Inhibition of monoamine oxidase by indole and benzofuran derivatives

Article abstract of DOI:10.1016/j.ejmech.2010.07.005

Monoamine oxidase (MAO) is an important drug target for the treatment of neurological disorders. A series of indole and benzofuran derivatives were synthesised and evaluated as inhibitors of the two MAO isoforms, MAO-A and MAO-B. In general, the derivatives were found to be selective MAO-B inhibitors with K_i values in the nanoMolar (nM) to microMolar (μM) concentration range. The most potent MAO-B inhibitor, 3,4-dichloro-N-(2-methyl- 1H-indol-5-yl)benzamide, exhibited a K_i value of 0.03 μM and was 99 fold more selective for the B isoform. We conclude that these indole and benzofuran derivatives are promising reversible MAO-B inhibitors with a possible role in the treatment of neurodegenerative diseases such as Parkinson's disease (PD).

Full text of DOI:10.1016/j.ejmech.2010.07.005

European Journal of Medicinal Chemistry 45 (2010) 4458e4466
Contents lists available at ScienceDirect
European Journal of Medicinal Chemistry
journal homepage: http://www.elsevier.com/locate/ejmech
Original article
Inhibition of monoamine oxidase by indole and benzofuran derivatives
*
Louis H.A. Prins, Jacobus P. Petzer, Sarel F. Malan
Pharmaceutical Chemistry, School of Pharmacy, North-West University, Private Bag X6001, Potchefstroom 2520, South Africa
a r t i c l e i n f o
a b s t r a c t
Article history:
Monoamine oxidase (MAO) is an important drug target for the treatment of neurological disorders. A
series of indole and benzofuran derivatives were synthesised and evaluated as inhibitors of the two MAO
isoforms, MAO-A and MAO-B. In general, the derivatives were found to be selective MAO-B inhibitors
Received 10 March 2010
Received in revised form
8 June 2010
Accepted 3 July 2010
Available online 31 July 2010
with Ki values in the nanoMolar (nM) to microMolar (
inhibitor, 3,4-dichloro-N-(2-methyl-1H-indol-5-yl)benzamide, exhibited a Ki value of 0.03
m
M) concentration range. The most potent MAO-B
M and was
m
99 fold more selective for the B isoform. We conclude that these indole and benzofuran derivatives are
promising reversible MAO-B inhibitors with a possible role in the treatment of neurodegenerative
diseases such as Parkinson’s disease (PD).
Keywords:
Monoamine oxidase (MAO)
Indole derivatives
Benzofuran derivatives
Parkinson’s disease (PD)
Molecular docking
LigandFit
Ó 2010 Elsevier Masson SAS. All rights reserved.
1. Introduction
Both these metabolic by-products are toxic and may contribute to
the pathogenesis of PD [7]. Increased MAO-B levels have also been
Parkinson’s disease (PD) is a condition associated with the
degeneration 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 (reduction in move-
ment amplitude), akinesia (absence of normal unconscious move-
ments) and other extrapyramidal effects. Since monoamine oxidase
A and B (MAO-A and -B) are involved in the metabolic degradation
of dopamine in the brain, inhibitors of these enzymes are consid-
ered useful for the treatment of PD [2].
The MAO isoform predominantly found in the human brain is
MAO-B [3]. It has also been demonstrated that brain MAO-B
activity, but not MAO-A activity, increases with aging [4]. As MAO-B
appears to be located in glial cells [5], this may be due to gliosis
associated with aging. Not only is MAO-B a major dopamine
metabolising enzyme but it is also involved in the formation of free
radicals and other potentially neurotoxic species. In the catalytic
cycle of MAO-B, 1 mol of hydrogen peroxide (H2O2) and dop-
aldehyde is produced for every mole of dopamine metabolised [6].
observed in plaque-associated astrocytes in the brains of Alz-
heimer’s disease (AD) patients. This increase in MAO-B activity
produces an elevation in hydroxyl radicals (ꢀOH), which has been
correlated with amyloid-
b (Ab) plaque formation. Hence, the
therapeutic potential of selective reversible MAO-B inhibitors does
not rely solely on their ability to increase the biological half-life of
dopamine (symptomatic effects) but also on their ability to
potentially slow PD progression (neuroprotective effects) and to
inhibit Ab plaque formation [4].
(R)-Deprenyl (selegiline) (1) (Fig. 1) was one of the first selective
MAO-B inhibitors to be identified [8] and has since been shown to
possess neuroprotective properties [9,10]. These properties may, in
part, be dependent on the ability of (R)-deprenyl to inhibit the
MAO-B catalysed formation of H2O2 and dopaldehyde in the brain
[11]. In 1981 it was found that an N-demethylated aminoindan
propargylamine derivative, AGN 1135, also was a potent and
selective inhibitor of MAO-B [12]. Unlike (R)-deprenyl, this
compound is not an amphetamine derivative and therefore did not
present with amphetamine associated sympathomimetic side-
effects. The R(þ)-enantiomer of AGN 1135, now called rasagiline
(2), is nearly three orders of a magnitude more potent than the S
(ꢁ)-enantiomer in inhibiting MAO-B [13]. Rasagiline is considered
useful as an adjuvant to levodopa for the treatment of PD, as this
compound enhances dopamine levels in the primate brain
following systemic administration of levodopa. Youdim and co-
* Corresponding author. School of Pharmacy, University of the Western Cape,
Private Bag X17, Bellville 7535, South Africa. Tel.: þ27 21 9593190; fax: þ27 21
9591588.
E-mail address: sfmalan@uwc.ac.za (S.F. Malan).
0223-5234/$ e see front matter Ó 2010 Elsevier Masson SAS. All rights reserved.
doi:10.1016/j.ejmech.2010.07.005

L.H.A. Prins et al. / European Journal of Medicinal Chemistry 45 (2010) 4458e4466
4459
(6eeh), which are structural analogues of compound 6b, were
N
synthesised and evaluated as MAO-A and -B inhibitors.
H3C
H
N
2. Results and discussion
H
Rasagiline (2)
(R)-Deprenyl (1)
2.1. Chemistry
O
The indole derivatives 6aed, 6f and 6h were synthesised by
reacting 5-amino-2-methylindole (5) with the appropriate
carboxylic acid in the presence of N-(3-dimethylaminopropyl)-N0-
ethylcarbodiimide (EDC) as dehydrating agent [25e27]. For the
synthesis of compounds 6e and 6g a similar procedure was fol-
lowed, but in this instance N,N0-dicyclohexylcarbodiimide (DCC)
was employed as dehydrating agent. The benzofuran derivatives
(8aed) were prepared according to the procedure reported by
Scriven and co-workers [28] by reacting 5-hydroxy-3H-benzo-
furan-2-one (7) with the appropriate carboxylic acid in the pres-
ence of DCC to yield the esters (Scheme 1). The structures of the
target compounds were confirmed by 1H NMR, 13C NMR, HRMS and
IR while the purity was evaluated by HPLC analysis.
N
N
N
H
H
H
PF 9601N (4)
Indole (3)
Fig. 1. Chemical structures of (R)-deprenyl (1), rasagiline (2), indole (3) and PF
9601N (4).
workers [14] showed that rasagiline also activates enzymes that
play a key role in cellular events including mitochondrial viability,
modulation of apoptotic processes and neuronal plasticity. These
effects may further contribute to the observed protective effects of
rasagiline.
2.2. Enzyme inhibition studies
Irreversible inhibition of MAO, however, have certain disad-
vantages including the fact that enzyme recovery requires the
synthesis of new enzyme, the possible loss of selectivity as
a result of repeated administration and an inhibition that is not
affected by changes in substrate concentration [15]. The rate of
enzyme biosynthesis, and hence recovery from irreversible
inhibition is relatively slow and differs substantially between
tissues and species. For example, a study by Fowler and co-
workers [16] showed that the turnover rate for the synthesis of
MAO-B in the human brain is close to 40 days. This was done by
measuring the half-time for brain MAO-B synthesis in PD and in
normal subjects after withdrawal from (R)-deprenyl. Reversible
inhibition may be of a competitive, mixed, non-competitive or
uncompetitive nature and possess the following advantages: the
enzyme recovers as the inhibitor is eliminated from the tissues,
the risk of loss of selectivity is far less because of a shorter
duration of action and, in the case of competitive reversible
inhibition, the inhibition is relieved when the substrate
concentration is increased [15].
From the above discussion it is clear that the MAO-B inhibitors,
(R)-deprenyl (1) and rasagiline (2), have an array of biological
attributes that make them good candidates for the treatment and
possible prevention of neurodegenerative disorders. Both of these
are however irreversible MAO-B inhibitors, which is undesirable
and current focus has shifted to the development of selective
reversible MAO-B inhibitors [17e19].
Several previous studies [20e22] have indicated that the indole
nucleus (3) may be useful as a scaffold for the design of MAO
inhibitors. For example, a variety of indolylmethylamine derivatives
have been shown to act as irreversible MAO inhibitors [23]. Of
importance is the observation that bulky C5 substituents generally
increase potency and selectivity for the MAO-B isoform. An example
of such a substituent is the benzyloxy moiety (for example structure
4) which is thought to enhance the selectivity towards MAO-B by
increasing the molecular lipophilicity [24]. In the present study, we
evaluated the possibility that the indole nucleus may be used in the
design of reversible MAO inhibitors bysynthesising a series of indole
(6aed) and structurally related benzofuran (8aed) derivatives and
evaluating them as inhibitors of recombinant human MAO-A and -B
(Scheme 1). As shown in Scheme 1, the indole and benzofuran
derivatives were substituted with a variety of C5 side chains. Based
on the results of this study, a second series of indole derivatives
The MAO-A and -B activity measurements were based on the
degree to which the MAO-A/B non-selective substrate, kynuramine
(Km ¼ 16.1
mM and 22.67 mM for human recombinant MAO-A and
-B, respectively), is oxidised by recombinant human MAO-A and -B
to yield 4-hydroxyquinoline [29]. Kynuramine is non-fluorescent
until undergoing oxidative deamination by MAO to produce the
fluorescent metabolite 4-hydroxyquinoline
(
lEx
¼
310 nm,
lEm ¼ 400 nm). Quantification of product formation was achieved
by comparing the fluorescence emission of the samples to that of
known amounts of authentic 4-hydroxyquinoline [29]. The IC50
values (concentration of the inhibitor that inhibits 50% of the
enzyme activity) for the test compounds were determined by
measuring the extent by which different concentrations of a test
H
O
R
NH2
N
i, ii, iii
+
+
HO
R
R
O
N
N
H
H
5
6
O
R
OH
O
iv
O
O
HO
Cl
O
O
O
7
8
Cl
R =
6b, 8b
6a, 8a
6c, 8c
6d, 8d
Cl
SO3H
N+
Cl
6g
6h
6e
6f
Scheme 1. Synthetic pathways to the indole (6aeh) and benzofuran (8aed) deriva-
tives: (i) EDC, dioxane/H2O (6aed); (ii) DCC, CH2Cl2 (6e, 6g); (iii) EDC, MeOH (6f, 6h)
(iv) DCC, DMAP, CH2Cl2.

4460
L.H.A. Prins et al. / European Journal of Medicinal Chemistry 45 (2010) 4458e4466
Table 1
inhibitor slow the rate of the MAO catalysed deamination of
kynuramine (Fig. 2). To allow for the calculation of the selectivity
index (SI), the Ki values for the inhibition of MAO-A and -B were
estimated from the experimentally determined IC50 values
according to the ChengePrusoff equation [30].
The results of the MAO inhibition studies with compounds 6aeh
and 8aed are presented in Table 1. With the exception of 6d and 8a
all of the derivatives wereselective inhibitors of the MAO-B isozyme.
Compound 8a was essentially non-selective. Interestingly, chlorine
substitution of the C5 phenyl side chain of the indoles and benzo-
furanes enhances both MAO-A and -B inhibition potencies since the
unsubstituted homologues were less potent MAO inhibitors than
the corresponding chlorine substituted compounds (for example
compare 6b with 6a). The most potent MAO-B inhibitor was
MAO inhibition by the indole (6aeh) and benzofuran (8aed) derivatives.
Compound
MAO-A inhibition
IC50 M)
MAO-B inhibition
IC50 M) Ki (m
M)b
SIc
(
m
Ki (
m
M)a
(
m
6a
6b
6c
6d
6e
6f
6g
6h
8a
8b
8c
8d
149.71 ꢂ 4.10
31.04 ꢂ 0.11
662.71 ꢂ 65.27
7.67 ꢂ 0.55
69.42 ꢂ 2.01
259.07 ꢂ 8.27
9.79 ꢂ 1.26
20.76 ꢂ 0.4
11.57 ꢂ 0.32
7.66 ꢂ 0.63
12.5 ꢂ 0.18
3.9 ꢂ 0.36
39.45
8.18
174.63
2.02
18.29
68.27
2.58
5.47
3.05
2.02
3.29
3.84 ꢂ 0.75
0.76 ꢂ 0.17
18.82 ꢂ 3.28
9.83 ꢂ 1.63
6.96 ꢂ 0.09
175.94 ꢂ 11.51
0.06 ꢂ 0.01
92.62 ꢂ 1.87
8.89 ꢂ 0.99
0.99 ꢂ 0.1
1.65
0.33
8.01
4.23
3.0
75.72
0.03
39.86
3.83
0.43
0.52
0.19
24
25
22
0.5
6
0.9
99
0.1
0.8
5
1.2 ꢂ 0.17
6
5
1.03
0.44 ꢂ 0.03
benzofuran 8d with a Ki value of 0.19 mM. Compound 8d was also the
a
The experimentally determined IC50 values were used to calculate the Ki values
according to the equation by Cheng and Prusoff: Ki ¼ IC50/(1 þ [S]/Km) with
most potent MAO-A inhibitor of the test compounds. The most
selective MAO-B inhibitor was indole 6b with an SI of 25. Compound
6b was also found to be a potent MAO-B inhibitor with a Ki value of
[S] ¼ 45
m
M and Km (Kynuramine) ¼ 16.1
mM [30].
b
The experimentally determined IC50 values were used to calculate the Ki values
according to the equation by Cheng and Prusoff: Ki ¼ IC50/(1 þ [S]/Km) with
0.33 mM. Two other inhibitors, 8b and 8c, also exhibited MAO-B
[S] ¼ 30
m
M and Km (Kynuramine) ¼ 22.67
mM [30].
inhibition potencies in the sub-microMolar range. Compared to
indole 6b, the other potent MAO-B inhibitors of the series, benzo-
furanes 8bed, displayed relatively low selectivity for MAO-B (5e6
fold). Based on high inhibition potency and MAO-B selectivity as
selection criteria, 6b was considered a promising lead compound for
the development of reversible MAO-B inhibitors.
In the second series, four structural derivatives (6eeh) of 6b
were synthesised and evaluated as inhibitors of recombinant
human MAO-A and -B. The results of the MAO inhibition studies
with compounds 6eeh are presented in Table 1. Compound 6g was
found to be the most potent MAO-B inhibitor with a Ki value of
c
Relative selectivity for MAO-B is defined by the ratio of Ki(MAO-A)/Ki(MAO-B)
for each compound.
weaker MAO-B inhibitor than 6a indicates that phenyl substitution
is more favourable for MAO-B inhibition than cyclohexyl substitu-
tion. Also, the weak inhibition of both MAO-A and -B by the sulfonyl
substituted homologue 6f supports the proposal that side chains
with a relatively high degree of lipophilicity generally enhances
affinity of inhibitors for the MAO-A and -B active sites [33,34].
2.3. Reversibility of inhibition
0.03 mM. Compound 6g was also the most selective MAO-B inhib-
itor with an SI of 99. Since 6g is the most potent and selective MAO-
B inhibitor identified in this study, it can be concluded that the
addition of an additional chlorine substituent to the phenyl ring at
C5 of the indole ring is a suitable strategy to enhance binding
affinity to the MAO-B active site. As mentioned above, chlorine
substitution at the phenyl ring of the C5 side chain in general
appears to enhance MAO-B inhibition activity of both the indoles
and benzofuranes. By employing molecular docking studies,
possible reasons for this observation are discussed in the next
section. Interestingly, 6h was found to inhibit MAO-A selectively (7
fold). This is in agreement with literature reports that positively
charged structures, such as methylene blue, display affinity for the
MAO-A active site [31,32]. The observation that 6e is a relatively
Another goal of the present study was to determine if the indole
and benzofuran derivatives investigated here act as reversible or
irreversible enzyme inhibitors. Since the most active derivatives are
MAO-B inhibitors with only moderate to weak MAO-A inhibition
potencies, the time-dependence of MAO-B inhibition was investi-
gated with two potent derivatives, indole 6g and benzofuran 8b.
Recombinant human MAO-B was pre-incubated with 6g and 8b for
various periods of time (0e60 min). For this purpose the concen-
trations of the inhibitors were equal to 2 fold their respective IC50
values for the inhibition of MAO-B. The residual MAO-B activity was
subsequently determined. As shown in Fig. 3, no time-dependent
reduction in the rate of MAO-B catalysed oxidation of kynuramine
is observed with both 6g and 8b. This indicated that the inhibition
is reversible, at least for the time period (60 min) evaluated. Curi-
ously, a marked increase of the MAO-B catalytic rate with increased
preincubation time is observed for 8b. One possible explanation for
this observation is that 8b, and probably the other benzofuran
derivatives, is unstable in the potassium phosphate buffer used for
the enzymatic reactions. The possibility exists that hydrolysis of the
lactone and/or ester functional group may occur with the resulting
loss of inhibition potential. Further studies are underway to
investigate this observation.
4
3
2
1
To determine the mode of inhibition, sets of LineweavereBurk
plots were constructed for the inhibition of MAO-B by 6g. As shown
in Fig. 4 the linear LineweavereBurk plots intersect on the y-axis for
6g, which indicates that the mode of inhibition is competitive and
therefore reversible.
-4
-3
-2
-1
Log[I]
0
1
2
3
2.4. Docking studies
Fig. 2. Sigmoidal doseeresponse curve of the rate of MAO-B catalysed 4-hydroxy-
quinoline formation versus the logarithm of concentration of inhibitor 6g (expressed in
nM). Rates are expressed as nmol 4-hydroxyquinoline formed/mg protein/min. All
determinations were performed in duplicate and are expressed as mean ꢂ SEM.
While the indole and benzofuran derivatives investigated here
were relatively weak MAO-A inhibitors, several derivatives were
found to be potent inhibitors of MAO-B. Of these, compound 6g,

L.H.A. Prins et al. / European Journal of Medicinal Chemistry 45 (2010) 4458e4466
4461
recombinant MAO-B in complex with safinamide (PDB code: 2V5Z)
[35], was selected as receptor model. The valences of the FAD co-
factor and the co-crystallised ligand were corrected and the model
was subjected to a three-step energy minimisation cascade while
the protein backbone was constrained. Minimisation of the
receptor proteins were considered necessary since the protein X-
ray structure might contain residual energetic tensions from the
crystallisation process. Following the energy minimisation cascade,
the backbone constraint was removed and the co-crystallised
ligand was deleted. With the exception of three highly conserved
water molecules in the active site, all crystallographic waters were
deleted from the protein model [35]. The structures of 6g and 8d
were built, geometry optimised within Discovery Studio and
docked into the protein model with LigandFit. Following the
docking, the orientations and conformations of the docked ligands
were further refined with the Smart Minimizer algorithm in
Discovery Studio. Ten possible binding orientations were
computed. To determine the accuracy of this docking protocol, the
co-crystallised ligand, safinamide, was redocked into the MAO-B
active site. This procedure was repeated three times and the best-
3
2
1
0
A
0
15
30
60
Incubation time (min)
7.5
5.0
2.5
0.0
B
ꢀ
ranked solutions of safinamide exhibited an RMSD value of 1.54 A
from the position of the co-crystallised ligand. Since RMSD values
ꢀ
smaller than 2.0 A generally indicate that the docking protocol is
capable of accurately predicting the binding orientation of the co-
crystallised ligand [36], this protocol was deemed to be suitable for
the docking of inhibitors into the active site of MAO-B. The docking
solutions were ranked according to their respective DockScore
values.
Examination of the best-ranked docking solution for 6g, reveals
that the relatively polar indole nucleus is located in the substrate
cavity, in the vicinity of the FAD co-factor, with the C5 side chain
extending towards the entrance cavity space of MAO-B (Fig. 5a).
This binding orientation is similar to that observed for the co-
crystallised ligand, safinamide, which also traverses both active site
cavities [35]. The polar propanamidyl moiety of safinamide is also
located in the substrate cavity while the apolar 3-fluorobenzyloxy
side chain is stabilised within the entrance cavity space. As shown
in Fig. 5a, the 3,4-dichlorophenyl moiety of 6g is possibly stabilised
by Van der Waals interactions in the hydrophobic entrance cavity
defined by Phe-103, Trp-119, Leu-164, Leu-167, Phe-168 and Ile-316
[29]. Since the addition of chlorine would enhance the lipophilicity
of a phenyl ring, these interactions may explain the observation
that chlorine substitution of the phenyl side chains of the indoles
and benzofuranes enhances MAO-B inhibition potency. Of impor-
tance is the observation that the amide carbonyl oxygen of 6g is
stabilised by hydrogen bond interaction with the phenolic
hydrogen of Tyr-326 [37]. In MAO-A, the residue that corresponds
to Tyr-326 (in MAO-B) is Ile-335 [38]. The resulting absence of these
stabilising interactions may, in part, explain the lower MAO-A
inhibition potencies of this inhibitor. Other significant interactions
which may stabilise the MAO-Be6g complex include a possible
0
15
30
60
Incubation time (min)
Fig. 3. The time-dependence of the inhibition of the recombinant human MAO-B by 6g
(panel A) and 8b (panel B). MAO-B was pre-incubated for different periods of time
(0e60 min) with the inhibitors and the initial rates (nmoles 4-hydroxyquinoline
formed/min/mg protein) were recorded.
was the most potent and selective MAO-B inhibitor. To gain insight
into the possible mode of binding of 6g in the active site of MAO-B,
molecular docking studies were performed. The most potent
benzofuran derivative, 8d, was also included in the docking study.
For this purpose, the LigandFit application within Discovery
StudioÒ 1.7 was used. The crystallographic structure of human
1.00
0.75
0.50
0.25
pep interaction [38] between the indole ring and the amide
functional group of the Gln-206 side chain with an interplane
ꢀ
distance of approximately 3.5 A and a possible hydrogen bonding
between the amide carbonyl oxygen of 6g and an active site water
molecule.
The best-ranked docking solution obtained for 8d in the MAO-B
active site is similar to that observed for 6g. The polar benzofuran
ring binds within the substrate cavity where it is stabilised by
hydrogen bonding between the lactam functional group of the
inhibitor and the phenolic hydrogen of Tyr-435 and two active site
water molecules (Fig. 5b). These are the principal interactions
which stabilises the benzofuran derivatives in the MAO-B active
site. The C5 side chain of 8d extends towards the entrance cavity
space of MAO-B where it is stabilised by hydrophobic interactions.
-0.02
0.00
0.02
1/[S]
0.04
0.06
Fig. 4. Lineweaver-Burk plots of the inhibition of recombinant human MAO-B by 6g.
The lines were constructed in the absence (filled squares) and presence of 0.015
(open squares), 0.03 M (filled circles) and 0.06 M (open circles) of 6g. The rate (V) is
expressed as nmol 4-hydroxyquinoline formed/min/mg protein.
mM
m
m

4462
L.H.A. Prins et al. / European Journal of Medicinal Chemistry 45 (2010) 4458e4466
Fig. 5. Stereo view of the predicted binding modes of 6g (Panel A) and 8d (Panel B) in the human MAO-B active site. Hydrogens are hidden, except those involved in hydrogen
bonds, and selected amino acid residues are displayed in grey. The FAD co-factor is displayed in magenta and hydrogen bonds shown as black dashes.
Similar to the analysis for 6g above, the observation that chlorine
substitution improves the MAO-B inhibition potencies of the ben-
zofuranes (8aed) may be due to enhancement of the lipophilicity of
the C5 side chain. In contrast to the amide carbonyl of indole 6g, the
ester carbonyl oxygen of 8d does not interact via hydrogen bonding
with Tyr-326. The reason for this is that 8d binds deeper (than 6g)
within the MAO-B active site in order to establish hydrogen
bonding with the Tyr-435 and the two active site water molecules.
It can therefore be concluded that hydrogen bonding with Tyr-326
is a major factor in stabilising the MAO-Be6g complex while polar
interactions with Tyr-435 and the waters in the vicinity of the FAD
co-factor are the major factors stabilising the MAO-Be8d complex.
This analysis may explain why 8d also exhibits affinity for the MAO-
A active site. In MAO-A, the residue that corresponds to Tyr-435 (in
MAO-B) is Tyr-444 [38]. Tyr-444 and water molecules present in
the vicinity of the FAD co-factor may stabilise 8d via hydrogen
bonding in the MAO-A active site (Fig. 6), similar to that observed
for the MAO-B active site. The proposal that benzofuran derivatives

L.H.A. Prins et al. / European Journal of Medicinal Chemistry 45 (2010) 4458e4466
4463
Fig. 6. Stereo view of the predicted binding mode of 8d in the human MAO-A active site. Hydrogens are hidden, except those involved in hydrogen bonds, and selected amino acid
residues are displayed in grey. The FAD co-factor is displayed in magenta and hydrogen bonds shown as black dashes.
are mainly stabilised by polar interactions in the substrate cavity
may also explain the relatively small difference (w2 fold) in the
MAO-B inhibition potencies of 8d and its shorter side chain length
homologue, 8b. Regardless of the difference in chain lengths, both
these homologues are expected to be stabilised to the same degree
in the substrate cavity of MAO-B.
hydrophobic pockets located towards the entrance of the active site
cavities of the two enzymes.
We conclude that the benzofuran nucleus, in general, is a good
scaffold for designing potent, relatively non-selective MAO inhibi-
tors and that the indole nucleus show promise as scaffold for
developing potent, MAO-B selective inhibitors.
3. Conclusions
4. Experimental
In the current study a series of indole and benzofuran deriva-
tives were synthesised and evaluated as inhibitors of MAO-A and
-B. In general, the derivatives were found to be selective for the
MAO-B isoform with 3 benzofuranes and 2 indoles exhibiting
inhibition potencies in the sub-microMolar range. Indole derivative
6g was found to be the most potent and selective MAO-B inhibitor
4.1. General methods
Thin-layer chromatography was performed on 0.20 mm thick
aluminium silica gel sheets (AlugramÒ SIL G/UV254, Kieselgel 60,
Macherey-NagelÒ Düren, Germany) with an appropriate mobile
phase and visualisation was achieved using UV light (254 and
366 nm) and iodine vapour. All melting points (mp) were obtained
using a StuartÒ Scientific SMP10 melting point apparatus and are
uncorrected. 1H and 13C NMR spectra were recorded using
a BrukerÒ Avance III 600 spectrometer at a frequency of 600 and
150 MHz, respectively. Chemical shifts are expressed in parts per
with a Ki value of 0.03 mM and a selectivity index of 99. As suggested
by the molecular docking studies, the selective binding of 6g to
MAO-B may be explained by hydrogen bond interaction between
the inhibitor and the phenolic hydrogen of Tyr-326. The corre-
sponding residue in MAO-A, Ile-335 [38], would not undergo this
interaction with 6g. The most potent MAO-B inhibitor among the
million (d) relative to the signal from tetramethylsilane (Me4Si),
benzofuran derivatives was 8d with a Ki value of 0.19
m
M and
added to the deuterated solvent DMSO-d6. Spin multiplicities are
given as s (singlet), d (doublet), dd (doublet of doublets) and m
(multiplet). IR spectra were recorded in KBr on a NicoletÒ NexusÔ
470-FT IR spectrometer over the range 400e4000 cmꢁ1 employing
the diffuse reflectance method. Direct insertion electron impact
ionization (EIMS) and high resolution mass spectra (HRMS) were
obtained on a DFS high resolution magnetic sector mass spec-
trometer (Thermo Electron Corporation).
a selectivity index of 5. Molecular docking suggests that 8d is sta-
bilised in the MAO-B active site principally by hydrogen bonding
with Tyr-435 and the waters in the vicinity of the FAD co-factor. The
suggestion that similar polar interactions are possible between 8d
and the active site of MAO-A may explain the relatively low isoform
selectivity of this inhibitor.
An important observation is that chlorine substitution at the C5
phenyl side chain of the indoles and benzofuranes enhances both
MAO-A and -B inhibition potencies. This may be due to the
enhancement of nonspecific hydrophobic interactions between the
inhibitors and the active sites of the MAO isozymes. Halogen
substitution of an aromatic ring may therefore be considered as
a general strategy to improve the binding affinities of ligands to
MAO-A and -B. An important consideration however, is that the
binding of these ligands should place the halogen in the
HPLC analyses were performed using an AgilentÒ 1100 series
HPLCequippedwithaquaternarygradientpump,autosampler, diode
array UV detector and ChemstationÒ data acquisition and analysis
software. AVenusil XBP C18 column (4.60 ꢃ 150 mm, 5
mm) was used
and elutionwas effected by means of a linear gradient starting at 30%
acetonitrile to 85% acetonitrile after 5 min and holding at 85%
acetonitrile until 10 min, where after the instrument was re-equili-
brated atthestartingconditions for5 min. Standard solutions (1mM)

4464
L.H.A. Prins et al. / European Journal of Medicinal Chemistry 45 (2010) 4458e4466
ofeachof theeighttest compoundswerepreparedinanalyticalgrade
acetonitrileandanalysed atwavelengthsof210, 254and 300 nmover
a 15 min time period. The mobile phase flow rate was 1 mL/min and
7.50e7.43 (m, 2H), 7.22 (s, 2H), 6.93 (d, J ¼ 16.03 Hz,1H), 6.10 (s,1H),
2.37 ppm (s, 3H); 13C NMR [150 MHz, DMSO-d6]:
d
¼ 163.0 (C), 137.8
(2 ꢃ CH), 136.8 (C), 134.2 (C), 133.5 (C), 131.5 (C), 131.3 (C), 129.6
(CH), 129.0 (CH), 127.7 (C), 126.5 (CH), 125.2 (CH), 114.0 (CH), 110.8
(CH), 110.3 (CH), 99.8 (CH), 13.9 ppm (CH3); IR (KBr): nmax ¼ 3410,
3236, 3049, 2853, 1655, 1595, 1485, 997, 984, 912, 884, 859, 800,
783, 745, 687, 671 cmꢁ1. Compound purity (HPLC): 98.0%. EIMS m/z
310 (Mꢀþ); HRMS calcd. 310.0873, found 310.0866.
the injection volume was 20 mL.
4.2. Synthesis
4.2.1. General synthetic method for indoles 6aed
5-Amino-2-methylindole (500 mg, 3.42 mmol) and EDC
hydrochloride (947 mg, 4.94 mmol) were dissolved in 20 mL
dioxane/water (1:1). The appropriate benzoic acid or cinnamic acid
(3.67 mmol) was added to the solution and the pH of the suspen-
sion adjusted to 5 with 2 M HCl (aq). The reaction mixture was
stirred at room temperature for 2 h where after the pH was
adjusted to 7 with 1 M NaOH (aq). Cooling the reaction with an
external ice bath resulted in the final products (6aed) precipitating
from the reaction mixture. The precipitates were collected by
filtration, giving the amide products in good yield, without the
need for further purification [27].
4.2.6. General synthetic method for indoles 6eeh
Compounds 6e and 6g: The appropriate benzoic acid (4.1 mmol)
was dissolved in 10 mL CH2Cl2 (anhydrous) where after DCC
(932 mg, 4.52 mmol) was dissolved in an additional 10 mL CH2Cl2
(anhydrous) and added to the benzoic acid solution. The mixture
was stirred on an external ice bath for 6 h and the N,N0-dicyclo-
hexylurea (DCU) produced during activation of the carboxylic acid,
was removed by filtration. 5-Amino-2-methylindole (600 mg,
4.1 mmol) was dissolved in 10 mL CH2Cl2 (anhydrous) and added to
the reaction mixture. After allowing the reaction to stir at room
temperature for 80 h, purification was performed with column
chromatography using diethyl ether as mobile phase.
4.2.2. N-(2-Methyl-1H-indol-5-yl)benzamide (6a)
The melting point of this compound corresponds to that
described in literature (192e192.5 ꢄC) [39]. Yield: 230 mg (27%);
Rf ¼ 0.47 (diethyl ether); mp: 190e193 ꢄC 1H NMR [600 MHz,
Compounds 6f and 6h: 5-Amino-2-methylindole (500 mg,
3.42 mmol) and the appropriate benzoic acid (3.49 mmol) were
dissolved in 40 mL methanol. EDC hydrochloride (720 mg,
3.49 mmol) was added and the reaction allowed to stir at room
temperature for 85 h. Compound 6h was collected from the reac-
tion mixture by filtration, without the need for further purification.
Compound 6f was purified by column chromatography with
ethanol (anhydrous) as mobile phase [25].
DMSO-d6]:
d
¼ 10.90 (s, 1H), 10.09 (s, 1H), 8.00 (s, 1H), 7.88 (s, 1H),
7.64e7.52 (m, 3H), 7.39e7.21 (m, 3H), 6.12 (s, 1H), 2.42 ppm (s, 3H);
13C NMR [150 MHz, DMSO-d6]:
d
¼ 164.0 (C), 137.0 (C), 136.0 (C),
134.0 (C), 131.6 (CH), 131.0 (C), 128.0 (3 ꢃ CH), 127.5 (C, CH), 115.0
(CH), 112.0 (CH), 110.0 (CH), 99.3 (CH), 13.5 ppm (CH3); IR (KBr):
nmax ¼ 3378, 3285, 2924, 2851, 1653, 1578, 1385, 912, 892, 856, 840,
797, 747, 713, 690 cmꢁ1. Compound purity (HPLC): 98.3%. EIMS m/z
250 (Mꢀþ); HRMS calcd. 250.1106, found 250.1102.
4.2.7. N-(2-Methyl-1H-indol-5-yl)cyclohexanecarboxamide (6e)
Yield: 258 mg (25%); Rf ¼ 0.66 (diethyl ether); mp: 198e200 ꢄC
1H NMR [600 MHz, DMSO-d6]:
d
¼ 10.74 (s, 1H), 9.49 (s, 1H), 7.70 (s,
4.2.3. 3-Chloro-N-(2-methyl-1H-indol-5-yl)benzamide (6b)
Yield: 369 mg (38%); Rf ¼ 0.62 (diethyl ether); mp: 178e182 ꢄC
1H), 7.11 (dd, J ¼ 8.44, 8.54 Hz, 2H), 6.02 (s, 1H), 2.33 (s, 4H),
1.78e1.63 (m, 2H), 1.44e1.38 (m, 2H), 1.28e1.09 ppm (m, 6H); 13C
1H NMR [600 MHz, DMSO-d6]:
d
¼ 10.86 (s, 1H), 10.13 (s, 1H),
NMR [150 MHz, DMSO-d6]:
d
¼ 173.6 (C), 136.0 (C), 132.7 (C), 131.3
8.01e8.00 (m, 1H), 7.94e7.90 (m, 1H), 7.82 (s, 1H), 7.66e7.61 (m,
1H), 7.58e7.52 (m,1H), 7.32e7.26 (m, 1H), 7.24e7.19 (m, 1H), 6.10 (s,
(C), 128.4 (C), 113.7 (CH), 110.0 (CH), 109.8 (CH), 99.1 (CH), 44.8 (CH),
29.3 (2 ꢃ CH2), 25.3 (3 ꢃ CH2), 13.4 ppm (CH3); IR (KBr):
nmax ¼ 3253, 3089, 2930, 2856, 1656, 1587, 1518, 1481, 1448, 897,
795, 770 cmꢁ1. Compound purity (HPLC): 97.7%. EIMS m/z 256
(Mꢀþ); HRMS calcd. 256.1576, found 256.1569.
1H), 2.36 ppm (s, 3H); 13C NMR [150 MHz, DMSO-d6]:
d
¼ 163.5 (C),
137.5 (C), 136.3 (CH), 133.3 (CH), 133.2 (CH), 131.0 (C), 130.5 (C),
130.3 (C), 128.4 (C), 127.3 (C), 126.3 (CH), 114.9 (CH), 111.5 (CH), 110.1
(CH), 99.3 (CH), 13.5 ppm (CH3); IR (KBr): nmax ¼ 3377, 3166, 2925,
2852, 1647, 1569, 1382, 913, 891, 869, 845, 799, 731, 675 cmꢁ1
.
4.2.8. 3-[(2-Methyl-1H-indol-5-yl)carbamoyl]benzene-1-sulfonic
acid (6f)
Compound purity (HPLC): 98.6%. EIMS m/z 284 (Mꢀþ); HRMS calcd.
284.0716, found 284.0711.
Yield: 1 g (94%); Rf ¼ 0.71 (ethanol); mp: 222e225 ꢄC 1H NMR
[600 MHz, DMSO-d6]:
d
¼ 10.85 (s, 1H), 10.15 (s, 1H), 8.22 (s, 1H),
4.2.4. (2E)-N-(2-Methyl-1H-indol-5-yl)-3-phenylprop-2-enamide
(6c)
7.92 (d, J ¼ 7.39 Hz, 1H), 7.84 (s, 1H), 7.79 (d, J ¼ 7.39 Hz, 1H), 7.47 (t,
1H), 7.26 (dd, J ¼ 9, 9 Hz, 2H), 6.09 (s, 1H), 2.36 (s, 3H), 1.05 ppm (t,
Yield: 859 mg (91%); Rf ¼ 0.49 (diethyl ether); mp: 198e200 ꢄC
1H); 13C NMR [150 MHz, DMSO-d6]:
d
¼ 164.8 (C), 148.3 (C), 136.2
(decomp). 1H NMR [600 MHz, DMSO-d6]:
d
¼ 10.88 (s, 1H), 10.02 (s,
(C), 135.1 (C), 133.2 (C), 130.8 (C), 128.4 (CH), 127.8 (CH), 127.6 (CH),
124.9 (C, CH), 115.0 (CH), 111.4 (CH), 110.1 (CH), 99.3 (CH), 13.5 ppm
(CH3); IR (KBr): nmax ¼ 3630, 3259, 3072, 2921, 1642, 1542, 1481,
1451, 1195, 1043, 917, 875, 803, 771, 752, 739, 676 cmꢁ1. Compound
purity (HPLC): 91.2%.
1H), 7.90 (s, 1H), 7.70e7.60 (m, 2H), 7.57 (d, J ¼ 15.63 Hz, 1H)
7.50e7.36 (m, 3H), 7.25 (dd, J ¼ 7.85, 8.21 Hz, 2H), 6.91 ppm (d,
J ¼ 15.66 Hz, 1H); 13C NMR [150 MHz, DMSO-d6]:
d
¼ 162.9 (C),
139.0 (CH), 136.3 (C), 134.2 (C), 133.0 (C), 131.1 (C), 129.5 (CH), 129.0
(CH), 128.9 (CH), 127.5 (C, 2 ꢃ CH), 123.0 (CH), 113.5 (CH),110.3 (CH),
109.8 (CH), 99.2 (CH), 13.4 ppm (CH3); IR (KBr): nmax ¼ 3423, 3288,
3052, 2854, 1651, 1592, 1450, 992, 894, 887, 866, 811, 799, 784, 775,
748 cmꢁ1. Compound purity (HPLC): 97.6%. EIMS m/z 276 (Mꢀþ);
HRMS calcd. 276.1263, found 276.1259.
4.2.9. 3,4-Dichloro-N-(2-methyl-1H-indol-5-yl)benzamide (6g)
Yield: 201 mg (15%); Rf ¼ 0.74 (diethyl ether); mp: 200 ꢄC 1H
NMR [600 MHz, DMSO-d6]:
d
¼ 10.88 (s, 1H), 10.19 (s, 1H), 8.21 (s,
1H), 7.86 (dd, J ¼ 8.17, 8.15 Hz, 2H), 7.82 (s, 1H), 7.25 (dd, J ¼ 8.57,
8.34 Hz, 2H), 6.10 (s, 1H), 2.36 ppm (s, 3H); 13C NMR [150 MHz,
4.2.5. (2E)-3-(3-Chlorophenyl)-N-(2-methyl-1H-indol-5-yl)prop-2-
DMSO-d6]:
d
¼ 162.6 (C), 136.4 (C), 135.8 (C), 133.9 (C), 133.4 (C),
enamide (6d)
131.2 (C), 130.7 (CH), 130.3 (C), 129.5 (CH), 128.4 (C), 127.9 (CH),
114.8 (CH), 111.5 (CH), 110.1 (CH), 99.3 (CH), 13.5 ppm (CH3); IR
(KBr): nmax ¼ 3304, 3090, 2922, 2854, 1928, 1855, 1824, 1794, 1741,
Yield: 773 mg (73%); Rf ¼ 0.51 (diethyl ether); mp: 155e160 ꢄ
C
1H NMR [600 MHz, DMSO-d6]:
(s, 1H), 7.70 (s, 1H), 7.63e7.57 (m, 1H), 7.54 (d, J ¼ 16.03 Hz, 1H),
d
¼ 10.86 (s, 1H), 10.00 (s, 1H), 7.90
1707, 1632, 1587, 914, 873, 849, 836, 786, 762, 751, 680 cmꢁ1
.

L.H.A. Prins et al. / European Journal of Medicinal Chemistry 45 (2010) 4458e4466
4465
Compound purity (HPLC): 97.7%. EIMS m/z 318 (Mꢀþ); HRMS calcd.
318.0327, found 318.0315.
128.3 (CH), 127.9 (CH), 123.9 (C), 122.0 (CH), 118.5 (CH), 116.7 (CH),
111.3 (CH), 33.3 ppm (CH2); IR (KBr): nmax ¼ 3059, 3026, 1825, 1708,
1478, 1307, 1261, 1233, 1149, 1067, 920, 879, 866, 801, 765, 738, 709,
686 cmꢁ1. Compound purity (HPLC): 97.3%.
4.2.10. 1-Methyl-N-(2-methyl-1H-indol-5-yl)-1l
4-pyridine-3-
carboxamide (6h)
Yield: 410 mg (55%); Rf ¼ not calculable; mp: 258e260 ꢄC
4.2.15. (E)-3-Chlorocinnamic acid 2-oxo-2,3-dihydro-benzofuran-
5-yl ester (8d)
(decomp). 1H NMR [600 MHz, DMSO-d6]:
d
¼ 11.05 (s, 1H), 10.86 (s,
1H), 9.81 (s, 1H), 9.12 (s, 2H), 8.22e8.19 (m, 1H), 7.91 (s, 1H), 7.35
(dd, J ¼ 8.58, 8.57 Hz, 2H), 6.11 (s, 1H), 4.45 (s, 3H), 2.38 ppm (s, 3H);
Yield: 367 mg (35%); Rf ¼ 0.62 (ethyl acetate/petroleum ether
(1:1)); mp: 85e95 ꢄC 1H NMR [600 MHz, DMSO-d6]:
d
¼ 7.77 (d,
13C NMR [150 MHz, DMSO-d6]:
d
¼ 159.3 (C), 146.4 (CH), 145.3 (CH),
J ¼ 15.99 Hz, 1H), 7.56e7.53 (m, 2H), 7.43e7.27 (m, 3H), 7.60 (dd,
143.2 (CH), 136.3 (C), 134.1 (C), 133.4 (C), 129.7 (C), 128.2 (C), 126.8
(CH), 114.4 (CH), 111.2 (CH), 109.9 (CH), 99.0 (CH), 47.8 (CH3),
13.0 ppm (CH3); IR (KBr): nmax ¼ 3349, 3079, 2860,1674, 1633,1592,
J ¼ 8.89, 8.60 Hz, 2H), 6.60 (d, J ¼ 16.00 Hz, 1H), 3.75 ppm (s, 2H);
13C NMR [150 MHz, DMSO-d6]:
d
¼ 173.7 (C), 165.1 (C), 152.1 (C),
146.9 (C), 145.4 (CH), 135.7 (C), 135.0 (C), 130.7 (CH), 130.3 (CH),
127.6 (CH), 127.1 (CH), 126.0 (CH), 124.0 (C), 122.0 (CH), 118.5 (CH),
111.4 (CH), 33.3 ppm (CH2); IR (KBr): nmax ¼ 3247, 3053, 1745, 1706,
1479, 1259, 1233,1201, 1082, 1045, 893, 860, 842, 787, 753, 715, 688,
674 cmꢁ1. Compound purity (HPLC): 90.4%.
1549, 1508, 1480, 1428, 1387, 904, 882, 856, 808, 745, 731, 693 cmꢁ1
.
4.2.11. General synthetic method for benzofuranes 8aed
The appropriate benzoic acid or cinnamic acid (4.54 mmol) was
dissolved in 10 mL CH2Cl2 (anhydrous) where after 4-dimethyla-
minopyridine (DMAP) (37 mg, 0.3 mmol) was added to the solution
at room temperature. 5-Hydroxy-3H-benzofuran-2-one (500 mg,
3.33 mmol) was dissolved in 5 mL N,N0-dimethylformamide (DMF)
(anhydrous) and added to the reaction mixture. The reaction was
cooled with an external ice bath for 5 min. DCC (687 mg,
3.33 mmol) was dissolved in 5 mL CH2Cl2 (anhydrous) and added to
the reaction while on ice. The reaction was stirred on ice for 5 min
and the DCU produced during activation of the carboxylic acid, was
removed by filtration. The external ice bath was removed and the
reaction was stirred at room temperature for 24 h. The pure
benzofuran derivatives (8aed) were obtained through column
chromatography with ethyl acetate/petroleum ether (1:1) as
mobile phase [28].
4.3. MAO activity measurements and inhibition
Human recombinant MAO-A and -B, expressed in baculovirus
infected BTI insect cells, were used to evaluate the indole (6aeh)
and benzofuran (8aed) derivatives as potential MAO inhibitors.
Both enzymes were purchased from SigmaeAldrichÒ and stored at
ꢁ70 ꢄC until use. All incubations were performed in potassium
phosphate buffer (100 mM, pH 7.4, made isotonic with KCl) and
contained kynuramine (45
mM and 30 mM for MAO-A and -B,
respectively), the enzyme (0.075 mg protein/mL), and various
concentrations of the test inhibitors, producing a final incubation
volume of 500
mL. Stock solutions of the test inhibitors were
prepared in dimethyl sulfoxide (DMSO) and added to the incuba-
tion mixtures to yield a final DMSO concentration of 4% (v/v). DMSO
concentrations higher than 4% are reported to inhibit MAO activity
[40]. Samples were incubated at 37 ꢄC for 20 min and the enzyme
4.2.12. Benzoic acid 2-oxo-2,3-dihydro-benzofuran-5-yl ester (8a)
Yield: 393 mg (46%); Rf ¼ 0.71 (ethyl acetate/petroleum ether
(1:1)); mp: 83e85 ꢄC 1H NMR [600 MHz, DMSO-d6]:
d
¼ 8.12e8.11
reactions were terminated by the addition of 200
mL sodium
(m, 2H), 7.59e7.57 (m, 1H), 7.46e7.30 (m, 3H), 7.08 (s, 1H), 7.05 (s,
hydroxide (2 N) and centrifuged at 16,000g for 10 min. The
supernatant fractions were removed and the concentrations of the
MAO generated product, 4-hydroxyquinoline, were measured
spectrofluorometrically at an excitation wavelength of 310 nm and
an emission wavelength of 400 nm using a VarianÒ Cary EclipseÒ
fluorescence spectrophotometer. The IC50 values were determined
from plots of the rate of MAO-A or -B catalysed 4-hydroxyquinoline
formation versus the logarithm of the inhibitor concentration. For
this purpose the PrismÒ 4.02 (GraphPad, Sorrento Valley, CA)
software package was employed. Ten different concentrations of
the test inhibitor, spanning three orders of magnitude were used to
construct the sigmoidal doseeresponse curve. The IC50 values are
reported as mean ꢂ standard error of the mean (SEM) of duplicate
determinations [29].
1H), 3.71 ppm (s, 2H); 13C NMR [150 MHz, DMSO-d6]:
d
¼ 173.7 (C),
165.3 (C),152.1 (C),147.1 (C),133.8 (CH),130.7 (CH),130.1 (CH),129.1
(C), 128.6 (CH), 128.5 (CH), 124.0 (C), 122.2 (CH), 118.7 (CH), 111.4
(CH), 33.3 ppm (CH2); IR (KBr): nmax ¼ 1805, 1738, 1479, 1264, 1138,
1120, 1052, 1023, 895, 875, 822, 809, 768, 713, 685, 669 cmꢁ1
.
Compound purity (HPLC): 96.4%. EIMS m/z 254 (Mꢀþ); HRMS calcd.
254.0579, found 254.0576.
4.2.13. 3-Chlorobenzoic acid 2-oxo-2,3-dihydro-benzofuran-5-yl
ester (8b)
Yield: 563 mg (59%); Rf ¼ 0.68 (ethyl acetate/petroleum ether
(1:1)); mp: 90e93 ꢄC 1H NMR [600 MHz, DMSO-d6]:
d
¼ 8.14 (m,
1H), 8.05e8.04 (m, 1H), 7.61 (m, 1H), 7.41e7.39 (m, 1H), 7.16e7.10
(m, 3H), 3.77 ppm (s, 2H); 13C NMR [150 MHz, DMSO-d6]:
d
¼ 173.6
(C), 164.1 (C), 152.3 (C), 146.9 (C), 134.8 (C), 133.8 (CH), 130.1 (C),
130.0 (CH), 129.8 (CH), 128.3 (CH), 124.1 (C), 122.1 (CH), 118.5 (CH),
111.4 (CH), 33.3 ppm (CH2); IR (KBr): nmax ¼ 1801, 1734, 1480, 1249,
4.4. Time-dependent inhibition studies
1136, 1106, 1073, 1053, 898, 877, 817, 780, 747, 708, 679, 660 cmꢁ1
.
Recombinant human MAO-B (0.03 mg protein/mL) was pre-
Compound purity (HPLC): 97.4%. EIMS m/z 288 (Mꢀþ); HRMS calcd.
incubated with 6g (0.12
m
M; 2 ꢃ IC50) or 8b (1.98 M; 2 ꢃ IC50) for
m
288.0189, found 288.0201.
periods of 0, 15, 30 and 60 min. These pre-incubations were carried
out at 37 ꢄC in potassium phosphate buffer (100 mM, pH 7.4, made
4.2.14. (E)-Cinnamic acid 2-oxo-2,3-dihydro-benzofuran-5-yl
ester (8c)
isotonic with KCl). Kynuramine (30 mM) was added to the pre-
incubations and the reactions were incubated for an additional
15 min at 37 ꢄC. The reactions were terminated with the addition of
sodium hydroxide (2 N) and the rates of 4-hydroxyquinoline
formation were measured as described above. The final enzyme
concentration in the incubations was 0.015 mg protein/mL and the
Yield: 470 mg (50%); Rf ¼ 0.66 (ethyl acetate/petroleum ether
(1:1)); mp: 75e80 ꢄC 1H NMR [600 MHz, DMSO-d6]:
d
¼ 7.75 (d,
J ¼ 15.97 Hz, 1H), 7.47e7.45 (m, 2H), 7.37e7.23 (m, 3H), 6.97 (dd,
J ¼ 8.65, 8.81 Hz, 2H), 6.49 (d, J ¼ 15.96 Hz, 1H), 3.63 ppm (s, 2H);
13C NMR [150 MHz, DMSO-d6]:
147.0 (CH), 133.9 (C), 130.9 (C), 129.0 (CH), 128.9 (CH), 128.7 (CH),
d
¼ 173.7 (C), 165.5 (C), 152.0 (C),
final concentrations of 6g and 8b were 0.06 mM and 0.99 mM,
respectively [41].

4466
L.H.A. Prins et al. / European Journal of Medicinal Chemistry 45 (2010) 4458e4466
[3] J. Saura, Z. Bleuel, J. Ulrich, A. Mendelowitsch, K. Chen, J.C. Shih, P. Malherbe,
4.5. Examining the mode of inhibition
M. Da Prada, J.G. Richards, Neuroscience 70 (1996) 755e774.
[4] C.J. Fowler, A. Wiberg, L. Oreland, J. Marcusson, B. Winblad, J. Neural Transm.
Gen. Sect. 49 (1980) 1e20.
[5] K.N. Westlund, R.M. Denney, R.M. Rose, C.W. Abell, Neuroscience 25 (1988)
439e456.
In order to determine the modes of inhibition of 6g, Line-
weavereBurk plots were constructed by measuring the initial rates
of MAO-B catalysed oxidation at four different kynuramine
concentrations (15e90
ence of three different concentrations of the inhibitor. The
concentration range chosen for 6g was 0.015e0.06 M. The reac-
tion conditions and rate measurements were carried out as
described above. Linear regression analysis was performed using
the SigmaPlot software package (Systat Software Inc.).
[6] A. Parkinson, B.W. Ogilvie, Biotransformation of xenobiotics. in: C.D. Klaassen
(Ed.), Casarett & Doull’s Toxicology: The Basic Science of Poisons. McGraw-Hill
Inc., New York, 2008, pp. 204e206.
mM) in the absence as well as in the pres-
[7] G. Cohen, J. Neural Transm. Suppl. 32 (1990) 229e238.
[8] J. Knoll, K. Magyar, Adv. Biochem. Psychopharmcol. 5 (1972) 393e408.
[9] R.E. Heikkila, L. Manzino, F.S. Cabbat, R.C. Duvoisin, Nature 311 (1984)
467e469.
m
[10] P.H. Yu, B.A. Davis, D.A. Durden, A. Barber, I. Terleckyj, W.G. Nicklas,
A.A. Boulton, J. Neurochem. 62 (1994) 697e704.
[11] C.W. Olanow, Ann. Neurol. Suppl. 32 (1992) S2eS9.
[12] J.P.M. Finberg, M. Tenne, M.B.H. Youdim, Br. J. Pharmacol. 73 (1981) 65e74.
[13] M.B.H. Youdim, A. Gross, J.P.M. Finberg, Br. J. Pharmacol. 132 (2001) 500e506.
[14] M.B.H. Youdim, W. Maruyama, M. Naoi, Drugs of Today 41 (2005) 369e391.
[15] K.F. Tipton, S. Boyce, J. O’Sullivan, G.P. Davey, J. Healy, Curr. Med. Chem. 11
(2004) 1965e1982.
[16] J.S. Fowler, N.D. Volkow, J. Logan, G.J. Wang, R.R. MacGregor, D. Schyler,
A.P. Wolf, N. Pappas, D. Alexoff, C. Shea, Synapse 18 (1994) 86e93.
[17] F. Ooms, R. Frédérick, F. Durant, J.P. Petzer, N. Castagnoli Jr., C.J. Van der Schyf,
J. Wouters, Bioorg. Med. Chem. Lett. 13 (2003) 69e73.
[18] F. Stocchi, L. Vacca, P. Grassini, M.F. De Pandis, G. Battaglia, C. Cattaneo,
R.G. Fariello, Neurology 67 (2006) S24eS29.
[19] M.J. Matos, D. Viña, C. Picciau, F. Orallo, L. Santana, E. Uriarte, Bioorg. Med.
Chem. Lett. 19 (2009) 5053e5055.
[20] D. Balsa, E. Fernández-Álvarez, K.F. Tipton, M. Unzeta, J. Neural Transm. Suppl.
32 (1990) 103e105.
[21] D. Balsa, E. Fernández-Álvarez, K.F. Tipton, M. Unzeta, Biochem. Soc. Trans. 19
(1991) 215e218.
[22] C. Fernández García, J.L. Marco, E. Fernández-Álvarez, Eur. J. Med. Chem. 27
(1992) 909e918.
[23] J.A. Morón, M. Campillo, V. Pérez, M. Unzeta, L. Pardo, J. Med. Chem. 43 (2000)
1684e1691.
[24] V. Pérez, J.L. Marco, E. Fernández-Álvarez, M. Unzeta, Br. J. Pharmacol. 127
(1999) 869e876.
[25] C.E. Müller, J. Sandoval-Ramírez, U. Schobert, U. Geis, W. Frobenius, K.-
N. Klotz, Bioorg. Med. Chem. 6 (1998) 707e719.
[26] D. van den Berg, K.R. Zoellner, M.O. Ogunrombi, S.F. Malan, G. Terre’Blanche,
N. Castagnoli Jr., J.J. Bergh, J.P. Petzer, Bioorg. Med. Chem. 15 (2007)
3692e3702.
[27] F. Suzuki, J. Shimada, S. Shiozaki, S. Ichikawa, A. Ishii, J. Nakamura, H. Nonaka,
H. Kobayashi, E. Fuse, J. Med. Chem. 36 (1993) 2508e2518.
[28] E.F.V. Scriven, Chem. Soc. Rev. 12 (1983) 129e161.
[29] L. Novaroli, M. Reist, E. Favre, A. Carotti, M. Catto, P.-A. Carrupt, Bioorg. Med.
Chem. 13 (2005) 6212e6217.
[30] Y.C. Cheng, W.H. Prusoff, Biochem. Pharmacol. 22 (1973) 3099e3108.
[31] R.R. Ramsay, C. Dunford, P.K. Gillman, Br. J. Pharmacol. 152 (2007) 946e951.
[32] P.K. Gillman, Can. J. Anaesth. 55 (2008) 311e312.
[33] J. Wouters, F. Ooms, S. Jegham, J. Koenig, P. George, F. Durant, Eur. J. Med.
Chem. 32 (1997) 721e730.
4.6. Ligand docking
Manipulation of the crystal structure of MAO-B (PDB code: 2V5Z)
[35] was performed with Discovery StudioÒ 1.7 (Accelrys Software
Inc., San Diego, CA). The valences of the co-crystallised ligand (safi-
namide) and FAD co-factor were firstly corrected and hydrogen
atoms were added. The receptor proteins were then typed by
applying the CHARMm forcefield and a three-step minimisation
protocol (steepest descent, conjugate gradient and adopted basis
Newton-Rapheson) was performed while the protein backbone was
kept rigid. During the minimisation cascade the Generalised Born
with Simple Switching implicit solvation model was used to account
for the effects of the aqueous environment. Minimisation of the
receptor proteins were considered necessary since the protein X-ray
structure might contain residual energetic tensions from the crys-
tallisationprocess.Safinamidewassubsequentlyeliminatedfromthe
energy-minimised receptor proteins and the backbone constraints
were removed. The following conserved active site waters were
retained: HOH 1155, 1169, 1346, 1351 (MAO-B, A-chain). The final
modified proteins were used as the starting models for the docking
simulation in Discovery StudioÒ. The structures of 6g and 8d, built
and geometry optimised within Discovery StudioÒ, were docked into
the protein model with LigandFit. Following the docking, the orien-
tations and conformations of the docked ligands were further refined
with the Smart Minimizer algorithm. All the parameters for the
docking run were set to their default values. A total of ten docking
solutions were allowed and the top-ranked docking conformations
were determined using the DockScore values.
[34] U. Thull, S. Kneubühler, P. Gaillard, P.-A. Carrupt, B. Testa, C. Altomare,
A. Carotti, P. Jenner, K.St.P. McNaught, Biochem. Pharmacol. 50 (1995)
869e877.
[35] C. Binda, J. Wang, L. Pisani, C. Caccia, A. Carotti, P. Salvati, D.E. Edmondson,
A. Mattevi, J. Med. Chem. 50 (2007) 5848e5852.
Acknowledgements
Financial support received from the National Research Foun-
dation (South Africa) is gratefully acknowledged.
[36] J. Boström, J.R. Greenwood, J. Gottfries, J. Mol. Graph. Model. 21 (2003)
449e462.
[37] F. Chimenti, R. Fioravanti, A. Bolasco, P. Chimenti, D. Secci, F. Rossi, M. Yanez,
F. Orallo, F. Ortuso, S. Alcaro, J. Med. Chem. 52 (2009) 2818e2824.
[38] S.-Y. Son, J. Ma, Y. Kondou, M. Yoshimura, E. Yamashita, T. Tsukihara, Proc.
Natl. Acad. Sci. U S A 105 (2008) 5739e5744.
[39] A.P. Terent’ev, M.N. Preobrazhenskaya, A.S. Bebkov, G.M. Sorokina, Zh.
Obshch. Khim. 29 (1959) 2541e2551;
Appendix. Supplementary data
Supplementary data associated with this article can be found in
the on-line version, at doi:10.1016/j.ejmech.2010.07.005.
Chem. Abstr. 54 (1960) 10991.
[40] C. Gnerre, M. Catto, F. Leonetti, P. Weber, P.-A. Carrupt, C. Altomare, A. Carotti,
B. Testa, J. Med. Chem. 43 (2000) 4747e4758.
References
[41] H. Inoue, K. Castagnoli, C.J. van der Schyf, S. Mabic, K. Igarashi, N. Castagnoli Jr.,
J. Pharmacol. Exp. Ther. 291 (1999) 856e864.
[1] W. Dauer, S. Przedborski, Neuron 39 (2003) 889e909.
[2] D.A. Di Monte, L.E. De Lanney, I. Irwin, Brain Res. 738 (1996) 53e59.