Inhibition of horseradish peroxidase catalytic activity by new 3-phenylcoumarin derivatives: Synthesis and structure-activity relationships

Article abstract of DOI:10.1016/j.bmc.2006.10.068

Twenty hydroxylated and acetoxylated 3-phenylcoumarins were synthesized, and the structure-activity relationships were investigated by evaluating the ability of these compounds to modulate horseradish peroxidase (HRP) catalytic activity and comparing the results to four flavonoids (quercetin, myricetin, kaempferol and galangin), previously reported as HRP inhibitors. It was observed that 3-phenylcoumarins bearing a catechol group were as active as quercetin and myricetin, which also show this substituent in the B-ring. The presence of 6,2′-dihydroxy group or 6,7,3′,4′-tetraacetoxy group in the 3-phenylcoumarin structure also contributed to a significant inhibitory effect on the HRP activity. The catechol-containing 3-phenylcoumarin derivatives also showed free radical scavenger activity. Molecular modeling studies by docking suggested that interactions between the heme group in the HRP active site and the catechol group linked to the flavonoid B-ring or to the 3-phenyl coumarin ring are important to inhibit enzyme catalytic activity.

Full text of DOI:10.1016/j.bmc.2006.10.068

Bioorganic & Medicinal Chemistry 15 (2007) 1516–1524
Inhibition of horseradish peroxidase catalytic activity by new
3-phenylcoumarin derivatives: Synthesis and structure–activity
relationships
Luciana M. Kabeya,a Anderson A. de Marchi,b Alexandre Kanashiro,a
Norberto P. Lopes,a Carlos H. T. P. da Silva,b Moˆnica T. Pupob,* and
Yara M. Lucisano-Valima,*
a
ˆ
ˆ
Departamento de Fı´sica e Quı´mica, Faculdade de Ciencias Farmaceuticas de Ribeira˜o Preto da Universidade de Sa˜o Paulo,
Av. Cafe´ s/no, Ribeira˜o Preto—SP, CEP 14040-903, Brazil
b
ˆ
Departamento de Ciencias Farmaceuticas, Faculdade de Ciencias Farmaceuticas de Ribeira˜o Preto da Universidade de Sa˜o Paulo,
ˆ
ˆ
ˆ
Av. Cafe´ s/no, Ribeira˜o Preto—SP, CEP 14040-903, Brazil
Received 13 July 2006; revised 18 October 2006; accepted 31 October 2006
Available online 2 November 2006
Abstract—Twenty hydroxylated and acetoxylated 3-phenylcoumarins were synthesized, and the structure–activity relationships
were investigated by evaluating the ability of these compounds to modulate horseradish peroxidase (HRP) catalytic activity and
comparing the results to four flavonoids (quercetin, myricetin, kaempferol and galangin), previously reported as HRP inhibitors.
It was observed that 3-phenylcoumarins bearing a catechol group were as active as quercetin and myricetin, which also show this
substituent in the B-ring. The presence of 6,20-dihydroxy group or 6,7,30,40-tetraacetoxy group in the 3-phenylcoumarin structure
also contributed to a significant inhibitory effect on the HRP activity. The catechol-containing 3-phenylcoumarin derivatives also
showed free radical scavenger activity. Molecular modeling studies by docking suggested that interactions between the heme group
in the HRP active site and the catechol group linked to the flavonoid B-ring or to the 3-phenyl coumarin ring are important to inhib-
it enzyme catalytic activity.
ꢀ 2006 Elsevier Ltd. All rights reserved.
1. Introduction
compound II + AH2 ! peroxidase + AHÅ + H2O,
Heme peroxidases are widespread in eukaryotes and
prokaryotes, and have a pivotal role in biology.1,2 De-
where ROOH is
a
peroxide and ROH the
spite the striking chemical diversity of these proteins,
the biological oxidation reactions they catalyze involve
similar high-oxidation state intermediates whose reactiv-
ity is modulated by the protein environment.3–6 The
following general mechanism was proposed for peroxi-
dase-catalyzed reaction:
corresponding two-electron reduction product. Com-
pounds I and II are reactive redox intermediates,
which are responsible for the one-electron oxidation
of substrates (AH2) to their corresponding radicals
(AHÅ).7,8
Both mammalian and plant peroxidases have been
reported to produce aromatic oxyl radicals from several
aromatic substrates and reactive oxygen species as part
of a defense mechanism against pathogens.9 These
organic and inorganic free radicals may also participate
in plant physiological processes, such as lignification,
cell wall biosynthesis, auxin catabolism, and wound
healing,10 and in mammalian cells as intracellular signal-
ing, in pro-inflammatory eicosanoids biosynthesis, and
microbial killing.11–13
peroxidase + ROOH $ compound I + ROH
compound I + AH2 ! compound II + AHÅ
Keywords: Catechol; Coumarin; Docking; Horseradish peroxidase.
*
Corresponding authors. Tel.: +55 16 36024710; fax: +55 16 36024879
(M.T.P.); tel.: +55 16 36024434; fax: +55 16 36024880 (Y.M.L.);
e-mail addresses: mtpupo@fcfrp.usp.br; yaluva@usp.br
0968-0896/$ - see front matter ꢀ 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2006.10.068

L. M. Kabeya et al. / Bioorg. Med. Chem. 15 (2007) 1516–1524
1517
However, such reactive species generated by peroxidase-
catalyzed reactions can also contribute to progressive
human tissue damage in chronic inflammatory diseases,
as atherosclerosis, asthma, and rheumatoid arthritis,
which makes peroxidases interesting pharmacological
targets.13,14 The use of phenolic compounds, widely
known as free radical scavengers, has been one of the
therapeutic strategies against inflammatory diseases.15
Suppression of prostaglandin synthesis by phenolic
compounds has been reported to be mediated by the
inhibition of cyclooxygenase peroxidative activity.16,17
ing were performed in order to propose a binding model
for the compounds and the HRP active site.
2. Results
The acetoxylated 3-phenylcoumarin derivatives were
synthesized by the Perkin–Oglialoro reaction of phenyl-
acetic acid derivatives with salicylaldehyde derivatives,
in Ac2O and CH3CO2K (Scheme 1).27 The correspond-
ing hydroxylated 3-phenylcoumarins were obtained by
deacetylation in the presence of aqueous HCl solution
and MeOH. Structures of the synthesized compounds
were established on the basis of their spectral data (see
Section 4.2 for details).
Horseradish peroxidase (HRP) is a well-known and
highly investigated member of the peroxidase superfam-
ily I.18 The HRP-catalyzed oxidation of luminol in the
presence of H2O2 has been extensively used in the
screening of new molecules with antioxidant activity,
especially due to its high sensitivity, simple testing pro-
cedures, and relatively low cost. It has also been a useful
tool in the evaluation of natural bioactive compounds,
products of chemical synthesis, and even of compounds
incorporated into pharmaceutical formulations.19–21
2.1. Effect in the HRP catalytic activity
The modulatory effect of all 3-phenylcoumarins synthe-
sized on the HRP catalytic activity was first screened at
50 lM (Fig. 2A). Most acetoxylated compounds were
not tested in higher concentrations, due to low solubility
in the reaction medium. Non-substituted coumarins (7,
8), the acetoxylated derivatives (except for compound
20), and those compounds not bearing a vicinal dihy-
droxy group (except for compound 15) were not signif-
icantly active, under the assessed conditions. The other
compounds (5, 15, 17, 19, 20) were further evaluated
at six concentrations (0–50 lM) and showed concentra-
tion-dependent inhibitory effects on HRP catalytic activ-
ity (Fig. 2B).
Although the antioxidant activity has been primarily
attributed to the presence of free hydroxy groups, a sig-
nificant antioxidant effect has also been reported for
compounds where these groups are acetylated.22,23 In
previous works, our research group had reported some
classes of natural compounds with antioxidant and
anti-inflammatory activities, such as flavonoids, couma-
rins, and sesquiterpene lactones.21,24,25 We also have
identified hydroxy and acetoxy groups as the most
promising structural features for the biological activities
investigated. In order to understand better the relative
contribution of these substituents to the antioxidant
activity and to the modulation of peroxidase catalytic
activity, twenty hydroxylated and acetoxylated 3-
phenylcoumarins were synthesized having structural fea-
tures of flavonoids previously reported as inhibitors of
HRP activity (Fig. 1 and Scheme 1).26 Their biological
effects as well as structure–activity relationships are dis-
cussed. Moreover, molecular modeling studies by dock-
The calculated IC50 values for the 3-phenylcoumarins
(Table 1) were compared to four flavonoids, previously
reported as HRP inhibitors.26 The results revealed sim-
ilar structure–activity relationships for both classes of
compounds. For example, the most active compounds
within each class bear a catechol group. Moreover,
removal of hydroxy groups in the quercetin B-ring
(yielding kaempferol and galangin) or in the 3-phenyl
ring of the coumarin structure (compare 17 to 11 and
13, Fig. 2A) reduced their inhibitory effects on the
HRP catalytic activity.
Hydroxy group locations on the 3-phenyl ring influ-
enced the coumarin activity (see compounds 11, 13,
and 15, Fig. 2A). Coumarin 15, bearing one hydroxy
group at position C-20, inhibited enzyme activity more
than its analogues 11 and 13, which bear a hydroxy
group at C-30 and C-40, respectively. In addition, acety-
lation of coumarin 19 catechol groups (yielding 20) also
reduced its inhibitory effect.
2.2. Radical scavenging potential
The radical scavenging abilities of all 3-phenylcouma-
rins synthesized based on the determination of drop in
the absorption of the stable free radical DPPH were
evaluated. Four compounds (5, 15, 17, 19) showed a sig-
nificant radical scavenging activity and the results were
expressed as IC50 values (Table 2). The other com-
pounds did not exhibit appreciable activity at a concen-
tration of 50 lM (data not shown).
Figure 1. Chemical structure of the flavonoids included in the
molecular modeling analysis by docking.26

1518
L. M. Kabeya et al. / Bioorg. Med. Chem. 15 (2007) 1516–1524
Scheme 1. Reagents and conditions: (i) 3,4-(methylenedioxy)phenylacetic acid, CH3CO2K, Ac2O, reflux; (ii) HCl 2 N, MeOH; (iii) phenylacetic acid
derivatives, CH3CO2K, Ac2O, reflux.
Figure 2. Modulatory effect of the 3-phenylcoumarin derivatives in the HRP catalytic activity, assessed by luminol-enhanced chemiluminescence
assay. (A) All compounds were tested at 50 lM. (B) Concentration-dependent plots of the 3-phenylcoumarins showing the highest inhibitory effects
in HRP activity. Data are expressed as means standard deviation of three independent experiments, with duplicate measurements.

L. M. Kabeya et al. / Bioorg. Med. Chem. 15 (2007) 1516–1524
Table 1. Inhibitory effect of 3-phenylcoumarins and flavonoids on
1519
HRP catalytic activity
Compounda
c
IC50 (lM)
X
X
X
Y
Y
X
X
Z
19
17
1.2 0.1
2.8 0.2
3.4 0.3
24.6 6.9
28.2 7.9
3.0 0.5
3.1 0.4
10.2 2.9
16.3 2.2
5
15
20
Quercetinb
Myricetinb
Kaempferolb
Galanginb
W
a Chemical structures are shown in Scheme 1 and Figure 1.
b Taken from the literature.26
c Data expressed as means standard deviation of three independent
experiments, with duplicate measurements. IC50 is the concentration
that inhibits HRP catalytic activity by 50%. Statistics: X, Y, Z, W:
values not sharing the same letter are significantly different from each
other (ANOVA and Tukey’s post hoc test).
Table 2. Radical scavenging potential of 3-phenylcoumarins
a
b
Compound
IC50 (lM)
X
Y
Z
Z
19
5
9.8 0.6
13.2 0.6
24.9 2.5
26.8 1.9
15
17
a Chemical structures are shown in Scheme 1.
b Data expressed as means standard deviation of three independent
experiments, with duplicate measurements. IC50 is the concentration
that promoted 50% of DPPH reduction. Statistics: X, Y, Z: values
not sharing the same letter are significantly different from each other
(ANOVA and Tukey’s post hoc test).
Compound 19, bearing two catechol groups, was the
most effective free radical scavenger within the set of
compounds tested in this study. Substitution of 30,40-di-
hydroxy group from compound 19 by 30,40-methylenedi-
oxy group (yielding 5) significantly reduced its activity.
In addition, compounds 15 and 17 were significantly less
active than 5 and 19. The former (15) has hydroxy
groups attached to C-6 and C-20, and the latter (17)
has hydroxy groups at positions C-6, C-30, and C-40.
2.3. Molecular modeling studies by docking
Figure 3. In (A), the top ranked solution of GOLD for quercetin
(carbon atoms in orange) with the HRP unbound structure (PDB code
1H5G; represented by a ribbon diagram) is superimposed to the HRP
crystal structure in a complex with benzhydroxamic acid (PDB code
2ATJ; ligand with carbon atoms in violet), as well as to the HRP
crystal structure in complex with acetate ion (PDB code 1H5A; ligand
with carbon atoms in cyan). In (B), the superposition of the top ranked
solutions for quercetin (carbon atoms in orange) and coumarin 17
(carbon atoms in magenta) with the HRP unbound structure (PDB
code 1H5G) is shown. In (C), the superposition of the top ranked
solutions for quercetin (carbon atoms in orange) and 5 (carbon atoms
in cyan) with the HRP unbound structure (PDB code 1H5G) is shown.
In all structures, the heme group and selected HRP residues are
represented with carbon atoms in green, and the hydrogen atoms are
omitted for clarity.
In order to propose a binding model for the HRP inhib-
itors under study, a flexible docking procedure was em-
ployed with 20 coumarins and 4 flavonoids, using the
˚
unbound HRP structure solved at 1.57 A resolution
(PDB code 1H5G) as the receptor.28 The structure of
HRP in a complex with benzhydroxamic acid, solved
29
˚
at 2.0 A resolution (PDB code 2ATJ), was not used
due to the close similarity between the bound and un-
bound structures.
In Figure 3A, the top ranked GOLD solution for quer-
cetin with the unbound HRP structure (PDB code
1H5G) is superimposed to the crystal structure of
HRP–benzhydroxamic acid complex (PDB code
2ATJ), as well as the HRP crystal structure in complex
with the acetate ion (PDB code 1H5A).28 In general, re-
sults suggest that the compounds here investigated could
bind to HRP in a way that their specific hydroxy or acet-
oxy groups would occupy the positions originally occu-
pied by the oxygen atoms of ligands in the

1520
L. M. Kabeya et al. / Bioorg. Med. Chem. 15 (2007) 1516–1524
crystallographic complexes, close to the heme group.
Catechol group of quercetin firstly interacts with the
conserved and catalytic His42 and Arg38 residues
(Fig. 3A).30 After the simulations, the positions of the
hydroxyls coincide with the oxygen atoms of benzhydr-
oxamic acid and acetate ion in their complexes.
In addition, the fused rings of quercetin were observed
to be flanked by Phe68 and Phe142 of HRP, suggesting
p-stacking interactions (face-to-face and edge-to-face).
These residues are conserved across other peroxidases,
and Phe68 plays a role in the HRP–benzhydroxamic
acid complex, shielding the inhibitor from the solvent
such as a cap, as represented in Figure 4.29 Furthermore,
the hydrophobic interactions due to the Phe68 and
Phe142 residues complement the main interaction ob-
served with the catalytic residues.
Additionally, quercetin B-ring has occupied the binding
site of the benzhydroxamic acid ring, after the simula-
tion. This flavonoid was found to be one of the most ac-
tive compounds assayed so far, suggesting that both
meta- and para-hydroxy groups of the B-ring (catechol
system) play an important role. Moreover, myricetin,
which has three hydroxy groups in the B-ring, was as ac-
tive as quercetin, despite containing one additional hy-
droxy group. Interestingly, the highly active compound
17 contains a catechol group in the benzene ring and a
hydroxy in the coumarinic system. After the docking
simulation, GOLD orients this molecule positioning
the benzene ring close to the heme group, with the hy-
droxy at the same binding site observed for the quercetin
catechol group, thus reinforcing the importance of the
hydroxy oriented in meta- and para-positions in the ben-
zene ring (Fig. 3B).
3. Discussion
In the present study, a series of twenty 3-phenylcouma-
rins was synthesized and evaluated as free radical scav-
engers and inhibitors of the HRP catalytic activity. As
shown in Tables 1 and 2, the catechol group is the main
structural requirement to achieve a high inhibitory effect
in both investigated activities. The experimental results
for the inhibitory effect of 3-phenylcoumarins on the
HRP catalytic activity were compared to those previous-
ly reported by our research group for flavonoids.26
Taken together, the results of the present investigation
highlighted the importance of similar structural features
between flavonoids and 3-phenylcoumarins for the HRP
inhibition, as the presence of the catechol group and the
number and position of hydroxy groups.
On the other hand, results have also indicated that, the-
oretically, the oxygen atoms of the benzodioxole ring in
some of the investigated compounds do not necessarily
mimic the hydroxy groups of the flavonoid B-ring.
Top-ranked solution for coumarin 5, one of the most ac-
tive compounds assayed, indicates an inverted orienta-
tion for this molecule inside the HRP active site, after
the simulation, with both hydroxy groups positioned
close to the heme group, as shown in Figure 3C.
Furthermore, molecular modeling studies by docking
with both classes of compounds revealed similar binding
models for the most active compounds in the HRP cat-
alytic site. In general, we found that the catechol group
of flavonoids and 3-phenylcoumarins was positioned
close to the heme group and the two hydroxyls coincide
with the position of the oxygen atoms in the bezhydrox-
amic acid and acetate molecules, whose crystallographic
orientations inside the HRP active site are known.28,29
Although we did not have an extended series of com-
pounds with HRP inhibitory activity, our attempt in
ascertaining the experimental results obtained was sup-
ported by the agreement of the top-ranked GOLD ori-
entations of the inhibitors with the crystallographic
ligands reported from PDB, regarding the oxygen atoms
that bind to HRP catalytic residues, as well as to an aro-
matic binding site.
The excellent radical scavenging ability of dihydroxy-
lated derivatives of 3-phenylcoumarins and flavonoids
may be explained by the fact that the ortho-dihydroxy
system is able to form a resonance-stable radical and
yield quinone or semi-quinone products, which are less
reactive. In addition, the better interaction of these mol-
ecules with the enzyme active site when compared to the
monophenolic compounds favors the scavenging of the
two electrons involved in the HRP catalytic cycle by
one catechol group.31 Bors et al.32 suggested that three
structural features are important determinants of the
free radical-scavenger potential of flavonoids: (a) the
O-dihydroxy (catechol) system in the B-ring; (b) the
2,3-double bond conjugated with a 4-oxo function;
and (c) the presence of 3- and 5-OH groups. Considering
these requirements, the present investigation suggests
Figure 4. Top ranked solution of GOLD for quercetin (carbon atoms
in orange) with the HRP non-bound structure (PDB code 1H5G;
represented by a surface) is superimposed to the HRP crystal structure
in a complex with benzhydroxamic acid (PDB code 2ATJ; ligand with
carbon atoms in magenta). Flavonoid fused rings are flanked by Phe68
and Phe142 residues in HRP.

L. M. Kabeya et al. / Bioorg. Med. Chem. 15 (2007) 1516–1524
1521
that the catechol group is the most significant determinant
of electron-donating activity and the possible competitive
inhibition of HRP catalytic activity. Accordingly, the
ability of catechol-containing 3-phenylcoumarins to
inhibit enzyme activity was comparable with the cate-
chol-containing flavonoids quercetin and myricetin.
rylhydrazyl radical (DPPH), horseradish peroxidase
(HRP) type VI-A (EC.1.11.1.7), and luminol (5-amine-
2,3-dihydro-1,4-phthalazinedione) were purchased from
Sigma Chemical Co. (St. Louis, MO, USA). Galangin
(3,5,7-trihydroxyflavone) was purchased from Sigma–Al-
drich Chemie (Steinheim, Germany). Dimethylsulfoxide
(DMSO; Merck-Schuchardt, Hohenbrunn, Germany),
ethanol 95% (F. Maia, Sa˜o Paulo, Brazil), and hydrogen
peroxide (H2O2; Labsynth, Diadema, Brazil) were the
other chemicals used.
The present work also identified structural features
other than the catechol group that contributed to the
inhibition of HRP activity, such as the 6,20-dihydroxyla-
tion (15) and the 6,7,30,40-tetraacetoxylation (20).
Regarding compound 15, we suggest that the spatial
proximity between the 20-hydroxy group and the lactone
carbonyl group probably mimics the catechol group
interactions with the HRP active site. However, the
inhibitory activity in the presence of the 20-hydroxy
group was significantly lower. Moreover, the inhibitory
effect of compound 15 was similar to that observed for
the acetoxylated compound 20.
4.2. Synthesis, characterization, and identification of
coumarins
Compounds 1, 2, 3, 4, and 7 were synthesized as previ-
ously described (Scheme 1).27 Purity of all synthesized
compounds was evaluated by thin-layer chromatogra-
phy, using hexane–EtOAc in different proportions
(adjusted to each compound) as the mobile phase.
Although numerous phenol derivatives may act as sub-
strates and donate hydrogen to HRP,31 some acetylated
compounds such as formylphenylacetic acid ethyl ester
and acetylated derivatives of hydroxamic acids33 have
also been reported to be metabolized by the enzyme, but
at lower rates when compared to their hydroxylated par-
ent compounds. Similar results were observed for the
acetoxylated coumarin 20, which had the same activity
pattern of its hydroxylated counterpart but with lower
intensity. The other acetoxylated 3-phenylcoumarin
derivatives did not show inhibitory effect on the HRP cat-
alytic activity, which suggests a specific structural require-
ment of acetoxylated compounds to this biological effect.
Equipments employed: Melting point: Fisher-Johns
12144 melting-point apparatus. 1H NMR: Bruker
DRX400 spectrometer at 400 and 100 MHz or Bru-
ker BPX300 spectrometer at 300 and 75 MHz. Chem-
ical shifts d in parts per million with SiMe4 as an
internal standard, coupling constants J = in hertz.
High-resolution mass spectra: HR-ESI-MS was per-
formed in the positive mode (UltrOTOF-Q system,
version 1.10, Bruker Daltonics, MA, USA). Samples
(0.1 lg/mL) dissolved in dichloromethane were intro-
duced in the mass spectrometer at 5 lL/min. MS
experiments were performed using standard isolation
and excitation procedures.34 Nitrogen was the nebu-
lizing and collision gas with the collision energy set
at 4 eV. An accurate-mass calibration was obtained
through a postacquisition application of a calibration
created by the ratio MS/MS of a monensin A
[Mꢀ18+Na]+ obtained under the same CID and cell
conditions or by an internal calibration using a
known ion in the spectra.
The free radical scavenger activity of phenolic com-
pounds has been reported to mediate the inhibition of
prostaglandin synthesis by cyclooxygenase, since this en-
zyme catalyzes the controlled peroxidation of arachi-
donic acid.16,17 Therefore, inhibition of cyclooxygenase
peroxidative activity by phenolic compounds seems to
be an important mechanism underlying their anti-in-
flammatory effect.
4.2.1. General procedure for the preparation of coumarin
8 and acetoxylated derivatives. Non-substituted couma-
rin 8 and acetoxylated derivatives (6, 10, 12, 14, 16,
18, 20) were synthesized under anhydrous conditions,
by using equipment previously dried at 60 ꢁC for at least
12 h and also at 300 ꢁC during few minutes immediately
before use.
In conclusion, the results reported here may be helpful
to understand better the interactions between 3-phen-
ylcoumarins and the HRP catalytic site. The structure–
activity relationships allowed identification of new
relevant structural features to the HRP catalytic activity
modulation. Furthermore, as the formation of free rad-
icals through peroxidase-catalyzed reactions is involved
in the development of chronic inflammatory diseases,
the present study can help in developing new molecules
to control inflammation and free radical-mediated tissue
injury.
The solution containing anhydrous CH3CO2K (289 mg,
2.94 mmol), the appropriate phenylacetic acid derivative
(1.67 mmol),
and
hydroxylated
benzaldehyde
(1.67 mmol) in Ac2O (1.2 mL) was refluxed (138 ꢁC)
with stirring during different periods of time. Except
for compound 16 (see item Section 4.2.1.6), the reaction
mixture of the other compounds was cooled, neutralized
with 10% aqueous NaHCO3, and extracted (3· 30 mL)
with EtOAc (8, 6, 18, 20) or CHCl3 (10, 12, 14). The
organic layers were pooled, washed with distilled water,
dried (anhydrous Na2SO4), filtered, and evaporated un-
der reduced pressure. The products were purified by
recrystallization in EtOH (8, 10, 12, 16, 14, 20) or
EtOH/EtOAc 2:1 (6) and dried.
4. Experimental
4.1. Chemicals
Kaempferol (3,40,5,7-tretrahydroxyflavone), myricetin
(3,30,40,5,50,7-hexahydroxyflavone), quercetin (3,30,40,5,
7-pentahydroxyflavone dihydrate), 1,1-diphenyl-2-pic-

1522
L. M. Kabeya et al. / Bioorg. Med. Chem. 15 (2007) 1516–1524
4.2.1.1. 3-Phenyl-coumarin (8). Prepared as described
7.51 (dl, J = 2.7 Hz, H-5), 7.21 (dd , J = 8.8, 2.7 Hz,
H-7), 7.66 (dl, J = 8.8 Hz, H-8), 7.27 (dl, J = 8.5 Hz,
H-30)a, 7.36 (ddd, J = 8.0, 7.7, 1.3 Hz, H-40)b, 7.08
(ddd, J = 7.7, 7.7, 1.0, Hz, H-50)b, 7.12 (ddd, J = 8.1,
0.9, 0.5 Hz, H-60)a, 2.32 (s, OAc), 2.28 (s, OAc). (a,b
May be interchangeable.) HR-ESI-MS: m/z = 361.0701
[M+Na]+ (calcd 361.0688).
in Section 4.2.1, using phenylacetic acid (228 mg) and 2-
hydroxy-benzaldehyde (0.18 mL), refluxed for 3 h.
Yield: 63 mg (17%). Mp: 130–133 ꢁC. 1H NMR
(400 MHz, CDCl3): 7.75 (s, H-4), 7.20–7.66 (m, H-5,
H-6, H-7, H-8, H-20, H-30, H-40, H-50 and H-60). HR-
ESI-MS: m/z = 223.0782 [M + H]+ (calcd 223.0759).
4.2.1.2. 6,7-Diacetoxy-3-[30,40-methylenedioxyphenyl]-
coumarin (6). Prepared as described in Section 4.2.1,
using 3,4-(methylenedioxy)phenylacetic acid (301 mg)
and 2,4,5-trihydroxy-benzaldehyde (257 mg), refluxed
for 2 h. Yield: 118 mg (19%). Mp: 186–189 ꢁC. 1H
NMR (300 MHz, CDCl3): 7.67 (s, H-4), 7.37 (s, H-5)a,
7.24 (s, H-8)a, 7.20 (d, J = 1.5 Hz, H-20), 6.89 (d,
J = 8.4 Hz, H-50), 7.17 (dd, J = 8.4, 1.5 Hz, H-60), 6.04
(s, H-70), 2.36 (s, OAc), 2.35 (s, OAc). (a May be inter-
changeable.) HR-ESI-MS: m/z = 405.0684 [M+Na]+
(calcd 405.0650).
4.2.1.7. 6-Acetoxy-3-[30,40-diacetoxyphenyl]-coumarin
(18). Prepared as described in Section 4.2.1, using 3,4-
dihydroxyphenylacetic acid (281 mg) and 2,5-dihy-
droxy-benzaldehyde (231 mg), refluxed for 5 h. After
extraction with EtOAc, a small amount of solid reaction
product remained in the organic layer, which was re-
moved by filtration and washing with EtOAc. The organic
layers resulting from these procedures were pooled,
washed with distilled water, dried (anhydrous Na2SO4),
filtered, and evaporated under reduced pressure. Couma-
rin 18 was obtainedas a pure product, and no further puri-
fication by recrystallization was carried out. Yield:
1
4.2.1.3. 6-Acetoxy-3-phenyl-coumarin (10). Prepared
as described in Section 4.2.1, using phenylacetic acid
(228 mg), and 2,5-dihydroxy-benzaldehyde (231 mg), re-
fluxed for 3.5 h. Yield: 140 mg (30%). Mp: 150–153 ꢁC.
1H NMR (400 MHz, CDCl3): 7.68 (s, H-4), 7.23 (d,
J = 2.5 Hz, H-5), 7.17 (dd, J = 8.8, 2.5 Hz, H-7), 7.30 (d,
J = 8.8 Hz, H-8), 7.60–7.63 (m, H-20 and H-60), 7.32–
7.41 (m, H-30, H-40 and H-50), 2.26 (s, OAc). HR-ESI-
MS: m/z = 303.0693 [M+Na]+ (calcd 303.0661).
181 mg (27%). Mp: 208–212 ꢁC. H NMR (300 MHz,
CDCl3): 7.79 (s, H-4), 7.62 (m, H-5 and H-7), 7.29 (d,
J = 8.9 Hz, H-8), 7.32 (d, J = 2.7 Hz, H-20), 7.38 (d,
J = 8.9 Hz, H-50), 7.27 (dd, J = 8.9, 2.7 Hz, H-60), 2.35
(s, OAc), 2.32 (s, OAc). HR-ESI-MS: m/z = 419.0787
[M+Na]+ (calcd 419.0743).
4.2.1.8. 6,7-Diacetoxy-3-[30,40-diacetoxyphenyl]-cou-
marin (20). Prepared as described in Section 4.2.1, using
3,4-dihydroxyphenylacetic acid (281 mg) and 2,4,5-tri-
hydroxy-benzaldehyde (257 mg), refluxed for 1 h. Yield:
4.2.1.4. 6-Acetoxy-3-[30-acetoxyphenyl]-coumarin (12).
Prepared as described in Section 4.2.1, using 3-
hydroxyphenylacetic acid (254 mg) and 2,5-dihydroxy-
benzaldehyde (231 mg), refluxed for 2 h. Yield: 193 mg
1
100 mg (13%). Mp: 185–188 ꢁC. H NMR (300 MHz,
CDCl3): 7.77 (s, H-4), 7.40 (s, H-5), 7.26 (s, H-8)a,
7.60 (m, H-20 and H-60), 7.28 (d, J = 9.1 Hz, H-50),
2.34 (s, OAc), 2.33 (s, OAc), 2.32 (s, OAc). a Signal coin-
cident with CDCl3. HR-ESI-MS: m/z = 477.0850
[M+Na]+ (calcd 477.0798).
1
(34%). Mp: 154–156 ꢁC. H NMR (400 MHz, CDCl3):
7.72 (s, H-4), 7.24 (d, J = 2.8 Hz, H-5), 7.19 (dd,
J = 8.8, 2.5 Hz, H-7), 7.31 (d, J = 9.1 Hz, H-8), 7.41
(dd, J = 2.3, 1.0 Hz, H-20), 7.09 (ddd, J = 8.1, 2.3,
1.0 Hz, H-40), 7.38 (t, J = 8.1 Hz, H-50), 7.50 (ddd,
4.2.2. General procedure for the preparation of the
hydroxylated compounds. Hydroxylated coumarins (5,
9, 11, 13, 15, 17, 19) were obtained by hydrolysis of their
acetoxylated counterparts. In general, the appropriate
acetoxylated coumarin mixed with 2 N aqueous HCl
and MeOH was refluxed (100 ꢁC) with stirring during
different periods of time. The resulting reaction mixtures
were cooled in an ice-bath and the reaction products,
obtained as solids, were filtered, washed with cold dis-
tilled water, and dried under vacuum.
J = 7.8, 1.8, 1.0 Hz, H-60), 2.26 (s, OAc)a, 2.27 (s, OAc)a.
a
(
May be interchangeable.) HR-ESI-MS: m/z =
361.0723 [M+Na]+ (calcd 361.0688).
4.2.1.5. 6-Acetoxy-3-[40-acetoxyphenyl]-coumarin (14).
Prepared as described in Section 4.2.1, using 4-hydroxy-
phenylacetic acid (254 mg) and 5-dihydroxy-benzalde-
hyde (231 mg), refluxed for 2 h. Yield: 162 mg (29%).
1
Mp: 199–200 ꢁC. H NMR (400 MHz, CDCl3): 7.75 (s,
H-4), 7.31 (d, J = 2.8 Hz, H-5), 7.25 (dd, J = 8.8,
2.8 Hz, H-7), 7.38 (d, J = 9.1 Hz, H-8), 7.73 (d,
J = 8.8 Hz, H-20 and H-60), 7.19 (d, J = 8.8 Hz, H-30
and H-50), 2.33 (s, 6-OAc)a, 2.34 (s, OAc)a. (a May be in-
terchangeable.) HR-ESI-MS: m/z = 361.0727 [M+Na]+
(calcd 361.0688).
4.2.2.1. 6,7-Dihydroxy-3-[30,40-methylenedioxyphenyl]-
coumarin (5). A mixture of coumarin 6 (75 mg), HCl
(10 mL) and MeOH (4 mL) was allowed to react for
5 h. Yield: 49 mg (84%). Mp: >300 ꢁC. 1H NMR
(400 MHz, CD3OD): 7.86 (s, H-4), 6.99 (s, H-5)a, 6.77
(s, H-8)a, 7.21 (d, J 1.8 Hz, H-20), 6.87 (d, J 8.1 Hz, H-
50), 7.17 (dd, J 8.1, 1.8 Hz, H-60), 5.98 (s, H-70). (a
May be interchangeable.) HR-ESI-MS: m/z = 299.0573
[M+H]+ (calcd 299.0556).
4.2.1.6.
6-Acetoxy-3-[20-acetoxyphenyl]-coumarin
(16). Prepared as described in Section 4.2.1, using
2-hydroxyphenylacetic acid (254 mg) and 2,5-dihy-
droxy-benzaldehyde (231 mg), refluxed for 1 h. After
the reaction period, distilled water was added to the
reaction mixture and the resulting solid filtered and
washed with distilled water. Yield: 192 mg (34%). Mp:
4.2.2.2. 6-Hydroxy-3-phenyl-coumarin (9). A mixture
of coumarin 10 (80 mg), HCl (10 mL) and, MeOH
(4 mL) was allowed to react for 2 h. Yield: 54 mg
(79%). Mp: 202–204 ꢁC. 1H NMR (400 MHz, CD3OD):
1
161–164 ꢁC. H NMR (400 MHz, CDCl3): 7.69 (s, H-2),

L. M. Kabeya et al. / Bioorg. Med. Chem. 15 (2007) 1516–1524
1523
7.99 (s, H-4), 7.06–7.09 (m, H-5 and H-7), 7.26 (ddd,
J = 7.6, 1.8, 0.8 Hz, H-8), 7.70–7.73 (m, H-20 and H-
60), 7.39–7.48 (m, H-30, H-40 and H-50). HR-ESI-MS:
m/z = 239.0736 [M+H]+ (calcd 239.0708).
were added to 0.1 M sodium phosphate buffer, pH 7.4.
After incubation for 3 min at 30 ꢁC, HRP solution
(0.2 IU/mL) was added to the reaction medium and
chemiluminescence production was measured in a lumi-
nometer (AutoLumat LB953, EG&G Berthold, Germa-
ny) during 20 min at 30 ꢁC. Luminol and the test-
compounds were prepared in DMSO, whereas H2O2
and HRP were prepared in 0.1 M phosphate buffer,
pH 7.4. DMSO final concentration in the reaction medi-
um was 1%. Values of the integrated area of chemilumi-
nescence were used to calculate percentages of inhibition
or increase promoted by each concentration of the
tested compounds. IC50 values were determined for
those compounds with significant inhibitory activity.
4.2.2.3. 6-Hydroxy-3-[30-hydroxyphenyl]-coumarin (11).
A mixture of coumarin 12 (85 mg), HCl (7 mL), and
MeOH (3 mL) was allowed to react for 4 h. Yield:
46 mg (72%). Mp: 267–268 ꢁC. 1H NMR (400 MHz,
CD3OD): 7.95 (s, H-4), 7.03 (d, J = 2.8 Hz, H-5), 7.05
(dd, J = 8.4, 2.8 Hz, H-7), 7.22 (dl, J = 8.4 Hz, H-8),
7.12–7.16 (m, H-20 and H-40), 7.25 (t, J = 7.6 Hz, H-
50), 6.82 (ddd, J = 8.1, 2.3, 1.0 Hz, H-60). HR-ESI-MS:
m/z = 255.0679 [M+H]+ (calcd 255.0657).
4.2.2.4. 6-Hydroxy-3-[40-hydroxyphenyl]-coumarin (13).
A mixture of coumarin 14 (90 mg), HCl (10 mL), and
MeOH (4 mL) was allowed to react for 1 h. Yield:
55 mg (81%). Mp: 289–292 ꢁC. 1H NMR (400 MHz,
CDCl3: CD3OD 8:2): 7.58 (s, H-4), 6.81 (d, J = 2.8 Hz,
H-5), 6.88 (dd, J 8.8, 2.8 Hz, H-7), 7.07 (d, J = 8.8 Hz,
H-8), 7.42 (d, J = 8.8 Hz, H-20 and H-60), 6.75 (d,
J = 8.8 Hz, H-30 and H-50). HR-ESI-MS: m/
z = 255.0673 [M+H]+ (calcd 255.0657).
4.4. Assay of DPPH radical scavenging
DPPH free radical scavenging ability of the 3-phenyl-
coumarin derivatives was evaluated according to a mod-
ified method of Blois.36 DPPH solution (final
concentration of 100 lM) in ethanol was diluted in
0.1 M sodium acetate buffer, pH 5.4 (6:4) prior to addi-
tion of the test-compounds (0–50 lM) in DMSO. The
contents were vigorously mixed and allowed to stand
at 25 ꢁC for 5 min. Absorbance was measured at
510 nm, before (100%) and after addition of the test-
compounds. Percentage of DPPH reduction by each
sample was calculated.
4.2.2.5.
(15).
6-Hydroxy-3-[20-hydroxyphenyl]-coumarin
mixture of coumarin 16 (110 mg), HCl
A
(10 mL), and MeOH (3 mL) was allowed to react for
2 h. Yield: 30 mg (36%). Mp: 269–270 ꢁC. 1H NMR
(400 MHz, CD3OD): 7.86 (s, H-4), 6.99 (d, J = 2.8 Hz,
H-5), 7.05 (dd, J = 8.8, 2.8 Hz, H-7), 7.24 (dl,
J = 8.8 Hz, H-8), 6.88 (dd, J = 7.5, 1.0 Hz, H-30), 7.21
(ddd, J = 8.3, 7.3, 1.8 Hz, H-40), 6.89 (dt, J = 7.5, 7.5,
1.0 Hz, H-50), 7.29 (dd, J = 7.8, 1.8 Hz, H-60). HR-
ESI-MS: m/z = 255.0659 [M+H]+(calcd 255.0657).
4.5. General docking strategy
Flexible docking simulations were performed using the
GOLD 3.0 software,37 which contains an implemented
genetic algorithm. Default parameters are originally
optimized for a set of 305 different complexes with
solved structures provided in the Protein Data Bank
(PDB). A population of 100 conformers, 100000 oper-
ations, 95 mutations, and 95 crossovers was used.
Docking calculations were performed inside a sphere
4.2.2.6. 6-Hydroxy-3-[30,40-dihydroxyphenyl]-couma-
rin (17). A mixture of coumarin 18 (100 mg), HCl
(8 mL), and MeOH (3 mL) was allowed to react for
1
˚
2.5 h. Yield: 64 mg (95%). Mp: 294–297 ꢁC. H NMR
with a 15 A radius and centered at the His42 Ne2 oxy-
gen of the horseradish peroxidase structure (PDB code
(400 MHz, CD3OD): 7.85 (s, H-4), 7.00–7.03 (m, H-5
and H-7), 7.19–7.21 (m, H-8), 7.21 (d, J = 2.3 Hz, H-
20), 6.82 (d, J = 8.3 Hz, H-50), 7.07 (dd, J = 8.3,
2.3 Hz, H-60). HR-ESI-MS: m/z = 271.0624 [M+H]+
(calcd 271.0607).
1H5G),28 which has been solved at the highest resolu-
˚
tion (1.57 A) amongst all horseradish peroxidases with
coordinates deposited in PDB. The residue chosen is
situated approximately at the center of the HRP ac-
tive site. Ten orientations of the highest score ob-
tained for each compound were thus selected
through the GoldScore. Based on this function,
GOLD classifies the orientations of the molecules by
a decreasing affinity order (the scores) with the recep-
tor binding site. Previous to the docking calculations
and after the removal of the crystallographic waters
inside the active site, residue side-chain hydrogen
atoms were added and oriented with constrained min-
imization procedures, where only these atoms were
kept free. Atomic changes and potentials of the AM-
BER force field were added to the protein structure
using the Insight II software.38
4.2.2.7. 6,7-Dihydroxy-3-(30,40-dihydroxyphenyl)-cou-
marin (19). A mixture of coumarin 20 (75 mg), HCl
(7 mL), and MeOH (3 mL) was allowed to react for
3 h. Yield: 35 mg (75%). Mp: >300 ꢁC. 1H NMR
(400 MHz, CD3OD): 7.79 (s, H-4), 6.98 (s, H-5)a,6.76
(s, H-8)a, 7.17 (d, J = 2.0 Hz, H-20), 6.80 (d,
J = 8.3 Hz, H-50), 7.02 (dd, J = 8.3, 2.0, H-60). (a May
be interchangeable.) HR-ESI-MS: m/z = 309.0401
[M+Na]+ (calcd 309.0375).
4.3. Evaluation of HRP catalytic activity
HRP activity was evaluated by a modification of the
method described by Krol et al.35 Values in parentheses
are final concentrations in a reaction volume of 0.5 mL.
Briefly, aliquots of H2O2 (50 lM), luminol (140 lM),
and each test-compound (0–50 lM) or DMSO (control)
Conformational search was performed with the 20 cou-
marins and 4 flavonoids, using the systematic method
for molecules with 5 or less rotatable bonds. The MON-
TE CARLO method was used for the most complex

1524
L. M. Kabeya et al. / Bioorg. Med. Chem. 15 (2007) 1516–1524
´
17. Garcıa-Argaez, A. N.; Apan, T. O. R.; Delgado, H. P.;
Velazquez, G.; Martınez-Vazquez, M. Planta Med. 2000,
´
molecules, with the MMFF molecular mechanics model.
Structures were then fully optimized in the gas phase at
B3LYP/6-31G* level, using the Spartan 04 1.0.3
software.39
´
´
´
66, 279.
18. Veitch, N. C. Phytochemistry 2004, 65, 249.
19. Dure, L. S.; Cormier, M. J. J. Biol. Chem. 1964, 239, 2351.
20. Marquele, F. D.; Di Mambro, V. M.; Georgetti, S. R.;
Casagrande, R.; Lucisano-Valim, Y. M.; Fonseca, M. J.
V. J. Pharm. Biomed. Anal. 2005, 39, 455.
Acknowledgments
21. Kanashiro, A.; Kabeya, L. M.; Grael, C. F. F.; Jordao, C.
O.; Azzolini, A. E. C. S.; Lopes, J. L. C.; Lucisano-Valim,
Y. M. J. Pharm. Pharmacol. 2006, 58, 853.
´
The authors thank Mr. Jose Carlos Tomaz and Mrs.
Claudia Castania de Macedo (Faculdade de Ciencias
´
ˆ
22. Kumar, A.; Singh, B. K.; Tyagi, R.; Jain, S. K.; Sharma,
S. K.; Prasad, A. K.; Raj, H. G.; Rastogi, R. C.;
Watterson, A. C.; Parmar, V. S. Bioorg. Med. Chem.
2005, 13, 4300.
23. Raj, H. G.; Parmar, V. S.; Jain, S. C.; Priyadarsini, K. I.;
Mittal, J. P.; Goel, S.; Das, S. K.; Sharma, S. K.; Olsen, C.
E.; Wengel, J. Bioorg. Med. Chem. 1999, 7, 2091.
24. Gobbo-Neto, L.; Santos, M. D.; Kanashiro, A.; Almeida,
M. C.; Lucisano-Valim, Y. M.; Lopes, J. L. C.; Souza, G.
E. P.; Lopes, N. P. Planta Med. 2005, 71, 3.
ˆ
Farmaceuticas de Ribeira˜o Preto da Universidade de
Sa˜o Paulo, Ribeira˜o Preto, SP, Brazil) for helpful tech-
nical assistance, and Conselho Nacional de Desenvolvi-
´
mento Cientıfico e Tecnologico (CNPq, Brazil, Grant
140462/2003-1) and Fundac¸a˜o de Amparo a Pesquisa
do Estado de Sa˜o Paulo (FAPESP, Brazil, Grants 98/
14107-0, 01/05617-9, and 02/06800-4) for financial sup-
port. Carlos H.T.P. da Silva acknowledges CAPES for
financial support.
´
`
25. Kabeya, L. M.; Kanashiro, A.; Azzolini, A. E. C. S.;
Soriani, F. M.; Lopes, J. L. C.; Lucisano-Valim, Y. M.
Res. Commun. Mol. Pathol. Pharmacol. 2002, 111, 103.
26. Moreira, M. R.; Polizello, A. C. M.; Kabeya, L. M.;
Kanashiro, A.; Lucisano-Valim, Y. M. Braz. J. Pharm.
Sci. 2003, 39, 87, 4th Congress of Pharmaceutical Sciences,
Ribeira˜o Preto, Brazil, June 29th–July 2nd, 2003. Univer-
sity of Sa˜o Paulo: Sa˜o Paulo, SP, 2003.
27. de Marchi, A. A.; Castilho, M. S.; Nascimento, P. G. B.;
Archanjo, F. C.; Del Ponte, G.; Oliva, G.; Pupo, M. T.
Bioorg. Med. Chem. 2004, 12, 4823.
28. Berglund, G. I.; Carlsson, G. H.; Smith, A. T.; Szoke, H.;
Henriksen, A.; Hadju, J. Nature 2002, 417, 463.
29. Henriksen, A.; Schuller, D. J.; Meno, K.; Smith, A. T.;
Welinder, K. G.; Gajhede, M. Biochemistry 1998, 37,
8054.
References and notes
1. Everse, J.; Everse, K. E.; Grisham, M. B. In Peroxidases
in Chemistry and Biology; CRC Press: Boca Raton, FL,
1991.
2. Poulos, T. L.; Fenna, R. E.. In Metals in Biological
Systems; Siegel, H., Ed.; Marcel Dekker: New York, 1994;
Vol. 30, pp 25–75.
3. Keilin, D.; Hartree, E. F. Biochem. J. 1951, 49, 88.
4. Nakajima, R.; Yamazaki, I. J. Biol. Chem. 1987, 262,
2576.
5. Acosta, M.; Arnao, M. B.; Hernandez-Ruiz, J.; Garcia-
Canovas, F. In Plant Peroxidases: Biochemistry and
Physiology; Welinder, K. G., Rassmussen, S. K., Penel,
C., Greppin, H., Eds.; University of Geneva: Geneva,
1993; pp 201–205.
30. Rodrıguez-Lopez, J. N.; Lowe, D. J.; Hernandez-Ruiz, J.;
Hiner, A. N. P.; Garcıa-Canovas, F.; Thorneley, R. N. F.
J. Am. Chem. Soc. 2001, 123, 11838.
6. Hernandez-Ruiz, J.; Arnao, M. B.; Hiner, A. N. P.;
Garcia-Canovas, F.; Acosta, M. Biochem. J. 2001, 354,
107.
´
A. J. Biolumin. Chemilumin 1998, 13, 75.
´
´
´ ´
31. Navas Dıaz, A.; Garcıa Sanches, F.; Gonzalez Garcıa, J.
7. George, P. Nature 1952, 169, 612–613.
32. Bors, W.; Heller, W.; Michel, C.; Saran, M. Methods
Enzymol. 1990, 186, 343.
33. Corbett, M. D.; Corbett, B. R. Chem. Biol. Interact. 1987,
63, 249.
34. Crotti, A. E. M.; Vessecchi, R.; Lopes, J. L. C.; Lopes, N.
P. Quı´m. Nova 2006, 29, 287.
35. Krol, W.; Czuba, Z.; Scheller, S.; Paradowski, Z.; Shani, J.
J. Ethnopharmacol. 1994, 41, 121.
36. Blois, M. S. Nature 1958, 181, 1199.
8. Poulos, T. L.; Kraut, J. J. Biol. Chem. 1980, 255, 8199.
9. Takahama, U.; Yoshitama, K. J. Plant Res. 1998, 111, 97.
10. Hiraga, S.; Sasaki, K.; Ito, H.; Ohashi, Y.; Matsui, H.
Plant Cell Physiol. 2001, 42, 462.
11. Babior, B. M. Am. J. Med. 2000, 109, 33.
12. Zhang, R.; Brennan, M.-L.; Shen, Z.; MacPherson, J. C.;
Schmitt, D.; Molenda, C. E.; Hazen, S. L. J. Biol. Chem.
2002, 277, 46116.
13. Klebanoff, S. J. J. Leukocyte Biol. 2005, 77, 598.
14. Obinger, C. Arch. Biochem. Biophys. 2006, 445, 197.
15. Middleton, E., Jr.; Kandaswami, C.; Theoharides, T. C.
Pharmacol. Rev. 2000, 52, 673.
37. Verdonk, M. L.; Cole, J. C.; Hartshorn, M. J.; Murray, C.
W.; Taylor, R. D. Proteins 2003, 52, 609.
38. Insight II user guide, v. 2000, Accelrys, CA, USA, 2000.
39. Spartan ’04 full, v.1.0.3; Tutorial and User’s Guide;
Wavefunction, Inc., USA, 2005.
´
16. Hoult, J. R. S.; Paya, M. Gen. Pharmacol. 1996, 27, 713.