Investigation of binding-site homology between mushroom and bacterial tyrosinases by using aurones as effectors

Article abstract of DOI:10.1002/cbic.201402003

Tyrosinase is a copper-containing enzyme found in plants and bacteria, as well as in humans, where it is involved in the biosynthesis of melanin-type pigments. Tyrosinase inhibitors have attracted remarkable research interest as whitening agents in cosmetology, antibrowning agents in food chemistry, and as therapeutics. In this context, commercially available tyrosinase from mushroom (TyM) is frequently used for the identification of inhibitors. This and bacterial tyrosinase (TyB) have been the subjects of intense biochemical and structural studies, including X-ray diffraction analysis, and this has led to the identification of structural homology and divergence among enzymes from different sources. To better understand the behavior of potential inhibitors of TyM and TyB, we selected the aurone family - previously identified as potential inhibitors of melanin biosynthesis in human melanocytes. In this study, a series of 24 aurones with different hydroxylation patterns at the A- and B-rings were evaluated on TyM and TyB. The results show that, depending on the hydroxylation pattern of A- and B-rings, aurones can behave as inhibitors, substrates, and activators of both enzymes. Computational analysis was performed to identify residues surrounding the aurones in the active sites of both enzymes and to rationalize the interactions. Our results highlight similarities and divergence in the behavior of TyM and TyB toward the same set of molecules. A lighter future: Aurones have been identified as inhibitors of melanin biosynthesis. In this study, 24 aurones were evaluated on mushroom and bacterial tyrosinases (TyM and TyB). The compounds behaved as inhibitors, substrates, or activators of both enzymes. Our results highlight similarities and differences in behavior between TyM and TyB with the same set of molecules.

Full text of DOI:10.1002/cbic.201402003

CHEMBIOCHEM
FULL PAPERS
DOI: 10.1002/cbic.201402003
Investigation of Binding-Site Homology between
Mushroom and Bacterial Tyrosinases by Using Aurones as
Effectors
Romain Haudecoeur,[a] Aurꢀlie Gouron,[b] Carole Dubois,[c] Hꢀlꢁne Jamet,[b] Mark Lightbody,[a]
Renaud Hardrꢀ,[c] Anne Milet,[b] Elisabetta Bergantino,[d] Luigi Bubacco,[d] Catherine Belle,[b]
Marius Rꢀglier,[c] and Ahcꢁne Boumendjel*[a]
Tyrosinase is a copper-containing enzyme found in plants and
bacteria, as well as in humans, where it is involved in the bio-
synthesis of melanin-type pigments. Tyrosinase inhibitors have
attracted remarkable research interest as whitening agents in
cosmetology, antibrowning agents in food chemistry, and as
therapeutics. In this context, commercially available tyrosinase
from mushroom (TyM) is frequently used for the identification
of inhibitors. This and bacterial tyrosinase (TyB) have been the
subjects of intense biochemical and structural studies, includ-
ing X-ray diffraction analysis, and this has led to the identifica-
tion of structural homology and divergence among enzymes
from different sources. To better understand the behavior of
potential inhibitors of TyM and TyB, we selected the aurone
family—previously identified as potential inhibitors of melanin
biosynthesis in human melanocytes. In this study, a series of 24
aurones with different hydroxylation patterns at the A- and B-
rings were evaluated on TyM and TyB. The results show that,
depending on the hydroxylation pattern of A- and B-rings,
aurones can behave as inhibitors, substrates, and activators of
both enzymes. Computational analysis was performed to iden-
tify residues surrounding the aurones in the active sites of
both enzymes and to rationalize the interactions. Our results
highlight similarities and divergence in the behavior of TyM
and TyB toward the same set of molecules.
Introduction
Tyrosinases (Ty, EC 1.14.18.1) are metalloenzymes found in
mammals, plants, fungi, and bacteria, and play critical roles in
melanin pigment biosynthesis.[1] Ty belongs to the type-3
copper-containing enzymes, characterized by a coupled bi-
nuclear active site that accepts phenols and catechols as sub-
strates (Scheme 1).[2,3] In humans, Ty is responsible for the two-
step oxidation of l-tyrosine to dopaquinone: 1) hydroxylation
of l-tyrosine to l-3,4-dihydroxyphenylalanine (l-DOPA), and
2) subsequent oxidation to ortho-quinone (dopaquinone). The
latter undergoes further transformations through several poly-
merization steps to form melanin, the primary determinant of
Scheme 1. Reactions catalyzed by tyrosinases.
skin color. Ty also plays protective roles against UV radiation
and oxidants, and in enzymatic hydrolysis, antimicrobials and
phagocytosis. In bacteria, Ty was reported to be involved in
several processes, such as pathogenic virulence and redox pro-
cesses.[4] In fungi and plants, Ty is a key enzyme in the brown-
ing process that occurs upon tissues damage and long term
storage of fruit and vegetables.
[a] Dr. R. Haudecoeur,+ M. Lightbody, Prof. A. Boumendjel++
Univ. Grenoble Alpes/CNRS, DPM UMR 5063
38041 Grenoble (France)
E-mail: Ahcene.Boumendjel@ujf-grenoble.fr
[b] A. Gouron,+ Dr. H. Jamet, Prof. A. Milet, Dr. C. Belle
Univ. Grenoble Alpes/CNRS, DCM UMR 5250
38041 Grenoble (France)
[c] Dr. C. Dubois,+ Dr. R. Hardrꢀ, Dr. M. Rꢀglier++
Aix-Marseille Universitꢀ, Centrale Marseille, CNRS, ISM2 UMR 7313
13397 Marseille (France)
In terms of structure–function relationship, low homology in
sequence alignments (Table S1, Figures S1–S2 in the Support-
ing Information) between tyrosinases from different origins,
suggests important differences in 3D structure, and therefore
in functional behavior. These structural differences were high-
lighted by recent solutions of the crystallographic structures of
different Ty forms. Because of the difficulties in producing
human Ty (TyH) in high quantities by heterologous expression
(see the Supporting Information), its 3D structure is still un-
[d] Dr. E. Bergantino, Prof. L. Bubacco
Department of Biology, University of Padova
Via Ugo Bassi 58b, Padova, 35121 (Italy)
[+] These authors contributed equally to this work.
[
++] Both senior investigators contributed equally to this work.
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/cbic.201402003.
ꢂ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemBioChem 0000, 00, 1 – 10
&
1
&
ÞÞ
These are not the final page numbers!

CHEMBIOCHEM
FULL PAPERS
www.chembiochem.org
known. In the last decade, the X-ray structures of two bacterial
Ty forms (TyB1 from Bacillus megaterium[5] and TyB2 from
Streptomyces castaneoglobisporus)[6] and two mushroom forms
(TyM1 from Agaricus bisporus[7] and TyM2 from Aspergillus
oryzae)[8] were solved. TyM1 crystallized as an H2L2 (two heavy,
two light subunits) tetramer, including the copper-containing
subunit H (392 residues), whereas TyM2 and the bacterial
counterparts TyB1 and TyB2 crystallized as homodimers (620,
568, and 273 residues, respectively). In these three enzymes,
the active site includes two copper atoms, each coordinated
by three histidine residues (Figure 1). In both TyM1 and TyM2,
with Tys from different sources (e.g., natural plant extracts, as
reported very recently).[17] In order to investigate this hypothe-
sis, we studied the behavior of a series of compounds with Ty
from mushroom and bacterial sources.
Previously, we reported that aurones (benzylidenebenzofur-
an-3(2H)-ones) are inhibitors of melanin biosynthesis in human
melanocytes.[18] Aurones are naturally occurring compounds of
the flavonoid family, distributed across the plant kingdom, and
their therapeutic potential is emerging.[19,20] The effects of
aurones on melanogenesis encouraged us to evaluate their
action on the catalysis of mushroom and bacterial tyrosinases.
For this, we compared the activities of A. bisporus tyrosinase
(TyM1) and Streptomyces antibioticus tyrosinase (TyB3). The
choice of TyB3 was motivated by our expertise in its expression
and purification.[21] In addition, TyB2 and TyB3 belong to the
same Streptomyces genus and share more than 82% identity
(Table S1).
Herein, we report the synthesis of a systematic library of
polyhydroxylated aurones and the in vitro biological evaluation
of this library against TyM1 and TyB3. The interactions between
both Ty forms with aurones were analyzed and rationalized by
computational analysis.
Results and Discussion
Figure 1. Superimposition of active sites of TyB1 (light gray), TyB2 (white),
TyM1 (dark grey) and TyM2 (black).
Synthesis
The synthesis of aurones was accomplished according to
Scheme 2.[20a] Typically, hydroxylated aurones were obtained
from hydroxybenzofuran-3(2H)-one derivatives (1–5) through
aldol condensation with commercially available benzaldehydes,
by one of three procedures: basic KOH conditions in methanol
at 658C, for aurones bearing up to one hydroxyl group per
ring (6b–8b, 6c–8c, 6d–8d; procedure A); more concentrated
KOH conditions in ethanol at 808C, for aurones bearing more
than one hydroxyl group per ring (6e–8e, 6 f–8 f, 9; procedure
one histidine residue (His85 and His94, respectively) is linked
by a thioether bond to the side chain of a cysteine residue
(Cys83 and Cys92, respectively), a feature not shared by TyB1
or TyB2 (where the active site is fairly well conserved). On the
other hand, the second coordination spheres seem to differ
considerably: only 58.3% identity for residues within 6 ꢃ of the
two copper atoms in superimposed structures of TyM1 and
TyB2 (Figure S2).
The medical and economical impact of tyrosinase inhibitors
are attested by the increasing number of compounds de-
scribed recently.[9] In humans, accumulation of high levels of
melanin causes a variety of disorders, such as cutaneous hyper-
pigmentation (solar lentigo, melasma, naevi, etc.) and ocular
retinitis pigmentosa.[10] Also, Ty might be involved in Parkin-
son’s disease development, through degradation of l-DOPA.[11]
In cosmetology, attention is focused on the development of
tyrosinase inhibitors as skin-whitening agents by silencing mel-
anin formation.[12,13] In the food industry, tyrosinase inhibition
(to reduce melanin synthesis) is being developed to control
browning of fruit and vegetables.[14–16]
Most tyrosinase inhibitors were developed by using com-
mercially available mushroom tyrosinase (TyM) as a model,
under the assumption of similarity between TyM and other
tyrosinases. This assumption is rather hazardous because of
the low sequence homology between Tys of different origin
and differences in the binding pocket topologies, as recently
reported for TyB[5,6] and TyM.[7,8] Consequentially, it can be hy-
pothesized that a particular inhibitor might behave differently
Scheme 2. Procedure A: a) KOH/H2O, MeOH, 658C, 1–18 h. Procedure B:
a) KOH/H2O, EtOH, 808C, 2–5 h; b) BBr3, CH2Cl2, 08C to RT, 24–72 h. Proce-
dure C: a) H2SO4, MeOH, 658C.
ꢂ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemBioChem 0000, 00, 1 – 10
&
2
&
ÞÞ
These are not the final page numbers!

CHEMBIOCHEM
FULL PAPERS
www.chembiochem.org
B); and acidic stoichiometric H2SO4 conditions in methanol at
658C, for 4,6,4’-trihydroxyaurone (9d; procedure C). Some 4,6-
dihydroxyaurones (9) were obtained from the corresponding
4,6-dimethoxyaurones (10), by demethylation with boron tri-
bromide in CH2Cl2.[20a]
475 nm against time (formation of DOPAchrome). For TyM1,
Km =(0.39ꢀ0.04) mm, vmax =(0.48ꢀ0.01) mmsꢁ1 (ref. [25]: Km =
(0.6ꢀ0.15) mm; vmax =0.38 mmsꢁ1) and for TyB3, Km =(2.4ꢀ
0.7) mm, vmax =(45ꢀ4) mmsꢁ1.
For substrate behavior, the enzymatic kinetics were deter-
mined by monitoring the disappearance of the large band pre-
sented by the aurones in the UV–visible spectra (~400 nm, ad-
justed for each substrate; Table S4). For activator activity, activi-
ties in absence or presence of aurones were compared at satu-
rating concentrations of l-DOPA (2.8 mm) followed by classical
analysis (plotting 1/vapp against reciprocal activator concentra-
tion).[26] For inhibitors, we verified first that Ty inhibition is con-
centration-dependent, and determined IC50 values at 1.25 mm
l-DOPA.
Structure–activity relationships
Our previous study with a few aurone members (6-hydroxyaur-
ones, Table 1, 8a–e) highlighted the extraordinary versatile ef-
fects of the aurone structure on TyM1 activity.[22] Although the
B-unsubstituted aurone (8a) was found to be inactive, 4’-hy-
droxyaurones (8d and 8e) behaved as alternative substrates
and were converted by TyM1 into ortho-quinone, thereby dem-
onstrating that aurones interact at their B-ring with the dinu-
clear copper–oxygen active site. In contrast, 3’- and 2’-hydrox-
yaurones (8b and 8c) behaved as activators. An important fea-
ture pointed out in our previous study is the likely occurrence
of multiple binding sites on TyM1 for aurones.[22] In the present
study, we compared the behavior of TyM1 and TyB3 toward
a larger library of aurones: those with diversely hydroxylated
A- and B-rings. Tyrosinase inactivation by resorcinols was re-
cently reported; this prompted us to include 6 f–9 f, which
contain resorcinol moieties at the B-ring.[23] For TyM1, we used
a commercially available A. bisporus Ty, purified by Q-sepharose
FF chromatography according to a known procedure.[24] The
S. antibioticus TyB3 was produced in the strain S. antibioticus
pIJ703 and purified according to a published procedure.[21] For
both enzymes, we determined the Michaelis–Menten kinetic
parameters for l-DOPA oxidation by plotting absorbance at
The previously reported values for aurones 8a–8e with
TyM1 (discussed above)[22] were fully reflected for other au-
rones. Regardless the A-ring substitution, the general behavior
associated with the B-ring substitution remained similar. In all
cases, an unsubstituted B-ring led to inactive compounds; 2’-
hydroxy- (6b–9b) and 3’-hydroxyaurones (6c–9c) were TyM1
activators (Table 2); 4’-hydroxy- (6d–9d) and 3’,4’-dihydrox-
yaurones (6e–9e) were identified as alternative TyM1 sub-
strates (Table 3).
The second important discovery was that aurones have dif-
ferent reactivities toward TyM1 and TyB3. The main differences
were observed with 2’-hydroxy- (6b–9b) and 3’-hydroxyaur-
ones (6c–9c). Although 6b–9b and 6c–9c exhibited activation
of TyM1, they behaved as poor inhibitors of TyB3 (Table 2). This
difference can be explained by the significant differences in
the structures of TyM1 and TyB3. Unlike TyBs, which are mono-
Table 1. Activity of aurones (A, activator; I, inhibitor; S, substrate; n.a., no activity; n.d., not determined).
TyM1
TyB3
TyM1
TyB3
TyM1
TyB3
TyM1
TyB3
TyM1
TyB3
TyM1
TyB3
6a
7a
8a
6b
7b
8b
6c
7c
8c
6d
7d
8d
6e
7e
8e
6 f
7 f
8 f
n.a.
n.a.
n.a.
n.a.
A
A
I*
I*
A
A
I*
I*
S***
S**
S***
S**
S**
S**
n.d.
n.d.
I***
I*
I***
I**
n.a.
n.a.
A
I*
A
I*
S***
S***
S*
S*
I*
I**
9a
9b
9c
9d
9e
9 f
n.a.
n.a.
A
I**
A
I**
S***
S****
S**
S****
I**
I***
**** Very high activity (Km or IC50 ꢂ1 mm). *** High activity (20 mmꢃKm or IC50 >1 mm). ** Moderate activity (200 mmꢃKm or IC50 >20 mm). * Low activity
(1 mmꢃKm or IC50 >200 mm).
ꢂ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemBioChem 0000, 00, 1 – 10
&
3
&
ÞÞ
These are not the final page numbers!

CHEMBIOCHEM
FULL PAPERS
www.chembiochem.org
Km =52.9 mm) to poor activity
(8e, Km =288 mm), whereas (as
for 9d), the 4,6-disubstituted
aurone (9e) was a very efficient
substrate for TyB3 (Km =1 mm).
Finally, 2’,4’-dihydroxyaurones
(resorcinol-based aurones, 6 f–
9 f) behaved as inhibitors of
both enzymes (Tables 1 and 4),
but with significant IC50 varia-
tion. For TyM1, 2’,4,4’-trihydrox-
yaurone (6 f) behaved as
Table 2. Hyperbolic activation (H.A.) values and IC50 determined for compounds 6b–9b and 6c–9c against
TyM1 and TyB3.
Compound
R4
R5
R6
R2’
R3’
TyM1 H.A. [%]
TyB3 IC50 [mm]
6b
7b
8b
9b
6c
7c
8c
9c
OH
H
H
OH
OH
H
H
OH
H
H
H
OH
H
H
H
H
OH
OH
H
H
OH
OH
OH
OH
OH
OH
H
H
H
H
H
H
H
H
OH
OH
OH
OH
200
120
120
180
170
150
240
150
400ꢀ20
450ꢀ100
>1000
150ꢀ25
>400
700ꢀ100
>1000
>200
H
OH
meric in solution, TyM1 is ob-
tained as a heterodimer of a
heavy subunit (containing the
active center) and a light subu-
nit.
Table 4. Inhibition constants and IC50 values with standard deviations determined for compounds 6 f–9 f
against TyM1 and TyB3.
Compound
R4
R5
R6
R2’
R4’
TyM1 IC50 [mm]
TyB3 IC50 [mm]
6 f
7 f
8 f
9 f
OH
H
H
H
OH
H
H
H
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
9ꢀ1
>1000
>1000
4ꢀ1
62ꢀ12
100ꢀ10
18ꢀ3
For the other aurones, the ac-
tivities were similar. A substrate
or inhibitor for TyM1 behaved
similarly toward TyB3. The differ-
ences lay in the range of activi-
OH
H
300ꢀ20
ties. For example, the 4’-hydroxyaurones (6d–9d) showed the
same reactivity for both Tys; they behaved as substrates for
both enzymes (Table 3). However some Km differences were ob-
served, depending on the A-ring substitution. Introduction of
a hydroxy substituent at the 5-position (7d) induced a negative
effect on Km; 7d was found to be a bad substrate for TyM1
(Km =170 mm) and TyB3 (Km =79 mm). Regarding hydroxylation
of positions 4 and 6, both hydroxy aurones constituted good
substrates for TyM1 (6d, Km =5 mm; 8d, Km =4.8 mm) and TyB3
(6d, Km =10 mm; 8d, Km =9 mm). Interestingly, the 4,6-dihy-
droxy aurone (9d) showed increased Km with TyM1 (Km =
18.2 mm), but decreased Km (by two orders of magnitude) with
TyB3 (Km =0.2 mm), compared to monosubstituted aurones 6d
and 8d. For 3’,4’-dihydroxy aurones (catechol-based aurones,
6e–9e), the hydroxylation effect was less clear. With TyM1,
6e–9e exhibited average (6e, Km =51 mm; 7e, Km =53 mm, 9e,
a good inhibitor (IC50 =9 mm) whereas the 2’,4,4’,6-tetrahydrox-
yaurone (9 f) showed weaker activity (IC50 =300 mm). Monosub-
stituted 5- and 6-aurones (7 f and 8 f) displayed almost no ac-
tivity (IC50 >1 mm). For TyB3, the tendency observed with sub-
strates 6d–9d was conserved. The best inhibitor was the 4,6-
disubstituted aurone (9 f, IC50 =18 mm), noticeably better than
the 4- and 6-substituted counterparts (6 f, IC50 =100 mm, 8 f,
IC50 =100 mm). Collectively, these results reveal distinct variabil-
ity in interaction of these aurones with TyM1 and TyB3.
Aurone–Ty complexes: interaction studies using quantum
and molecular mechanics studies
To rationalize the interactions between the aurones and the
two Ty enzymes, dynamics coupling quantum and molecular
mechanics (QM/MM) were performed. Calculations were con-
Table 3. Michaelis constants and maximum reaction rates with standard deviations determined for compounds 6d–9d and 6e–9e against TyM1 and TyB3
(n.d., not determined).
Compound
R4
R5
R6
R3’
R4’
TyM1
vapp [mmminꢁ1
TyB3
vapp [mmminꢁ1
Km [mm]
]
Km [mm]
]
6d
7d
8d
9d
6e
7e
8e
9e
OH
H
H
OH
OH
H
H
OH
H
H
H
OH
H
H
H
H
OH
OH
H
H
OH
OH
H
H
H
H
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
5ꢀ1
0.046ꢀ0.003
0.56ꢀ0.09
3.0ꢀ0.3
2.4ꢀ0.6
7.1ꢀ0.6
25ꢀ3
10ꢀ3
79ꢀ14
9ꢀ2
0.10ꢀ0.01
0.42ꢀ0.06
0.63ꢀ0.05
6.7ꢀ0.8
n.d.
170ꢀ50
4.8ꢀ0.1
18.2ꢀ0.1
51ꢀ12
0.20ꢀ0.08
n.d.
n.d.
300ꢀ200
1.0ꢀ0.3
53ꢀ5
n.d.
H
OH
288ꢀ10
52.9ꢀ0.3
91ꢀ8
6.1ꢀ0.7
0.36ꢀ0.05
39ꢀ5
ꢂ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemBioChem 0000, 00, 1 – 10
&
4
&
ÞÞ
These are not the final page numbers!

CHEMBIOCHEM
FULL PAPERS
www.chembiochem.org
ducted with 6 f, 8 f, and 9 f (Table 4), which showed the great-
est contrast in IC50 values. Estimation of the binding free ener-
gies of aurone–Ty complexes was computed with the inclusion
of solvation effects through Poisson–Boltzmann/surface area
calculations, namely QM/MM-PBSA.[27–29] This approach provid-
ed valuable information for better identification of the residues
surrounding the aurones in both Tys. Similar MM evaluation
was performed on the docked positions of thujaplicins with
a homology model of TyM.[30] In our case, aurones 6 f, 8 f, and
9 f are competitive inhibitors, so their interaction with the
copper atoms of the active site had to be incorporated into
the theoretical calculations for QM analysis.
For TyM1, calculations were performed with the recent X-ray
structure.[7] In the absence of an X-ray structure of TyB3, these
calculations were performed by using crystallographic data for
TyB2 (>82% identity to TyB3; Table S1).[6] For the aurone, the
choice and the construction of the initial position in the active
site are detailed in Computational Methods (below). QM/MM
geometry optimizations of the aurone–Ty complexes were per-
formed. For each enzyme, the optimized structures for the
three inhibitors in the active site were found to be very close,
and this cannot explain the experimentally observed differen-
ces between the three ligands in terms of activity. Then protein
flexibility was considered, by computing short QM/MM dynam-
ics for the six complexes. These simulations did not identify
major structural differences at the active site and similar
copper–copper (3.5–3.9 ꢃ), copper–ligand (1.9–2.1 ꢃ), and
copper–water (2.0–2.4 ꢃ) distances were observed. But several
differences were observed for the interactions with neighbor-
ing active-site residues. Specifically, if no major differences
were identified between TyB2 complexes, we were able to
highlight key interactions for TyM1; this allowed efficient dis-
crimination of the three studied inhibitors.
For TyM1, the calculated binding free energies (QM/MM-
PBSA) matched the experimental order of inhibition efficiency
(6 f>9 f>8 f, Figure 2). Several components of these energies
(Table S2) were analyzed to identify residues contributing the
most. Some residues in the second coordination sphere of the
active site (2.5–4 ꢃ from the ligand) led to favorable van der
Waals effects. The major contributions for these interactions
were from Phe264 and Val283, but Asn260, Val248, and His263
also had an effect (Figure 2). Compounds 6 f and 9 f seemed
to be more buried in the binding pocket than 8 f. For 6 f, the
most important feature was Phe264, rotation of which places
the plane of the aromatic ring parallel to the five-membered
ring of the aurone, whereas it is perpendicular to five-mem-
bered rings of 8 f and 9 f. The weak interactions with Val248
seemed to contribute more in the case of 9 f than for 6 f and
8 f. Residues Asn260, His263, and Arg268 displayed electrostat-
ic effects on the ligands. In the case of 8 f, very weak interac-
tions were found between Arg268 and the hydroxyl group at
position 6 (6.7 ꢃ), and the inhibitor was experimentally deter-
mined to be inactive (DGcalcd =6.1 kcalmolꢁ1). For compound
9 f, Arg268 seemed to interact more strongly with the phenol
group at position 4 of the aurone (2.7 ꢃ), in agreement with its
inhibitory activity (IC50 =300 mm; DGcalcd =3.9 kcalmolꢁ1). In
both complexes, the arginine was not shifted and led to di-
Figure 2. Snapshots from the MD trajectory of TyM1–aurone complexes.
hedral angles Ca-Cb-Cg-Cd of ꢁ1608 to ꢁ1808. Interestingly, in
the case of 6 f, Arg268 underwent a rotation (Ca-Cb-Cg-Cd 130
to 1408) and shifted to build a hydrogen bond with the car-
bonyl of aurone and some weaker electrostatic interactions
with the phenol at position 4 (3.9 ꢃ). This motion was allowed
because of the absence of 6-hydroxyl group in the case of 6 f,
whereas 8 f and 9 f bear this substituent. Additionally, the argi-
nine was stabilized in this position through the formation of
an extra hydrogen bond with the main chain of adjacent
Leu275 (1.9 ꢃ). All these elements contributed to a strength-
ened inhibitory effect of TyM1 (IC50 =9 mm; DGcalcd =ꢁ5.7 kcal
molꢁ1).
ꢂ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemBioChem 0000, 00, 1 – 10
&
5
&
ÞÞ
These are not the final page numbers!

CHEMBIOCHEM
FULL PAPERS
www.chembiochem.org
For TyB2, although binding affinities DGcalcd seemed to
follow the experimental order of inhibition efficiency, high
standard deviation values prevented us from drawing a clear
rational conclusion from these calculations (Table S3). Addition-
ally, no major differences were identified in the interactions be-
tween the three inhibitors and the neighboring active-site resi-
dues, even after QM/MM dynamics. However the calculations
enabled us to identify the hot-spot residues for TyB2 and to
highlight divergences from TyM1. In TyB2, the main van der
Waals and electrostatic contributions arose from residues Ile42,
Asn191, and Trp183. The aurones also seemed to stand at the
edge of the binding pocket (5 ꢃ from Val195; Figure 3). Hence,
their A-rings allowed much less interaction with the protein
residues than for TyM1, and, logically, van der Waals contribu-
tions in TyB2 were significantly lower.
A brief comparison between TyB2 and TyM1 crystallographic
structures provided a first glimpse of similarities and differen-
ces between their active sites. The residues surrounding the
aurones in TyB2 and TyM1 were identified after performing
QM/MM dynamics on aurone–Ty complexes. Superimposition
of the structures revealed that residues at 6 ꢃ from the two
copper centers and aurones in the structures exhibit only
28.6% identity (see the Supporting Information). The analysis
of interactions between Ty and aurones highlighted the simi-
larly positioned hot-spot residues in TyM1 and TyB2, especially
Ile42/Val283, Val195/Phe264, Asn191/Asn260, and Trp183/
Val248 (for TyB2/TyM1). Thus, their contributions to the respec-
tive interaction patterns appeared to diverge. Finally, the
extensively interacting residue Arg268 of TyM1 was missing in
the case of TyB2. This divergence constitutes one of the keys
for affinity variability for the three aurones 6 f, 8 f, and 9 f.
Conclusions
Aurones were investigated as ligands of tyrosinases from two
sources (mushroom and bacteria) with the aim of better un-
derstanding the behavior of both enzymes toward the same
compounds. The synthesis of 24 diversely hydroxylated au-
rones allowed us to identify common and divergent reactivity
profiles toward bacterial and mushroom tyrosinases, TyB3 and
TyM1. Substrate, activator, and inhibitor activities were found.
The behavior and efficiency strongly depend on the A- and B-
ring hydroxylation patterns, as well as on the enzyme origin.
2’,4’-Dihydroxyaurones (resorcinol-like aurones) were identified
as inhibitors of both TyM1 (IC50 =9–1000 mm) and TyB3 (IC50 =
18–100 mm), with a high degree of potency. The QM/MM calcu-
lations on the complexes between inhibitors derived from au-
rones and TyM1 or TyB2 allowed us to identify key interactions
and to highlight divergences between the two enzymes. In
TyM1, we identified some residues to explain the differences in
inhibition efficiency, such as Arg268, which can form a hydro-
gen bond with the carbonyl of some aurones. The second
sphere of coordination of the active site is very poorly con-
served between TyM1 and TyB2. Analysis of the computed
interaction energies showed that the residues of the binding
pockets interact differently with aurones.
These results suggest that different modes of binding and
action of aurones occur, and that the different behaviors could
be relevant to human tyrosinase. The main conclusion reached
in this study is that inhibitors discovered by using TyM are not
automatically effective for Tys from other origins, and that cau-
tion should thus be exercised.
Experimental Section
1
Chemistry: H and 13C NMR spectra were recorded on an AC-400
instrument (Bruker; 400 MHz for 1H, 100 MHz for 13C). Chemical
shifts are reported in ppm relative to Me4Si (internal standard).
Electrospray ionization (ESI) mass spectra were acquired at the An-
alytical Department of Grenoble University on an Esquire 300 Plus
Figure 3. Snapshots from the MD trajectory of TyB2–aurone complexes.
ꢂ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemBioChem 0000, 00, 1 – 10
&
6
&
ÞÞ
These are not the final page numbers!

CHEMBIOCHEM
FULL PAPERS
www.chembiochem.org
instrument (Bruker Daltonics) with a nanospray inlet. Combustion
analyses were performed at the Analytical Department of Grenoble
University; all tested compounds had a purity of at least 95%.
Silica gel F-254 plates (0.25 mm; Merck) were used for thin-layer
chromatography (TLC), and silica gel 60 (200–400 mesh; Merck)
was used for flash chromatography. All solvents were distilled prior
to use. Unless otherwise stated, reagents were obtained from com-
mercial sources and were used without further purification. The
synthesis of 1–5, 6 f, 7d, 8a–8d, 9a–9b, 9d–9 f, and 10a–10d is
as previously reported.[20a]
(Z)-2-(2-hydroxybenzylidene)-5-hydroxybenzofuran-3(2H)-one
(7b): The compound was prepared according to procedure A,
starting from 5-hydroxybenzofuran-3(2H)-one (2) and 2-hydroxy-
benzaldehyde, and was purified by column chromatography on
silica gel eluted with CH2Cl2/MeOH (9:1) to yield a pure yellow
solid (88%). M.p. >2308C (decomposition); 1H NMR (400 MHz,
[D6]DMSO): d=10.36 (brs, 1H; OH), 9.77 (brs, 1H; OH), 8.12 (d, J=
7.7 Hz, 1H; H6’), 7.39 (d, J=8.6 Hz, 1H; H6), 7.28 (t, J=7.7 Hz, 1H;
H4’), 7.21 (d, J=8.6 Hz, 1H; H7), 7.19 (s, 1H; ꢁCH=), 7.02 (s, 1H;
H4), 6.95 ppm (m, 2H; H3’, H5’); 13C NMR (100 MHz, [D6]DMSO): d=
183.7, 159.1, 157.4, 153.8, 146.7, 131.7, 131.1, 125.5, 121.4, 119.7,
118.8, 115.8, 113.9, 107.7, 106.0 ppm; MS (ESI): m/z 255 [M+H]+,
277 [M+Na]+.
Preparation of (Z)-2-benzylidenebenzofuran-3(2H)-one deriva-
tives (procedure A): Aqueous potassium hydroxide (50%,
1.5 mLmmolꢁ1) and a benzaldehyde derivative (1.5 equiv) were
added to a solution of a benzofuran-3(2H)-one derivative in metha-
nol (15 mLmmolꢁ1). The solution was then refluxed until TLC
showed complete disappearance of the starting material (1–18 h).
After cooling, the mixture was concentrated under reduced pres-
sure, then the residue was diluted in water (50 mLmmolꢁ1), and
aqueous hydrochloric acid (1m) was added (to pH 2–3). The mix-
ture was extracted with ethyl acetate or dichloromethane and
washed with water and brine. The combined organic layers were
dried over magnesium sulfate and filtered, and the filtrate was con-
centrated under reduced pressure to give the crude compound.
(Z)-2-(3-hydroxybenzylidene)-5-hydroxybenzofuran-3(2H)-one
(7c): The compound was prepared according to procedure A, start-
ing from 5-hydroxybenzofuran-3(2H)-one (2) and 3-hydroxybenzal-
dehyde, and was purified by column chromatography on silica gel
eluted with CH2Cl2/MeOH (9:1) to yield a pure yellow solid (73%).
M.p. >2308C (decomposition); 1H NMR (400 MHz, [D6]DMSO): d=
9.80 (brs, 1H; OH), 9.68 (brs, 1H; OH), 7.39 (m, 3H; H6, H2’, H6’),
7.29 (t, J=7.8 Hz, 1H; H5’), 7.23 (dd, J1 =8.8 Hz, J2 =2.6 Hz, 1H;
H7), 7.03 (d, J=2.6 Hz, 1H; H4), 6.86 (dd, J1 =7.8 Hz, J2 =1.9 Hz,
1H; H4’), 6.78 ppm (s, 1H; ꢁCH=); 13C NMR (100 MHz, [D6]DMSO):
d=183.9, 159.3, 157.6, 153.9, 147.0, 133.1, 129.9, 125.8, 122.6,
121.2, 117.5, 117.4, 113.8, 112.0, 107.8 ppm; MS (ESI): m/z 255
[M+H]+, 277 [M+Na]+.
Preparation of (Z)-2-benzylidenebenzofuran-3(2H)-one deriva-
tives (procedure B): Aqueous potassium hydroxide (50%,
5 mLmmolꢁ1) and a benzaldehyde derivative (2 equiv) were added
to benzofuran-3(2H)-one derivative in ethanol (4 mLmmolꢁ1). The
solution was then refluxed until TLC showed complete disappear-
ance of the starting materials (2–5 h). After cooling, the mixture
was concentrated under reduced pressure, then the residue was
diluted in water (50 mLmmolꢁ1), and aqueous hydrochloric acid
(1m) was added (to pH 2–3). The mixture was extracted with ethyl
acetate or dichloromethane and washed with water and brine. The
combined organic layers were dried over magnesium sulfate and
filtered, and the filtrate was concentrated under reduced pressure
to give the crude compound.
(Z)-2-(3,4-hydroxybenzylidene)-5-hydroxybenzofuran-3(2H)-one
(7e): The compound was prepared according to procedure B, start-
ing from 5-hydroxybenzofuran-3(2H)-one (2) and 3,4-dihydroxy-
benzaldehyde, and was purified by column chromatography on
silica gel eluted with CH2Cl2/MeOH (9:1) to yield a pure yellow
solid (15%). M.p. >3008C (decomposition). 1H NMR (400 MHz,
[D6]DMSO): d=9.73 (brs, 2H; OH), 9.29 (brs, 1H; OH), 7.47 (d, J=
2.0 Hz, 1H; H2’), 7.34 (d, J=8.8 Hz, 1H; H6), 7.29 (dd, J1 =8.4 Hz,
J2 =2.0 Hz, 1H; H6’), 7.20 (dd, J1 =8.8 Hz, J2 =2.7 Hz, 1H; H7), 7.00
(d, J=2.7 Hz, 1H; H4), 6.84 (d, J=8.4 Hz, 1H; H5’), 6.73 ppm (s, 1H;
ꢁCH=); 13C NMR (100 MHz, [D6]DMSO): d=183.4, 158.8, 153.7,
148.4, 145.6, 145.5, 125.3, 125.0, 123.4, 121.7, 118.1, 116.0, 113.6,
113.4, 107.6 ppm; MS (ESI): m/z 271 [M+H]+, 293 [M+Na]+.
(Z)-2-(3,4-dihydroxybenzylidene)-4-hydroxybenzofuran-3(2H)-
one (6e): The compound was prepared according to procedure B,
starting from 4-hydroxybenzofuran-3(2H)-one (1) and 3,4-dihydrox-
ybenzaldehyde, and was purified by column chromatography on
silica gel eluted with CH2Cl2/MeOH (9:1) to yield a pure yellow
solid (43%). M.p. >2608C (decomposition); 1H NMR (400 MHz,
[D6]DMSO): d=11.04 (brs, 1H; OH), 9.68 (brs, 1H; OH), 9.30 (brs,
1H; OH), 7.53 (t, J=8.2 Hz, 1H; H6), 7.46 (d, J=1.9 Hz, 1H; H2’),
7.26 (dd, J1 =8.4 Hz, J2 =1.9 Hz, 1H; H6’), 6.84 (m, 2H; H7, H5’),
6.63 (d, J=8.2 Hz, 1H; H5), 6.63 ppm (s, 1H; ꢁCH=); 13C NMR
(100 MHz, [D6]DMSO): d=181.0, 165.7, 156.9, 147.9, 145.5, 144.8,
138.1, 124.5, 123.4, 117.8, 116.0, 111.5, 110.3, 109.4, 102.3 ppm; MS
(ESI): m/z 269 [MꢁH]ꢁ.
(Z)-2-(2,4-hydroxybenzylidene)-5-hydroxybenzofuran-3(2H)-one
(7 f): The compound was prepared according to procedure B, start-
ing from 5-hydroxybenzofuran-3(2H)-one (2) and 2,4-dihydroxy-
benzaldehyde, and was purified by column chromatography on
silica gel eluted with CH2Cl2/MeOH (9:1) to yield a pure yellow
solid (88%). M.p. >3008C (decomposition); 1H NMR (400 MHz,
[D6]DMSO): d=10.05 (brs, 3H; OH), 7.98 (d, J=8.5 Hz, 1H; H6’),
7.34 (d, J=8.8 Hz, 1H; H6), 7.17 (s, 1H; ꢁCH=), 7.16 (dd, J1 =8.8 Hz,
J2 =2.7 Hz, 1H; H7), 6.99 (d, J=2.7 Hz, 1H; H4), 6.37 ppm (m, 2H;
H3’, H5’); 13C NMR (100 MHz, [D6]DMSO): d=182.9, 161.8, 160.2,
158.4, 153.6, 144.8, 132.6, 124.8, 122.0, 113.6, 110.5, 108.4, 107.9,
107.4, 102.4 ppm; MS (ESI): m/z 271 [M+H]+, 293 [M+Na]+.
(Z)-2-benzylidene-5-hydroxybenzofuran-3(2H)-one (7a): The
compound was prepared according to procedure A, starting from
5-hydroxybenzofuran-3(2H)-one (2) and benzaldehyde, and was
purified by column chromatography on silica gel eluted with
CH2Cl2/MeOH (9:1) to yield a pure yellow solid (10%). M.p. 216–
(Z)-2-(3,4-dihydroxybenzylidene)-6-hydroxybenzofuran-3(2H)-
one (8e, sulfuretin): The compound was prepared according to
procedure B, starting from 6-hydroxybenzofuran-3(2H)-one (3) and
3,4-dihydroxybenzaldehyde, and was purified by column chroma-
tography on silica gel eluted with CH2Cl2/MeOH (9:1) to yield
1
2188C; H NMR (400 MHz, [D6]DMSO): d=9.83 (brs, 1H, OH), 7.98
1
(d, J=7.3 Hz, 2H; H2’, H6’), 7.49 (m, 3H; H3’, H4’, H5’), 7.41 (d, J=
8.8 Hz, 1H; H6), 7.23 (dd, J1 =8.8 Hz, J2 =2.6 Hz, 1H; H7), 7.03 (d,
J=2.6 Hz, 1H; H4), 6.89 ppm (s, 1H; ꢁCH=); 13C NMR (100 MHz,
[D6]DMSO): d=185.5, 160.9, 155.5, 148.7, 133.6, 132.8, 131.5, 130.5,
127.4, 122.7, 115.4, 113.2, 109.3 ppm; MS (ESI): m/z 239 [M+H]+.
a pure yellow solid (26%). H NMR (400 MHz, [D6]DMSO): d=9.69
(brs, 3H; OH), 7.58 (d, J=8.4 Hz, 1H; H4), 7.44 (d, J=1.9 Hz, 1H;
H2’), 7.24 (dd, J1 =8.2 Hz, J2 =1.9 Hz, 1H; H6’), 6.83 (d, J=8.2 Hz,
1H; H5’), 6.72 (d, J=1.7 Hz, 1H; H7), 6.68 (dd, J1 =8.4 Hz, J2 =
1.7 Hz, 1H; H5), 6.62 ppm (s, 1H; ꢁCH=); 13C NMR (100 MHz,
ꢂ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemBioChem 0000, 00, 1 – 10
&
7
&
ÞÞ
These are not the final page numbers!

CHEMBIOCHEM
FULL PAPERS
www.chembiochem.org
[D6]DMSO): d=181.1, 167.4, 166.3, 148.0, 145.6, 145.5, 125.7, 124.5,
123.4, 117.9, 116.0, 113.1, 112.9, 111.8, 98.3 ppm; MS (ESI): m/z 271
[M+H]+.
tonated form while the other hydroxy groups of the aurone were
protonated. This hypothesis was tested, and the results are given
in Supporting Information (Figures S3 and S4). Then, MM minimiza-
tion with AMBER10 was carried out with the protein and the fixed
ligand.[34] For copper ions, histidine residues, and the hydroxyl
group linked to the copper atoms in the active site, specific param-
eters were used.[32] QM/MM calculations were performed with the
Gaussian03 package[35] for QM and with Tinker 4.2[36] for MM. Both
mechanical and electrostatic embedding (direct polarization)
schemes were used to describe the interactions between the QM
and the MM parts, thus ensuring a good description of the effects
of the environment of the active site. The QM part contained the
two coppers (in triplet state), the hydroxyl ion, the ligand, a water
molecule close to the copper, the side chains of the six histidines
linked to the coppers, and (for TyM1) Cys83, which forms a covalent
thioether bond with His85. The rest of the protein and the solvent
were computed at the MM level. The partition between QM and
MM parts led to the cut of six or seven CaꢁCb bonds of the histi-
dines (and of Cys83). The link atom scheme was used to treat the
QM/MM boundary. The QM part was computed with the B3LYP/6-
31G* level of theory and the MM part with Amber 99SB force-field
parameters. Geometry optimizations were first performed only
with residues at 4 ꢃ of the QM part and allowed to move (other
residues were frozen). Afterwards, QM/MM Born-Oppenheimer mo-
lecular dynamics were run in the NVT ensemble at 298 K by using
the Berendsen algorithm. The time step was increased by using
mass-scaling molecular dynamics. The hydrogen mass was set at
10 amu in order to “freeze” the XꢁH vibrations and allow us to
employ of 3 fs time steps for the dynamics.[37] For each complex,
a 4 ps dynamic simulation was performed. After 1.5 ps of equilibra-
tion, ten snapshots were selected to calculate the binding affinities
by using the QM/MM-PBSA approach (Equations (1) and (2)).
(Z)-2-(2,4-dihydroxybenzylidene)-6-hydroxybenzofuran-3(2H)-
one (8 f): The compound was prepared according to procedure B,
starting from 6-hydroxybenzofuran-3(2H)-one (3) and 2,4-dihydrox-
ybenzaldehyde, and was purified by column chromatography on
silica gel eluted with CH2Cl2/MeOH (9:1) to yield a pure red solid
(72%). M.p. >2308C (decomposition); 1H NMR (400 MHz,
[D6]DMSO): d=10.24 (brs, 3H; OH), 7.97 (d, J=8.5 Hz, 1H; H6’),
7.57 (d, J=8.4 Hz, 1H; H4), 7.05 (s, 1H; ꢁCH=), 6.74 (d, J=1.7 Hz,
1H; H7), 6.68 (dd, J1 =8.4 Hz, J2 =1.7 Hz, 1H; H5), 6.38–6.42 ppm
(m, 2H; H3’, H5’); 13C NMR (100 MHz, [D6]DMSO): d=181.0, 167.2,
166.3, 160.8, 159.1, 145.1, 132.4, 125.5, 113.2, 112.8, 110.5, 108.3,
105.8, 102.2, 98.4 ppm; MS (ESI): m/z 269 [MꢁH]ꢁ.
(Z)-2-(3-hydroxybenzylidene)-4,6-dihydroxybenzofuran-3(2H)-
one (9c): The compound was prepared according to procedure B,
starting from 4,6-dihydroxybenzofuran-3(2H)-one (4) and 3-hydrox-
ybenzaldehyde, and was purified by column chromatography on
silica gel eluted with CH2Cl2/MeOH (9:1) to yield a pure yellow
solid (49%). M.p. >2308C (decomposition); 1H NMR (400 MHz,
[D6]DMSO): d=11.03 (brs, 2H; OH), 9.63 (brs, 1H; OH), 7.28 (m,
3H; H2’, H5’, H6’), 6.80 (ddd, J1 =7.5 Hz, J2 =2.1 Hz, J3 =1.5 Hz, 1H;
H4’), 6.49 (s, 1H; ꢁCH=), 6.19 (d, J=1.6 Hz, 1H; H7), 6.07 ppm (d,
J=1.6 Hz, 1H; H5); 13C NMR (100 MHz, [D6]DMSO): d=179.0, 167.8,
167.6, 158.5, 157.5, 147.7, 133.4, 129.8, 121.9, 116.9, 116.5, 108.3,
102.5, 97.8, 90.5 ppm; MS (ESI): m/z 269 [MꢁH]ꢁ.
Enzymatic assays: A. bisporus TyM1 (6.4 mg; Sigma–Aldrich) was
dissolved in phosphate buffer (5 mL, 50 mm, pH 7.0) and purified
by Q-Sepharose FF chromatography with an NaCl gradient (0–
1.0m).[24] The purity of tyrosinase was checked with SDS-PAGE (pu-
rified tyrosinase exhibits two bands: Mw ~14 and 45 kDa). Tyrosi-
nase activity was measured spectroscopically with l-DOPA as the
substrate. TyB3 was prepared from liquid cultures of S. antibioticus
harboring the pIJ703 expression plasmid, and purified according to
published procedures.[21] Through all purification steps, protein sol-
utions were kept on ice to prevent loss of enzymatic activity. The
purity of the sample was indicated by the presence of a single
band in the SDS-PAGE gel. All compounds were dissolved in 10%
DMSO, and the DMSO stock solutions were diluted in phosphate
buffer (pH 7.0). Tyrosinase (6 U) was preincubated with the com-
pounds in phosphate buffer (50 mm, pH 7.0) for 5 min at 258C,
then l-DOPA (2 mm) was added to the reaction mixture. The
enzyme reaction was monitored by measuring the absorbance at
475 nm (DOPAchrome) for 5 min.
DG ¼ GcomplexꢁGproteinꢁGligand
ð1Þ
ð2Þ
GðXÞ ¼ EQM=MMðXÞ þ GPBðXÞ þ GSAðXÞ
Having compared three structurally close ligands binding the same
protein, changes in solute entropy were not included in the calcu-
lations. For the QM/MM scoring function, only the ligand was treat-
ed at the QM level. The PBSA part was computed with AMBER10.
The charges used for these calculations were the charges of the
Amber 99SB force field for the protein, the Mulliken charges for
the coppers, and the hydroxyl ion obtained after our QM/MM sim-
ulations. For the ligands, charges were derived with a RESP fit on
HF/6-31g* calculations.[38] The decomposition of the QM/MM inter-
action in a van der Waals contribution and an electrostatic contri-
bution are given in the Supporting Information (Tables S2 and S3).
Computational methods: Calculations were performed with the
recently resolved mushroom tyrosinase from A. bisporus (PDB ID:
2Y9X; TyM1)[7] and the bacterial tyrosinase from S. castaneoglobis-
porus (PDB ID: 2AHK; TyB2).[6] The initial PDB structures were sol-
vated in a TIP3P water rectangular box (10 ꢃ larger than the solute
in every direction). All histidines were protonated. To construct the
initial position of the aurone in the active Ty sites, we employed
the optimized structures of TyM1 and TyB2 complexed with depro-
tonated HOPNO (2-hydroxypyridine-N-oxide), an efficient competi-
tive inhibitor of TyM1[31] and TyB3[32] in a monodentate binding
mode. As the enzymatic study[22] demonstrate that aurones are
more likely to interact with the active site at the B-ring, we aligned
this ring with HOPNO by using the “pair fitting” tool of PYMOL.[33]
The 4’-hydroxy group of each aurone was superimposed on the
oxygen atom of the NO group of HOPNO, linked to one copper
atom of Ty.[31,32] We considered the 4’-hydroxy group in its depro-
Acknowledgements
The authors are grateful to ANR (Agence Nationale pour la Re-
cherche) for financial support (Labex Arcane (ANR-11-LABX-0003-
01) and Blanc program 2Cu-TargMelanin (ANR-09-BLAN-0028-01/
02/03), CECIC for computer facilities and the European COST Pro-
gram in the framework of which this work was carried out (COST
action CM1003 WG 2). The work has been partially financed by
the grant “Human recombinant tyrosinase: a testing bench for
new drugs to treat melanin-related disorders” (Progetto di Ricerca
di Ateneo CPDA110789/11) from the University of Padova to E.B.
ꢂ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemBioChem 0000, 00, 1 – 10
&
8
&
ÞÞ
These are not the final page numbers!

CHEMBIOCHEM
FULL PAPERS
www.chembiochem.org
[21] L. Bubacco, M. van Gastel, E. J. J. Groenen, E. Vijgenboom, G. W. Canters,
J. Biol. Chem. 2003, 278, 7381–7389.
Keywords: aurones
· bacterial tyrosinases · enzymes ·
[22] C. Dubois, R. Haudecoeur, M. Orio, C. Belle, C. Bochot, A. Boumendjel, R.
Hardrꢀ, H. Jamet, M. Rꢀglier, ChemBioChem 2012, 13, 559–565.
[23] M. R. L. Stratford, C. A. Ramsden, P. A. Riley, Bioorg. Med. Chem. 2013, 21,
1166–1173.
[24] S. Yamazaki, S. Itoh, J. Am. Chem. Soc. 2003, 125, 13034–13035.
[25] Z. Yang, F. Wu, Biotechnology 2006, 5, 344–348.
inhibitors · mushroom tyrosinases · QM/MM
[1] C. Belle in Encyclopedia of Metalloproteins (Eds.: R. H. Kretsinger, V. N.
Uversky, E. A. Permyakov), Springer, New York, 2013, pp. 574–579.
[2] E. I. Solomon, U. M. Sundaram, T. E. Machonkin, Chem. Rev. 1996, 96,
2563–2606.
[26] A. Cornish-Bowden, Fundamentals of Enzyme Kinetics, Portland, London,
1995.
[3] M. Rolff, J. Schottenheim, H. Decker, F. Tuczek, Chem. Soc. Rev. 2011, 40,
4077–4098.
[27] P. A. Kollman, I. Massova, C. Reyes, B. Kuhn, S. Huo, L. Chong, M. Lee, T.
Lee, Y. Duan, W. Wang, O. Donini, P. Cieplak, J. Srinivasan, D. A. Case,
T. E. Cheatham, Acc. Chem. Res. 2000, 33, 889–897.
[4] P. M. Plonka, M. Grabacka, Acta Biochim. Pol. 2006, 53, 429–443.
[5] M. Sendovski, M. Kanteev, V. S. Ben-Yosef, N. Adir, A. Fishman, J. Mol.
Biol. 2011, 405, 227–237.
[28] A. Ciancetta, S. Genheden, U. Ryde, J. Comput.-Aided Mol. Des. 2011, 25,
729–742.
[6] Y. Matoba, T. Kumagai, A. Yamamoto, H. Yoshitsu, M. Sugiyama, J. Biol.
Chem. 2006, 281, 8981–8990.
[29] M. Retegan, A. Milet, H. Jamet, J. Chem. Inf. Model. 2009, 49, 963–971.
[30] S. Takahashi, T. Kamiya, K. Saeki, T. Nezu, S.-i. Takeuchi, R. Takasawa, S.
Sunaga, A. Yoshimori, S. Ebikura, T. Abe, S.-i. Tanuma, Bioorg. Med.
Chem. 2010, 18, 8112–8118.
[7] W. T. Ismaya, H. J. Rozeboom, A. Weijn, J. J. Mes, F. Fusetti, H. J. Wichers,
B. W. Dijkstra, Biochemistry 2011, 50, 5477–5486.
[8] N. Fujieda, S. Yabuta, T. Ikeda, T. Oyama, N. Muraki, G. Kurisu, S. Itoh, J.
Biol. Chem. 2013, 288, 22128–22140.
[31] E. Peyroux, W. Ghattas, R. Hardrꢀ, M. Giorgi, B. Faure, A. J. Simaan, C.
Belle, M. Rꢀglier, Inorg. Chem. 2009, 48, 10874–10876.
[9] T.-S. Chang, Int. J. Mol. Sci. 2009, 10, 2440–2475.
[10] K. R. Alexander, P. E. Kilbride, G. A. Fishman, M. Fishman, Vision Res.
1987, 27, 1077–1083.
[11] K. Kalinderi, L. Fidani, Z. Katsarou, S. Bostantjopoulou, Int. J. Clin. Pract.
2011, 65, 1289–1294.
[12] J. M. Gillbro, M. J. Olsson, Int. J. Cosmet. Sci. 2011, 33, 210–221.
[13] N. Smit, J. Vicanova, S. Pavel, Int. J. Mol. Sci. 2009, 10, 5326–5349.
[14] M. Friedman, J. Agric. Food Chem. 1996, 44, 631–653.
[15] M. R. Loizzo, R. Tundis, R. Menichini, Compr. Rev. Food Sci. Food Saf.
2012, 11, 378–398.
[16] T. F. M. Kuijpers, H. Gruppen, S. Sforza, W. J. H. van Berkel, J.-P. Vincken,
FEBS J. 2013, 280, 6184–6195.
[17] T. F. M. Kuijpers, T. van Herk, J.-P. Vincken, R. H. Janssen, D. L. Narh,
W. J. H. van Berkel, H. Gruppen, J. Agric. Food Chem. 2014, 62, 214–221.
[18] S. Okombi, D. Rival, S. Bonnet, A.-M. Mariotte, E. Perrier, A. Boumendjel,
J. Med. Chem. 2006, 49, 329–333.
[32] C. Bochot, E. Favre, C. Dubois, B. Baptiste, L. Bubacco, P.-A. Carrupt, G.
Gellon, R. Hardrꢀ, D. Luneau, Y. Moreau, A. Nurisso, M. Rꢀglier, G. Serra-
trice, C. Belle, H. Jamet, Chem. Eur. J. 2013, 19, 3655–3664.
[33] W. Delano, The PyMOL Molecular Graphics System, DeLano Scientific, Palo
Alto, CA, 2006.
[34] D. A. Case, T. A. Darden, T. E. Cheatham III, C. L. Simmerling, J. Wang,
R. E. Duke, R. Luo, M. Crowley, R. C. Walker, W. Zhang, K. M. Merz, B.
Wang, S. Hayik, A. Roitberg, G. Seabra, I. Kolossvꢄry, K. F. Wong, F. Paesa-
ni, J. Vanicek, X. Wu, S. R. Brozell, T. Steinbrecher, H. Gohlke, L. Yang, C.
Tan, J. Mongan, V. Hornak, G. Cui, D. H. Mathews, M. G. Seetin, C. Sagui,
V. Babin, P. A. Kollman, AMBER 10, University of California, San Francisco,
2008.
[35] M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb, J. R.
Cheeseman, J. A. Montgomery Jr., T. Vreven, K. N. Kudin, J. C. Burant,
J. M. Millam, S. S. Iyengar, J. Tomasi, V. Barone, B. Mennucci, M. Cossi, G.
Scalmani, N. Rega, G. A. Petersson, H. Nakatsuji, M. Hada, M. Ehara, K.
Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O.
Kitao, H. Nakai, M. Klene, X. Li, J. E. Knox, H. P. Hratchian, J. B. Cross, V.
Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E. Stratmann, O. Yazyev,
A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski, P. Y. Ayala, K. Moro-
kuma, G. A. Voth, P. Salvador, J. J. Dannenberg, V. G. Zakrzewski, S. Dap-
prich, A. D. Daniels, M. C. Strain, O. Farkas, D. K. Malick, A. D. Rabuck, K.
Raghavachari, J. B. Foresman, J. V. Ortiz, Q. Cui, A. G. Baboul, S. Clifford,
J. Cioslowski, B. B. Stefanov, G. Liu, A. Liashenko, P. Piskorz, I. Komaromi,
R. L. Martin, D. J. Fox, T. Keith, M. A. Al-Laham, C. Y. Peng, A. Nanayak-
kara, M. Challacombe, P. M. W. Gill, B. Johnson, W. Chen, M. W. Wong, C.
Gonzalez, J. A. Pople, Gaussian 03, Revision A1, Gaussian, Inc., Walling-
ford CT, 2004.
[19] a) A. Boumendjel, Curr. Med. Chem. 2003, 10, 2621–2630; b) C. Zwergel,
F. Gaascht, S. Valente, M. Diederich, D. Bagrel, G. Kirsch, Nat. Prod.
Commun. 2012, 7, 389–394; c) R. Haudecoeur, A. Boumendjel, Curr.
Med. Chem. 2012, 19, 2861–2875.
[20] a) R. Haudecoeur, A. Ahmed-Belkacem, W. Yi, A. Fortunꢀ, R. Brillet, C.
Belle, E. Nicolle, C. Pallier, J.-M. Pawlotsky, A. Boumendjel, J. Med. Chem.
2011, 54, 5395–5402; b) H. Cheng, L. Zheng, Y. Liu, S. Chen, X. Cheng,
X. Lu, Z. Zheng, G.-C. Zhou, Eur. J. Med. Chem. 2010, 45, 5950–5957;
c) C.-Y. Lee, E.-H. Chew, M.-L. Go, Eur. J. Med. Chem. 2010, 45, 2957–
2971; d) H.-M. Sim, C.-P. Wu, S. V. Ambudkar, M.-L. Go, Biochem. Pharma-
col. 2011, 82, 1562–1571; e) B. P. Bandgar, S. A. Patil, B. L. Korbad, S. C.
Biradar, S. N. Nile, C. N. Khobragade, Eur. J. Med. Chem. 2010, 45, 3223–
3227; f) K. N. Tiwari, J.-P. Monserrat, A. Hequet, C. Ganem-Elbaz, T. Cres-
teil, G. Jaouen, A. Vessiꢁres, E. A. Hillard, C. Jolivalt, Dalton Trans. 2012,
41, 6451–6457; g) M. Roussaki, S. Costa Lima, A.-M. Kypreou, P. Kefalas,
A. Cordeiro da Silva, A. Detsi, Int. J. Med. Chem. 2012, 2012, 196921;
h) S. Y. Shin, M. C. Shin, J.-S. Shin, K.-T. Lee, Y. S. Lee, Bioorg. Med. Chem.
Lett. 2011, 21, 4520–4523; i) M. Ono, R. Ikeoka, H. Watanabe, H. Kimura,
T. Fuchigami, M. Haratake, H. Saji, M. Nakayama, Bioorg. Med. Chem.
Lett. 2010, 20, 5743–5748; j) S. Shrestha, S. Natarajan, J.-H. Park, D.-Y.
Lee, J.-G. Cho, G.-S. Kim, Y.-J. Jeon, S.-W. Yeon, D.-C. Yang, N.-I. Baek,
Bioorg. Med. Chem. Lett. 2013, 23, 5150–5154.
[36] J. W. Ponder, Tinker 4.2 ed., Washington University, Saint Louis, 2004.
[37] H. Zheng, S. Wang, Y. Zhang, J. Comput. Chem. 2009, 30, 2706–2711.
[38] W. D. Cornell, P. Cieplak, C. I. Bayly, P. A. Kollman, J. Am. Chem. Soc.
1993, 115, 9620–9631.
Received: February 3, 2014
Published online on && &&, 0000
ꢂ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemBioChem 0000, 00, 1 – 10
&
9
&
ÞÞ
These are not the final page numbers!

FULL PAPERS
R. Haudecoeur, A. Gouron,
A. Boumendjel*
A lighter future: Aurones have been
identified as inhibitors of melanin bio-
synthesis. In this study, 24 aurones were
evaluated on mushroom and bacterial
tyrosinases (TyM and TyB). The com-
pounds behaved as inhibitors, sub-
strates, or activators of both enzymes.
Our results highlight similarities and dif-
ferences in behavior between TyM and
TyB with the same set of molecules.
&& – &&
Investigation of Binding-Site
Homology between Mushroom and
Bacterial Tyrosinases by Using
Aurones as Effectors
ꢂ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemBioChem 0000, 00, 1 – 10
&10
&
ÞÞ
These are not the final page numbers!