ORIGINAL PAPER

Enzymatic synthesis of oligoesculin: structure and biologicalactivities characterizations

Julie Anthoni • Catherine Humeau • Elaine Rose Maia •

Latifa Chebil • Jean-Marc Engasser • Mohamed Ghoul

Received: 25 February 2010 / Revised: 17 May 2010 / Accepted: 21 May 2010 / Published online: 20 June 2010

� Springer-Verlag 2010

Abstract The oligomerization of esculin, catalyzed by

the laccase from Trametes versicolor, was realized in an

attempt to improve the properties of this glycosidic cou-

marin. MALDI-TOF analyses showed a degree of oligo-

merization up to 9, whereas NMR spectra revealed the

formation of C–C and C–O bridges, which involve both

the phenolic and the glucosidic part of the coumarin. The

solubility of oligomers is 189-fold higher than esculin’s

solubility. Moreover, antioxidant properties of oligomers

were correlated with their mass; the more the mass, the

more the xanthine oxidase inhibitory activity and the rad-

ical scavenging activity are important. These experimental

results were completed by in silico structural investiga-

tions, which suggested the preferential formation of C8–C8

linkages between esculin units during the oligomerization

reaction.

Keywords Oligomerization � Esculin �Laccase from Trametes versicolor � Solubility �Antioxidant activity � Molecular modeling

Introduction

Esculin is a glycosidic coumarin belonging to a group of

phenolic compounds widely distributed in natural plants.

Coumarins are naturally occurring derivatives of benzo-

pyrene. They are used in beverages [1], food [2], perfumes

[3], cosmetics [4] and as additives in polymers for their

fluorescence, photodimerization, photocleavage or liquid

crystalline properties. They have recently attracted much

attention due to their antiradical [5], xanthine oxidase

inhibitory [6], anti-inflammatory [7], anticarcinogenic [8]

and anti-HIV [9] activities. The magnitude of coumarin

activities depends upon the degree and the localization of

hydroxyl, glycosyl or methoxy groups on their backbone.

Esculin or aesculin (Fig. 1) is a monosaccharide of

esculetin characterized by a fine blue fluorescent solution.

It is extracted from the Aesculus hippocastanum L. (Hip-

pocastanaceae), or horse-chestnut tree. It exhibits an

UV-B-protective effect and antiflogistic, cytostatic and

antimutagenic properties [10]. Contrary to the majority of

phenolic species, esculin is soluble in water and relatively

stable in this medium.

The enzymatic polymerization of several phenolic spe-

cies like hydroxybenzene [11–13], cresol [13, 14], cardanol

[15], catechol [16], catechin [17] or rutin [18–20] was

already reported. In some cases, these modifications led to

a higher solubility in water, thermal stability and an

increase in antioxidant and xanthine oxidase inhibitory

activities, and in other cases, a loss of these activities was

observed. This behavior is due to the fact that such activ-

ities are correlated with the structure and the functional

groups of the molecules. Different studies have mentioned

that an ortho-catechol moiety [21], a 1,2-pyrone in B ring

[22], an O-glucopyranoside group and more globally

hydroxyl functions [23] increase the antioxidant activity of

J. Anthoni � C. Humeau � L. Chebil � J.-M. Engasser �M. Ghoul (&)

Laboratoire d’Ingenierie des Biomolecules, Ecole Nationale

Superieure d’Agronomie et des Industries Alimentaires,

Institut National Polytechnique de Lorraine ENSAIA-INPL,

54505 Vandoeuvre les Nancy, France

e-mail: mohamed.ghoul@ensaia.inpl-nancy.fr

E. R. Maia

Laboratorio de Estudos Estruturais Moleculares (LEEM),

Instituto de Quımica, Universidade de Brasılia, Campus Darcy

Ribeiro, C.P. 4478, Brasılia 70904-970, DF, Brazil

e-mail: emaia@unb.br

123

Eur Food Res Technol (2010) 231:571–579

DOI 10.1007/s00217-010-1298-3

coumarins. Moreover, a substitution in position 8 (hydroxyl

group or glucosyl group), a methyl substitution in positions

3, 4 and 8 or a double bond on positions 4, 3 increases the

xanthine oxidase inhibitory activity [23]. For the oligomers

having interesting properties, their use in food formulation

is easier and more efficient than for monomers.

In spite of the interesting properties of esculin, the

polymerization of this compound was not yet studied. This

knowledge could favor the development of oligomers’ use

as antioxidant, antimicrobial or colorant for several appli-

cations in food and cosmetics areas. Moreover, availability

of oligomers with enhanced activities and stability can

facilitate their use in a wide range of operating conditions

(pH, temperature, etc). The aim of this paper is to inves-

tigate the relationships between the structure of the syn-

thesized esculin oligomers and their solubility, antioxidant

and xanthine oxidase inhibitory activities. In order to

attempt this goal, the spectroscopic analyses of UV, FTIR,

MALDI-TOF and NMR, and molecular modeling approa-

ches were used. Properties’ results (stacking, solubility and

antioxidant activity) were correlated with structural

descriptors, which were calculated by molecular modeling

under vacuum and in solvent environment.

Results and discussion

Production of esculin oligomers

The oligomerization of esculin was performed in a meth-

anol/water medium with an initial pH of 7.0, at 20 �C, by

laccase from Trametes versicolor. After 72 h of incubation,

the Mw of esculin oligomers and the residual concentration

of esculin were, respectively, equal to 1800 g mol-1

(corresponding to a degree of oligomerization of 5) and

0.7 g L-1 (Fig. 2). These values of Mw were reached after

12 h of reaction, whereas the consumption of esculin was

stabilized after 48 h of reaction. The polydispersity was

stabilized around 0.6 after 12 h of reaction. The limit value

of Mw is due neither to the limitation of the substrate,

which is not totally consumed, nor to the denaturation of

enzyme, which remains active after 12 h. After 12 h of

reaction, esculin concentration continued to diminish

showing that the enzyme was always active, but the size of

the oligomers has reached its ceiling. Such behavior could

be due to a lower reactivity of large oligomers comparing

to smaller ones. The enzyme continues to catalyze the

formation of esculin radical species, but these radicals are

supposed to react preferentially with each other, rather than

with large oligomers, which are less reactive.

To establish the structure of esculin oligomers, MALDI-

TOF, UV, FTIR and NMR analyses were realized both on the

bulk medium and on the enriched fractions of oligomers.

Structural investigation by MALDI-TOF, UV and FTIR

The determination of the absolute masses of oligomers by

MALDI-TOF showed that, under the used conditions, the

longest oligomer was a nonamer (Fig. 3). The gap of mass

due to the addition of one esculin monomer indicated the

abstraction of two hydrogen atoms, showing that the con-

nection mode between esculin units was a simple bridge.

The UV–visible spectra of esculin and reaction medium

containing the mixture of oligoesculins revealed two peaks

around 204 and 336 nm. However, oligomer peaks were

broader than those of esculin, which could be attributed to

conjugated oligomeric structure [19]. The FTIR spectra of

esculin and oligoesculin were similar; no new character-

istic band was observed.

Structural investigation by NMR and molecular

modeling

The bulk medium after 24 h of incubation and oligo-

mers’ enriched fractions were analyzed by NMR. The

Fig. 2 Kinetic study of the oligomerization reaction of esculin

catalyzed by laccase from Trametes versicolor, in methanol/water

(30:70, v/v), at 20 �C. Open square concentration of esculin, blackcircle weight-average mass

Fig. 1 Structure and chemical numbering system of esculin

572 Eur Food Res Technol (2010) 231:571–579

123

1H-NMR spectra revealed broad signals in both aromatic

and sugar zones and the apparition of new peaks around

7.5 ppm (Fig. 4). These peaks could correspond to a

downfield shift of the 5 H signal, due to the formation of

a C4–O7 bridge, or an upfield shift of the 4 H signal due

to the formation of a C3–C bridge. This profile suggests

that several types of bridges should occur during the

oligomerization of esculin, involving both the aromatic

and the sugar parts of this compound. Moreover, NMR

spectra indicated that, on the one hand, the presence of

residual esculin (340 g mol-1) even in the retentat

obtained with a 50 kDa membrane and, on the other

hand, a signal shift in the aromatic region for sample

issued from 1 kDa membrane retentat. The presence of

residual esculin for high cut-off membrane and the signal

shift suggests a stacking phenomenon between aromatic

rings. The stacking of phenolic species due to hydro-

phobic interactions is well known [24] and has already

been observed in the case of quercetin by Desentis-

Mendoza et al. [18].

This stacking phenomenon was studied by molecular

modeling using systems of four and of six molecules of

esculin immersed in a mixture of methanol/water cell. At

the equilibrium, the system having four esculin molecules

was clearly piled up and positioned head to tail (Fig. 5).

The solvent molecules formed a dense network around

esculin molecules, stabilized by at least eight hydrogen

bonds formed among esculins’ hydroxyl or carbonyl

groups and water or methanol unities. The distances

between the center of mass of their aromatics rings

were 4.9 A, in average. Besides, the aromatic rings

stayed closer when system had six esculin molecules

(\d[ = 4.3 A).

Solubility study

The solubility of esculin was 3.70 g L-1 in water and

5.96 g L-1 in methanol/water (30:70, v/v, 20 �C). The

solubility of esculin oligomers was around 700 g L-1 in

water and 0.4 g L-1 in methanol/water. In order to

understand such variations of solubility and especially the

low solubility of oligomers in methanol/water mixture,

the interactions between esculin, diesculin (O7–C4) and the

solvent (water or methanol/water) were studied by molec-

ular modeling and then compared. The stable conformation

of esculin was shown to form around seven to nine

H-bonds with water molecules and eight to eleven H-bonds

in the case of a methanol/water medium. These observations

could explain why esculin is slightly less soluble in water

than in methanol/water medium. For esculin dimer (based

on a O7–C4 linkage), twenty-two H-bonds (Fig. 6) can be

formed with water molecules, while only a few H-bonds

were observed with methanol/water medium (around four

H-bonds). This behavior is consistent with experimental

results and those obtained by Saidman et al. [25].

Study of xanthine oxidase inhibitory activity

Figure 7 shows xanthine oxidase inhibition activity of

esculin and its oligomers, assessed by measuring uric acid

formation from xanthine oxidase. IC 50 of oligomers

appeared to be lower than IC 50 of esculin (770 lM). The

IC50 fell when increasing Mw and stabilized around 150 lM.

The increase in the xanthine oxidase inhibition by oligomers

is reported by Kurisawa et al. in the case of poly(cathechin)

[19]. These authors attributed this increase to the effective

multivalent interactions between xanthine oxidase and the

Fig. 3 MALDI-TOF positive

ion spectrum ([M ? Na or K]?)

in the reflectron mode of esculin

oligomerization media and

isotopic distribution of tetramer.

Dihydroxybenzoic acid was

used as matrix

Eur Food Res Technol (2010) 231:571–579 573

123

polymeric chains of poly(catechin). The inhibition of xan-

thine oxidase activity by coumarins with different substitu-

tions is well investigated by Chang and Chiang [23] and Lin

et al. [26]. These authors have shown that the xanthine oxi-

dase inhibitory activity is strongly affected by both the

position and the nature of substitution on the benzene and

a-pyrone rings. The presence of hydroxyl groups on position

6 and 7 enhances the activity of coumarins against xanthine

oxidase. In the case of esculin, the position 6 is already

substituted by the sugar part. So, the observed increase in the

Fig. 4 1H-NMR spectra

(600 MHz by cryoprobe) of

media containing esculin and its

oligomers’ enriched fraction.

Esculin (3 g L-1)

oligomerization was catalyzed

by laccase from Trametes

versicolor (3 U mL-1) in

methanol/water (30:70, v/v),

at 20 �C

Fig. 5 Graphical representation

of a system comprising 4 and 6

esculin monomers in a

methanol/water cell (30:70,

v:v). For esculin molecules,

carbon atoms are displayed in

blue and oxygen atoms in red.

For clarity, solvent oxygen

atoms are colored in green and

methanol carbon atoms in blue.

a Snapshot of hydrogen bonds

distribution among water or

methanol solvent molecules and

esculin; b Distances between

aromatic ring center of mass of

four esculin molecules;

c Distances between aromatic

ring center of six esculin

molecules

574 Eur Food Res Technol (2010) 231:571–579

123

activity for the esculin oligomers means that position 7 is not

involved in the linkage between esculin units. Lin et al. [26]

investigated by molecular modeling the interaction between

xanthine oxidase and several coumarin derivatives. They

observed that there is a great interaction between coumarins

and molybdopterin region of xanthine oxidase. They also

reported that 6 and 7 hydroxyl groups on benzene ring play a

major role in coumarins affinity toward xanthine oxidase

binding site.

The molecular modeling study realized in this work

indicated that one probable linkage between esculin units

to form oligomers could be C8–C8. This result is in

agreement with the observation of Kurisawa et al. [19] but

not with those of Chang and Chiang [23] who observed a

decrease in the activity when a substitution is occurred on

C8. However, these authors did not study oligomers; they

only compared the activities of different coumarins

exhibiting different substitutions.

Study of antiradical activity

The antiradical activity for each oligomeric fraction was

evaluated. Results are reported on Fig. 8.

The IC50 of esculin (9 mM) is 18-fold higher than the

IC50 of oligomers (0.5 mM). However, the degree of

polymerization does not affect this activity. Different

authors reported that the absence of the catechol moiety

[21] and/or the hydroxyl groups [21, 22] led to a decrease

in the antiradical activity. Results obtained in the present

study show that these essential groups are not affected

during esculin oligomerization. Moreover, the high scav-

enging activity of oligomers indicates a high capacity to

stabilize radical species, which can be evaluated by the

bond dissociation energy (BDE) [27].

To verify these assumptions, molecular systems formed

by esculin and esculin dimers were studied, by molecular

modeling, under vacuum. Four different dimer structures

corresponding to four linkage modes were modeled: O7–C4,

C4–C5, C4–O20 and C8–C8 (Fig. 9).

The dimers composed by O7–C4 and C4–C5 linkages

were chosen because they come from the association of the

most stable radicals (Table 1).

NMR analysis of oligomers’ enriched fractions showed

that linkage between aromatic part and glycosidic part of

esculin units could occur. That’s why the dimer C4–O20

was modeled. Finally, the dimer C8–C8 was studied

because all essential groups remain free. The dimers were

minimized as described in methodology. Their BDE was

estimated and compared to that of esculin (Table 2).

At least one BDE value of the dimers (C4–C5),

(C8–C8) and (C4–O20) is lower than that of esculin. The

dimer C8–C8 is characterized by two low BDE values,

which suggest a high scavenging activity. The dimer

Fig. 6 Snapshot of esculin dimer (based on a O7–C4 linkage) in

water. An extensive hydrogen bonds network stabilizes diesculin into

water. Aromatic rings still oriented in a P-stacking interaction

Fig. 7 Influence of weight-average mass of esculin oligomers on

their xanthine oxidase inhibition activity. Esculin activity is indicated

as reference

Fig. 8 Influence of weight-average mass of esculin oligomers on

their antiradical activity determined by DPPH test. Esculin activity is

indicated as reference

Eur Food Res Technol (2010) 231:571–579 575

123

O7–C4 seems to be the less probable structure because

its BDE is higher than that of esculin. The high antiox-

idant activity measured for esculin oligomers suggests

that the dimer C8–C8 is the most probable and the dimer

O7–C4 the less one. This assumption is in accordance

with data reported by Ncanana et al. [28], Intra et al. [29]

and Mita et al. [12] who observed a majority of C–C

linkages, in the case of totarol, tetrahydro-2-naphtol and

phenol polymerization.

Experimental section

Materials

Laccase from Trametes versicolor (E.C. 1.10.3.2., 21.4

U mg-1) and esculin sesquihydrate ([98%) were

purchased from Fluka, and xanthine and xanthine oxidase

(1.17.3.2., grade IV from bovine milk, 0.2 U mg-1 protein)

from Sigma. All solvents were HPLC grade from VWR.

Oligomerization reaction

Reactions were carried out on a Chemspeed ASW1000

automated synthesizer (Chemspeed Ltd). Eight parallel

reactions were realized simultaneously. Esculin (3 g L-1)

was suspended in 27-mL vessels in 10 ml of methanol/

water (30:70, v/v). Laccase solution (3 U mL-1) was added

to the mixture. The reactions were stirred at 600 rpm, for

24 h, at 293 K (or 20 �C).

SEC analysis

Relative masses of oligomers were evaluated online by size

exclusion chromatography (SEC) (HPLC LaChrom

(Merck), UV 280 nm LaChrom L-7400, Tosoh TSK Gel a3000 column, at 333 K). DMF with 1% LiBr was used as

the mobile phase, at a flow rate of 1 mL min-1. Polysty-

rene standards were used to define molecular mass cali-

bration. Average molecular masses were defined by

(Fig. 10), where ni represents the number of oligomer

molecules having a mass of Mi at a polymerization degree

of i and wi the weight fraction of chain at a polymerization

degree of i. The quantity of oligomers was expressed in

arbitrary units of area (a.u.). Error values were determined

by three repetitions of the same reaction.

Fig. 9 Schematic

representation of esculin

oligomerization catalyzed by

laccase and molecular structure

of four esculin dimers studied

by molecular simulations.

(R: glucose)

Table 1 Heat of formation of esculin and its radicals in different positions, obtained from molecular minimization in vacuum, calculated by

AM1 semi-empirical quantum method

Esculin Radical species: abstraction of a hydrogen atom from

C3 C4 C5 O7 C8

DH (kcal.mol-1) -350.17 -295.49 -298.34 -295.84 -321.74 -292.37

DHesculin-DHradical = BDEa (kcal.mol-1) 54.68 51.83 54.33 28.43 57.80

a Bond dissociation energy

Table 2 Bond dissociation energy (BDE) of the O7-H bond of

esculin and its dimers, obtained from molecular minimization under

vacuum, calculated by AM1 semi-empirical quantum method

Esculin Diesculin with linkage between

O7–C4 C4–C5 C8–C8 C4–O20

DH (kcal.mol-1) -350.17 -677.30 -688.15 -690.59 -681.71

DDH (kcal.mol-1)

BDE1 28.43 30.54 30.65 25.98 30.65

BDE2 22.34 27.43 27.46

576 Eur Food Res Technol (2010) 231:571–579

123

Property determination

Scavenging activity and xanthine oxidase inhibitory

activity of different pools of oligomers were analyzed.

Fractions enriched with oligomers of various mass were

prepared from 24-h reaction medium (methanol/water,

30:70, v/v, 293 K), by successive filtration processes on a

50 kDa membrane to remove the enzyme and then suc-

cessively on 8, 5 and 3 kDa membranes by a tangential

diafiltration process (v/v). Then, the fractions were

lyophilized.

Free radical scavenging power was evaluated by the

DPPH test as described by Burda et Oleszek [24]. One

milliliter of solution of esculin or oligomers at different

concentrations in methanol was added to 2 mL of DPPH at

10 mg L-1in methanol/water (80:20, v/v). A reference

sample was prepared with 1 ml of methanol in 2 mL of

DPPH solution. Esculin and oligomers’ absorbance for

each concentration was evaluated at 527 nm. After 15 min,

at 296 K, the absorbance at 527 nm was measured. The

antiradical activity was calculated as a percentage of DPPH

discoloration using the following equation (Fig. 11).

The antiradical activity of the molecules is expressed as

the final concentration that leads to the half-maximal

DPPH discoloration (IC50) and calculated by exponential

regression analysis. Each absorbance was measured in

triplicate.

A second property tested the capacity to inhibit the

activity of xanthine oxidase. All solutions were prepared in

phosphate buffer (pH 7.5, 50 mM), except for esculin,

which was initially dissolved in a minimum of DMSO and

then rapidly diluted in buffer. Test solutions were prepared

by adding 1600 lL of buffer, 300 lL of solution of esculin

or its oligomers (concentration varying from 10-5 to 12

10-6 M and calculated from Mw), 1000 ll of a solution of

xanthine (0.15 mM, initially dissolved in a minimum of

NaOH 2% and then rapidly diluted in buffer) and 100 lL

of a solution of xanthine oxidase (0.2 U mL-1). The

reaction was monitored for 6 min at 295 nm. Two test

mixtures were prepared, the first containing no phenolic

species to measure the total production of uric acid and the

second containing no enzyme to measure the absorbance of

phenols at 290 nm for the range of concentrations. The

capacity of the molecules to inhibit the xanthine oxidase is

expressed as the final concentration that leads to the half-

maximal enzyme velocity (IC50) and calculated by stan-

dard curve regression analysis. Each absorbance was

measured in triplicate.

MALDI analysis

Absolute masses were determinate by MALDI-TOF. The

concentration of DHB matrix is 5 mg mL-1 in acetonitrile/

water (30:70, v/v) with 0.1% TFA.

After 24 h of reaction, samples were automatically

taken from reaction media and diluted twofold in aceto-

nitrile/water (30:70, v/v) with 0.1% TFA. Solutions (3 lL)

were spotted online on MALDI target with DHB (1 lL)

and crystallized at room temperature. All samples were

spotted automatically in triplicate, on the Chemspeed

ASW1000 (Chemspeed Ltd, August, Switzerland)

platform.

MALDI-TOF–MS was performed on Ultraflex (Bru-

ker) in reflectron operation mode. All spectra were

obtained in positive ion mode. Ionization was performed

with 337 nm pulsed nitrogen laser and accelerated under

20 kV with time-delayed extraction. Spectra were

obtained by accumulating of at least 300 shots. Spectra

were calibrated with statistic calibration of FlexAnalysis

software (Bruker).

FTIR analysis

The IR analyses were achieved by ATR-FT-IR spectros-

copy using a FT-IR spectrometer Tensor 27 (Bruker). The

solutions containing oligomers were diluted with water/

methanol (70:30, v/v) and then deposited on the ATR unit.

Solvents were evaporated under vacuum at 323 K to give a

film, which was analyzed.

UV analysis

The UV spectra of esculin and oligomers solutions in

methanol were determined on a UV6000LP spectrometer

(Spectra System, Thermofinnigan).

NMR analysis

The 1H NMR spectra of reaction media were recorded on a

Bruker 600 MHz spectrometer with a cryoprobe at low

concentration of esculin (0.2 g L-1) and 0.6 U mL-1 of

laccase in CD3OD/D2O (30:70, v/v), at 300 K.Fig. 10 Equations 1, 2, 3

Eur Food Res Technol (2010) 231:571–579 577

123

A diffusion-ordered spectroscopy (DOSY) NMR tech-

nique was also used to aid the deconvolution of the system.

A diffusion delay time of 100 ms was applied, in CD3OD/

D2O (30:70, v/v), at 293 K.

Esculin 1H NMR

(D2O/CD3OD, 70:30, v/v): d = 3.52 (2d JH30/H40/H50 H40),3.62 (m JH20/H30/H40/H50/H60 H20,H30,H50), 3.77 (d,d JH50/H60/H60

H60 5.8 ? 12.5 Hz), 3.95 (dd JH60/H60/H50 H60 2.0 ?

12.5 Hz), 5.06 (d JH10/H20 H10b 7.5 Hz), 6.31 (d JH3/H4 H3

9.6 Hz), 6.91 (s H8), 7.37 (s H5), 7.93 (d JH4/H3 H4 9.6 Hz)

ppm.

Molecular modeling simulations

Semi-empirical quantum Austin model (AM1) method

Enthalpic study was carried out using the semi-empirical

AM1 method [30], with HyperChem software [31]. Opti-

mization was made with conjugated gradient (algorithm

Polak-Ribiere) with a self-consistent field (SCF) of

1.0 9 10-3 kcal mol-1 A-1, under vacuum. The bond

dissociation energy (BDE) was calculated as the energy of

the radical resulting from hydrogen atom abstraction minus

the energy of the neutral molecule. For the dimers, the

BDE was calculated with the radical species formed in

position 7. For dimers, which present two hydroxyls in

position 7, two radical species and so two BDE were

obtained.

Molecular mechanics and molecular dynamics approach

The molecular systems have been investigated essentially

by molecular mechanics (MM) and molecular dynamics

(MD) methods. The simulations were performed using the

Accelrys program-package Materials Studio (MS) soft-

ware, version 4.2 [32], running on a bi-processor AMD

Dual Core 280, with a memory capacity of 2.4 GHz and

Insight II programs version 2005, on an O2 SGI worksta-

tion. All calculations were performed using Discover with

CFF force field [33, 34], through the Steepest Descent and

Conjugate Gradient (Polak-Ribiere) algorithms. InsightII

S/W and SGI 3D functionalities were essentially used for

an easy construction and initial equilibration of the solvent

boxes and their sizes.

All monomeric and dimeric esculins had their confor-

mational space and their relative orientations, one mono-

mer to another and one dimer to another, firstly studied

under vacuum. MD trajectories were carried out at 600 K,

during 2 ns. After fluctuations analysis and behavior

understanding, the most stable conformations of each sys-

tem were considered for solvent simulations.

Solvent molecules were introduced into a box, around

the central molecules, monomeric or dimeric entities. Fix

constraints were imposed to the atomic coordinates of

those entities, in order to avoid unexpected structural dis-

tortions, due to solvent molecular bumps. The dynamics

trajectories run, at 300 K, during 600 ps, with a time step

of 1 fs, under periodic boundary conditions, using a 14 A

spherical cut-off for non-bound interactions and using the

NVT thermodynamic ensemble. The Verlet-Leapfrog

algorithm was employed for the integration of the equa-

tions of motion. The last frame of each dynamic period was

submitted to energy optimizations until a maximum

derivative of 1.0 9 10-1 kcal mol-1 A-1 to permit solvent

adjustments. After steric hindrance corrections, the con-

straints were removed and the systems were freely reop-

timized, before to be submitted to final dynamics

simulations at 600 K. The temperature was decreased until

300 K. The lowest energy conformation of each molecular

system was, then, minimized, the box size and solvent

equilibration were verified again, and energetic optimiza-

tions were performed until a maximum derivative of

1.0 9 10-4 kcal mol-1 A-1.

For water solvent simulations, a box containing 1370

water molecules and one esculin monomer, and a box

containing 1418 water molecules and one diesculin mole-

cule (O7–C4 linkage) were analyzed. To be closed to the

operating conditions, cell boxes of 50.0 A 9 50.0 A 9 70.0 A

side length containing 938 molecules or 4200 atoms, for a

system composed by methanol/water and 4 esculin mole-

cules; or a box of 55.0 A 9 55.0 A 9 80.0 A side length

containing a system composed by methanol/water and 6

esculin molecules; or a box containing 395 methanol

molecules and 2073 water molecules plus 2 molecules of

diesculin; another cell system having 998 molecules or

3104 atoms, for a system composed by methanol/water,

one diesculin and one esculin molecules. So, many dif-

ferent systems were submitted to dynamics procedure until

to permit us to evaluate the behavior of esculin into water

or into methanol/water (30:70, v/v) mixture.

Conclusions

Structural analyses of esculin oligomers, synthesized by

laccase from Trametes versicolor, showed that esculin

Fig. 11 Equation 4

578 Eur Food Res Technol (2010) 231:571–579

123

units are linked by a simple bridge, which can occur on

both sugar part and phenolic part of the coumarin.

Molecular modeling results suggest the preferential for-

mation of C8–C8 linkages during esculin oligomerization.

Esculin oligomers are characterized by a very high solu-

bility in water and a very poor one in methanol/water

mixture. The higher solubility has been correlated with the

number of hydrogen bonds between oligomers and the

solvent. The oligomers show a very high antioxidant

activity compared with esculin.

References

1. Glenn RG (2007) Functional foods, vol 3. FIS Publishing (ed),

104 p

2. Hiramoto T, Saiki K, Masumura S, Shimizu T, Yamashita T,

Kaneko N, Maruta Y (2002) US Patent No 6475544

3. Soeda M, Umehara H, Seiga K, Yoshida T (1997) JP Patent No

09003054

4. Sung Woo K (2004) KR20040097115

5. Yu J, Wang L, Walzem RL, Miller EG, Pike LM, Patil BS (2005)

J Agric Food Chem 53(6):2341–2343

6. Chang WS, Chiang HC (1969) Anticancer Res 15(5B):1969–

1973

7. Kontogiorgis C, Hadjipavlou-Litina D (2003) J Enzym Inhib Med

Chem 18(1):63–69

8. Cai Y, Baer-Dubowska W, Ashwood-Smith M, DiGiovanni J

(1997) Carcinogenesis 18(1):215–222

9. Yu D, Suzuki M, Xie L, Morris-Natschke SL, Lee KH (2003)

Med Res Rev 23(3):322–345

10. Wright WC (2002) Artemisia, ed Taylor & Francis, 359 p

11. Akita M, Tsutsumi D, Kobayashi M, Kise H (2001) Biosci Bio-

technol Biochem 65(7):1581–1588

12. Mita N, Tawaki SI, Uyama H, Kobayashi S (2003) Macromol

Biosci 3:253–257

13. Uyama H, Kurioka H, Kobayashi S (1997) Polym J 29(2):190–

192

14. Ayyagari M, Akkara JA, Kaplan DL (1998) Solvent-enzyme-

polymer interactions in the molecular-weight control of

poly(m-cresol) synthesis in nonaqueous media. In: Gross, RA,

Kaplan DL, Swift G (eds) Enzymes in polymer synthesis. ACS

symposium series 684: Washington pp 112–124

15. Kim YH, An ES, Song BK, Kim DS, Chelikani R (2003) Bio-

technol Lett 25:1521–1524

16. Dubey S, Singh D, Misra RA (1998) Enzyme Microb Techno

23:432–437

17. Kurisawa M, Chung JE, Uyama H, Kobayashi S (2003) Macro-

mol Biosci 3:758–764

18. Desentis-Mendoza RM, Hernandez-Sanchez H, Moreno A, Rojas

del CE, Chel-Guerrero L, Tamariz J, Jaramillo-Flores ME (2006)

Biomacromolecules 7(6):1845–1854

19. Kurisawa M, Chung JE, Uyama H, Kobayashi S (2003) Bio-

macromolecules 4:1394–1399

20. Anthoni J, Linneton F, Wieruzeski JM, Magdalou J, Engasser JM,

Chebil L, Humeau C, Ghoul M (2008) Rasayan J Chem 1(4):718–

731

21. Kaneko T, Baba N, Matsuo M (2003) Chemico-Biological Interac

142(3):239

22. Zhang HY, Wang LF (2004) THEOCHEM 673(1–3):199

23. Chang WS, Chiang HC (1995) Anticancer Res 15:1969–1974

24. Burda S, Oleszek W (2001) J Agric Food Chem 49:2774–2779

25. Saidman E, Yurquina A, Rudyk R, Molina MAA, Ferretti FH

(2002) J Mol Struct 585:1–13

26. Tsai Lin HC, SH Chen CS, Chang YC, Lee CM, Lai ZY, Lin CM

(2008) Biochem Pharmacol 75:1416–1425

27. Trouillas P, Fagnere C, Lazzaroni R, Calliste C, Marfak A, Du-

roux J-L (2004) Food Chem 88(4):571–582

28. Ncanana S, Baratto L, Roncaglia L, Riva S (2007) Adv Synth

Catal 349:1507–1513

29. Intra A, Nicotra S, Riva S, Danieli B (2005) Adv Synth Catal

347(7–8):973–977

30. Dewar MJS, Zoebisch EG, Healy EF, Stewart JJP (1985) J Am

Chem Soc 107:3902–3909

31. CyberChem, Inc. 1115 NW 4th Street Suite 2, Gainesville, FL

32601 USA

32. Materials Studio software and Insight II software. Accelrys, 9685

Scranton Road, San Diego, CA 92121-3752, USA

33. Hwang MJ, Stockfisch TP, Hagler AT (1994) J Am Chem Soc

116:2515–2525

34. Sun H, Mumby SJ, Maple JR, Hagler AT (1994) J Am Chem Soc

116:2978–2987

Eur Food Res Technol (2010) 231:571–579 579

123