield: 76%, purity: 89%, KA-F-NH2;
yield: 80%, purity: 91%, KA-F-NH2; yield: 80%, purity: 90%.
18. Kermasha, S.; Goetghebeur, M.; Monfette, A.; Metche, M.; Rovel, B.
Phytochemistry 1993, 2, 349.
In summary, we prepared 20 kinds of KA-AA-NH2 derivatives
and screened their tyrosinase inhibitory activities. Among them,
KA-F-NH2, KA-Y-NH2, KA-W-NH2, and KA-H-NH2 showed much
higher tyrosinase inhibitory activity and their enhanced inhibitory
activity was maintained for over 3 months. We also confirmed that
the KA-F-NH2 reduced the amount of dopachrome production dur-
ing the melanin formation. Finally, we suggest a tyrosinase inhibi-
tion mechanism of KA-AA-NH2 based on the possible hydrophobic
interactions between the side chain of KA-AA-NH2 and tyrosinase
active site by a docking program. Further studies are on progress
to confirm the tyrosinase inhibitory activity of KA-AA-NH2 in the
cell system.
19. Li, B.; Huang, Y.; Paskewitz, S. M. FEBS Lett. 2006, 580, 1877.
20. Park, Y. D.; Kim, S. Y.; Lyou, Y. J.; Lee, D. Y.; Yang, J. M. Biochem. Cell Biol. 2006,
84, 112.
21. Shin, N. H.; Ryu, S. Y.; Choi, E. J.; Kang, S. H.; Chang, I. M.; Min, K. R.; Kim, Y.
Biochem. Biophys. Res. Commun. 1998, 243, 801.
22. Bernard, P.; Berthon, J. Y. Int. J. Cosmet. Sci. 2000, 22, 219.
23. Kim, Y. M.; Yun, J.; Lee, C. K.; Lee, H.; Min, K. R.; Kim, Y. J. Biol. Chem. 2002, 18,
16340.
24. Klabunde, T.; Eicken, C.; Sacchettini, J. C.; Krebs, B. Nat. Struct. Biol. 1998, 12,
1084.
25. Steiner, R. A.; Kooter, I. M.; Dijkastra, B. Biochemistry 2002, 41, 7955.
26. Matoba, Y.; Kumagai, T.; Yamamoto, A.; Yoshitsu, H.; Sugiyama, M. J. Biol. Chem.
2006, 13, 8981.
27. Naka, H.; Kondo, Y.; Usui, S.; Hashimoto, Y.; Uchiyama, M. Adv. Synth. Catal.
2007, 349, 595.
Acknowledgments
This work was supported by a grant from the Korea Health Z1
R&D Project, Ministry & Welfare, Republic of Korea (A050432).
28. Khatib, S.; Nerya, O.; Musa, R.; Tamir, S.; Peter, T.; Vaya, J. J. Med. Chem. 2007,
50, 2676.