MAO Inhibitory Activity of 2-Arylbenzofurans versus 3-Arylcoumarins: Synthesis, invitro Study, and Docking Calculations

Article abstract of DOI:10.1002/cmdc.201300048

Monoamine oxidase (MAO) is an important drug target for the treatment of neurological disorders. Several 3-arylcoumarin derivatives were previously described as interesting selective MAO-B inhibitors. Preserving the trans-stilbene structure, a series of 2-arylbenzofuran and corresponding 3-arylcoumarin derivatives were synthesized and evaluated as inhibitors of both MAO isoforms, MAO-A and MAO-B. In general, both types of derivatives were found to be selective MAO-B inhibitors, with IC₅₀ values in the nano- to micromolar range. 5-Nitro-2-(4-methoxyphenyl)benzofuran (8) is the most active compound of the benzofuran series, presenting MAO-B selectivity and reversible inhibition (IC₅₀=140nM). 3-(4′-Methoxyphenyl)-6-nitrocoumarin (15), with the same substitution pattern as that of compound 8, was found to be the most active MAO-B inhibitor of the coumarin series (IC₅₀=3nM). However, 3-phenylcoumarin 14 showed activity in the same range (IC₅₀=6nM), is reversible, and also severalfold more selective than compound 15. Docking experiments for the most active compounds into the MAO-B and MAO-A binding pockets highlighted different interactions between the derivative classes (2-arylbenzofurans and 3-arylcoumarins), and provided new information about the enzyme-inhibitor interaction and the potential therapeutic application of these scaffolds.

Full text of DOI:10.1002/cmdc.201300048

CHEMMEDCHEM
FULL PAPERS
DOI: 10.1002/cmdc.201300048
MAO Inhibitory Activity of 2-Arylbenzofurans versus 3-
Arylcoumarins: Synthesis, in vitro Study, and Docking
Calculations
Giulio Ferino,[a] Enzo Cadoni,[a] Maria Jo¼o Matos,*[b] Elias Quezada,[b] Eugenio Uriarte,[b]
Lourdes Santana,[b] Santiago Vilar,[c] Nicholas P. Tatonetti,[c] Matilde YꢀÇez,[d] Dolores ViÇa,[d]
Carmen Picciau,[e] Silvia Serra,[e] and Giovanna Delogu*[e]
Monoamine oxidase (MAO) is an important drug target for the
treatment of neurological disorders. Several 3-arylcoumarin de-
rivatives were previously described as interesting selective
MAO-B inhibitors. Preserving the trans-stilbene structure,
a series of 2-arylbenzofuran and corresponding 3-arylcoumarin
derivatives were synthesized and evaluated as inhibitors of
both MAO isoforms, MAO-A and MAO-B. In general, both types
of derivatives were found to be selective MAO-B inhibitors,
with IC50 values in the nano- to micromolar range. 5-Nitro-2-(4-
methoxyphenyl)benzofuran (8) is the most active compound
of the benzofuran series, presenting MAO-B selectivity and re-
versible inhibition (IC50 =140 nm). 3-(4’-Methoxyphenyl)-6-nitro-
coumarin (15), with the same substitution pattern as that of
compound 8, was found to be the most active MAO-B inhibitor
of the coumarin series (IC50 =3 nm). However, 3-phenylcoumar-
in 14 showed activity in the same range (IC50 =6 nm), is reversi-
ble, and also severalfold more selective than compound 15.
Docking experiments for the most active compounds into the
MAO-B and MAO-A binding pockets highlighted different inter-
actions between the derivative classes (2-arylbenzofurans and
3-arylcoumarins), and provided new information about the
enzyme–inhibitor interaction and the potential therapeutic ap-
plication of these scaffolds.
Introduction
Parkinson’s disease (PD) is a condition associated with degen-
eration of the dopamine-containing nigrostriatal neurons.[1]
The resulting depletion of striatal dopamine is responsible for
the characteristic PD symptoms such as bradykinesia (slowness
of movement), rigidity (stiffness), hypokinesia (decreased
movement amplitude), akinesia (absence of normal uncon-
scious movements) and other extrapyramidal effects. Because
monoamine oxidases A and B (MAO-A and MAO-B) are in-
volved in the metabolic degradation of dopamine in the brain,
inhibitors of these enzymes are considered useful for the treat-
ment of PD.[2]
MAOs are FAD-dependent enzymes found in the outer mito-
chondrial membrane in neuronal, glial, and other mammalian
cells.[3] MAO-A and MAO-B were identified and distinguished
by their amino acid sequences,[4] three-dimensional structure,[5]
tissue distribution,[6] inhibitor selectivity,[7] and substrate prefer-
ences.[8] The similarities between the genes that encode the
MAO-A and MAO-B isoforms suggest that both are derived
from a common ancestral gene. MAO-A and MAO-B share 70%
overall primary sequence identity, yet this value is higher for
certain regions, such as the pentapeptide (Ser-Gly-Gly-Cys-Tyr)
that allows covalent attachment of FAD through cysteine. The
MAO-A isoform has higher affinity for serotonin and noradre-
naline, whereas the MAO-B isoform preferentially deaminates
b-phenylethylamines and benzylamine.[9] However, dopamine
and tyramine are common substrates for both isoforms. Due
to their general affinity for different substrates, these enzymes
are involved in different pathologies such as anxiety and de-
pression (MAO-A), and Alzheimer’s disease (AD) and PD (MAO-
B). These physiological properties therefore determine the clin-
ical interest of MAO inhibitors. Selective MAO-A inhibitors
(MAOI-A), such as clorgyline (irreversible) and moclobemide
(reversible), are used in the treatment of neurological disor-
ders,[10] whereas selective and irreversible MAO-B inhibitors
(MAOI-B), such as selegiline and rasagiline, are used in the
treatment of neurodegenerative diseases.[11] Several studies
[a] Dr. G. Ferino, Prof. E. Cadoni
Department of Chemical and Geological Sciences
University of Cagiari, S.S. 554, 09042 Monserrato Cagliari (Italy)
[b] Dr. M. J. Matos, Dr. E. Quezada, Prof. E. Uriarte, Prof. L. Santana
Department of Organic Chemistry, Faculty of Pharmacy
University of Santiago de Compostela
15782 Santiago de Compostela (Spain)
E-mail: mariacmatos@gmail.com
[c] Dr. S. Vilar, Dr. N. P. Tatonetti
Department of Biomedical Informatics
Columbia University Medical Center, New York, NY 10032 (USA)
[d] M. YꢀÇez, D. ViÇa
Department of Pharmacology
CIMUS, University of Santiago de Compostela
15782 Santiago de Compostela (Spain)
[e] Dr. C. Picciau, Dr. S. Serra, Prof. G. Delogu
Department of Life Sciences and Environment
Section of Pharmaceutical Sciences
University of Cagliari, 09124 Cagliari (Italy)
E-mail: delogug@unica.it
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 0000, 00, 1 – 12
&
1
&
ÞÞ
These are not the final page numbers!

CHEMMEDCHEM
FULL PAPERS
www.chemmedchem.org
point to the neuroprotective role of MAOI-B in PD and their
modifying effect by delaying progression of the disease. Rasa-
giline, a selective MAOI-B, has been reported to retard the fur-
ther deterioration of cognitive functions, displaying an impor-
tant neuroprotective activity against the consequences of oxi-
dative stress in patients.[12] Irreversible inhibition of MAO, how-
ever, has certain disadvantages including the fact that enzyme
recovery requires the synthesis of new enzyme, the possible
loss of selectivity as a result of repeated administration, and in-
hibition that is unaffected by changes in substrate concentra-
tion.[13]
Figure 1. General structures of the described benzofuran and coumarin
derivatives.
conditions and characterization data for the new compounds
are described below in the Experimental Section. Benzofuran
derivatives 1–10 were synthesized according to the protocol
outlined in Scheme 1. The key step for the formation of the
All of these aspects have led to an intensive search for novel
MAOIs. Additionally, reports of the crystal structures of both
MAO isoforms by Binda et al. has provided relevant informa-
tion about the selective interactions and the pharmacophoric
requirements needed for the design of potent and selective in-
hibitors.[14–16]
Coumarins are a large family of compounds of natural and
synthetic origin that display a variety of pharmacological prop-
erties. Due to their structural variability they occupy an impor-
tant place in the realm of natural products and synthetic or-
ganic chemistry. Several studies have paid special attention to
their antioxidative and enzyme inhibitory properties.[17] Recent
findings revealed that affinity and selectivity for MAO-A and
MAO-B can be efficiently modulated by appropriate substitu-
tions on the coumarin moiety; the 3/4 and 6/7 positions are
particularly suitable to these modifications.[18–22] Our research
group has also reported high MAO-B selectivity with 6- or 8-
substituted-3-arylcoumarin scaffolds. A variety of functional
groups of various size, electronic, and lipophilic properties
were introduced in both aromatic rings, concluding that
a small substitution at C6 or C8 of the coumarin core is impor-
tant if a phenyl group is located at C3. Substituents on the 3-
phenyl group also play a crucial role in activity and selectivity:
meta and para substitutions are the most favorable for desired
activity.[23–28]
The benzofuran scaffold is ubiquitous in the area of pharma-
cologically active agents and isolated natural products. Natural
products bearing a 2-substituted moiety exhibit a broad range
of pharmacological activities such as antimicrobial,[29,30] anti-in-
flammatory,[31] antifungal,[32] antipsychotic,[33] antilipidemic,[34]
analgesic,[35] cytotoxic,[36] and central nervous system stimu-
lant.[37] The benzofuran nucleus has also been recently de-
scribed as a good scaffold for the design of potent and rela-
tively nonselective MAOIs.[38] Attending to the previously de-
scribed considerations, in the present work a series of 2-aryl-
benzofurans and 3-arylcoumarins, preserving the trans-stilbene
nucleus in their structure, were synthesized and their MAO ac-
tivity studied in order to compare experimental results
(Figure 1).
Scheme 1. Reagents and conditions: a) NaBH4, EtOH, 08C!RT, ~2 h;
b) PPh3·HBr, CH3CN, 828C, 2 h; c) toluene, NEt3, 1108C, 2 h; d) DCC, DMSO,
110 8C, 24 h; e) NaH, Ac2O, RT, 20 h.
benzofuran moiety was achieved by an intramolecular Wittig
reaction between ortho-hydroxybenzyltriphosphonium salts
IIa–IIg and the appropriate benzoyl chlorides.[39] The desired
Wittig reagent was readily prepared from the conveniently
substituted
ortho-hydroxybenzyl
alcohol
Ia–Ig
and
PPh3·HBr.[40–43] The benzofuran structures[44–49] were confirmed
1
by H NMR, 13C NMR, mass spectrometry, and elemental analy-
ses.
Coumarin derivatives 11–17 were synthesized according to
the protocol outlined in Scheme 1. Perkin condensation of vari-
ous commercially available ortho-hydroxybenzaldehydes with
a conveniently substituted arylacetic acid, using N,N’-dicyclo-
hexylcarbodiimide (DCC) as dehydrating agent, afforded 3-aryl-
coumarins 11–14, 16, and 17. In fact, these reaction conditions
have been found to be appropriate if the reagents contain
Results
Chemistry
Compounds 1–17 were efficiently synthesized according to the
synthetic strategy outlined in Scheme 1. The general reaction
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 0000, 00, 1 – 12
&
2
&
ÞÞ
These are not the final page numbers!

CHEMMEDCHEM
FULL PAPERS
www.chemmedchem.org
methyl, methoxy, ethoxy, or halogen groups in their struc-
tures.[23,28,50–53] However, nitro-substituted compound 15 was
obtained via modified Perkin conditions using a substituted 2-
hydroxy-5-nitrobenzaldehyde and 4-methoxyphenylacetic acid
in the presence of sodium hydride in acetic anhydride for 20 h.
1H NMR, 13C NMR, mass spectrometry, and elemental analyses
confirmed the coumarin structures.
Table 1. IC50 values and hMAO-B selectivity indexes for new compounds
and reference inhibitors on the activity of human recombinant MAO iso-
forms expressed in baculovirus-infected BTI insect cells.
Compd
IC50 [mm][a]
SI[c]
hMAO-A
hMAO-B
1
*
*
–
2
3
4
5
*
*
*
*
**
–
>46[c]
>76[c]
–
2.18ꢀ0.14
1.31ꢀ0.09
*
Pharmacology: inhibition of MAO
6
7
8
9
10
11
12
13
14
15
16
17
*
*
*
*
*
*
*
*
**
**
–
–
The potential effects of the newly synthesized compounds on
hMAO activity were investigated by measuring the production
of hydrogen peroxide from para-tyramine, using the Amplex
Red MAO assay kit (Molecular Probes Inc., Eugene, OR, USA)
and MAO isoforms in microsomes prepared from insect cells
(BTI-TN-5B1-4) infected with recombinant baculovirus contain-
ing cDNA inserts for hMAO-A or hMAO-B (Sigma–Aldrich Quꢂ-
mica S.A., Alcobendas, Spain). The inhibition of hMAO activity
was evaluated using the above method following the general
procedure previously described by us.[19] The test compounds
did not show any interference with the reagents used for the
biochemical assay. The control activity of hMAO-A and hMAO-
B using para-tyramine as the common substrate was 165ꢀ
2 pmol para-tyramine oxidized to para-hydroxyphenylacetalde-
hyde per min (n=20).
0.14ꢀ0.01
0.22ꢀ0.01
11.76ꢀ0.79
1.63ꢀ0.11
0.072ꢀ0.005
0.013ꢀ0.0009
0.006ꢀ0.0004
0.003ꢀ0.0002
1.06ꢀ0.07
0.208ꢀ0.014
0.017ꢀ0.0019
7.54ꢀ0.36
21.07ꢀ1.47
>714[c]
>455[c]
>8.5[c]
>61[c]
>1389[c]
>7692[c]
>16667[c]
390
*
1.17ꢀ0.08[b]
31.11ꢀ2.10[b]
*
29
>481[c]
selegiline
iproniazide
isatin
68.73ꢀ4.21[b]
6.56ꢀ0.76
**
4.043
0.87
>4.7[c]
[a] Each IC50 value is the mean ꢀSEM of five experiments (n=5). *Inactive
at 100 mm (highest concentration tested); the compounds precipitate at
higher concentrations. **Enzymatic activity inhibited by ~45–50% at
100 mm; the compounds precipitate at higher concentrations. [b] Statisti-
cal significance: p<0.01 versus the corresponding IC50 values obtained
against MAO-B, as determined by ANOVA/Dunnett’s test. [c] SI: hMAO-B
selectivity index=IC50(hMAO-A)/IC50(hMAO-B); values obtained under the as-
sumption that the corresponding IC50 value against MAO-A or MAO-B is
the highest concentration tested (100 mm).
The results of the hMAO-A and hMAO-B inhibition studies
with our compounds are listed in Table 1 together with the
MAO-B selectivity index. Enzyme assays revealed that most of
the test compounds were moderate to potent hMAO-B inhibi-
tors from low micromolar to nanomolar concentrations, show-
ing selectivity toward hMAO-B.
Reversibility experiments were performed to evaluate the
type of inhibition for derivatives 8, 14, and 15, the most active
compounds in the series (Table 2). An effective dilution
method was used,[54] and selegiline (irreversible inhibitor) and
isatin (reversible inhibitor) were taken are standards.[55]
Table 2. Reversibility results of hMAO-B inhibition by derivatives 8, 14,
and 15 and reference inhibitors.
Compd
Slope (AUF/t) [%][a]
8 (150 nm)
14 (5 nm)
15 (3 nm)
selegiline (20 nm)
isatin (33 nm)
39.4ꢀ2.7
68.9ꢀ4.60
12.6ꢀ0.8
8.5ꢀ0.60
67.8ꢀ7.5
Molecular docking studies with hMAO-B
Docking calculations were carried out to obtain detailed infor-
mation on the preferred orientations adopted by the two stud-
ied scaffolds in the hMAO-B binding pocket upon stable com-
plex formation. As the presence of structural water molecules
inside the binding pocket could modify the ligand–protein in-
teraction patterns,[56–59] the crucial role played by these mole-
cules inside the MAO-B catalytic site was also studied.
[a] Percentage values represent the mean ꢀSEM of five experiments (n=
5) relative to control; data show recovery of hMAO-B activity after dilu-
tion.
Protein targets
For this reason, we chose to perform different docking simu-
lations by changing the number of structural water molecules
retained inside the receptor binding site. The aim was to deter-
minate the differences and similarities between the two most
potent coumarin and benzofuran derivatives in their binding
pattern recognition, as well as to evaluate possible alternative
binding solutions adopted by the ligands inside the enzyme.
We created six alternative protein targets based on the crystal
structure of hMAO-B in complex with 4-carboxaldehyde-7-(3-
chlorobenzyloxy)coumarin (c17) used in our previous studies
(PDB ID: 2V60).[14,28,60,61]
During the protein setup, all water molecules further than 5 ꢃ
from the co-crystallized ligand were gradually deleted. In this
way nine water molecules were initially retained; those mole-
cules seem to be stabilized into the binding site by an exten-
sive hydrogen bond network involving, besides the water mol-
ecules, the FAD cofactor, co-crystallized coumarin, and receptor
residues. The receptor thus prepared was used as the first
target for docking calculations. For convenience we refer to
this target as MAO-B-9W.
The second target, called MAO-B-8W, was prepared by re-
moving the waterfront to FAD, in view of the high value
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 0000, 00, 1 – 12
&
3
&
ÞÞ
These are not the final page numbers!

CHEMMEDCHEM
FULL PAPERS
www.chemmedchem.org
(37.68) of the PDB B-factor, which is an indicator of thermal
motion about an atom. By deleting two other crystal water
molecules placed in the highly flexible beginning of the “en-
trance cavity” close to the outer surface region, MAO-B-6W
was created. The water located between Tyr326 and Ile199
(which separate the “entrance cavity” from the “substrate
cavity”, forming a bottleneck along the entire binding site) was
removed to form MAO-B-5W. By leaving only the water mole-
cule that establishes a hydrogen bond with the co-crystallized
ligand, and by removing all the structural waters inside the re-
ceptor, the last two targets, called MAO-B-1W and MAO-B-0W,
respectively, were prepared (see the Experimental Section for
water PDB identification numbers).
Table 3. RMSD values between the most stable poses of co-crystallized
coumarin c17 and MAO-B protein targets.[a]
Target
RMSD [ꢃ]
Binding modes 8
Binding modes 15
MAO-B-9W
MAO-B-8W
MAO-B-6W
MAO-B-5W
MAO-B-1W
MAO-B-0W
2.50
0.38
0.42
0.42
0.51
0.73
8a
15a
15a, 15b
15a
15b
15b
8a, 8b
8b, 8a
8b
8b
8b, 8a
15a, 15b
[a] Number of ligand binding modes retrieved for benzofuran 8 and cou-
marin 15 into the six different MAO-B protein targets.
results in an RMSD value as high as 2.50 ꢃ. Figure 2 shows the
MAO-B binding site with the structural water molecules used
to create our targets.
Docking protocol
Based on the good results obtained by our group and pub-
lished recently,[28,61] we decided to adopt the Quantum Me-
chanical Polarized Ligand Docking (QPLD), implemented in
Schrçdinger suite software, as the protocol for the simula-
tions.[62]
By applying ab initio methodologies, an accurate treatment
of the ligand charges within the protein environment was
made possible by means of density functional theory (DFT) cal-
culations. In this way the polarization effect of the ligand
charges by the receptor was taken into account. A post-dock-
ing process for the QPLD output complexes, scored by
Emodel, was also operated using the Prime MM-GBSA tool.[63]
Each ligand–protein complex, with the full quantum me-
chanical (QM)-treated ligand charges, was minimized including
the flexibility of the binding pocket defined by a distance of
5 ꢃ from the ligand. Furthermore, the free energy of binding,
calculated including ligand and receptor strain energies, was
estimated by Prime with Equation (1):
DGbind ¼ EDGbind ¼ EcomplexꢁðEligand þ Ereceptor
Þ
ð1Þ
Figure 2. Comparison of co-crystallized coumarin c17 (carbon atoms in grey)
in the hMAO-B binding pocket (PDB ID: 2V60) with the most stable binding
poses of c17 retrieved for the six targets after docking calculations. The
binding receptor surface is displayed in grey mesh style. The c17 carbon
ligand atoms are colored according to target: yellow for MAO-B-9W, purple
for MAO-B-8W, white for MAO-B-6W, turquoise for MAO-B-5W, blue for MAO-
B-1W, and orange for MAO-B-0W. Water molecules retained inside the vari-
ous targets are labeled as the PDB identification numbers (see Experimental
Section for details) and are represented as colored spheres. The hydrogen
bond between co-crystallized c17 and water molecule 1193 is displayed as
a red dotted line. Water molecule 1401 shows the highest PDB B-factor
value (37.68). The FAD cofactor is depicted in ball-and-stick format (carbon
atoms in grey).
in which DGbind is the free energy of binding, Ecomplex is the
energy of the minimized complex, Eligand is the unbounded
minimized ligand energy, and Ereceptor is the energy derived
from the minimized receptor. All energies are expressed in kcal
molꢁ1. Each ligand–protein complex was rescored according to
the DGbind values.
Docking protocol validation
To assess whether the number of retained structural water
molecules influences the effectiveness of our docking protocol,
we re-docked the co-crystallized coumarin c17 inside the bind-
ing pocket of the different targets and calculated the root-
mean-square deviation (RMSD) of heavy atoms. As listed in
Table 3, the results obtained for receptors from eight to zero
water molecules can be considered practically equivalent, with
the best RMSD observed for MAO-B-8W, with a value of 0.38 ꢃ.
Different behavior was observed for MAO-B-9W, in which the
presence of the water molecules with a high PDB B-factor
placed in front of FAD turns c17 in an uncorrected pose, which
We focused our attention on the most active benzofuran 8
and coumarin 15 analogues, which showed MAO-B inhibitory
activity (IC50) values of 140 and 3 nm, respectively. Both struc-
tures are substituted at the para position of the phenyl ring
with a methoxy group, while a nitro group occupies position 5
of benzofuran and position 6 of the coumarin moieties. There-
fore, both structures present the same substituents in equiva-
lent positions. As expected, prediction of binding recognition,
as well as the possibility to obtain an alternative pose for li-
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 0000, 00, 1 – 12
&
4
&
ÞÞ
These are not the final page numbers!

CHEMMEDCHEM
FULL PAPERS
www.chemmedchem.org
gands 8 and 15, depends heavily on the diverse protein tar-
gets (Table 3).
the FAD cofactor. Cys172 contributes to the stabilization of
complexes, acting as a hydrogen bond donor with the carbon-
yl oxygen atom. Moreover, in MAO-B-9W, Tyr326 is involved in
an edge-to-face p-stacking interaction with 3-phenylcoumarin.
Notably, in MAO-B-6W the absence of the two water molecules
close to Tyr326 does not allow establishment of p-stacking in-
teractions, due to the slight shift of the residue. In MAO-B-9W
an additional hydrogen bond involving the water molecule
below FAD and the nitro group of 15 was observed. Further-
more, the highly hydrophobic environment of the “entrance
cavity” contributes to the stabilization of the ligands, as the
methoxy of the para-methoxy-3-phenyl group undergoes
stable hydrophobic interactions with Leu164, Trp119, Pro104,
and Ile316. Tyr326, Ile199, Phe168, and Leu171 also contribute
significantly to hydrophobic stabilization of the 3-aryl ring (Fig-
ure 3A).
Binding mode analysis for coumarin 15
As listed in Table 3, a unique pattern recognition is recovered
for compound 15 in MAO-B-9W, -6W, -5W, and -1W. However,
analysis of the binding modes revealed that coumarin 15 can
bind targets in two alternative modes—15a and 15b—as
shown in Figure 3. In 15a, the pose adopted for MAO-B-9W
and -6W, the para-methoxy-3-phenyl ring occupies the “en-
trance cavity”, while the coumarin ring is well accommodated
by the catalytic cleft, leaving the 6-nitro substituents facing
A different behavior (15b) is observed in MAO-B-5W and
-1W: the para-methoxy-3-phenyl ring is directed toward FAD,
with the lactone moiety turned by ~1808. Due to the absence
of waters in the “entrance cavity”, the oxygen atom of the car-
bonyl group interacts with both Tyr326 and Gln206 by hydro-
gen bonds, providing stability to the complex (Figure 3B).
Regarding MAO-B-8W and -0W binding modes, our findings
revealed that both 15a and 15b patterns are present. In the
15b pattern for MAO-B-0W, a weak hydrogen bond between
the methoxy moiety of the coumarin and Tyr435 was ob-
served, whereas for MAO-B-8W, the hydroxy group of Tyr326
directed toward the water molecule of the “entrance cavity”
acts as hydrogen bond donor at a distance of 1.76 ꢃ, thereby
preventing a hydrogen bond interaction with the coumarin de-
rivative (Figure 3B).
With an aim to identify the most energetically favored bind-
ing pose, we compared the DGbind values retrieved from Prime
calculations. As listed in Table 4, in both targets 15a proved to
be the most favorable binding mode, with DDGbind values of
5.73 and 7.44 kcalmolꢁ1 in MAO-B-8W and MAO-B-0W, respec-
tively.
Table 4. Prime binding free energy differences between binding modes
15a and 15b in MAO-B-8W and MAO-B-0W.[a]
Target
DGbind 15a
DGbind 15b
DDGbind
MAO-B-8W
MAO-B-0W
ꢁ80.24
ꢁ75.91
ꢁ74.51
ꢁ68.47
5.73
7.44
[a] All values expressed in kcalmolꢁ1
.
Figure 3. A) Comparison of 15a binding modes for the 3-phenylcoumarin
into MAO-B-9W, MAO-B-8W, MAO-B-6W, and MAO-B-0W. Ligands (in tube
format), interacting residues (wire), and water molecules (spheres) have
carbon atoms colored by the same scheme: green for MAO-B-9W, purple for
MAO-B-8W, yellow for MAO-B-6W, and orange for MAO-B-0W. The FAD cofac-
tor is depicted in ball-and-stick format; hydrogen bonds are displayed as
yellow dotted lines. B) Comparison of 15b binding modes for the 3-phenyl-
coumarin into MAO-B-8W, MAO-B-5W, MAO-B-1W, and MAO-B-0W. Ligands
(in tube format), interacting residues (wire), and water molecules (spheres)
have carbon atoms colored by the same scheme: purple for MAO-B-8W, tur-
quoise for MAO-B-5W, blue for MAO-B-1W, and orange for MAO-B-0W. The
FAD cofactor is depicted in ball-and-stick format; hydrogen bonds are dis-
played as yellow dotted lines.
Binding mode analysis for benzofuran 8
Slightly different binding features were revealed in analysis of
the results obtained for the 2-phenylbenzofuran derivative rel-
ative to the 3-phenylcoumarin analogue (Figure 4). In MAO-B-
9W only one pose (8a) was recovered after docking calcula-
tions; the nitro benzofuran moiety of 8 binds the “entrance
cavity” of the receptor with the furan oxygen atom directed
toward Tyr326, while the para-methoxy-2-phenyl portion of
the molecule is stably linked to the substrate cleft. On one
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 0000, 00, 1 – 12
&
5
&
ÞÞ
These are not the final page numbers!

CHEMMEDCHEM
FULL PAPERS
www.chemmedchem.org
interactions with protein residues described for the 15a bind-
ing mode. Due to the increased distance from the furan
oxygen atom (3.8 ꢃ), no hydrogen bond with Cys172 was ob-
served. Tyr326 still provides favorable edge-to-face p-stacking
interaction energies with the benzofouran ring.
The presence of the above-described distinct 8a and 8b
binding poses was revealed when benzofuran 8 was docked
into MAO-B-8W, -6W, and -0W receptor targets. By the analysis
of DGbind we were able to assess which of the 8a and 8b bind-
ing modes was the most favored for compound 8. Table 5
shows that 8a seems to be the preferred pose (as well as in
Table 5. Prime binding free energy differences between binding modes
8a and 8b in MAO-B-8W, -6W, and -0W.[a]
Target
DGbind 8a
DGbind 8b
DDGbind
MAO-B-8W
MAO-B-6W
MAO-B-0W
ꢁ77.66
ꢁ71.49
ꢁ64.87
ꢁ73.76
ꢁ72.77
ꢁ71
3.9
1.28
6.13
[a] All values expressed in kcalmolꢁ1
.
MAO-B-W9, where no alternative poses were retrieved; see
Table 3) only in MAO-B-8W, for which a DDGbind value of 3.9
kcalmolꢁ1 was obtained. Indeed, with loss of the two water
molecules facing Tyr326, 8b becomes the most favored pose,
with a DDGbind value of 1.28 kalmolꢁ1, likewise in -0W, in which
8b was shown to be the most stable pattern, with DDGbind
6.13 kalmolꢁ1.
=
As consequence we can assume that with the progressive
decrease in structural water molecules, 8a decreases its affinity
for the receptor. Moreover, the dependence of docked 2-phe-
nylbenzofuran on the number of retained water molecules can
also be appreciated by alignment of the six targets. In Fig-
ure 4A, a certain target-dependent shift of the para-methoxy-
2-phenyl ring can be observed.
Figure 4. A) Comparison of 8a binding modes for the 2-phenylbenzofuran
into MAO-B-9W, MAO-B-8W, MAO-B-6W, and MAO-B-0W. Ligands (in tube
format), interacting residues (wire), and water molecules (spheres) have
carbon atoms colored by the same scheme: green for MAO-B-9W, purple for
MAO-B-8W, yellow for MAO-B-6W, and orange for MAO-B-0W. The FAD cofac-
tor is depicted in ball-and-stick format; the hydrogen bond is displayed as
a yellow dotted line. B) Comparison of 8b binding modes for the 2-phenyl-
benzofuran into MAO-B-8W, MAO-B-6W, MAO-B-5W, MAO-B-1W, and MAO-B-
0W. Ligands (in tube format), interacting residues (wire), and water mole-
cules (spheres) have carbon atoms colored by the same scheme: purple for
MAO-B-8W, yellow for MAO-B-6W, turquoise for MAO-B-5W, blue for MAO-B-
1W, and orange for MAO-B-0W. The FAD cofactor is depicted in ball-and-
stick format.
Comparative docking: hMAO-B versus hMAO-A
We docked compound 14 into hMAO-B and hMAO-A to further
study the different behavior against both isoenzymes. The
binding site in the hMAO-B labeled as 8W that presents the
lower RMSD values for the co-crystallized ligand (RMSD=0.38)
was used in the docking protocol. The binding poses for com-
pound 14 in hMAO-B are in agreement with the previous bind-
ing modes described for compounds 8 and 15. The best pose
according to Prime binding free energy score was considered
representative of the calculation (Figure 5A). We also docked
compound 14 into the hMAO-A crystal structure (PDB ID:
2Z5X) using two protocols: 1) retaining water molecules within
5 ꢃ from the co-crystallized ligand and 2) with no water mole-
cules in the protein pocket. In the first protocol with water
molecules in the pocket, compound 14 occupied an area simi-
lar to that of the co-crystallized ligand harmine, but the bind-
ing mode shown by compound 14 did not yield hydrogen
bond contacts with water molecules or the protein (see Fig-
ure 5B). This fact could explain the lack of hMAO-A activity.
hand, the 3-phenyl ring forms favorable p–p stacking interac-
tions with Tyr326, while on the other, the water molecule with
a high B-factor (placed in front of FAD) interacts with the me-
thoxy oxygen atom via a hydrogen bond at a distance of
1.8 ꢃ.
Also in MAO-B-5W and -1W, no alternative poses were iden-
tified for compound 8. Although at variance with the recogni-
tion pattern observed for MAO-B-9W, the benzofuran analogue
binds the targets with an inverted configuration (pose 8b). No-
tably, the para-methoxy-2-phenyl ring, situated into the “en-
trance cavity”, is perfectly superposed with that of coumarin
analogue found in 15a, also retaining favorable hydrophobic
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 0000, 00, 1 – 12
&
6
&
ÞÞ
These are not the final page numbers!

CHEMMEDCHEM
FULL PAPERS
www.chemmedchem.org
Discussion
2-Arylbenzofurans 1–10 were ef-
ficiently synthesized by reaction
of ortho-hydroxybenzyltriphos-
phonium salts and variously sub-
stituted benzoyl chlorides. Like-
wise, the corresponding 3-aryl-
coumarins 11–17 were efficiently
synthesized by Perkin condensa-
tion of various commercially
available ortho-hydroxybenzalde-
hydes with a conveniently sub-
stituted arylacetic acid.
Substituents and their posi-
tions in the 2-arylbenzofuran or
3-arylcoumarin nucleus were se-
lected based on previous results,
which had shown very high
MAOI activity for some deriva-
tives. Although substitution at
position 5 of the 2-arylbenzofur-
an lead to inactive compounds,
there is an exception for com-
pound 8 (5-nitro-2-(4-methoxy-
phenyl)benzofuran), which re-
sulted in the most active com-
pound of the benzofuran series.
Corresponding 6-nitro-3-(4-me-
Figure 5. Comparative docking results in hMAO-B and hMAO-A: A) Representative binding mode for compound
14 (green carbon atoms) into MAO-B-8W. The hydrogen bond with Cys172 is displayed as a yellow dotted line.
B) Comparison of the binding modes for compounds 14 (pink carbons) and the co-crystallized harmine (green car-
bons) in hMAO-A, retaining water molecules within 5 ꢃ from the co-crystallized ligand in the docking calculation.
C) Binding mode for compound 14 (pink carbons) in hMAO-A with no water in the cavity. To adopt the mentioned
pose, compound 14 would have to displace a water molecule present in the crystal structure, represented in red
(water molecules were not considered in the docking calculation). D) Pose retrieved for compound 15 (grey car-
bons) in hMAO-A with no water in the pocket. The hydrogen bond between the ligand and residue Tyr197 is dis-
played as a yellow dotted line.
thoxyphenyl)coumarin
(com-
pound 15) also showed interest-
ing MAO-B inhibitory activity in
However, this is not the case for the co-crystallized harmine,
the conformation of which is stabilized by two hydrogen
bonds with an important contribution to complex formation.
Following the second protocol with no water molecules in the
hMAO-A cavity, the poses obtained in hMAO-A for compound
14 showed a preference for the ligand to be placed in
a deeper region (see Figure 5C). The displacement or reorgani-
zation of some water molecules by compound 14 would be
a necessary step to bind to the protein using the described
pose. However, this step could be energetically unfavorable for
the ligand binding potential, representing an important obsta-
cle for the ligand–protein interaction. To further study this hy-
pothesis, we docked compound 15 into the enzyme, as this
compound showed some hMAO-A activity. The results are in
accord with the poses determined for compound 14. However,
in the case of compound 15, the nitro group at position R1
could establish similar hydrogen bond network interactions as
those of the water molecule (HOH739) in the pocket showing
a hydrogen bond with the Tyr197 residue (Figure 5D). In this
case, polar rather than hydrophobic substituents seem to inter-
act strongly, simulating the interactions between water and
the protein.
the low nanomolar range, and similar activity was shown by
compound 14. Replacement of nitro by bromo substituents
offers advantages from a toxicological point of view. Reversibil-
ity of the inhibition shown by compound 14 is similar to that
described for isatin. Compound 8 was also shown to be a rever-
sible. However, data obtained for compound 15 showed that
its degree of reversibility is much lower than that of the
above-mentioned compounds.
Substitution at position 7 of the benzofuran nucleus or at
position 8 of the coumarin ring gave derivatives with an inter-
esting profile as MAO inhibitors. Comparing both series, the 3-
arylcoumarin derivatives showed better inhibitory profiles than
2-arylbenzofuran derivatives in almost all cases. An exception
was verified for compounds 9 and 16. Both present the same
substitution pattern, and the benzofuran derivative proved to
be a better MAOI-B than the corresponding coumarin deriva-
tive (IC50 =220 nm and 1 mm, respectively). As had been ob-
served for 3-arylcoumarin derivatives, substitutions at the meta
and para positions on the phenyl ring led to the most potent
derivatives, also in the case of 2-arylbenzofurans.
It is important to highlight the activity profile of compound
14. It presented MAO-B activity similar to that of compound
15, yet is more selective. In fact, compound 14 is threefold
more active than selegiline, and is more than fourfold more se-
lective than this reference compound.
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 0000, 00, 1 – 12
&
7
&
ÞÞ
These are not the final page numbers!

CHEMMEDCHEM
FULL PAPERS
www.chemmedchem.org
CDCl3 as solvent. Chemical shifts (d) are expressed in parts per mil-
lion (ppm) using TMS as an internal standard. Coupling constants J
are expressed in hertz (Hz). Spin multiplicities are given as s (sin-
glet), d (doublet), dd (doublet of doublets), and m (multiplet). Mass
spectrometry was carried out with a Kratos MS-50 or a Varian MAT-
711 spectrometer. Elemental analyses were performed by using
a PerkinElmer 240B microanalyzer and are within ꢀ0.4% of calcu-
lated values in all cases. The analytical results indicate ꢂ98%
purity for all compounds. Flash chromatography (FC) was per-
formed on silica gel (Merck 60, 230–400 mesh); analytical TLC was
performed on pre-coated silica gel plates (Merck 60 F254). Organic
solutions were dried over anhydrous Na2SO4. Concentration and
evaporation of the solvent after reaction or extraction was carried
out on a rotary evaporator (Bꢄchi Rotavapor) operating under re-
duced pressure.
Depending on the target used to perform docking, two al-
ternative MAO-B binding recognition patterns were observed
for 3-arylcoumarins and 2-arylbenzofurans. Based on the analy-
sis of the free energy of binding, as well as on the study of
enzyme–inhibitor interactions, 15a and 8b seem to be the
most energetically favored poses. As expected, the selection of
structural water molecules that must be considered in creating
the docking targets is not a trivial process, particularly if the
purpose for docking is the rationalization of the pharmacologi-
cal assays of these compound classes. We also developed
a comparative study between hMAO-A and hMAO-B, taking
into account compounds 14 and 15. Both isoenzymes present
some differences in some residues in the pocket that affect the
selectivity in binding of the studied compounds. Docking bind-
ing modes are in agreement with previous results recently
published.[64] However, other possible binding modes have also
been reported for benzofurans and coumarins.[65,66] As an ex-
ample, Bonivento et al.[65] reported a MAO-B protein structure
with a benzofuranyl–imidazoline derivative placed in the en-
trance cavity. The different nature of some substituents in the
benzofuran and coumarin nucleus could strongly influence the
orientation of the ligands in the protein interaction binding
process. In our opinion, the QPLD-Prime protocol was well
suited for the study of MAO systems, taking into consideration
the helpful contribution from the QM calculations operated by
QPLD and from the rescoring stage applied using Prime.
Synthesis
General procedure for the preparation of 2-hydroxybenzylalco-
hols: NaBH4 (6.60 mmol) was added to a stirring solution of 2-hy-
droxybenzaldehyde (6.60 mmol) in EtOH (20 mL) in an ice bath.
The reaction mixture was stirred at room temperature for 1 h. After
that, the solvent was removed, 1n aqueous HCl (40 mL) was
added to the residue and extracted with Et2O. The solvent was
evaporated under vacuum to give the desired compounds Ia–
Ig.[67,68]
General procedure for the preparation of 2-hydroxybenzyltri-
phenylphosphonium bromide: A mixture of 2-hydroxybenzyl alco-
hol (24.6 mmol) and PPh3·HBr (24.6 mmol) in CH3CN (50 mL) was
stirred under reflux for 2 h. The solid that formed was filtered and
washed with CH3CN to give the desired compounds IIa–IIg.[40–43]
Conclusions
5-Methyl-2-hydroxyphenylphosphonium bromide (IIc): White
solid. Yield: 79%; mp: 217–2188C; 1H NMR (300 MHz, CDCl3): d=
3.77 (s, 3H, CH3), 4.80 (d, J=14.1 Hz, 2H, CH2), 6.35 (s, 1H, H-6),
6.52 (d, J=1.5 Hz, 1H, H-4), 6.89 (d, J=1.5 Hz, 1H, H-3), 7.74 (bs,
1H, OH), 7.82–7.90 (m, 12H, ArH), 7.95–7.98 ppm (m, 3H, ArH);
13C NMR (75 MHz, CDCl3): d=23.24 (CH3) 34.90 (CH2), 114.72, 120.10
(3C), 123.84, 126.62, 127.30, 130.89 (6C), 133.43, 134.15 (6C), 135.50
(3C), 154.52 ppm.
A general and efficient synthesis of a new series of 2-arylben-
zofurans and 3-arylcoumarins was developed in this study. De-
termination of hMAO isoform activity was carried out, and the
majority of the compounds exhibited selectivity for the hMAO-
B isoenzyme with high affinity in the micro- and nanomolar
concentration ranges. Compounds 8 and 15 proved to be the
most active of both series. Compound 14, in which the nitro
substituent is replaced by a bromine atom, also exhibited in-
teresting activity and advantages from a selectivity and reversi-
bility point of view. In addition, compounds 8 and 14 were
shown to be reversible inhibitors. 3-Arylcoumarins are general-
ly more active than 2-arylbenzofurans. Molecular docking stud-
ies were performed to establish the nature of the interaction
between the studied compounds and hMAO-B/A, leading to
a rationalization of the structure–activity relationship for the
synthesized series. The results encouraged us to further ex-
plore the potential of these chemical families as drug candi-
dates for the treatment of PD.
3-Ethoxy-2-hydroxyphenylphosphonium bromide (IIg): White
solid. Yield: 97%; mp: 156–1578C; 1H NMR (300 MHz, CDCl3): d=
1.52 (t, J=5.6 Hz, 3H, CH2CH3), 4.35 (q, J=5.6 Hz, 2H, CH2CH3), 4.8
(d, J=14.1 Hz, 2H, CH2), 6.68 (d, J=7.7 Hz, 1H, H-6), 6.0 (bs, 1H,
OH), 6.70 (d, J=7.7 Hz, 1H, H-4), 7.15 (t, J=7.7 Hz, 1H, H-5), 7.58–
7.78 (m, 12H, ArH), 7.82–7.92 ppm (m, 3H, ArH); 13C NMR (75 MHz,
CDCl3): d=15.2 (CH2CH3), 34.5 (CH2), 62.5 (CH2CH3), 114.0, 119.3,
120.0 (3C), 126.8, 131.0 (6C), 132.1, 133.5 (6C), 134.0 (3C), 145.0,
149.3 ppm.
General procedure for the preparation of 2-phenylbenzofuran:
A
mixture of 2-hydroxybenzyltriphenylphosphonium bromide
(1.10 mmol) and benzoyl chloride (1.11 mmol) in a mixed solvent
(toluene 20 mL and Et3N 0.5 mL) was stirred under reflux for 2 h.
The precipitate was removed by filtration. The filtrate was concen-
trated, and the residue was purified by silica gel chromatography
(hexane/EtOAc 9:1) to give the desired compounds 1–10.[44–49]
Experimental Section
General methods
Starting materials and reagents were obtained from commercial
suppliers (Sigma–Aldrich) and were used without further purifica-
tion. Melting points (mp) are uncorrected and were determined
with a Reichert Kofler thermopan or in capillary tubes in a Bꢄchi
7-Ethoxy-2-(4’-methoxyphenyl)benzofuran (10): White solid.
Yield: 54%; mp: 55–568C; H NMR (300 MHz, CDCl3): d=1.76 (t, J=
7.0 Hz, 3H, CH3), 3.87 (s, 3H, OCH3), 4.30 (q, J=7.0 Hz, 2H, CH2),
6.54 (d, J=6.9 Hz, 1H, H-6), 6.60 (s, 1H, H-3), 6.70 (d, J=8.9 Hz, 2H,
H-3’, H-5’), 7.11 (m, 2H, H-4, H-5), 7.81 ppm (d, J=8.9 Hz, 2H, H-2’,
H-6’); 13C NMR (75 MHz, CDCl3): d=14.9, 55.2, 64.4, 99.9, 107.6,
1
1
510 apparatus. H NMR (300 MHz) and 13C NMR (75.4 MHz) spectra
were recorded with a Bruker AMX spectrometer using [D6]DMSO or
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 0000, 00, 1 – 12
&
8
&
ÞÞ
These are not the final page numbers!

CHEMMEDCHEM
FULL PAPERS
www.chemmedchem.org
112.9, 114.0 (2C), 123.1, 123.4, 126.4 (2C), 131.2, 144.0, 144.3, 156.0,
159.8 ppm; MS (EI, 70 eV): m/z (%): 268 (30) [M]+, 240 (11), 225
(14); Anal. calcd for C17H16O3: C 76.10, H 6.01, found: C 76.15, H
6.10.
microplate fluorescence reader (FLX800, BioTek Instruments Inc.,
Winooski, VT, USA) based on the fluorescence generated
(lex 545 nm, lem 590 nm) over a 15 min period, in which the fluo-
rescence increased linearly.
General procedure for the preparation of 3-phenylcoumarins
11–14, 16 and 17: To a solution of ortho-hydroxybenzaldehyde
(7.34 mmol) and the corresponding phenylacetic acid (9.18 mmol)
in DMSO (15 mL), N,N’-dicyclohexylcarbodiimide (11.46 mmol) was
added. The mixture was heated at 1108C for 24 h. Then, ice
(100 mL) and AcOH (10 mL) were added to the reaction mixture.
After keeping it at room temperature for 2 h, the mixture was ex-
tracted with Et2O (3ꢅ25 mL). The organic layers were combined
and washed with sodium bicarbonate solution (50 mL, 5%) and
water (20 mL). The solvent was then evaporated under vacuum,
and the dry residue was purified by flash chromatography
(hexane/EtOAc 9:1) to give the desired compounds 11–14, 16, and
17.[23,28,50–53]
Control experiments were carried out simultaneously by replacing
the test drugs (new compounds and reference inhibitors) with ap-
propriate dilutions of the vehicles. In addition, the potential capaci-
ty of the above test drugs to modify the fluorescence generated in
the reaction mixture due to non-enzymatic inhibition (e.g., for di-
rectly reacting with Amplex Red reagent) was determined by
adding these drugs to solutions containing only the Amplex Red
reagent in a sodium phosphate buffer.
The specific fluorescence emission (used to obtain the final results)
was calculated after subtraction of the background activity, which
was determined from vials containing all components except the
hMAO isoforms, which were replaced by a sodium phosphate
buffer solution.
General procedure for the preparation of 3-(4’-methoxyphenyl)-
6-nitrocoumarin (15): To a dry 100 mL round-bottomed flask, the
Determination of inhibition mode
conveniently
substituted
2-hydroxy-5-nitrobenzaldehyde
(42.5 mmol), 4-methoxyphenylacetic acid (42.5 mmol) and acetic
anhydride (40.1 mL, 0.43 mol) were added. Then, NaH (60% disper-
sion in mineral oil) (42.5 mmol) was added in small aliquots. After
the dissolution of the reagents, an spontaneous (after 2–5 min)
precipitation process was observed. The reaction mixture was
stirred for 20 h and then water (7 mL) was added. Following the
addition of 43 mL AcOH, the mixture was cooled to 48C for 4 h.
The resulting precipitate was filtered and washed with cold glacial
AcOH. The AcOH was then removed as an azeotrope upon addi-
tion of 250 mL toluene and rotary evaporation to dryness. The pro-
cess was repeated three times. The final residue was dried under
vacuum and purified by FC (hexane/EtOAc 9:1) to give compound
To evaluate whether compounds 8, 14, and 15 are reversible or ir-
reversible hMAO-B inhibitors, a dilution method was used.[54]
A
100ꢅ concentration of the enzyme used in the above described
experiments was incubated with a concentration of inhibitor equiv-
alent to 10-fold its IC50 value. After 30 min, the mixture was diluted
100-fold into reaction buffer containing Amplex Red reagent,
horseradish peroxidase, and p-tyramine, and the reaction was
monitored for 15 min. Reversible inhibitors show linear progress
with a slope equal to ~91% of the slope of the control sample,
whereas irreversible inhibition reaches only ~9% of this slope.
Control tests were carried out by pre-incubating and diluting the
enzyme in the absence of inhibitor.
1
15. Pale-yellow solid. Yield: 65%; mp: 62–638C; H NMR (300 MHz,
CDCl3): d=3.80 (s, 3H, -OCH3), 7.05 (d, J=9.5 Hz, 2H, H-3’, H-5’),
7.62–7.69 (m, 3H, H-2’, H-6’, H-8), 8.36 (s, 1H, H-4), 8.73 ppm (d, J=
2.9 Hz, 2H, H-5, H-7); 13C NMR (75 MHz, CDCl3): d=55.9 (OCH3),
116.4 (2C), 119.7, 127.9, 128.6, 129.2, 132.2, 132.5 (2C), 134.3, 139.1,
139.4, 151.5, 155.8, 161.2 ppm; MS (EI, 70 eV): m/z (%): 298 (17)
[M+H]+, 297 (100) [M]+, 223 (30), 167 (48), 152 (15), 137 (11), 121
(31), 98 (14), 84 (11), 71 (12), 69 (19), 65 (19), 63 (10), 57 (31), 55
(16). Anal. calcd for C16H11NO5: C 64.65, H 3.73, found: C 64.67, H
3.76.
Molecular docking
Ligand docking: The Protein Preparation tool of Maestro 9.1[69]
was used to prepare the high-resolution crystal structure of hMAO-
B (PDB ID: 2V60) for docking calculations. According to the PDB
identification numbers, the following water molecules were re-
tained to create the six targets of receptor. MAO-B-9W: 1193, 1201,
1207, 1249, 1273, 1274, 1358, 1359, 1401; MAO-B-8W: 1193, 1201,
1207, 1249, 1273, 1274, 1358, 1359; MAO-B-6W: 1193, 1207, 1249,
1273, 1358, 1359; MAO-B-5W: 1193, 1207, 1249, 1358, 1359; MAO-
B-1W: 1193. In the targets refinement stage, the OPLS_2005 force
field was applied to minimize hydrogen atoms previously added,
allowing relaxation of the hydrogen bond network. Hydroxy
groups and amide groups of asparagine and glutamine residues
were reoriented. An optimization of protonation states of histidine
was also performed. The ligands were prepared by applying the
MMFFs force field.[70]
Determination of MAO isoforms in vitro activity
The effects of the new synthesized compounds on hMAO isoform
enzymatic activity were evaluated by a fluorimetric method follow-
ing the experimental protocol previously described by us.[19] Briefly,
0.1 mL of sodium phosphate buffer (0.05m, pH 7.4) containing the
test drugs (new compounds or reference inhibitors) in various con-
centrations and adequate amounts of recombinant hMAO-A or
hMAO-B required and adjusted to obtain in our experimental con-
ditions the same reaction velocity, [hMAO-A: 1.1 mg protein; specif-
ic activity: 150 nmol of p-tyramine oxidized to p-hydroxyphenylace-
taldehydeminꢁ1 (mg protein)ꢁ1; hMAO-B: 7.5 mg protein; specific
In the QPLD protocol, Glide Standard Precision level (SP) was used
to perform the initial docking, generating the initial charges with
semiempirical methods and retaining eight poses for each ligand.
The QM treatment of ligand charges were calculated by Jaguar
using the 6-31G*/LACVP* basis set and B3LYP density functional.
activity: 22 nmol of p-tyramine transformedminꢁ1 (mg protein)ꢁ1
]
were incubated for 15 min at 378C in a flat black-bottomed 96-
well microtest plate, placed in the dark fluorimeter chamber. After
this incubation period, the reaction was started by adding (final
concentrations) 200 mm Amplex Red reagent, 1 UmLꢁ1 horseradish
peroxidase and 1 mm p-tyramine. The production of H2O2 and, con-
sequently, of resorufin was quantified at 378C in a multidetection
Glide at Extra Precision level (XP) re-docks the ligands with im-
proved charges, returning the best four poses for each ligand. The
improved partial charges obtained from the QPLD were main-
tained during the Prime MM-GBSA rescoring stage instead of
OPLS_2005 force field (default setting). No constraints on the flexi-
ble residues were applied.
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 0000, 00, 1 – 12
&
9
&
ÞÞ
These are not the final page numbers!

CHEMMEDCHEM
FULL PAPERS
www.chemmedchem.org
[26] M. J. Matos, S. Vꢀzquez-Rodriguez, E. Uriarte, L. Santana, D. ViÇa, Bioorg.
Med. Chem. Lett. 2011, 21, 4224–4227.
[27] M. J. Matos, D. ViÇa, P. Janeiro, F. Borges, L. Santana, E. Uriarte, Bioorg.
Med. Chem. Lett. 2010, 20, 5157–5160.
[28] D. ViÇa, M. J. Matos, G. Ferino, E. Cadoni, R. Laguna, F. Borges, E. Uriarte,
L. Santana, ChemMedChem 2012, 7, 464–470.
[29] R. Basawaraj, Y. S. Agasimundin, Indian J. Heterocycl. Chem. 2002, 12, 1–
4.
Similar docking treatment was applied for the hMAO-A crystal
structure with the ligand harmine (PDB ID: 2Z5X). We created two
different targets of the hMAO-A receptor according to the water
molecules retained in the pocket: 1) only water molecules within
5 ꢃ from the co-crystallized ligand were retained, 2) no water mol-
ecules were retained.
[30] R. Basawaraj, G. Parameshwarappa, S. S. Sangapure, Indian Drugs 2007,
44, 8–12.
Acknowledgements
[31] L. Santana, M. Teijeira, E. Uriarte, C. Teran, B. LiÇares, R. Villares, R.
Laguna, E. Cano, Eur. J. Pharm. Sci. 1999, 7, 161–166.
[32] M. Masubuchi, H. Ebiike, K. Kawasaki, S. Sogabe, K. Morikami, Y. Shira-
tori, S. Tsujii, T. Fujii, K. Sakata, M. Hayase, H. Shindoh, Y. Aoki, T. Ohtsu-
ka, N. Shimma, Bioorg. Med. Chem. 2003, 11, 4463–4478.
[33] I. Van Wijingaarden, C. G. Kruse, J. A. M. Van der Heyden, M. T. M. Tulp, J.
Med. Chem. 1988, 31, 1934–1940.
[34] D. T. Witiak, H. N. Newman, G. K. Poochikian, W. Loh, S. Sankarappa,
Lipids 1976, 11, 384–391.
[35] D. B. A. Kumar, G. K. Prakash, B. P. Nandeshwarappa, B. S. Sherigara,
K. M. Mahadevan, Indian J. Pharm. Sci. 2006, 68, 809–814.
[36] W. J. Song, X. D. Yang, X. H. Zeng, X. L. Xu, G. L. Zhang, H. B. Zhang, RSC
Adv. 2012, 2, 4612–4615.
[37] D. Dauzonne, J. M. Gillardin, F. Lepage, R. Pointetl, S. Risk, G. Lamottel,
P. Demerseman, Eur. J. Med. Chem. 1995, 30, 53–59.
[38] L. H. A. Prins, J. P. Petzer, S. F. Malan, Eur. J. Med. Chem. 2010, 45, 4458–
4466.
[39] A. Hercouet, M. Le Corre, Tetrahedron Lett. 1979, 20, 2145–2148.
[40] M. Ono, H. Kawashima, A. Nonaka, T. Kaway, M. Haratake, H. Mori, M. P.
Kung, H. F. Kung, H. Saji, M. Nakayama, J. Med. Chem. 2006, 49, 2725–
2730.
[41] M. Ono, M. P. Kung, C. Hou, H. F. Kung, Nucl. Med. Biol. 2002, 29, 633–
642.
[42] J. Gubin, P. Chatelain, J. Lucchetti, G. Rosseels, H. Inion, US Patent
US5223510A1, 1993; [Chem. Abstr. 117:26336].
[43] K. Hagihara, H. Kashima, K. Lida, J. Enokizono, S. Uchida, H. Nonaka, M.
Kurokawa, J. Shimada, Bioorg. Med. Chem. Lett. 2007, 17, 1616–1621.
[44] G. W. Kabalka, L. Wang, R. M. Pagni, Tetrahedron 2001, 57, 8017–8028.
[45] W. Guiyong, Y. Weiyu, S. Hong, H. Yong, Green Chem. 2012, 14, 580–
585.
[46] N. Ortega, B. Beiring, S. Urban, F. Glorius, Tetrahedron 2012, 68, 5185–
5192.
[47] X. Jiang, W. Liu, W. Zhang, F. Jiang, Z. Gao, H. Zhuang, L. Fu, Eur. J. Med.
Chem. 2011, 46, 3526–3530.
[48] M. L. N. Rao, D. K. Awasthi, J. B. Talode, Tetrahedron Lett. 2012, 53,
2662–2666.
[49] L. M. Geary, P. G. Hultin, Eur. J. Org. Chem. 2010, 5563–5573.
[50] M. J. Matos, S. Vazquez-Rodriguez, F. Borges, L. Santana, E. Uriarte, Tetra-
hedron Lett. 2011, 52, 1225–1227.
We are grateful to Fondazione Banco di Sardegna, MIUR, and
Universitꢁ di Cagliari (National Project “Sintesi e stereocontrollo
di molecole organiche mediante metodologie innovative”) for
partial financial support. M.J.M. thanks the Fundażo para
a CiÞncia e Tecnologia for a grant (SFRH/BD/61262/2009). S.V.
thanks the Angeles AlvariÇo program from Xunta de Galicia
(Spain).
Keywords: 2-arylbenzofurans · 3-arylcoumarins · docking ·
inhibitors · monoamine oxidase · reversibility
[1] W. Dauer, S. Przedborski, Neuron 2003, 39, 889–909.
[2] D. A. Di Monte, L. E. De Lanney, I. Irwin, Brain Res. 1996, 738, 53–59.
[3] K. F. Tipton, Cell Biochem. Funct. 1986, 4, 79–87.
[4] A. W. J. Bach, N. C. Lan, D. L. Johnson, C. W. Abell, M. E. Bembenek, S. W.
Kwan, P. H. Seeburg, J. C. Shih, Proc. Natl. Acad. Sci. USA 1988, 85,
4934–4938.
[5] J. Wouters, Curr. Med. Chem. 1998, 5, 137–162.
[6] J. Grimsby, N. C. Lan, R. Neve, K. Chen, J. C. Shih, J. Neurochem. 1990,
55, 1166–1169.
[7] A. S. Kalgutkar, N. Castagnoli, Jr., Med. Res. Rev. 1995, 15, 325–388.
[8] R. M. Geha, I. Rebrin, K. Chen, J. C. Shih, J. Biol. Chem. 2001, 276, 9877–
9882.
[9] J. Ma, M. Yoshimura, E. Yamashita, A. Nakagawa, A. Ito, T. Tsukihara, J.
Mol. Biol. 2004, 338, 103–114.
[10] M. V. Rudorfer, V. Z. Potter, Drugs 1989, 37, 713–718.
[11] H. H. Fernandez, J. J. Chen, Clin. Neuropharmacol. 2007, 30, 150–168.
[12] H. Zheng, M. Fridkin, M. B. H. Youdim, Neurochem. Res. 2010, 35, 2117–
2123.
[13] K. F. Tipton, S. Boyce, J. O’Sullivan, G. P. Davey, J. Healy, Curr. Med. Chem.
2004, 11, 1965–1982.
[14] C. Binda, P. Newton-Vinson, F. Hubalek, D. E. Edmondson, A. Mattevi,
Nat. Struct. Mol. Biol. 2002, 9, 22–26.
[15] C. Binda, M. Li, F. Hubelek, N. Restelli, D. E. Edmondson, A. Mattevi, Proc.
Natl. Acad. Sci. USA 2003, 100, 9750–9755.
[16] L. De Colibus, M. Li, C. Binda, A. Lustig, D. E. Edmondson, A. Mattevi,
Proc. Natl. Acad. Sci. USA 2005, 102, 12684–12689.
[51] S. Vilar, E. Quezada, C. Alcaide, F. Orallo, L. Santana, E. Uriarte, QSAR
Comb. Sci. 2007, 26, 317–332.
[52] E. Quezada, G. Delogu, C. Picciau, L. Santana, G. Podda, F. Borges, V.
Garcꢂa-Morales, D. ViÇa, F. Orallo, Molecules 2010, 15, 270–279.
[53] M. S. Phansalkar, K. K. Deshmukh, S. L. Kelkar, M. S. Wadia, Indian J.
Chem. B 1987, 26, 562–563.
[54] R. A. Copeland, Evaluation of Enzyme Inhibitors in Drug Discovery, Wiley
Interscience, Hoboken, 2005.
[17] F. Borges, F. Roleira, N. Milhazes, E. Uriarte, L. Santana, Front. Med. Chem.
2009, 4, 23–85.
[18] L. Santana, E. Uriarte, H. Gonzꢀlez-Dꢂaz, G. Zagotto, R. Soto-Otero, E.
Mꢆndez-Alvarez, J. Med. Chem. 2006, 49, 1118–1124.
[19] L. Santana, H. Gonzꢀlez-Dꢂaz, E. Quezada, E. Uriarte, M. YꢀÇez, D. ViÇa, F.
Orallo, J. Med. Chem. 2008, 51, 6740–6751.
[55] M. Gerlach, P. Riederer, M. B. Youdim, Eur. J. Pharmacol. 1992, 226, 97–
108.
[56] S. B. de Beer, N. P. Vermeulen, C. Oostenbrink, Curr. Top. Med. Chem.
2010, 10, 55–66.
[20] M. Catto, O. Nicolotti, F. Leonetti, A. Carotti, A. Favia, R. Soto-Otero, E.
Mꢆndez-Alvarez, A. Carotti, J. Med. Chem. 2006, 49, 4912–4925.
[21] C. Gnerre, M. Catto, F. Leonetti, P. Weber, P. Carrupt, C. Altomare, A. Car-
otti, B. Testa, J. Med. Chem. 2000, 43, 4747–4758.
[57] Z. Li, T. Lazaridis, Phys. Chem. Chem. Phys. 2007, 9, 573–581.
[58] R. L. Mancera, Curr. Opin. Drug Discovery Dev. 2007, 10, 275–280.
[59] O. Villacanas, S. Madurga, E. Giralt, I. Belda, Curr. Comp. Aided Drug Des.
2009, 5, 145154.
[60] G. Delogu, C. Picciau, G. Ferino, E. Quezada, G. Podda, E. Uriarte, D.
ViÇa, Eur. J. Med. Chem. 2011, 46, 1147–1152.
[61] S. Serra, G. Ferino, M. J. Matos, S. Vazquez-Rodriguez, G. Delogu, D.
ViÇa, E. Cadoni, L. Santana, E. Uriarte, Bioorg. Med. Chem. Lett. 2012, 22,
258–261.
[22] F. Chimenti, D. Secci, A. Bolasco, P. Chimenti, A. Granese, O. Befani, P.
Turini, S. Alcaro, F. Ortuso, Bioorg. Med. Chem. Lett. 2004, 14, 3697–
3703.
[23] M. J. Matos, D. ViÇa, E. Quezada, C. Picciau, G. Delogu, F. Orallo, L. San-
tana, E. Uriarte, Bioorg. Med. Chem. Lett. 2009, 19, 3268–3270.
[24] M. J. Matos, D. ViÇa, C. Picciau, F. Orallo, L. Santana, E. Uriarte, Bioorg.
Med. Chem. Lett. 2009, 19, 5053–5055.
[25] M. J. Matos, C. Terꢀn, Y. Pꢆrez-Castillo, E. Uriarte, L. Santana, D. ViÇa, J.
Med. Chem. 2011, 54, 7127–7137.
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 0000, 00, 1 – 12
&10
&
ÞÞ
These are not the final page numbers!

CHEMMEDCHEM
FULL PAPERS
www.chemmedchem.org
[62] Schrçdinger Suite 2011: QM Polarized Ligand Docking protocol (QPLD);
Glide version 5.7, Jaguar version 7.8, and QSite version 5.7, Schrçdinger
LLC, New York, NY (USA), 2011.
[63] Prime version 3.0, Schrçdinger LLC, New York, NY (USA), 2011.
[64] M. J. Matos, S. Vilar, R. M. Gonzalez-Franco, E. Uriarte, L. Santana, C.
Friedman, N. P. Tatonetti, D. ViÇa, J. A. Fontenla, Eur. J. Med. Chem.
2013, 63, 151–161.
[68] P. Gradeff, S. Murayama, US Patent US4351962A 1982 09 28, 1982;
[Chem. Abstr. 97:215746].
[69] Schrçdinger Suite 2011: Protein Preparation Wizard; Epik version 2.2,
Impact version 5.7, Prime version 3.0, Schrçdinger LLC, New York, NY
(USA), 2011.
[70] LigPrep version 2.5, Schrçdinger LLC, New York, NY (USA), 2011.
[65] D. Bonivento, E. M. Milczek, G. R. McDonald, C. Binda, A. Holt, D. E. Ed-
mondson, A. Mattevi, J. Biol. Chem. 2010, 285, 36849–36856.
[66] O. M. Abdelhafez, K. M. Amin, M. M. Abdalla, R. Z. Batran, Neurochem.
Int. 2013, 62, 198–209.
Received: January 31, 2013
Revised: April 2, 2013
[67] C. Meier, E. De Clercq, J. Balzarini, Eur. J. Org. Chem. 1998, 837–846.
Published online on && &&, 0000
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 0000, 00, 1 – 12
&
11
&
ÞÞ
These are not the final page numbers!

FULL PAPERS
G. Ferino, E. Cadoni, M. J. Matos,*
E. Quezada, E. Uriarte, L. Santana,
S. Vilar, N. P. Tatonetti, M. YꢀÇez, D. ViÇa,
C. Picciau, S. Serra, G. Delogu*
A tale of two scaffolds: Herein we de-
scribe the synthesis, in vitro inhibition
of MAO-A and MAO-B, reversibility stud-
ies, and docking calculations for two
groups of 2-arylbenzofurans and 3-aryl-
coumarins. A comparative study of both
scaffolds was performed.
&& – &&
MAO Inhibitory Activity of 2-
Arylbenzofurans versus 3-
Arylcoumarins: Synthesis, in vitro
Study, and Docking Calculations
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 0000, 00, 1 – 12
&12
&
ÞÞ
These are not the final page numbers!