Chalcones as potent tyrosinase inhibitors: The importance of a 2,4-substituted resorcinol moiety

Article abstract of DOI:10.1016/j.bmc.2004.10.010

Chalcones with tetra-substituted hydroxyl groups were synthesized and tested as tyrosinase inhibitors towards designing novel whitening agents, resulting with the most potent inhibitor, with IC₅₀ of 0.02 μM concn. Compounds, which inhibit tyrosinase, could be effective as depigmenting agents. We have introduced a group of mono-, di-, tri- and tetra-substituted hydroxychalcones as effective tyrosinase inhibitors, showing that the most important factor determining tyrosinase inhibition efficiency is the position of the hydroxyl group(s) rather their number. The aim of the present study was to investigate the contribution of the different functional groups of the tetrahydroxychalcones to their inhibitory potency, with a view to optimizing the design of whitening agents. Four tetrahydroxychalcones were evaluated, the commercially available Butein and other three were synthesized, and their inhibitory effect on tyrosinase was tested. Results showed that a 2,4-substituted resorcinol subunit on ring B contributed the most to inhibitory potency. Changing the resorcinol substitute to position 3,5- or placing it on ring A significantly diminished the inhibitory effect of the compounds. A catechol subunit on ring A acted as a metal chelator (in the presence of copper ions) and as a competitive inhibitor (in the presence of tyrosinase), while a catechol on ring B oxidized to o-quinone (in the presence of both copper ions and tyrosinase). Three of the compounds also demonstrated antioxidant activity, which may contribute to the prevention of pigmentation. An examination of correlations between inhibitory activity and physical properties of the chalcones tested (such as dissociation energy and molecular planarity) showed positive correlation with the moment dipole value in the Y-axis, which may be used as an indicator of the inhibitory potential of new molecules. The present study revealed two very active tyrosinase inhibitors, 2,4,3′,4′- hydroxychalcone and 2,4,2′,4′-hydroxychalcone (with IC₅₀ of 0.2 and 0.02 μM, respectively). Structure-related activity studies added some understanding of the role and contribution of different functional groups associated with tyrosinase inhibitors.

Full text of DOI:10.1016/j.bmc.2004.10.010

Bioorganic & Medicinal Chemistry 13 (2005) 433–441
Chalcones as potent tyrosinase inhibitors: the importance
of a 2,4-substituted resorcinol moiety
Soliman Khatib,a,b Ohad Nerya,a Ramadan Musa,a Maayan Shmuel,a Snait Tamira,b
and Jacob Vayaa,b,
*
aLaboratory of Natural Medicinal Compounds, Migal-Galilee Technological Center, PO Box 831, Kiryat Shmona 11016, Israel
bDepartment of Biotechnology and Environmental Science, Tel-Hai Academic College, Israel
Received 27 June 2004; accepted 6 October 2004
Available online 28 October 2004
Abstract—Compounds, which inhibit tyrosinase, could be effective as depigmenting agents. We have introduced a group of mono-,
di-, tri- and tetra-substituted hydroxychalcones as effective tyrosinase inhibitors, showing that the most important factor determin-
ing tyrosinase inhibition efficiency is the position of the hydroxyl group(s) rather their number. The aim of the present study was to
investigate the contribution of the different functional groups of the tetrahydroxychalcones to their inhibitory potency, with a view
to optimizing the design of whitening agents. Four tetrahydroxychalcones were evaluated, the commercially available Butein and
other three were synthesized, and their inhibitory effect on tyrosinase was tested. Results showed that a 2,4-substituted resorcinol
subunit on ring B contributed the most to inhibitory potency. Changing the resorcinol substitute to position 3,5- or placing it on ring
A significantly diminished the inhibitory effect of the compounds. A catechol subunit on ring A acted as a metal chelator (in the
presence of copper ions) and as a competitive inhibitor (in the presence of tyrosinase), while a catechol on ring B oxidized to o-qui-
none (in the presence of both copper ions and tyrosinase). Three of the compounds also demonstrated antioxidant activity, which
may contribute to the prevention of pigmentation. An examination of correlations between inhibitory activity and physical prop-
erties of the chalcones tested (such as dissociation energy and molecular planarity) showed positive correlation with the moment
dipole value in the Y-axis, which may be used as an indicator of the inhibitory potential of new molecules. The present study
revealed two very active tyrosinase inhibitors, 2,4,30,40-hydroxychalcone and 2,4,20,40-hydroxychalcone (with IC50 of 0.2 and
0.02lM, respectively). Structure-related activity studies added some understanding of the role and contribution of different func-
tional groups associated with tyrosinase inhibitors.
Ó 2004 Elsevier Ltd. All rights reserved.
1. Introduction
reported as potent in tyrosinase inhibition,3 and
their structure–activity relationships and inhibition
mechanisms have been examined.4 It was suggested that
4-substituted resorcinol-type inhibitors bind to the
enzyme binuclear active site. The catechol structure,
with two OH groups at o-positions, may behave as a
chelator to the copper ions in the tyrosinase.1 These
polyphenols generally compete in inhibition with
tyrosinase.
Tyrosinase is
a
copper-containing enzyme widely
distributed in nature, catalyzes two key reactions in
the melanin biosynthesis pathway. Its inhibition is one
of the major strategy in developing new whitening
agents. Many tyrosinase inhibitors (Table 1) are
polyphenol derivatives of flavonoids or of trans-stilbene
(t-stilbene), such as resveratrol and its derivatives, which
have been investigated intensively.1,2 They are usually
constructed from one of two distinct substructures,
which dictate their mechanism of tyrosinase inhibition:
containing either a 4-substituted resorcinol moiety,
or catechol. The 4-substituted resorcinol group has been
trans-Stilbene derivatives contain two aromatic rings,
separated by two carbon atoms, connected by a double
bond in trans configuration. Another group of
compounds, with a similar structure to t-stilbene, is
the chalcones, which are widely distributed in higher
plants.5,6 Chalcones also consist of two aromatic rings
in trans configuration, separated by three carbons, of
which two are connected by double bond and the third
is a carbonyl group. Recently, we have demonstrated
Keywords: Tyrosinase; Inhibitor; Chalcone; SAR; Melanocyte.
*
Corresponding author. Tel.: +972 4 6953512; fax: +972 4 6944980;
e-mail: vaya@migal.org.il
0968-0896/$ - see front matter Ó 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2004.10.010

434
S. Khatib et al. / Bioorg. Med. Chem. 13 (2005) 433–441
Table 1. Tyrosinase inhibitors—data from the literature
Inhibitor
Structure
Substrate
Tyrosinase inhibition (IC50, lM)
HO
OH
O
Artocarbene19
L-Tyrosine
2.45
OH
OH
HO
Chlorophorin19
L-Tyrosine
2.6
OH
OH
OH
HO
Norartocarpanone19
L-Tyrosine
L-Tyrosine
1.76
0.91
HO
O
OH
O
OH
HO
4-Propylresorcinol4
O
O
OH
Glabridin7
L-Tyrosine
L-DOPA
0.09
OH
HO
HO
CN
3,4-Dihydroxybenzonitrile20
45
O
CH OH
2
Kojic Acid20
L-DOPA
22
HO
O
L-Tyrosine
8.66
that chalcones, like t-stilbene, are potential inhibitors of
tyrosinase7 and the position of the hydroxyl(s) attached
to the chalcone rings are of major importance in that
activity.8 In the present study we designed, synthesized
and tested new chalcones, constructed from substituted
resorcinol at one aromatic ring and a catechol moiety at
the second aromatic ring (A or B) of the molecule. Our
hypothesis was that a molecular structure containing,
within the same compound, a resorcinol moiety (with
the ability to penetrate to the enzyme active site and
inhibit it), together with a catechol moiety (which, if
properly designed, may chelate the copper ions present
in the active site), would exert potent tyrosinase
inhibition. In addition, these compounds contain
phenolic hydroxyls, which can prevent pigmentation
resulting from auto-oxidative processes. The relation
between the structure and different physical properties
of these chalcones to their tyrosinase inhibitory effects,
as well as their inhibition of melanin formation in
human melanocyte cells, were investigated, and the
contribution of each functional group to that activity
was characterized.
2. Results
2.1. 3,4,20,40-Tetrahydroxy-t-stilbene as tyrosinase
inhibitor
In order to test our hypothesis on tyrosinase inhibitors,
which combine the function of being a competitive
inhibitor (which many of the resorcinol derivatives
are) with that of a chelating agents, we chose a t-stil-
bene that contains resorcinol on one aromatic ring
and a catechol moiety on the other: 3,4,20,40-tetrahy-
droxy t-stilbene (compound I). As shown in Table 2,
compound I exhibited higher tyrosinase inhibitory
activity than other known inhibitors9 with IC50 of
1.5lM.
2.2. Chalcones as tyrosinase inhibitors
Having demonstrated that t-stilbene with the two sub-
units is indeed active, we then synthesized chalcones
containing both resorcinol and catechol. Unlike t-stil-
bene, the two aromatic rings in chalcones are not sym-

S. Khatib et al. / Bioorg. Med. Chem. 13 (2005) 433–441
435
Table 2. The inhibitory effect of t-stilbene and chalcones on mushroom tyrosinase activities
Compound
Structure
L-Tyrosine (IC50, lM)
L-Dopa (IC50, lM)
OH
OH
OH
3,4,2,4-trans-Stilbene I
1.5
—
HO
HO
OH
OH
3,4,20,40-HC (Butein) II
29.3
>100
A
B
OH
O
HO
HO
OH
2,4,30,40-HC III
3,5,30,40-HC IV
0.2
7.5
HO
O
OH
HO
HO
HO
31.68
>1000
OH
O
HO
OH
2,4,20,40-HC V
0.02
90
OH
O
A solution of tyrosinase (50units/mL), tested compound (0.005–50lM) and L-tyrosine (1mM) or L-DOPA (6mM) in 200lL PBS (50mM, pH6.5)
was prepared and its optical density at 492nm was measured after 20min.
metric, thus chalcones containing the catechol func-
tional group either on ring A or ring B were tested (com-
pounds II–IV) (Table 2).
depicts the change in OD at 475nm as a function of time
due to the formation of o-quinone in the absence (con-
trol) or the presence of compounds II–V. As expected,
in the presence of tyrosinase inhibitors, the lag phase
was prolonged from 440s in the control to 787s with
the addition of compound IV. Furthermore, compounds
The inhibition of the tyrosinase was tested on both en-
zyme activities stages: L-tyrosine as substrate to L-DOPA
and the oxidation of the later to o-quinone. Compounds
II–V demonstrated higher tyrosinase inhibitory activity
with the first stage than that of the second stage. Butein
(compound II), with a catechol subunit on ring B and
resorcinol on ring A (positions 20 and 40), inhibited the
first stage with an IC50 of 29.3lM and presented poor
inhibition of the second stage. The inhibitory activity
of compound III (2,4,30,40-HC), the opposite structure
to Butein, increased inhibition approximately 150-fold,
with IC50 of 0.2lM at the first stage, and with IC50 of
7.5lM for the second stage. In compound IV (3,5,30,40-
HC), the OH groups in the resorcinol moiety on ring B
are located on positions 3 and 5 (compared with posi-
tions 2 and 4 in compound III), keeping the catechol sub-
unit at the same location as in compound III. This
change decreased the tyrosinase inhibitory effect to about
the same degree as that obtained with Butein (IC50 of
31.7lM), with practically no inhibition of the second
stage.
0.4
0.3
0.2
0.1
0.0
0
100 200 300 400 500 600 700 800 900 1000 1100 1200
time (sec)
When the chalcone molecule was constructed with
resorcinol moieties on both rings A and B, at positions
2, 4 and 20, 40 (compound V), tyrosinase inhibitory
activity increased to IC50 of 0.02lM for the first stage,
and to 90lM for the second stage.
Figure 1. Lag period of L-dopaquinone formation. L-Tyrosine (1mM)
was incubated with tyrosinase in the absence (j-control) or with the
addition of chalcones II–V, each at twice its IC50 value; Butein (d,
compound II) 50lM, 2,4,30,40-HC ( , compound III) 0.4lM, 3,5,30,40-
*
HC (h, compound IV) 50lM and 2,4,20,40-HC (m, compound V)
0.04lM, and the change in the OD at 475nm were recorded as a
First stage activity of tyrosinase is characterized by a lag
time, followed by an increase in reaction rate.2 Figure 1
function of time. Ethanol was used as
means SE of three separate experiments.
a control. Values are

436
S. Khatib et al. / Bioorg. Med. Chem. 13 (2005) 433–441
III–V also reduced the reaction rate (the slope of the
curve) of the o-quinone formation by 80–60% versus
control. Contrary to the above effect, Butein decreased
the lag time to 220s a phenomena, which will be dis-
cussed further in Discussion section.
2.3. The chelating ability of compounds II–V
The interactions of each chalcone with Cu+2, at pH6.5,
was assessed by UV/vis spectroscopy, measuring batho-
chromic shifts as a result of complex formation between
1.4
1.2
1.0
0.8
0.6
0.4
0.2
0.0
0.9
0.7
0.5
0.3
0.1
-0.1
250 300 350 400 450 500 550 600
200 250 300 350 400 450 500 550 600
λ(nm)
(A)
(B)
λ(nm)
0.5
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0.0
-0.1
0.3
0.1
-0.1
250 300 350 400 450 500 550 600
200 250 300 350 400 450 500 550 600
(C)
(D)
λ (nm)
λ (nm)
100
%
269.4
7437 696
271.4
4887894
0
100
271.3
12260713
%
m/z
284
0
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
(E)
Figure 2. Copper ions chelating ability of chalcones II–V. The UV–vis spectra of Butein II (A), 2,4,30,40-HC (III) (B), 3,5,30,40-HC (IV) (C) and
2,4,20,40-HC (V) (D) at a 50lM solution of phosphate-buffered saline (PBS) each, at pH = 6.5 (thick solid line). Spectrum recorded again with the
addition of 100lM CuSO4 (thin solid line), and again with the addition of EDTA 250lM (dotted line). The mass spectrum of Butein (II) before (E,
below) and after the addition of copper ions (E, above) using negative ion electrospray ionization method.

S. Khatib et al. / Bioorg. Med. Chem. 13 (2005) 433–441
Table 4. The radical scavenging capacity of chalcones
437
the copper ion and the tested compounds, and by LC–
MS analysis, following changes in m/z values. A solution
of each compound (50lM) in PBS buffer was scanned
with or without CuSO4 (100lM), and the spectra ob-
tained were compared. The characteristic maximum at
382nm of Butein shifted to 388nm with the addition of
copper ions (Fig. 2A); on addition 250lM EDTA, the
shifted kmax remained at 388nm. LC–MS of Butein
showed molecular ion at m/z 271 (Mꢀ1, negative electro-
spray ionization), whereas upon injection of the reaction
mixture after the addition of copper ions, the molecular
ion shifted by two mass units giving m/z of 269 (Mꢀ1)
(Fig. 2E). Similar results were obtained with chalcone
III, showing kmax of 378nm; with the addition of copper
ions the kmax shifted to 382nm (Fig. 2B), but it returned
to its original value (378nm) with the addition of EDTA
(250lM). LC–MS analysis of the mixture of the chalcone
III with copper ions did not show any change in the m/z
value as the reaction proceeded. Upon the addition of
copper ions to chalcone IV, kmax at 317nm shifted to give
a shoulder at 400nm, which was re-absorbed to kmax of
317 with the addition of EDTA (Fig. 2C). The MS anal-
ysis of compound IV with copper ions did not show any
change in the mass of the peak of the parent compound.
The UV/vis spectrum of chalcone V was not affected by
the addition of copper ions (Fig. 2D).
Compounds
DPPH reduction ratio
(DPPH/antioxidant (mole/mole))
Ethanol
0
Trolox
0.32
7.87
5.46
3.4
Butein, 3,4,20,40-HC (II)
2,4,30,40-HC (III)
3,5,30,40-HC (IV)
2,4,20,40-HC (V)
Not active
Trolox (positive control), 2,4,30,40-HC (III), 3,5,30,40-HC (IV) and
2,4,20,40-HC (V), each at 10lM, and Butein (II) at 2lM in methanol
were added to 63lM DPPH. The absorption at 515nm was recorded
after 420min incubation at 37°C. Ethanol was used as control.
changes in absorption at 515nm after 7h incubation,
due to elimination of the DPPH radical. The reactions
were carried out under excess DPPH, in order to deter-
mine the number of electrons (the ratio of mole DPPH/
mole antioxidant) each compound may donate to
DPPH (Table 4). Results demonstrate that Butein do-
nated the highest number of electron to the DPPH
(8). Compounds III, IV and trolox donated 5.5, 3.4
and 0.32 electrons, respectively. Compound V failed to
donate any electron to DPPH under these experimental
conditions.
2.4. The effect of chalcones III–V on melanin biosynthesis
in human melanocytes
3. Discussion
Table 3 presents the effect of glabridin, a known tyrosin-
ase inhibitor,10 in comparison to chalcones III–V, on
melanin biosynthesis in melanocyte cultures. Glabridin
(10lM) decreased melanin synthesis to about 41% and
had no cytotoxic effect. Compound III (5lM) inhibited
melanin synthesis to about 63%, with 88% cell survival.
Compound V (5lM) inhibited melanin synthesis to
about 55% and had no cytotoxic effect. Butein (5lM)
was cytotoxic (20% survival, data not shown).
In the present article we aimed to explore the depigmen-
tation potential of hydroxylated chalcones with four
substituted OH groups. These were studied as new
inhibitors of melanin formation, in relation to their
structure. The new compounds included chalcones
constructed from a resorcinol moiety on one aromatic
ring and a catechol substructure on the other, to
evaluate the possibility of combining two types of
inhibitory mechanisms in a single molecule, that is,
binding to the enzyme active site and thus being compet-
itive to tyrosine and chelating the copper ion(s) within
the active site. As a model system, the validity of such
a combination was first examined with the t-stilbene
class of compounds.
2.5. Radical scavenging capacity of compounds II–V
Phenols may exert antioxidant properties, due to their
ability to donate an electron or chelate metals. In the
present study, chalcones II–V represent polyphenols
and thus may perform antioxidant activity which pre-
vents or delays pigmentation resulting form auto-oxida-
tion. The antioxidant activity of chalcones II–V was
tested by measuring their ability to donate an electron
to the free radical compound DPPH, monitoring
Two-well-known tyrosinase inhibitors of the t-stilbene
class are resveratrol (40,3,5-trihydroxy-t-stilbene), found
in grapes, with IC50 of 54.6lM, and oxyresveratrol
(20,40,3,5-tetrahydroxy-t-stilbene), extracted from Morus
alba, exerting stronger tyrosinase inhibition with IC50 of
1.2lM.9 Neither resveratrol nor oxyresveratrol contain
an aromatic ring with a catechol subunit. In accordance
to our hypothesis, a model molecule of the t-stilbene
class, containing both subunits was selected, exerting
tyrosinase inhibition with IC50 of 1.5lM, superior to
resveratrol and similar to that of oxyresveratrol.
These results are in agreement with previous studies,
suggesting that the number of the hydroxyl groups
attached to t-stilbenes have an important role in
determining their activity.3,4,9 Compound I is the first
compound of this t-stilbene class containing both
resorcinol and catechol moieties to be tested as tyrosin-
ase inhibitor.
Table 3. The inhibitory effect of glabridin and chalcones III and V on
melanin biosynthesis in G361 human melanocytes
Compounds (concn)
Melanin inhibition (%) Survival (%)
Glabridin 10lM
41
7
8
94.6
9
2,4,30,40-HC (III) 5lM 62.5
87.8 11
93.3 12
2,4,20,40-HC (V) 5lM
55 14
Melanocyte cells grown in McCoys medium (Bet-HaeÕemek) were
incubated with compounds added 24h after cell seeding. The cells were
harvested after 4–5days, and treated with 1N NaOH. Melanin content
was measured in a spectrophotometer at 405nm and protein content
was measured, using Bradford reagent at 595nm.

438
S. Khatib et al. / Bioorg. Med. Chem. 13 (2005) 433–441
Butein (compound II), a commercial compound with a
catechol subunit on ring B (positions 3 and 4) and
resorcinol on ring A (positions 20 and 40) was found to
inhibit the first stage of tyrosine oxidation with IC50 of
29.3lM. Compound III has the opposite structure to
Butein. This compound markedly enhanced the inhibi-
tory effect at the first stage (IC50 of 0.2lM). These
results emphasize the major role of 2,4-substituted
resorcinol substructure on ring B. Replacing the resorc-
inol substituted positions from 2 and 4 (in compound
III) to positions 3 and 5 (compound IV—a novel chal-
cone), and keeping the catechol subunit on ring A at
the same location in both compounds III and IV,
resulted in a decreased inhibitory potency to about the
same degree as Butein (IC50 of 31.68lM), with practi-
cally no inhibition at the second stage. These results
demonstrate that resorcinol construction on ring B is
not sufficient: it requires this resorcinol moiety to be at
position 2 and 4, rather than at position 3 and 5.
inhibitors.11 In the present study, compounds III–V pro-
longed the lag phase, while Butein decreased it by half.
This finding may suggest that chalcones III, IV and V,
with four OH groups, do not mimic diphenol action in
reducing copper (II) to copper (I) and regenerating it,
and thus shorten the lag time. On the other hand, Butein
may acts as reducing agent, donates electrons and con-
verts to quinone. This possibility was further studied
by injecting the reaction mixture to LC–MS and analyz-
ing the products formed, and by calculating differences
in the heat of formation. The results of both tests are
discussed below (Table 5).
The depigmenting action of kojic acid, used in cosmet-
ics, is attributed to its ability to chelate copper ion in
the tyrosinase active site.12 Similarly, chalcones II–IV
demonstrated chelating ability, which may contribute
to their inhibitory potency. In contrast, compound V
did not chelate copper ions, suggesting the necessity of
a catechol structure for chelation properties. The addi-
tion of EDTA to Butein did not reverse the reaction
as it did in compound III and IV, suggesting that copper
ions oxidized Butein to o-quinone, an assumption sup-
ported by the LC–MS analysis.
Substitution of the chalcone molecule with resorcinol
moieties on both rings A and B, at positions 2 and 4
and 20 and 40 (V) resulted with the formation of the most
effective tyrosinase inhibitor known to us so far, with
IC50 of 0.02lM. Comparing the tyrosinase inhibitory
activity of chalcone (III) with t-stilbene (I), both sharing
same functional groups, 30,40-substituted catechol and
2,4-substituted resorcinol, revealed that the chalcone is
7.5 times more active than the corresponding t-stilbene,
emphasizing the enhanced potential of the chalcone sub-
class as new whitening agents.
Tyrosinase was added to Butein, instead of copper ions
(in both cases an immediate colour change to brown-red
anticipated). An LC–MS analysis of the reaction mix-
ture showed that with the enzyme addition to Butein,
a new compound was formed with reduction of two
mass units. This result may confirm that Butein was
either oxidized to o-quinone by tyrosinase or copper
ions, or it was converted to aurone derivatives, a reac-
tion which may occur spontaneously.13,14 These possi-
bilities were further investigated. Thus, under stirring
Butein in ethanol during a week at room temperature
in open glass no change occurred (based on LC/MS/
MS analysis). FTIR analysis of the reaction product
showed a single sharp peak at 1626cmꢀ1 for the
carbonyl group, while that of Butein has three peaks
Monophenolase activity of tyrosinase is characterized
by a lag time, influenced by various factors such as sub-
strate and enzyme concentration, enzyme source, pH of
the medium, presence of a hydrogen donor such as L-
DOPA or other reducing agents or transition metal
ions.2 With the addition of exogenous reductants (ascor-
bate, hydroxylamine or hydroquinone), this lag period
can be shortened and re-extended with the addition of
Table 5. The relation between tyrosinase inhibitory potency of chalcones II–V and their physical parameters
Compound (IC50, lM)
Planarity
Dipole moment (Debye) X-, Y- and Z-axes
DDHf, kcal/mol (bond dissociation energy)
X
Y
Z
Total
3.18
II (29.3)
+
2.02
2.46
0
4-OH = 32.5
3-OH = 30.2
20-OH = 42.2
40-OH = 38.9
4-OH = 35.1
2-OH = 34.2
30-OH = 31.2
40-OH = 30.5
3-OH = 36.2
5-OH = 36.0
30-OH = 31.3
40-OH = 30.7
4-OH = 35.5
2-OH = 34.6
20-OH = 44.7
40-OH = 38.6
III (0.2)
IV (31.68)
V (0.02)
ꢀ
ꢀ
+
1.27
2.64
3.02
3.83
2.74
4.78
—
2.08
4.54
4.77
5.65
—
2.87
0
Potency was correlated to the planarity of the molecule [(+) planar, (ꢀ) not planar], to their total moment dipole (l) or their X-, Y- or Z-axes values
and to the O–H bond dissociation energy (DDHf). These parameters were obtained by generating semi-empirical calculations, using MOPAC program
and PM3 Hamiltonian.

S. Khatib et al. / Bioorg. Med. Chem. 13 (2005) 433–441
OH
439
OH
O
O
OH
O
O
Cu+2
or tyrosinase
OH
HO
HO
Figure 3. Oxidation of Butein to o-diquinone in the present of copper ions or tyrosinase.
at 1635, 1597 and 1558cmꢀ1 (data not shown), all
together correspond to formation of quinone rather
then aurone derivatives. We may thus conclude that a
catechol subunit on ring A chelates copper ions (as ob-
served in compounds III and IV), while a catechol sub-
unit on ring B is oxidized by copper ions to its o-quinone
derivatives (compound II, Fig. 3). Desirable skin-light-
ening agents should inhibit the synthesis of melanin in
melanosomes by acting specifically to reduce tyrosinase
activity. They should exhibit low cytotoxicity, and be
nonmutagenic.15 In the present study the two most ac-
tive inhibitory chalcones (compounds III and V) were
further tested for their ability to inhibit melanin synthe-
sis in human melanocytes. Both were active, with low
toxicity: compound III, at 5lM, decreased melanin
synthesis to about 63%, and compound V, at 5lM,
decreased it to about 55%. These results further encour-
age the possibility of using chalcones as new depigmenta-
tion agents.
oxidant; while in compound V the lowest DDHf was
34.5kcal/mole, which resulted in an absence of antioxi-
dant activity (testing a donation of hydrogen atom to
DPPH).
The moment dipole (l) of a molecule can be a relevant
parameter for evaluating the interaction efficiency of lig-
and/inhibitor with an enzyme active site. A correlation
of the total (l) of each of the four chalcones under study
with their tyrosinase inhibition was also shown to be not
indicative. Thus, Butein differed significantly in its total
(l) from compound IV, despite their similar tyrosinase
inhibition. We then separated the total (l) values into
their three X, Y and Z-axes. Such a process is pertinent,
since ligands are oriented inside an enzyme active site in
a relatively rigid position, and free rotation of the ligand
molecule is usually not possible. This separation re-
vealed a positive correlation between tyrosinase inhibi-
tion potency and the (l) of the compounds with their
Y-axis (Table 5). Thus, Butein and compound IV have
similar inhibitory activity and similar (l) values, (2.46
and 2.74Debye), while compounds III and V have high-
er (l) values in the Y-axis (3.83 and 4.78Debye, respec-
tively), and correspondingly higher inhibitory potency
(IC50 of 0.2.and 0.02lM). For drawing conclusions, this
last finding has to be supported by many additional
examples of compounds.
Browning of food, beverage and skin could be the result
of an enzymatic reaction of polyphenols oxidases and/or
a result of the auto-oxidation of polyunsaturated fatty
acids and proteins. Chalcones containing phenolic and
catecholic derivatives may exhibit antioxidant activity
as a result of their ability to donate an electron (or
hydrogen atom) and/or chelate transition metals, such
as copper or ferrous ions, and thereby eliminating reac-
tive oxygen and nitrogen species (ROS and RNS) and
decay-free radical propagation reactions: both reactions,
when strong, are known to induce melanin synthesis.16
In the present study, chalcones II–V demonstrated anti-
oxidant activity, which may contribute to the inhibition
of melanin biosynthesis in cells.
The collected data from our studies analyzing the rela-
tionship between the chemical structures of chalcones
and their tyrosinase inhibition potency, led us to design
and synthesis the 20,40,2,4-tetrahydoxychalcone, the
most powerful inhibitor known to date. The presence
of two functional resorcinol units, both in 2- and 4-posi-
tions, found to contribute the most to chalcone for
tyrosinase inhibition potency while the presence of cate-
chol moiety either on ring A or on ring B did not con-
tribute much to the inhibitory potency.
We then tried to search for some explanations for the
data gathered, especially trying to correlate tyrosinase
inhibition by chalcones with physical parameters associ-
ated with chalcone structure. Such parameters were their
molecular planarity, O–H bond dissociation energy, and
their moment dipole (l), using MOPAC program and
PM3 semi-empirical Hamiltonian. A summary of the re-
sults is shown in Table 5. As expected, chalcones with a
OH group at position 20 (ring A) are planar molecules,
due to a hydrogen bond formation between the 20-OH
group and the adjacent carbonyl group. No correlation
was anticipated between the planarity of the molecules
and their inhibitory potencies. The heat of formation
differences (DDHf)17,18 between each possible phenoxyl
radical and its parent chalcones (bond dissociation en-
ergy) correlated with antioxidant activity, but not with
tyrosinase inhibition potency (Table 5). Compounds
II, III and IV all contain a catechol moiety, and thus
have at least one O–H with low dissociation energy,
which can relatively, easily donate a hydrogen atom
(DDHf ꢁ 30kcal/mole) and thus act as a potent anti-
4. Materials and methods
4.1. Chemicals and reagents
Tyrosinase (EC1.14.18.1, Sigma Product T7755 with an
activity of 6680units/mg). Chalcones were synthesized
according to procedure described below. Butein (com-
pound II) and other reagents and solvents were pur-
chased from Aldrich–Sigma and used as received
unless otherwise noted.
4.2. Inhibition of tyrosinase activity
Potassium phosphate buffer (0.07mL, 50mM) at pH6.5,
0.03mL tyrosinase (333units/mL) and 2lL of the tested
compounds (0.5–500lM), dissolved in absolute ethanol

440
S. Khatib et al. / Bioorg. Med. Chem. 13 (2005) 433–441
were inserted into 96-well plates. After 5min incubation
at room temperature, 0.1mL L-tyrosine (2mM) or
12mM L-DOPA were added and incubated for addi-
tional 20min. The optical density of the samples at
492nm were measured (Elisa SLT Lab instruments Co.
A-5082) relative to control containing ethanol (2lL)
and without inhibitor, demonstrating a linear colour
change with time during the 20min of the experiment.
linked to an HP ChemStation data system. Helium
was used as carrier gas, at a flow rate of 0.5mL/min,
pressure 10.4psi and at a linear velocity of 26.8cm/s.
The MS transfer line was maintained at 280°C. The
injector was set at 300°C, the detector at 280°C and
the column heated at a gradient starting at 130°C,
increasing to 250°C at 10°/min and then at 5°/min to
300°C and maintained for an additional 15min at
300°C. LC–MS measurements was performed by using
LC–MS (Waters 2790 HPLC, with Waters photodiode
array detector 996 and micromass Quattro ultima MS).
4.3. Inhibition of melanin synthesis in human melanocytes
Melanocytes (G361) cells were seeded in 24-well plates,
and grown in McCoys medium (Bet-HaeÕemek). 24h
after seeding, the chalcones were added to a final con-
centration of 5lM (total volume 0.5mL). After four
or five days the cells were harvested, using trypsin, and
washed once with PBS. The cells in each well were dis-
solved in 1mL 1N NaOH. Melanin content was meas-
4.7. Synthesis of 2,4,30,40-tetrahydroxychalcone
(compound III)
30,40-Dimethoxyacetophenone: To a stirred mixture of
30,40-dihydroxyacetophenone (100mg, 0.66mmole) and
anhydrous K2CO3 (5g, 36mmole) in dry acetone
(10mL) 1mL, 16mmole MeI was added. The mixture
was heated at reflux for 45min, then cooled to room
temperature, filtered and evaporated under reduced
pressure. The residue was dissolved in CH2Cl2, washed
by two portions of water, dried over anhydrous Na2SO4,
and evaporated under reduced pressure. The product
was purified by flash chromatography (silica, hexane/
ethylacetate 4:1, v/v). Removal of the solvent gave a
93% yield of the product, 30,40-dimethoxyacetophenone
(110mg, 0.61mmole). The product was detected by
HPLC as one peak, with a retention time of 5.2min,
and identified by GC–MS (m/z = 180 M+; 165
M+ꢀCH3; 137 M+ꢀCOCH3).
ured in
a spectrophotometer at 405nm. Protein
content was measured, using Bradford reagent at
595nm.
4.4. Radical donation capacity of chalcones
The radical donation ability of the chalcones was meas-
ured, with trolox as a positive control, using a mixture
with 2,2-diphenyl-1-picrylhydrazyl (DPPH): 9lL of the
compounds (compounds III–V and trolox, 2mM each
and compound II—Butein—0.4mM), dissolved in etha-
nol, was added to a 63lM DPPH solution dissolved in
methanol (63lM). The final volume of the reaction
was 1.8mL. The absorbance at 515nm was measured
at different time interval until the reaction reached a pla-
teau. The donation capacity of the product was then cal-
culated as the molar ratio of a mole of DPPH/mole of
the compound.
2,4-Dimethoxybenzaldehyde: using the same procedure
as above afforded 100mg of 2,4-dimethoxybenzaldehyde
(0.60mmole, 85% yield). The product purity was deter-
mined by HPLC as one peak with a retention time of
6.3min, and identified by GC–MS with m/z = 166 M+;
149 M+ꢀM+ꢀCH3ꢀ2H; 120 M+ꢀCH3ꢀCOCH3ꢀ2H.
4.5. The ability of chalcones to chelate copper ions
Stock solution of each chalcone (2mM) was prepared in
ethanol. Then 50lM solution was prepared in a cuvette
containing PBS (50mM, pH = 6.5) and the sample
absorbance was recorded in the 200–800nm range. A re-
peated scan was recorded after addition of 100lM
CuSO4 and a third spectrum after the addition of EDTA
(250lM).
2,4,30,40-Tetramethoxychalcone: To a stirred solution of
30,40-dimethoxyacetophenone (24mg 0.133mmole) and
2,4-dimethoxybenzaldehyde (25mg, 0.15mmole) in
2mL ethanol at 0°C was added a solution of NaOH
(40%) in water (150lL). The mixture was stirred over-
night at 4°C, and then extracted with two portions of
ethylacetate and the organic layer was separated and
evaporated under reduced pressure. The residue after
evaporation was purified by flash chromatography
(silica, hexane/ethylacetate 7:3 v/v). Removal of the sol-
vent afforded 30,40,2,4-tetramethoxychalcone, 31mg,
0.095mmole (70% yield). The HPLC of the product
has a retention time of 8.85min.
4.6. HPLC, GC–MS and LC–MS analysis
HPLC analysis was performed on an instrument
connected to a diode array detector (HP-1100) and
equipped with a reverse phase column (C-8) (125mm
length; 4mm diameter) of 5lM particles. The mobile
phase was a mixture of acetonitrile and water with a
flow of 1mL/min and a gradient starting with 20%
acetonitirile in water and increasing in concentration
to 80% acetonitrile for 10min, and then to 97% for an
additional 10min. GC–MS analysis was performed by
means of an HP gas chromatograph, Model 5890 Series
II (Waldbronn, Germany), fitted with an HP-5 trace
analysis capillary column (column 0.32mm I.D. 0.25m
film thickness, 5% phenyl methyl silicone), with a mass
selective detector, Model 5972 (Waldbronn, Germany),
2,4,30,40-Tetrahydroxychalcone (compound III): Boron
tribromide BBr3 (100lL) was added to a stirred solution
of 30,40,2,4-tetramethoxychalcone (23mg, 0.07mmole)
in dried dichloromethane (10mL) at 0°C. The mixture
was stirred overnight at 4°C, extracted with two por-
tions of ethylacetate and evaporated under reduced
pressure. The product was purified by reverse phase
chromatography (C18, acetonitrile/water 1:4, v/v) and
the solvent was removed by lyophilizer. Two products
were isolated.

S. Khatib et al. / Bioorg. Med. Chem. 13 (2005) 433–441
441
(1) 2,4,30,40-Trihydroxymonomethoxychalcone (2mg,
0.007mmole—10% yield), which has a retention time
of 5.7min by HPLC, and identified by LC–MS in direct
injection method (ESꢀ; 285.3 MꢀH+) and by UV/vis
by flash chromatography (silica, hexane/ethylacetate
50:50, v/v) and by additional chromatography, using a
reverse phase column (semi-preparative column), result-
ing in two pure compounds.
spectrum (kmax of 374nm, ðeðmMꢀ1 cmꢀ1 ¼ 33:0Þ).
Þ
(1) One with a retention time of 7.5min which was iden-
tified by LC–MS (ESꢀ; m/z of 285.3 Mꢀ1), and by UV/
(2) 2,4,30,40-Tetrahydroxychalcone (4mg, 0.01mmole—
21% yield), with a retention time of 4.18min in HPLC,
and identified by LC–MS in direct injection method
(ESꢀ; m/z of 271.1 Mꢀ1) and by UV/vis spectrum with
vis spectrum (kmax of 390nm ðeðmMꢀ1
¼ 24:32Þ) as
cmꢀ1
Þ
20,40,2,4-trihydroxy monomethoxychalcone (10mg,
0.035mmole) (39% yield) and
kmax 378nm ðeðmMꢀ1 cmꢀ1 ¼ 22:12Þ.
Þ
(2) a second product with a retention time of 5.74min in
HPLC, which was identified by LC–MS (ESꢀ) with m/z
of 271.7 (Mꢀ1) and UV/vis spectrum with kmax of
4.8. Synthesis of 3,5,30,40-tetrahydroxychalcone
(compound IV)
390nm ðeðmMꢀ1
¼ 17:32Þ as 20,40,2,4-tetrahydroxy-
cmꢀ1
Þ
3,5-Dimethoxybenzaldehyde: the same procedure was
used as with 30,40-dimethoxyacetophenone to afford
124mg product (0.74mmole—90% yield). The product
has a retention time of 6.8min in HPLC, and was iden-
tified by GC–MS (m/z = 166 M+; 135 M+ꢀOCH3; 122
M+ꢀCH3ꢀCOCH3).
chalcone (10mg, 0.037mmole—41% yield).
No attempt was made to improve the yields in the above
synthesis.
References and notes
3,5,30,40-Tetramethoxychalcone: The same procedure as
with 30,40,2,4-tetramethoxychalcone was used to give
44mg product (0.134mmole—70% yield). This product
has a retention time of 9.15min in HPLC, and was iden-
tified by proton NMR (1H NMR) (CDCl3, 200mHz):
1. Briganti, S.; Camera, E.; Picardo, M. Pigment Cell Res.
2003, 16, 101–110.
2. Seo, S. Y.; Sharma, V. K.; Sharma, N. J. Agric. Food.
Chem. 2003, 51, 2837–2853.
3. Shimizu, K.; Kondo, R.; Sakai, K.; Takeda, N.; Naga-
hata, T.; Oniki, T. Lipids 2001, 36, 1321–1326.
4. Shimizu, K.; Kondo, R.; Sakai, K. Planta Med. 2000, 66,
11–15.
5. Star, A. E.; Mabry, T. J. Phytochemistry 1971, 10, 2812–
2817.
6. Stevens, J. F.; Taylor, A. W.; Nickerson, G. B.; Ivancic,
M.; Henning, J.; Haunold, A.; Deinzer, M. L. Phytochem-
istry 2000, 53, 759–775.
7. Nerya, O.; Vaya, J.; Musa, R.; Izrael, S.; Ben-Arie, R.;
Tamir, S. J. Agric. Food. Chem. 2003, 51, 1201–1207.
8. Nerya, O.; Musa, R.; Khatib, S.; Tamir, S.; Vaya, J.
Phytochemistry 2004, 65, 1389–1395.
9. Kim, Y. M.; Yun, J.; Lee, C. K.; Lee, H.; Min, K. R.;
Kim, Y. J. Biol. Chem. 2002, 277, 16340–16344.
10. Vaya, J.; Belinky, P. A.; Aviram, M. Free Radic. Biol.
Med. 1997, 23, 302–313.
11. Sanchez-Ferrer, A.; Rodriguez-Lopez, J. N.; Garcia-
Canovas, F.; Garcia-Carmona, F. Biochim. Biophys. Acta
1995, 1247, 1–11.
12. Battaini, G.; Monzani, E.; Casella, L.; Santagostini, L.;
Pagliarin, R. J. Biol. Inorg. Chem. 2000, 5, 262–
268.
13. Shimokoriyama, S.; Geissman, A. T. Chemistry 1960, 25,
1956–1959.
14. Jimenez-Consuelo, M.; Miranda, M. A.; Tormos, R.
Tetrahedron 1997, 53, 14729–14736.
15. Dooley, T. P. J. Dematol. Treat. 1997, 7, 188–200.
16. Kubo, I.; Chen, Q. X.; Nihei, K. I. Food Chem. 2003, 81,
241–247.
17. Van Acker, S. A.; de Groot, M. J.; van den Berg, D. J.;
Tromp, M. N.; Donne-Op den Kelder, G.; van der Vijgh,
W. J.; Bast, A. Chem. Res. Toxicol. 1996, 9, 1305–1312.
18. Vaya, J.; Mahmood, S.; Goldblum, A.; Aviram, M.;
Volkova, N.; Shaalan, A.; Musa, R.; Tamir, S. Phyto-
chemistry 2003, 62, 89–99.
19. Shimizu, K.; Kondo, R.; Sakai, K.; Lee, S. H.; Sato, H.
Planta Med. 1998, 64, 408–412.
20. Ley, J. P.; Bertram, H. J. Bioorg. Med. Chem. 2001, 9,
1879–1885.
dppm; 7.66 (m, 2H); 7.51 (d, 1H); 7.48 (d, 1H); 6.89 (d,
1H); 6.76 (2s, 2H); 6.50 (s, 1H); 3.95 (s, 6H); 3.82 (s, 6H).
3,5,30,40-Tetrhydroxychalcone (compound IV): The
same procedure as with 30,40,2,4-tetrahydroxychalcone
was used to give 30,40,3,5-tetrahydroxychalcone (3mg,
0.01mmole—25% yield) which has a retention time of
3.99min in HPLC, and was identified by LC–MS in direct
injection mode (ESꢀ; m/z of 271.1 Mꢀ1) and by UV/vis
spectrum with kmax of 330nm ðeðmMꢀ1 cmꢀ1 ¼ 9:16Þ.
Þ
4.9. Synthesis of 2,4,20,40-tetrahydroxychalcone
(compound V)
20,40-Dimethoxyacetophenone: the same procedure was
used as for 30,40-dimethoxy acetophenone to afford
100mg product (0.56mmole—85% yield). The product
has a retention time of 6.7min in HPLC, and was iden-
tified by GC–MS (m/z = 180 M+, 165 M+ꢀCH3, 122
M+ꢀCH3ꢀCOCH3).
2,4,20,40-Tetramethoxychalcone: The same procedure
was used as for 30,40,2,4-tetramethoxychalcone, except
that the NaOH solution (40% in water) was added at
room temperature and the reaction mixture was stirred
overnight at room temperature to give 10mg 20,40,2,4-
tetramethoxychalcone (0.03mmole—50% yield). The
product has a retention time of 9.34min in HPLC, and
was identified by proton NMR (CDCl3, 200mHz): dppm
7.87 (d, 1H); 7.70 (d, 1H); 7.48 (dd, 2H); 6.49 (m, 4H);
;
3.83 (3s, 12H).
2,4,20,40-Tetrhydroxychalcone (compound V): The same
procedure as for 30,40,2,4-tetrahydroxychalcone was
used, except that the reaction mixture was stirred over-
night at room temperature. The product was purified