Synthesis, structure-activity relationships and molecular modeling studies of new indole inhibitors of monoamine oxidases A and B

Article abstract of DOI:10.1016/j.bmc.2008.09.072

New monoamine oxidase inhibitors were synthesized as indole analogues of a previously reported pyrrole series. Several compounds were potent MAO-A (12, 17, 19-22, 31, 36, and 37) or MAO-B (14, 20, 24, 38, 44, and 46) inhibitors, and had K_i values in the nanomolar concentration range. In particular, 22 (K_i = 0.00092 μM, and SI = 68,478) was exceptionally potent and selective as MAO-A inhibitor. In molecular modeling studies, compounds 22, 24, 44, and 46 positioned the indole ring into an aromatic cavity of MAO-A, and established π-π stacking interactions with Tyr407, Tyr444, and FAD cofactor. However, only compound 22 was able to form hydrogen bonds with FAD, a finding which was in accordance with its potent anti-MAO-A activity. Conversely, 22/MAOB complex was highly unstable during the MD simulation.

Full text of DOI:10.1016/j.bmc.2008.09.072

Bioorganic & Medicinal Chemistry 16 (2008) 9729–9740
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
Synthesis, structure–activity relationships and molecular modeling studies
of new indole inhibitors of monoamine oxidases A and B
a
b,
b
Giuseppe La Regina a, Romano Silvestri a, , Valerio Gatti , Antonio Lavecchia , Ettore Novellino ,
*
*
Olivia Befani c, Paola Turini c, Enzo Agostinelli c
a Istituto Pasteur – Fondazione Cenci Bolognetti, Dipartimento di Chimica e Tecnologie del Farmaco, Sapienza Università di Roma, Piazzale Aldo Moro 5, I-00185 Roma, Italy
b Dipartimento di Chimica Farmaceutica e Tossicologica, Università di Napoli Federico II, Via Domenico Montesano 49, I-80131 Napoli, Italy
c Dipartimento di Scienze Biochimiche Rossi-Fanelli and Istituti di Biologia e Patologia Molecolare del CNR, Sapienza Università di Roma, Piazzale Aldo Moro 5, I-00185 Roma, Italy
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 20 June 2008
Revised 25 September 2008
Accepted 30 September 2008
Available online 2 October 2008
New monoamine oxidase inhibitors were synthesized as indole analogues of a previously reported pyr-
role series. Several compounds were potent MAO-A (12, 17, 19–22, 31, 36, and 37) or MAO-B (14, 20, 24,
38, 44, and 46) inhibitors, and had Ki values in the nanomolar concentration range. In particular, 22
(Ki = 0.00092
ular modeling studies, compounds 22, 24, 44, and 46 positioned the indole ring into an aromatic cavity of
MAO-A, and established stacking interactions with Tyr407, Tyr444, and FAD cofactor. However, only
lM, and SI = 68,478) was exceptionally potent and selective as MAO-A inhibitor. In molec-
p–p
Keywords:
Amine oxidase
Monoamine oxidase type A
Monoamine oxidase type B
Indole
compound 22 was able to form hydrogen bonds with FAD, a finding which was in accordance with its
potent anti-MAO-A activity. Conversely, 22/MAOB complex was highly unstable during the MD
simulation.
Ó 2008 Elsevier Ltd. All rights reserved.
Structure–activity relationships
Molecular modeling
Molecular dynamics
1. Introduction
tion (i.e., a MAO-A (2) plus a MAO-B (lazabemide, (4) reversible
inhibitors) to treat therapy-resistant depression.5 Ladostigil
Monoamine oxidases (MAOs) catalyze the oxidative catabolism
of biogenic amines.1 MAO-A and MAO-B are the isoforms of MAOs
present in most mammal tissues; they are often bound to the outer
mitochondrial membrane, but differ with respect to amino acid
sequence, distribution in body’s tissues, and substrate/inhibitor
specificity.2 MAO-A preferentially deaminates serotonin (5-HT),
adrenaline (A) and noradrenaline (NA), and is selectively inhibited
by clorgyline (1) and moclobemide (2). MAO-B catalyzes the
oxidative deamination of b-phenylethylamine and benzylamine,
and is selectively inhibited by selegiline (L-deprenyl, 3). Dopamine
(DA) in vitro, and tyramine are deaminated by both isoforms, but
human DA is preferentially deaminated by MAO-B.
MAO-A and MAO-B are attractive targets for therapeutic inter-
vention. MAO-A inhibitors are prescribed for the treatment of
mental depression and anxiety.3 MAO-B inhibitors are used with
(5, R-isomer) combines the structural features of rivastigmine (ace-
tylcholinesterase inhibitor) and rasagiline (MAO-B inhibitor), and
is potentially useful for Alzheimer’s (AD) disease, Lewy Body dis-
ease, and eventually PD.6 M-30 (6) is a brain selective agent char-
acterized by the presence of both propargyl anti-MAO and iron-
chelating moieties; it might serve as drug for neurodegenerative
disorders, such as PD and AD, in which oxidative stress and iron
alteration are implicated.7
Previously, we designed a new class of potent and selective pyr-
role MAO inhibitors by means of a simple model derived from the
structures 2 and 3.8 The synthesis of new pyrrole analogues
allowed to determine the structure–activity relationships (SARs)
and features of the MAO-A/B selectivity, and led to disclose highly
selective MAO-B (7, SI = 0.0057) and MAO-A (8, SI = 12,500)
inhibitors.9
L
-DOPA and/or DA agonists in the symptomatic treatment of Par-
As a progress of this research project, we here describe the
synthesis of new indole derivatives. The indole nucleus is the
key feature of potent anti-MAO agents (i.e., the MAO-B inhibitor
PF9601N (9) shows antioxidant/neuroprotective properties in an
experimental model of PD10,11; 10 is a potent MAO-A inhibitor
kinson’s disease (PD).4 MAO inhibitors are also used in combina-
* Corresponding authors. Tel.: +39 06 4991 3800; fax: +39 06 491 491 (R.S.);
tel./fax: +39 081 678 613 (A.L.).
E-mail addresses: romano.silvestri@uniroma1.it (R. Silvestri), lavecchi@unina.it
(A. Lavecchia).
of
a member a series of indolymethylamines derived from
tryptamine).12
0968-0896/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2008.09.072

9730
G. La Regina et al. / Bioorg. Med. Chem. 16 (2008) 9729–9740
According to our previous findings,8,9 the MAO-A/B selectivity
R2
N
of 11–50 was dependent on small modifications of the chemical
structure. Molecular modeling studies carried out using our previ-
ously reported model,9 allowed to clarify the binding mode of the
most active inhibitors 22, 24, 44, and 46 (see Chart 1).
a
OH
R3
N
R1
N
R1
O
O
11-13, 15, 16, 18, 19,
21-23, 25, 26, 41, 43-46
2. Results
2.1. Chemistry
R2
N
b or c
12, 16, 19, 22, 25
Reaction of indole-2-carboxylic acid with appropriate amines
at room temperature in the presence of (benzotriazol-1-
yloxy)tris(dimethylamino)phosphonium
R3
N
R1
O
hexafluorophosphate
14, 17, 20, 24, 27
(BOP) and triethylamine in anhydrous DMF afforded amides 11–
13, 15, 16, 18, 19, 21–23, 25, 26, 41, and 43–46. Compounds 12, 16,
19, 22, and 25 were alkylated with iodomethane by phase-transfer
reaction in the presence of tetrabutylammonium hydrogen sulfate
(TBAHS) in 50% sodium hydroxide solution and dichloromethane
to give 14, 17, 20, 24, and 27, respectively.
R2
N
d
11-14, 16, 17, 19, 20,
22, 24, 25, 41, 43-46
R3
N
Lithium aluminum hydride reduction of amides 11–14, 16, 17,
19, 20, 22, 24, 25, 41, and 43–46 in refluxing tetrahydrofuran fur-
nished amines 28–31, 33, 34, 36–42, and 47–50, respectively
(Scheme 1). Commercially available amines 51–54 were trans-
formed into the corresponding carbamates 55–58 by reaction with
ethyl chlorofomate in the presence of triethylamine in anhydrous
tetrahydrofuran. 55–58 were reduced to amines 59–62 with lith-
ium aluminum hydride (Scheme 2).
R1
28-31, 33, 34,
36-40, 42, 47-50
R
1 = H, CH3; R2 = H, CH3; R3 = alkyl chain.
Scheme 1. Synthesis of Derivatives 11–50. Reagents and reaction conditions: (a)
amine, BOP, triethylamine, anhydrous DMF, 25 °C, 4 h; (b) (14, 17, 20, and 24),
iodomethane (6 equiv), TBAHS, 50% NaOH, dichloromethane, room temperature,
25 °C, 3 h; (c) (27) iodomethane (12 equiv), TBAHS, 50% NaOH, dichloromethane,
room temperature, 25 °C, 48 h; (d) lithium aluminum hydride, anhydrous THF,
reflux, overnight.
2.2. Biology
Bovine brain mitochondria isolated according to Basford13 were
used as source of the two MAO isoforms. Compounds 1–3, 7,8 and
88 were used as reference drugs. MAO-A and MAO-B activities
were determined by a fluorometric assay, using kynuramine as a
Supplementary data). The inhibitory activities (Ki) and A-selectiv-
ity (SI) of compounds 11–50 are depicted in Table 1.
All compounds were reversible inhibitors; the enzyme activity
(approx. 95–100%) was restored after 24 h dialysis at 5 °C (dialysis
was performed in the presence of 0.1 M potassium phosphate buf-
fer at pH 7.2). Because of the high affinity of the inhibitors, no dis-
sociation of the enzyme–inhibitor complex was detected during
substrate, in the presence of their specific inhibitors (
M for MAO-A, and clorgyline 1
M for MAO-B).14 The four final
concentrations ranged from 5 M to 0.1 mM. Dixon plots showed
that the inhibition was not competitive (an example is shown in
L-deprenyl
1
l
l
l
Cl
O
N
Cl
H
N
Cl
H
N
N
N
NH2
N
O
O
O
Cl
Moclobemide (2)
Selegiline (L-deprenyl) (3)
Clorgyline (1)
Lazabemide (4)
A-selective, reversible
B-selective, irreversible
B-selective, reversible
A-selective, irreversible
Reference Model
R2
N
O
N
R
N
N
O
N
R1
HO
HN
X
Ladostigil (5)
M-30 (6)
7: R1 = H, R2 = CH3;
8: R1 = CH3, R2 = H.
X = electron-rich zone
non-selective, irreversible
O
non-selective, irreversible
O
R2
11-50:
H
H
N
R4
N
N
R1-R3 = H, CH3;
N
R1
11-50
N
H
N
n
R
4 = Ph, cyclohexyl;
X
R3
n = 0-4; X = O, H2.
PF9601N (9)
10
Chart 1. Structure of references 1–10 and new MAO inhibitors 11–50.

G. La Regina et al. / Bioorg. Med. Chem. 16 (2008) 9729–9740
9731
Compound 22 is a potent and selective MAO-A inhibitor, and
derivative 24 is a potent but not selective inhibitor of the MAO-B.
Carboxamides 44 and 46 were potent but not selective MAO inhib-
itors. Worthy to note, these compounds are (S)-enantiomers, while
in the pyrrole series9 the highest inhibition was due to the corre-
sponding (R)-enantiomers. Reduction of carboxamides 44 and 46
to the corresponding amines 48 and 50 resulted in reduction of
both MAO-A and MAO-B inhibitory activity.
H
N
a
X
X
NH2
O
Y
Y
O
55-58
51-54
H
N
X
b
Y
2.4. Molecular modeling
59-62
X-ray crystal structures of rat MAO-A (PDB code: 1O5W)15 and
human MAO-B (PDB code: 1OJC)16,17 were used for docking exper-
iments.9 The rat, human and bovine sequences at the catalytic site
show high level of identity.18 In particular, the rat and bovine se-
quences of MAO-A are characterized by 85.3% of identity and
95.4% of homology at the binding site, while the human and bovine
sequences of MAO-B have 91.3% of identity and 97.1% of homology.
The residues of the active site are largely conserved across the
MAO isoforms and only one mutation is found at the catalytic site
of the human and bovine MAO-B (Ile199 is replaced by Phe199).
Consequently, in order to simulate the active site of the bovine
MAO-B, the 1OJC crystal structure was virtually mutated, replacing
the Ile199 with a Phe residue.
51, 55, 59: X = phenyl, Y = (CH2)3; 52, 55, 60: X =
phenyl, Y = ( CH2)4; 53, 57, 61: X = cyclohexyl, Y = (R)-
CH(CH3); 54, 58, 62: X = cyclohexyl, Y = (S)-CH(CH3).
Scheme 2. Synthesis of derivatives 59–82. Reagents and reaction: (a) ethyl
chloroformate, triethyamine, anhydrous THF, 25 °C, 2 h; (b) lithium aluminum
hydride, anhydrous THF, reflux, overnight.
the enzymatic activity assay, and the inhibition was apparently
irreversible. In these experimental conditions the substrate did
not compete with the inhibitor. Accordingly, a decrease of Vmax
was observed, while the Km value was unchanged.
Four water molecules, labeled as WAT23, WAT82, WAT102, and
WAT160 according to the numbering reported in the human MAO-
B crystallographic structure (PDB code: 1OJC),16,17 were located
near the FAD cofactor, this latter being covalently bound to
Cys397. WAT102 and WAT23 were already detected in the crystal-
lographic analysis of hMAO-B:16,17 WAT160 was found below the
2.3. Discussion
The majority of tested compounds inhibited MAO-A at sub-
micromolar concentration, and compounds 12, 17, 19–22, 31, 36,
and 37 had Ki values in the nanomolar range. Several compounds
inhibited MAO-B at micromolar concentration, and five com-
pounds (14, 20, 24, 38, 44, and 46) had Kis in the low nanomolar
range. The selectivity indexes (SIs) ranged from 68,478 (22) to
0.034 (24).
FAD pyrimidine ring, fixed by the
p-systems of the aromatic side
chains of Tyr398 and Tyr435, as well as by the central heterocyclic
conjugated ring of FAD. WAT82 was positioned about 3.70 Å away
from the
a carbons of Ile198 and Gln206. Accordingly, we consid-
Generally, the carboxamides were potent MAO-A inhibitors.
N-methyl, N-(3-phenyl)propyl carboxamide (22) was found an
ered these water molecules as integral components of the protein
structure during the docking simulations.
exceptionally potent MAO-A inhibitor (Ki(MAO-A) = 0.00092
worthy to note, this compound was 8-fold more potent than corre-
sponding pyrrole analogue (Ki(MAO-A) = 0.007
M).8 Introduction
of a methyl group at position 1 of the indole of 22 provided 24, a
compound endowed with higher MAO-B inhibitory activity but
low selectivity. Reduction of the carbonyl functionality to a methy-
lene of either 22 or 24 resulted in reduction of anti-MAO-A activity
and selectivity (compounds 38 and 39). N-methylation of carbox-
amides 11, 18, and 21 improved the anti-MAO-A activity (compare
l
M);
Docking studies focused on the most active and/or selective
inhibitors 22, 24, 44, and 46, using the automated docking program
GOLD 3.1.19–24 GOLD provides accurate prediction of ligand bind-
ing conformation, while the GOLD scoring function (i.e., GOLD fit-
ness) is not parameterized for small molecule binding affinities.
Recently, Verdonk et al.25 reported GOLDscore as a modified GOLD
scoring function (is equal to the GOLD fitness minus the ligand
intramolecular terms). GOLDscore provides a better correlation
with the experimental binding affinities. Indeed, correlation statis-
tics from GOLDscore are comparable to those obtained from Chem-
score function, which is parameterized against experimental
binding affinities. In order to assess the reliability of predicted
binding models, the estimated GOLDscore values for the top-ranked
solutions of 22, 24, 44, and 46 were compared with the experimen-
tal MAO inhibitory activities (pKis). GOLDscore function imple-
mented in GOLD surprisingly approximated in a qualitative sense
the experimental binding affinities (Table 2).
l
11 with 12 (Ki(MAO-A) = 0.0035
lM), 18 with 19 (Ki(MAO-A) =
0.004 M), and 21 with 22 (Ki(MAO-A) = 0.00092
l
lM)). The anti-
MAO-A activity of 16 increased 12 times by introduction of a
methyl group at position 1 of the indole (compare 16 with 17
(Ki(MAO-A) = 0.006 lM)).
Reduction of the carbonyl functionality of 11–27 resulted in a
dramatic drop of MAO-A inhibition, and only 33 and 36 retained
the activity of the parent compounds 16 and 19, respectively. As
MAO-B inhibitor, amine 28 (Ki(MAO-B) = 0.32
as carboxamide 11 (Ki(MAO-B) = 0.23 M), and amines 36
(Ki(MAO-B) = 0.02 M) and 38 (Ki(MAO-B) = 0.025 M) were more
potent than carboxamides 19 and 22, respectively. Replacement of
the pyrrole nucleus of 7 with the indole (33) led to 35 times abate-
ment of anti-MAO-B activity and loss of selectivity.
l
M) was as active
To evaluate the dynamic stability of the predicted ligand/en-
zyme interactions, the complexes obtained from docking were sub-
mitted to molecular dynamics (MD) simulation for 600 ps at
constant temperature (300 K). The relative dynamic stability of
each ligand docked conformation was monitored by the root-
mean-squared deviations (rmsd) of the ligands relative to their
initial docked orientation, and by the enzyme–ligand H-bond
distance time series along the complete 600 ps MD trajectory.
Docking of compounds 22 (GOLDscore = 54.95 kJ/mol), 24
(GOLDscore = 53.31 kJ/mol), 44 (GOLDscore = 43.00 kJ/mol), and 46
(GOLDscore = 44.58 kJ/mol) to the rat MAO-A clearly showed one
prevailing pose into the active site (Fig. 1a–d). The top-ranking re-
sults from GOLD runs placed the indole ring of the ligands within
l
l
l
The most potent MAO-A (22, Ki(MAO-A) = 0.00092
lM) and
MAO-B (24, Ki(MAO-B) = 0.0015 M) inhibitors had the same
l
spacer group. Reference inhibitors 7 and 89 also have the same
spacer group which differs from the spacer group of compounds
22 and 24 in (i) linker chain length (propyl for indoles 22 and 24,
and methylene for pyrroles 7 and 8), (ii) functional group (indoles
22 and 24 are carboxamides, and pyrroles 7 and 8 are amines).

9732
G. La Regina et al. / Bioorg. Med. Chem. 16 (2008) 9729–9740
Table 1
Structures and monoamine oxidase inhibitory activities of derivatives 11–50
R2
N
R2
N
R2
N
*
N
R1
N
R1
N
R1
n
n
O
CH3
O
11-27
28-40
41
R2
N
R2
N
R2
N
*
*
*
N
R1
N
R1
N
O
CH3
CH3
CH3
R1
42
43-46
47-50
Compound
R1
R2
n
*
Ki (
lM) MAO-A
Ki (l
M) MAO-B
SIa
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
1b
H
H
CH3
CH3
H
H
CH3
H
H
CH3
H
H
CH3
H
CH3
H
CH3
CH3
H
CH3
CH3
H
CH3
H
CH3
H
CH3
CH3
H
CH3
H
CH3
H
CH3
CH3
H
CH3
CH3
CH3
CH3
H
0
0
0
0
1
1
1
2
2
2
3
3
3
3
4
4
4
0
0
0
0
1
1
1
2
2
2
3
3
4
—
—
—
—
——
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
R,S
R,S
R
0.13
0.0035
10
0.015
0.03
0.07
0.006
0.17
0.004
0.0065
0.005
0.00092
0.1
0.044
2.2
10
1.0
10
0.01
0.01
0.006
2.1
0.06
7.2
0.23
0.094
>100
0.021
1.8
26.8
>10
1.4
4.0
22.9
160.8
1.1
25
5.69
160
68478
>1000
0.034
4.5
>10
8.3
0.032
100.0
7500.0
1833.3
>47
11.6
1.3
>62.5
20.0
3571.4
0.34
10.4
>10
2
>10
29.8
1.3
131.4
0.27
590.0
0.5
0.5
0.12
1.6
0.96
0.19
0.1
0.037
0.8
H
63
>100
0.0015
10.0
>100
8.3
0.32
1.0
75.0
11.0
>100
0.7
9.7
>100
0.02
25
CH3
CH3
H
H
CH3
H
H
CH3
CH3
H
H
CH3
H
H
CH3
H
CH3
H
H
H
H
H
CH3
CH3
H
1.6
0.001
0.007
0.073
0.089
10
0.25
10
0.77
0.012
0.70
0.054
0.01
0.06
9.8
0.065
0.054
11.5
0.025
0.93
>100
0.5
>100
23
0.016
92
0.015
H
H
CH3
CH3
CH3
CH3
CH3
CH3
CH3
CH3
—
—
—
—
—
S
R
S
R
S
R
S
—
—
—
—
—
5.9
0.03
5.3
0.1
H
CH3
CH3
—
—
—
1.5
1074.1
>87
0.25
0.0057
12,500.0
58
>100
0.97
0.02
300
2b
3b
3.8
3.5
0.024
7b
—
—
8b
Data represent mean values for at least three separate experiments each performed in duplicate. Standard errors were within 2%.
a
SI, selectivity index = Ki(MAO-B)/Ki(MAO-A) ratio.
b
Lit.8.
an ‘aromatic cage’, so that it forms
p
–
p
stacking interactions with
compounds. The high MAO-A affinity and A/B selectivity of 22
might therefore be accounted by the formation of these two H-
bonds. Worthy to note, the binding mode of 24, 44, and 46 strongly
resembled that of 22, with the exception of the indole ring which
was rotated of 180° so that it projected the methyl group on the
opposite side of the cofactor (Fig. 1b and d).
Tyr407 and Tyr444 side chains, as well as with the isoalloxazine
FAD ring, while the aromatic or cyclohexyl rings were located in
a hydrophobic core delimited by residues Phe208, Val210, Ile325,
and Leu337. In particular, the aromatic ring of 22 and 24 estab-
lished a
p-stacking interaction with Phe208 (Fig. 1a and b). With
the only exception of 22, which formed two H-bonds with N5
nitrogen and C@O4 oxygen of FAD by its indole NH (Fig. 1a), GOLD
did not detect any H-bond in the enzymatic cleft for the remaining
To explore the dynamic stability of 22, 24, 44, and 46 in com-
plex with MAO-A, rmsd values for the enzyme Ca atoms during
the production phase (300 ps) relative to the starting structures

G. La Regina et al. / Bioorg. Med. Chem. 16 (2008) 9729–9740
9733
Table 2
during the entire MD simulation. These findings suggest that the
superior MAO-A inhibitory activity of 22 compared with 24, 44,
and 46 is regulated by its ability to form close and productive
interactions with FAD.
Docking scores of compounds 22, 24, 44, and 46 by GOLDscore in the GOLD software,
and experimental MAO-A and MAO-B inhibitory activities (pKi)
GOLDscoreb (kJ/mol)
pKi (lM)
a
Ligand
Enzyme
Ntot
The MAO-B active site consists of two cavities, the substrate
cavity in front of the flavin and the entrance cavity located under-
neath the protein surface and closed by the loop formed by resi-
dues 99–112. Compounds 22, 24, 44, and 46 easily fit into the
MAO-B active site with two prevailing binding modes (Fig. 3a–d).
The best scored docking poses of 22 (GOLDscore = 22.00 kJ/mol),
44 (GOLDscore = 33.45 kJ/mol) and 46 (GOLDscore = 29.98 kJ/mol)
occupied both the cavities (Fig. 3a, c, and d).
22
24
44
46
MAO-A
MAO-A
MAO-A
MAO-A
5
6
9
3
54.95
53.31
43.00
44.58
9.0
8.8
7.9
7.3
22
24
44
46
MAO-B
MAO-B
MAO-B
MAO-B
4
10
4
22.00
43.56
33.45
29.98
4.2
8.8
7.8
7.8
5
a
Ntot is the total number of docking poses generated by GOLD after 20 runs.
GOLDscore represents the GOLD fitness minus intramolecular terms.25
In particular, the indole system was hosted in the entrance
cavity made up by lipophilic residues Pro104, Phe168, Leu164,
Leu171, Phe199, Ile326, whereas the phenyl or cyclohexyl rings
occupied the substrate cavity and were in direct contact with
the isoalloxazine FAD ring, Tyr398, Tyr435, Phe343, and Tyr60.
Worthy to note, the phenyl ring of 22, differently from that
of 44 and 46, adapted itself just underneath the enzyme ‘aro-
matic cage’ and seemed unable to establish any charge-transfer
interaction with the enzyme. All three compounds established a
H-bond with Gln206 NH2 by their amide C@O oxygen. In addi-
tion, compounds 22 and 44 had their NH indole involved in a
H-bond with Tyr326 OH oxygen, whereas compounds 44 and
46 engaged a H-bond with WAT82 by their amide C@O oxygen.
As far as regarded compound 24, GOLD detected a different
binding mode (GOLDscore = 43.56 kJ/mol), which closely resem-
bled that of 22, 24, 44, and 46 into the MAO-A binding cleft
(Fig. 3b). The indole was in front of the FAD isoalloxazine ring,
while the phenyl ring was lodged in a hydrophobic pocket
framed by residues Phe168, Trp119, Phe103, Pro104, Ile316,
and Phe199; no intermolecular H-bonds between ligand and en-
zyme were observed.
b
were calculated. The rmsd plots (data not shown) indicated that
the conformations of the four complexes achieved equilibrium
after about 100 ps and that their average rmsd were 1.60 Å,
1.65 Å, 1.70 Å, and 1.75 Å, respectively. This behavior reveals that
the overall architectures of the macromolecular complexes were
preserved for the whole duration of the simulations. Moreover,
the H-bonds between the indole NH of 22 and N5 nitrogen and
C@O4 oxygen of FAD were maintained throughout the MD simula-
tion with a frequency of formation of 87% and 65%, respectively
(Fig. 2). A weak H-bond between the amide C@O oxygen of 22
and Gln215 NH was also observed (47%).
MD trajectories of 24 and 44 showed two stable H-bonds
formed between the C@O oxygen of 24 and the OH hydrogen of
Tyr298 (82%), and between the indole NH of 44 and the C@O oxy-
gen of Asn181 (70%). No intermolecular H-bonds were detected by
analyzing the MD trajectory of 46/MAO-A complex. The p–p stack-
ing interactions involving the ligand phenyl and benzo-fused in-
dole rings and Phe208, Tyr444, and Tyr407 were also preserved
Figure 1. Binding modes of compounds 22 (cyan, a), 24 (magenta, b), 44 (spring green, c) and 46 (orange, d) into the MAO-A binding cavity. For clarity, only interacting
residues are displayed. Ligands, FAD cofactor (yellow) and interacting key residues (white) are represented as stick models, while the enzyme (violet) as ribbons. The van der
Waals volume of FAD is displayed as a transparent yellow surface. H-bonds are shown as green lines.

9734
G. La Regina et al. / Bioorg. Med. Chem. 16 (2008) 9729–9740
ligand was mainly stabilized by hydrophobic and
interactions: the N-methyl indole moiety was embedded in the
‘aromatic cage’, where it established stacking and T-shaped
interactions with Phe343, Tyr60, and Tyr435; the phenyl moiety
made a strong interaction with Phe199, which turned out
p
Àp stacking
pÀp
pÀp
very stable throughout the whole MD simulation.6b Although
the H-bond between the indole NH of 44 and Tyr326 OH oxygen
was broken during the simulation, both 44 and 46 preserved
their H-bonds with both Gln206 and WAT82. In contrast, 22/
MAO-B complex showed a remarkably different behavior. In fact,
during the MD simulation, 22 was highly unstable, and moved
away from the FAD cofactor after a few hundreds of ps. The
most important interactions with the enzyme residues failed
and the final Ca atom rmsd values were equal to 3.1 Å. This im-
plies that compound 22 compared with 24, 44, and 46 undergoes
larger conformational changes after complex formation, which
reflects its very weak MAO-B binding affinity.
3. Conclusions
New monoamine oxidase inhibitors were synthesized as indole
analogues of a previously reported pyrrole series. Several com-
pounds were potent MAO-A (12, 17, 19–22, 24, 31, 36, and 37) or
MAO-B (14, 20, 24, 38, 44, and 46) inhibitors, and had Ki values
in the nanomolar concentration range. In particular, 22
Figure 2. MD snapshot of 22/MAO-A complex. Only aminoacids located within a
5 Å distance of the bound ligand are displayed and labeled. Carbon atoms of 22 and
cofactor FAD are cyan and yellow, respectively. The van der Waals volume of FAD is
displayed as a transparent yellow surface. The hydrogen bonds discussed in the text
are depicted as dashed green lines.
(Ki = 0.00092 lM, and SI = 68,478) was exceptionally potent and
selective as MAO-A inhibitor. Molecular modeling studies showed
that compounds 22, 24, 44, and 46 positioned the indole ring into
an ‘aromatic cage’ of MAO-A, and formed
p p stacking interac-
À
MD studies of 22, 24, 44, and 46/MAO-B complexes were car-
ried out in order to assess their dynamic stability. The average
tions with Tyr407 and Tyr444 side chains, as well as with the iso-
alloxazine FAD ring. MD simulations showed that compound 22
formed stable H-bonds with FAD, whereas compounds 24 and 44
were involved in H-bonds with Tyr298 and Asn181, respectively.
The potent MAO-A inhibitory activity of 22 compared with 24,
rmsd of the backbone Ca atoms with respect to the initial struc-
ture revealed a quite stable behavior in the 24/MAO-B (1.2 Å),
44/MAO-B (1.0 Å), and 46/MAO-B (1.1 Å) complexes. Analysis of
the MD trajectories of 24/MAO-B complex showed that the
Figure 3. Binding modes of compounds 22 (cyan, a), 24 (magenta, b), 44 (spring green, c) and 46 (orange, d) into the mutated MAO-B binding cavity. For clarity, only
interacting residues are displayed. Ligands, FAD cofactor (yellow) and interacting key residues (white) are represented as stick models, while the enzyme (green) as ribbons.
Structural water molecules are represented as red balls. The van der Waals volume of FAD is displayed as a transparent yellow surface. H-bonds are shown as yellow lines.

G. La Regina et al. / Bioorg. Med. Chem. 16 (2008) 9729–9740
9735
44, and 46 was then regulated by its ability to establish close and
productive interactions with FAD. Compounds 22, 24, 44, and 46
fitted into the MAO-B active site with two prevailing binding
modes. For 22, 44, and 46 the indole nucleus was hosted in the en-
trance cavity made up by lipophilic residues, whereas the phenyl
or cyclohexyl rings occupied the substrate cavity which is directly
in contact with the isoalloxazine FAD ring. Compound 22 was
highly unstable during the MD simulation into the MAO-B, thus
accounting for its low MAO-B affinity.
4.1.7. N-(2-Phenylethyl)-1H-indole-2-carboxamide (18)
Prepared as 11 using 2-phenylethylamine. Yield 77%, mp 188–
190 °C (from ethanol). 1H NMR (DMSO-d6): d 2.85 (t, J = 7.6 Hz,
2 H), 3.48–3.53 (m, 2H), 6.99–7.03 (m, 1H), 7.06–7.07 (m, 1H),
7.13–7.20 (m, 2H), 7.23–7.30 (m, 4H), 7.39–7.41 (dd, J = 0.82 and
7.1 Hz, 1H), 7.58 (d, J = 8.0 Hz, 1H), 8.56 (br s, disappeared on treat-
ment with D2O, 1H), 11.53 ppm (br s, disappeared on treatment
with D2O, 1H). IR:
m .
1630, 3257 cmÀ1
4.1.8. N-Methyl,N-(2-phenylethyl)-1H-indole-2-carboxamide
(19)
4. Experimental
4.1. Chemistry
Prepared as 11 (RS 2450) using N-methyl-N-(2-phenyl-
ethyl)amine. Yield 66%, mp 125–128 °C (from ethanol). 1H NMR
(CDCl3): d 3.04 (s, 2H), 3.30 (s, 3H), 3.89 (s, 2H), 6.82 (s, 1H),
7.12–7.16 (m, 1H), 7.24–7.35 (m, 6H), 7.45 (d, J = 8.0 Hz, 1H),
7.66 (d, J = 8.0 Hz, 1H), 9.82 ppm (br s, disappeared on treatment
4.1.1. Materials and methods
Melting points (mp) were determined on a Büchi 510 apparatus
and are uncorrected. Infrared spectra (IR) were acquired with Per-
kin-Elmer 1310 and SpectrumOne spectrophotometers. Band posi-
tion and absorption ranges are given in cmÀ1. Proton nuclear
magnetic resonance (1H NMR) spectra were recorded on Bruker
AM-200 (200 MHz) and Bruker Avance 400 MHz FT spectrometers
in the indicated solvent. Chemical shifts are expressed in d units
(ppm) from tetramethylsilane. Column chromatographies were car-
ried out with alumina (Merck, 70–230 mesh) and silica gel (Merck,
70–230 mesh). Aluminum oxide TLC cards (Fluka, aluminum oxide
precoated aluminum cards with fluorescent indicator at 254 nm)
and silica gel TLC cards (Fluka, silica gel precoated aluminum cards
with fluorescent indicator at 254 nm) were used for thinlayer
chromatography (TLC). Developed plates were visualized with a
Spectroline ENF 260C/F UV apparatus. Organic solutions were dried
over anhydrous sodium sulfate. Evaporation of the solvents was
carried out on Büchi Rotavapor R-210 equipped with Büchi V-850
vacuum controller and Büchi V-700 (ꢀ5 mbar) and V-710 (ꢀ2 mbar)
vacuum pumps. Elemental analyses of compounds 11–50 were
found within 0.4% of the theoretical values (Table 1S in Supplemen-
tary data).
with D2O, 1H). IR:
m .
1596, 3247 cmÀ1
4.1.9. N-(3-Phenylpropyl)-1H-indole-2-carboxamide (21)
Prepared as 11 using 3-phenylpropylamine. Yield 76%, mp 176–
178 °C (from ethanol). 1H NMR (DMSO-d6): d 1.83–1.86 (m, 2H),
2.62–2.66 (m, 2H), 3.28–3.33 (m, 2H), 7.02 (t, J = 7.3 Hz, 1H), 7.11
(d, J = 2.0 Hz, 1H), 7.14–7.19 (m, 2H), 7.22–7.30 (m, 4H), 7.42 (d,
J = 8.2 Hz, 1H), 7.59 (d, J = 8.0 Hz, 1H), 8.49 (t, J = 5.6 Hz, disap-
peared on treatment with D2O, 1H), 11.55 ppm (br s, disappeared
on treatment with D2O, 1H). IR:
m .
1619, 3289 cmÀ1
4.1.10. N-Methyl,N-(3-phenylpropyl)-1H-indole-2-carboxamide
(22)
Prepared as 11 using 59. Yield 84%, mp 147–150 °C (from etha-
nol). 1H NMR (CDCl3): d 2.06 (s, 2H), 2.41–2.75 (m, 3H), 3.19–3.39
(m, 2H), 3.71 (s, 2H), 7.11–7.63 (m, 10H), 9.74 ppm (br s, disap-
peared on treatment with D2O, 1H). IR:
m .
1619, 3289 cmÀ1
4.1.11. N-(3-Phenylpropyl)-1-methyl-1H-indole-2-carboxamide
(23)
Prepared as 11 using 1-methyl-1H-indole-2-carboxylic acid and
3-phenylpropylamine. Yield 75%, mp 80–84 °C (from ethanol). 1H
NMR (DMSO-d6): d 1.83–1.87 (m, 2H), 2.65 (t, J = 7.7 Hz, 2H),
3.26–3.31 (m, 2H), 3.97 (s, 3H), 7.07–7.31 (m, 8H), 7.51 (d,
J = 8.1 Hz, 1H), 7.62 (d, J = 7.8 Hz, 1H), 8.52 ppm (br s, disappeared
4.1.2. General procedure for the synthesis of compounds 11–13,
15, 16, 18, 19, 21–23, 25, 26, 41, and 43–46. Example: N-Phenyl-
1H-indole-2-carboxamide (11)
m .
1624, 3297 cmÀ1
BOP reagent (8.22 g, 0.0186 mol) was added to a mixture of in-
dole-2-carboxylic acid (3.00 g, 0.0186 mol), aniline (3.46 g,
3.38 mL, 0.0372 mol) and triethylamine (5.64 g, 7.77 mL,
0.056 mol) in anhydrous DMF (20 mL). The reaction mixture was
stirred at 25 °C for 4 h, diluted with water and extracted with ethyl
acetate. The organic layer was washed with 1 N HCl, brine and
dried. Evaporation of the solvent furnished 11 (4.02 g, 91%), mp
190–192 °C (from ethanol). Lit.26 192–193 °C.
on treatment with D2O, 1H). IR:
4.1.12. N-(4-Phenylbutyl)-1H-indole-2-carboxamide (25)
Prepared as 11 using 4-phenylbutylamine. Yield 85%, mp 119–
121 °C (from ethanol). 1H NMR (DMSO-d6): d 1.43–1.61 (m, 5H),
2.49–2.66 (m, 3H), 7.00 (t, J = 7.5 Hz, 1H), 7.14–7.19 (m, 5H), 7.23–
7.27 (m, 2H), 7.40 (d, J = 8.4 Hz, 1H), 7.57 (d, J = 8.0 Hz, 1H), 8.45
(br s, disappeared on treatment with D2O, 1H), 11.52 ppm (br s, dis-
appeared on treatment with D2O, 1H). IR:
m .
1626, 3272 cmÀ1
4.1.3. N-Methyl,N-phenyl-1H-indole-2-carboxamide (12)
Prepared as 11 using N-methylaniline. Yield 65%, mp 175–
177 °C (from ethanol). Lit.27 173 °C.
4.1.13. N-Methyl,N-(4-phenylbutyl)-1H-indole-2-carboxamide
(26)
Prepared as 11 using 60. Yield 62% mp 131–133 °C (from etha-
nol). 1H NMR (DMSO-d6): d 1.58–1.62 (m, 4H), 2.60 (s, 3H), 3.27–
3.56 (m, 4H), 6.81–6.90 (m, 1H), 7.03 (t, J = 7.5 Hz, 1 H), 7.14–
7.19 (m, 5H), 7.25 (t, J = 7.4 Hz, 1H), 7.42 (d, J = 8.3 Hz, 1H), 7.60
(d, J = 7.9 Hz, 1H), 11.53 ppm (br s, disappeared on treatment with
4.1.4. N-Phenyl-1-methyl-1H-indole-2-carboxamide (13)
Prepared as 11 using 1-methyl-1H-indole-2-carboxylic acid and
aniline. Yield 61%, mp 149–151 °C (from ethanol). Lit.27 147 °C.
D2O, 1H). IR:
m .
1599, 3260 cmÀ1
4.1.5. N-Benzyl-1H-indole-2-carboxamide (15)
Prepared as 11 using benzylamine. Yield 88%, mp 222–224 °C
(from ethanol). Lit.28 220 °C.
4.1.14. (R,S)-N-(
a
-Phenylethyl)-1H-indole-2-carboxamide (41)
-phenylethylamine. Yield 50%, mp
Prepared as 11 using (R,S)-
a
176–178 °C (ethanol). 1H NMR (DMSO-d6): d 1.65 (d, J = 7.0 Hz,
3H), 5.32–5.39 (m, 1H), 7.17 (t, J = 7.5 Hz, 1H), 7.32 (t, J = 8.1 Hz,
1H), 7.37 (t, J = 7.2 Hz, 1H), 7.41 (s, 1H), 7.47 (t, J = 7.6 Hz, 2H),
7.55–7.58 (m, 3H), 7.76 (d, J = 8.0 Hz, 1H), 8.94 (br s, disappeared
4.1.6. N-Benzyl,N-methyl-1H-indole-2-carboxamide (16)
Prepared as 11 using N-methylbenzylamine. Yield 70%, mp
195–198 °C (from ethanol). The quoted 1H NMR spectral data
was in agreement with values reported in Lit.29

9736
G. La Regina et al. / Bioorg. Med. Chem. 16 (2008) 9729–9740
on treatment with D2O, 1H), 11.68 ppm (br s, disappeared on treat-
ment with D2O, 1H). IR:
1645, 3252, 3410 cmÀ1
6.60 (s, 1H), 7.15–7.38 (m, 8H), 7.62 ppm (d, J = 7.9 Hz, 1H).
IR:
1624 cmÀ1
m
.
m
.
4.1.15. (R)-N-(a-Cyclohexylethyl),N-methyl-1H-indole-2-
carboxamide (43)
4.1.23. N-Methyl,N-(4-phenylbutyl)-1-methyl-1H-indole-2-
carboxamide (27)
It was prepared as 11 using 61 Yield 60%, mp 150–154 °C (from
ethanol). 1H NMR (CDCl3): d 0.94–1.19 (m, 6H), 1.28–1.42 (m, 2H),
1.48–1.71 (m, 6H), 2.86 (s, 1H), 3.16 (s, 2H), 4.51–4.52 (m, 1H), 6.79
(s, 1H), 7.04 (t, J = 7.4 Hz, 1H), 7.19 (t, J = 7.8 Hz, 1H), 7.34 (d,
J = 8.3 Hz, 1H), 7.57 (d, J = 8.0 Hz, 1H), 9.29 ppm (br s, disappeared
Iodomethane (4.25 g, 1.9 mL, 0.03 mol) was added to an ice-
cooled mixture of 22 (1.50 g, 0.005 mol), tetrabutylammonium
hydrogen sulfate (1.7 g, 0.005 mol), dichlorometane (25 mL) and
50% NaOH solution (16.6 mL). After stirring at 25 °C for 24 h, iodo-
methane (4.25 g, 1.9 mL, 0.03 mol) was added and the reaction
mixture was stirred for additional 24 h. Water was added while
stirring and the mixture extracted with chloroform. The organic
layer was separated, washed with brine and dried. Removal of
the solvent gave a residue which was purified by silica gel column
chromatography (ethyl acetate/n-hexane = 3:1 as eluent) to fur-
nish 27 (1.10 g, 69%) as an oil. 1H NMR (CDCl3): d 1.55–1.73 (m,
4H), 2.61–2.72 (m, 2H), 3.13 (s, 3H), 3.59–3.62 (m, 2H), 3.82 (s,
3H), 6.60 (s, 1H), 7.13–7.23 (m, 4H), 7.30–7.36 (m, 3H), 7.39 (d,
on treatment with D2O, 1H). IR:
m .
1590, 3249 cmÀ1
4.1.16. (S)-N-( -Cyclohexylethyl),N-methyl-1H-indole-2-
a
carboxamide (44)
Prepared as 11 using 62. Yield 42%, mp 160–163 °C (from etha-
nol). 1H NMR (CDCl3): d 0.86–1.39 (m, 6H), 1.40–1.49 (m, 2H),
1.63–1.71 (m, 6H), 2.96 (s, 1H), 3.24 (s, 2H), 4.42–4.67 (m, 1H),
6.61 (s, 1H), 7.01 (t, J = 7.5 Hz, 1H), 7.15 (t, J = 7.5 Hz, 1H), 7.40
(d, J = 8.3 Hz, 1H), 7.58 (d, J = 7.9 Hz, 1H), 9.74 ppm (br s, disap-
J = 7.0 Hz, 1H), 7.64 ppm (d, J = 7.9 Hz, 1H). IR:
m .
1623 cmÀ1
peared on treatment with D2O, 1H). IR:
m .
1590, 3249 cmÀ1
4.1.24. General Procedure for the Synthesis of Compounds 28–
31, 33, 34, 36–40, 42, and 47–50. Example. N-(1H-Indol-2-
ylmethyl)-N-phenylamine (28)
4.1.17. (R)-N-( -Cyclohexylethyl)-N-methyl-1-methyl-1H-
indole-2-carboxamide (45)
a
Prepared as 11 using 1-methyl-1H-indole-2-carboxylic acid and
61. Yield 59%, oil. 1H NMR (CDCl3): d 0.67–0.73 (m, 1H), 1.02–1.31
(m, 7H), 1.42–1.82 (m, 6H), 2.97 (s, 3H), 3.83 (s, 3H), 3.96–4.54 (m,
1H), 6.53–6.62 (m, 1H), 7.16–7.19 (m, 1H), 7.28–7.33 (m, 1H),
A solution of 11 (2.00 g, 0.008 mol) in anhydrous THF (30 mL)
was added dropwise to an ice-cooled suspension of LiAlH4
(1.14 g, 0.03 mol) in the same solvent (38 mL). The reaction mix-
ture was refluxed overnight. After cooling on an ice-bath the mix-
ture was made basic with 10% NaOH solution and extracted with
ethyl acetate. The organic layer was washed with brine and dried.
Evaporation of the solvent furnished pure 28 (1.76 g, 99%). Mp 59–
62 °C (from cyclohexane). 1H NMR (DMSO-d6): d 4.36 (s, 2H), 6.05
(br s, disappeared on treatment with D2O, 1H), 6.31 (s, 1H), 6.53 (t,
J = 7.2 Hz, 1H), 6.65 (d, J = 7.8 Hz, 2H), 6.92 (t, J = 7.4 Hz, 1H), 6.98–
7.07 (m, 3H), 7.30 (d, J = 8.1 Hz, 1H), 7.41 (d, J = 7.8 Hz, 1H),
7.38–7.41 (m, 1H), 7.65–7.67 ppm (m, 1H). IR:
m .
1621 cmÀ1
4.1.18. (S)-N-(
a-Cyclohexylethyl)-N-methyl-1-methyl-1H-
indole-2-carboxamide (46)
Prepared as 11 using 1-methyl-1H-indole-2-carboxylic acid and
62. Yield 59%, oil. 1H NMR (CDCl3): d 0.64–0.71 (m, 1H), 0.99–1.27
(m, 7H), 1.45–1.82 (m, 6H), 2.94 (s, 3H), 3.80 (s, 3H), 3.93–4.52 (m,
1H), 6.51–6.59 (m, 1H), 7.13–7.16 (m, 1H), 7.26–7.28 (m, 1H),
11.00 ppm (br s, disappeared on treatment with D2O, 1H). IR:
m
7.36–7.38 (m, 1H), 7.62–7.64 ppm (m, 1H). IR:
m .
1622 cmÀ1
3054, 3402 cmÀ1
.
4.1.19. General procedure for the synthesis of compounds 14,
17, 20, 24. Example. N-Methyl, N-phenyl-1-methyl-1H-indole-2-
carboxamide (14)
4.1.25. N-(1H-Indol-2-ylmethyl)-N-methyl-N-phenylamine (29)
Prepared as 28 using 12. Yield 74%, oil. 1H NMR (DMSO-d6): d
3.03 (s, 3H), 4.65 (s, 2H), 6.20 (s, 1H), 6.52–6.55 (m, 1H), 6.65 (t,
J = 7.6 Hz, 1H), 6.83 (d, J = 8.1 Hz, 1H), 6.94 (t, J = 7.8 Hz, 1H), 7.03
(t, J = 7.8 Hz, 1H), 7.17 (t, J = 8.0 Hz, 2H), 7.33 (d, J = 8.0 Hz, 1H),
7.42 (d, J = 7.4 Hz, 1H), 10.99 ppm (br s, disappeared on treatment
Iodomethane (1.68 g, 0.74 mL, 0.012 mol) was added to an ice-
cooled mixture of 12 (0.32 g, 0.002 mol), tetrabutylammonium
hydrogen sulfate (0.68 g, 0.002 mol), dichlorometane (10 mL),
and 50% NaOH solution (6.66 mL). The reaction was stirred at
25 °C for 3 h. Water was added while stirring and the mixture ex-
tracted with dichlorometane. The organic layer was separated,
washed with brine, and dried to furnish 11 (0.76 g, 58%), mp 80–
82 °C (from ethanol). Lit.27 84 °C.
with D2O, 1H). IR:
m .
3399 cmÀ1
4.1.26. N-(1-Methyl-1H-indol-2-ylmethyl)-N-phenylamine (30)
Prepared as 28 using 13. Yield 43%, mp 76–78 °C (ethanol). 1H
NMR (DMSO-d6): d 3.75 (s, 3H), 4.43 (d, J = 5.5 Hz, 2H), 6.08 (br s,
disappeared on treatment with D2O, 1H) 6.41 (s, 1H), 6.54–6.58
(m, 1H), 6.72 (d, J = 6.8 Hz, 2H), 7.00 (t, J = 7.5 Hz, 1H), 7.03–7.13
(m, 3H), 7.41 (d, J = 8.2 Hz, 1H), 7.48 ppm (br s, disappeared on
4.1.20. N-Benzyl,N-methyl-1-methyl-1H-indole-2-carboxamide
(17)
Prepared as 14 using 16. Yield 90%, oil. 1H NMR (CDCl3): d 3.12
(s, 3H), 3.90 (s, 3H), 4.84 (s, 2H), 6.67 (s, 1H), 7.14 (t, J = 7.3 Hz, 1H),
treatment with D2O, 1H). IR:
m .
3399 cmÀ1
7.25–7.42 (m, 7H), 7.60 ppm (s, 1H). IR:
m .
1625 cmÀ1
4.1.27. N-(1-Methyl-1H-indol-2-ylmethyl)-N-methyl-N-
phenylamine (31)
4.1.21. N-Methyl,N-(2-phenylethyl)-1-methyl-1H-indole-2-
carboxamide (20)
Prepared as 28 using 14. Yield 62%, mp 131–136 °C (n-hexane).
1H NMR (CDCl3): d 2.91 (s, 3 H), 3.70 (s, 3H), 4.59 (s, 2H), 6.39 (s,
1H), 6.80 (t, J = 7.2 Hz, 1H), 6.90 (d, J = 8.3 Hz, 2H), 7.10 (t,
J = 7.5 Hz, 1H), 7.19 (t, J = 6.7 Hz, 1H), 7.25–7.32 (m, 3H), 7.55 (d,
J = 7.8 Hz, 1H).
Prepared as 14 using 19. Yield 44%, mp 94–96 °C (from ethanol).
1H NMR (CDCl3): d 2.92–3.09 (m, 5H), 3.55–3.76 (m, 3H), 3.85 (t,
J = 7.1 Hz, 2H), 6.52 (s, 1H), 7.02–7.34 (m, 8H), 7.61 (d, J = 7.9 Hz,
1H). IR:
m .
1619 cmÀ1
4.1.28. N-Benzyl-N-(1H-indol-2-ylmethyl)-N-methylamine (33)
Prepared as 28 using 16. Yield 45%, oil. 1H NMR (CDCl3): d 2.25
(s, 3H), 3.61 (s, 2H), 3.74 (s, 2H), 6.39 (s, 1H), 7.09 (t, J = 8.0 Hz, 1H),
7.2 (t, J = 8.0 Hz, 1H), 7.25–7.32 (m, 1H), 7.34–7.38 (m, 5H), 7.57 (d,
4.1.22. N-Methyl,N-(3-phenylpropyl)-1-methyl-1H-indole-2-
carboxamide (24)
Prepared as 14 using 22. Yield 97%, oil. 1H NMR (CDCl3): d 2.04
(m, 2H), 2.55–2.73 (m, 2H), 3.16 (s, 3H), 3.67 (s, 2H), 3.84 (s, 3H),

G. La Regina et al. / Bioorg. Med. Chem. 16 (2008) 9729–9740
9737
J = 7.8 Hz, 1H), 8.69 ppm (br s, disappeared on treatment with D2O,
1H). IR:
3130 cmÀ1
8.47 ppm (br s, disappeared on treatment with D2O, 1H). IR:
3458 cmÀ1
m
m
.
.
4.1.29. N-Benzyl-N-(1-methyl-1H-indol-2-ylmethyl)-N-
methylamine (34)
4.1.37. (S)-N-(
ylmethyl)amine (48)
a-Cyclohexylethyl)-N-(1H-indol-2-
Prepared as 28 using 17. Yield 85%, oil. 1H NMR (CDCl3): d 2.23
(s, 3H), 3.60 (s, 2H), 3.71 (s, 2H), 3.83 (s, 3H), 6.46 (s, 1H), 7.13 (t,
J = 7.4 Hz, 1H), 7.2 (t, J = 8.2 Hz, 1H), 7.35–7.38 (m, 6H), 7.62 ppm
(d, J = 7.9 Hz, 1H).
Prepared as 28 using 44. Yield 91%, oil. 1H NMR (CDCl3): d 0.83–
0.94 (m, 5H), 1.11–1.44 (m, 4H), 1.66–1.80 (m, 4H), 2.13 (s, 3H),
2.18–2.22 (m, 1H), 2.33–2.37 (m, 1H), 3.65 (d, J = 14.3 Hz, 1H),
3.76 (d, J = 14.2 Hz, 1H), 6.31 (s, 1 H), 7.08 (t, J = 7.2 Hz, 1H), 7.14
(t, J = 8.0 Hz, 1H), 7.34 (d, J = 7.1 Hz, 1H), 7.53 (d, J = 7.8 Hz, 1H),
4.1.30. N-(1H-Indol-2-ylmethyl)-N-methyl-N-(2-
phenethyl)amine (36)
8.47 ppm (br s, disappeared on treatment with D2O, 1H). IR:
m
3458 cmÀ1
.
Prepared as 28 using 19. Yield 45%, oil. 1H NMR (CDCl3): d 2.36
(s, 3H), 2.67 (t, J = 7.2 Hz, 2 H), 2.81 (t, J = 7.1 Hz, 2H), 3.70 (s, 2H),
6.29 (s, 1H), 7.05–7.20 (m, 5H), 7.26–7.32 (m, 3H), 7.51 (d,
J = 7.7 Hz, 1H), 8.1 ppm (br s, disappeared on treatment with
4.1.38. (R)-N-(
a-Cyclohexylethyl)-N-(1H-indol-2-ylmethyl)-N-
methylamine (49)
Prepared as 28 using 45. Yield 89%, oil. 1H NMR (CDCl3): d
0.71–0.78 (m, 2H), 0.90–0.93 (m, 3H), 1.04–1.36 (m, 5H), 1.56–
1.71 (m, 4H), 2.03 (s, 3H), 2.31–2.37 (m, 1H), 3.63–3.65 (m,
2H), 3.74 (s, 3H), 6.31 (s, 1H), 6.98 (t, J = 7.4 Hz, 1H), 7.09 (t,
J = 7.6 Hz, 1H), 7.38 (d, J = 8.2 Hz, 1H), 7.46 ppm (d, J = 7.8 Hz,
1H).
D2O, 1H). IR:
m .
3412 cmÀ1
4.1.31. N-(1-Methyl-1H-indol-2-ylmethyl)-N-methyl-N-(2-
phenethyl)amine (37)
Prepared as 28 using 20. Yield 68%, oil. 1H NMR (CDCl3): d 2.31
(s, 3H), 2.73 (t, J = 7.8 Hz, 2H), 2.83 (t, J = 7.2 Hz, 2H), 3.62 (s, 3H),
3.67 (s, 2H), 6.39 (s, 1H), 7.10 (t, J = 7.8 Hz, 1H), 7.16 (d,
J = 8.5 Hz, 2H), 7.20–7.23 (m, 3H), 7.25–7.31 (m, 2H), 7.59 ppm
(d, J = 7.3 Hz, 1H).
4.1.39. (S)-N-(a-Cyclohexylethyl)-N-(1H-indol-2-ylmethyl)-N-
methylamine (50)
Prepared as 28 using 46. Yield 78%, oil. 1H NMR (CDCl3): d
0.70–0.75 (m, 2H), 0.93–0.95 (m, 3H), 1.10–1.31 (m, 5H), 1.61–
1.74 (m, 4H), 2.06 (s, 3H), 2.33–2.37 (m, 1H), 3.64–3.73 (m,
2H), 3.78 (s, 3H), 6.33 (s, 1H), 7.08 (t, J = 8.1 Hz, 1H), 7.18 (t,
J = 8.1 Hz, 1H), 7.29 (d, J = 74 Hz, 1H), 7.56 ppm (d, J = 7.8 Hz,
1H).
4.1.32. N-(1H-Indol-2-ylmethyl)-N-methyl-N-(2-
phenylpropyl)amine (38)
Prepared as 28 using 22. Yield 52%, oil. 1H NMR (CDCl3): d 1.84–
1.92 (m, 2H), 2.25 (s, 3H), 2.50 (t, J = 7.4 Hz, 2H), 2.67 (t, J = 7.7 Hz,
2H), 3.68 (s, 2H), 6.36 (s, 1H), 7.11 (t, J = 7.9 Hz, 1H), 7.16–7.24 (m,
4H), 7.29–7.36 (m, 3H), 7.58 (d, J = 7.8 Hz, 1H), 8.65 ppm (br s, dis-
4.1.40. General procedure for the synthesis of compounds 55–
60. Example: ethyl N-(3-phenylpropyl)carbamate (55)
Ethyl chloroformate (0.78 g, 0.67 mL, 0.0072 mol) in anhydrous
THF (15 mL) was added to a solution of 3-phenylpropylamine
(0.98 g, 1.04 mL, 0.0072 mol) and triethylamine (0.73 g, 1.00 mL,
0.0072 mol) in the same solvent (38 mL). The reaction mixture
was stirred at 25 °C for 2 h. Water and ethyl acetate were added
and the organic layer was separated, washed with brine, and dried.
Evaporation of the solvent furnished pure 55 (1.42 g, 95%) as an oil.
The quoted 1H NMR spectral data was in agreement with values re-
ported in Lit.30
appeared on treatment with D2O, 1H). IR:
m .
3414 cmÀ1
4.1.33. N-(1-Methyl-1H-indol-2-ylmethyl)-N-methyl-N-(2-
phenylpropyl)amine (39)
Prepared as 28 using 24. Yield 74%, oil. 1H NMR (CDCl3): d 1.81–
1.89 (m, 2H), 2.24 (s, 3H), 2.47 (t, J = 7.1 Hz, 2H), 2.64 (t, J = 7.8 Hz,
2H), 3.64 (s, 2H), 3.84 (s, 3H), 6.37 (s, 1H), 7.09–7.26 (m, 4H), 7.22–
7.35 (m, 3H), 7.33 (d, J = 8.2 Hz, 1H), 7.59 ppm (d, J = 7.8, 1H).
4.1.34. N-(1H-Indol-2-ylmethyl)-N-(4-phenylbutyl)amine (40)
Prepared as 28 using 25. Yield 47%, oil. 1H NMR (CDCl3): d 1.48–
1.54 (m, 2H), 1.56–1.62 (m, 2H), 2.54 (t, J = 7.5 Hz, 2H), 2.63 (t,
J = 7.1 Hz, 2H), 3.93 (s, 2H), 6.32 (s, 1H), 6.95 (t, J = 7.4 Hz, 1H),
7.04 (t, J = 7.5 Hz, 1H), 7.14–7.18 (m, 3H), 7.24–7.28 (m, 2H), 7.33
(d, J = 7.9 Hz, 1H), 7.45 (d, J = 7.8 Hz, 1H), 8.33 (br s, disappeared
on treatment with D2O, 1H), 11.07 ppm (br s, disappeared on treat-
4.1.41. Ethyl N-(3-phenylbutyl)carbamate (56)
Prepared as 55 using 4-phenylbutylamine. Yield 96%, oil. 1H
NMR: d 1.24 (t, J = 7.1 Hz, 3H), 1.49–1.54 (m, 2H), 1.61–1.68
(m, 2H), 2.63 (t, J = 7.4, 2H), 3.18 (q, J = 6.5 Hz, 2H), 4.11 (q,
J = 7.1 Hz, 2H), 4.61 (br s, disappeared on treatment with D2O,
ment with D2O, 1H). IR:
m
3414 cmÀ1
.
1H), 7.15–7.20 (m, 3H), 7.25–7.29 ppm (m, 2H). IR:
m 1710,
2993, 3338 cmÀ1
.
4.1.35. (R,S)-N-(1H-Indol-2-ylmethyl)-N-(a-phenylethyl)amine
(42)
4.1.42. Ethyl (R)-(
Prepared as 55 using (R)-
mp 46–48 °C (from cyclohexane). Lit.31 49–50 °C.
a
-cyclohexylethyl)carbamte (57)
Prepared as 28 using 41. Yield 47%, oil. 1H NMR (DMSO-d6): d
1.27 (d, J = 6.6 Hz, 3H), 3.35 (br s, disappeared on treatment with
D2O, 1H), 3.57–3.73, (m, 2H), 3.71 (q, J = 6.6 Hz, 1H), 6.18 (s, 1H),
6.89–6.93 (m, 1H), 6.97–7.01 (m, 1H), 7.22–7.24 (m, 1H), 7.29–
7.37 (m, 5H), 7.40 (d, J = 7.7 Hz, 1H), 10.86 ppm (br s, disappeared
a-cyclohexylethylamine. Yield 89%,
4.1.43. 4.1.43.Ethyl (S)-(a-cyclohexylethyl)carbamate (58)
Prepared as 55 using (S)-a-cyclohexylethylamine. Yield 89%, mp
on treatment with D2O, 1H). IR:
m
3410, 3415 cmÀ1
.
46–48 °C (from cyclohexane). Lit.31 49–50 °C.
4.1.36. (R)-N-(
ylmethyl)amine (47)
a
-Cyclohexylethyl)-N-(1H-indol-2-
4.1.44. N-(3-Phenylpropyl)-N-methylamine (59)
Prepared as 28 using 55. Yield 84%, oil. The quoted 1H NMR
spectral data was in agreement with values reported in Lit.32
Prepared as 28 using 43. Yield 96%, oil. 1H NMR (CDCl3): d 0.83–
0.96 (m, 5H), 1.15–1.44 (m, 4H), 1.66–1.81 (m, 4H), 2.13 (s, 3H),
2.18–2.21 (m, 1H), 2.32–2.39 (m, 1H), 3.65 (d, J = 14.1 Hz, 1H),
3.77 (d, J = 14.1 Hz, 1H), 6.31 (s, 1H), 7.07 (t, J = 7.1 Hz, 1H), 7.12
(t, J = 7.9 Hz, 1H), 7.35 (d, J = 7.1 Hz, 1H), 7.53 (d, J = 7.8 Hz, 1H),
4.1.45. N-(3-Phenylbutyl)-N-methylamine (60)
Prepared as 28 using 56. Yield 81%, oil. The quoted 1H NMR
spectral data was in agreement with values reported in Lit.33

9738
G. La Regina et al. / Bioorg. Med. Chem. 16 (2008) 9729–9740
4.1.46. (R)-N-(
a
-Cyclohexylethyl)-N-methylamine (61)
4.3. Molecular modeling
Prepared as 28 using 57. Yield 59%, oil. The quoted 1H NMR
spectral data was in agreement with values reported in Lit.34
4.3.1. Computational chemistry
Molecular modeling and graphics manipulations were per-
formed using Molecular Operating Environment (MOE)36 and
UCSF-CHIMERA37 software packages, running on a Silicon Graphics
Tezro R16000 workstation. Energy minimizations and MD simula-
tions were realized by employing the AMBER 9 program,38 select-
ing the parm99 force field.39
4.1.47. (S)-N-(a-Cyclohexylethyl)-N-methylamine (62)
Prepared as 28 using 58. Yield 58%, oil. The quoted 1H NMR
spectral data was in agreement with values reported in Lit.35
4.2. Biology
4.3.2. Ligand and protein setup
4.2.1. Mitochondria preparation
Model building and geometry optimizations of compounds
22, 24, 44, and 46 were accomplished with the MMFF94X force
field40–44 available within MOE. The crystal structure of rat
MAO-A (entry code: 1O5W)15 and human MAO-B (entry code:
1OJC)16,17 recovered from Brookhaven Protein Database45 were
used for the docking experiments. The choice of using 1OJC
structure was guided by the quality of the crystallographic data
and the fact that N-(2-aminoethyl)-p-chlorobenzamide was (at
the time of the simulation) the only cocrystallized noncovalent
ligand crossing the entire binding site of MAO-B. As a conse-
quence, the side chain of key residue Ile199 is oriented such
as the ‘entrance’ and ‘substrate’ cavities are fused. To simulate
the active site of the bovine MAO-B, the 1OJC model, containing
the four conserved water molecules (WAT23, WAT82, WAT102,
and WAT160), was virtually mutated replacing the Ile199 with
a Phe residue. The side chain of Phe199 was rotated into the
‘open’ conformation in such a way to allow for larger inhibitors
to span both cavities, as Hubàlek and coworkers suggested.46,47
Hydrogen atoms were added to both 1O5W and 1OJC in agree-
ment with the ionization state existing at physiological pH. Par-
tial atomic charges were computed by MOE using the Amber99
force field. Since the X-ray structures might contain residual
energetic tensions from the crystallization process, both struc-
tures were subjected to a preliminary constrained energy mini-
mization of those residues out of a radius of 15 Å from the N5
of the isoalloxazine ring in order to restore the natural planarity
of the FAD ring and relax the active site amino acids. Subse-
quently, the resulting energy-minimized structures were de-
prived of the covalent ligands (clorgyline for 1O5W and N-(2-
aminoethyl)-p-chlorobenzamide for 1OJC) and used as starting
models for docking simulations.
Mitochondria were prepared according to Basford.13 Re-
agents: Medium A contained 0.4 M sucrose, 0.001 EDTA, 0.02%
polyethersulfone (PES) or heparin and pH was adjusted to
6.8–7.0 with KOH; Medium F made was made up of the Med-
ium A to which Ficoll was added to a final concentration of 8%.
Calf or beef brains were removed from the animals within 5–
10 min after their death. The brains were immediately placed
in cold Medium A, stored on ice, and then transported to the
laboratory. In a cold room, at 5 °C, the cerebral hemispheres
were removed from the brains and the meninges were taken
up with forceps. The gray matter was scraped from the cortices
using a dull spatula. Two brains yield corresponded to about
100 g of wet tissue, which was homogenized in Medium A
(2 mL/g of wet tissue). The homogenate was kept at pH 7.0
by adding some drops of Tris-buffer 2 M; 1 mg of
e-aminoca-
proic acid/g of tissue was added and then the mixture was stir-
red at 0–4 °C for 15 min. The suspension was diluted with
Medium A (20 mL/g of the original tissue), centrifuged first at
184g for 20 min and then at 1153g for 20 min, without transfer-
ring of the supernatant. The residue R1 was discarded while the
supernatant S1 was centrifuged at 12,000g for 15 min, to yield a
crude mitochondria pellet R2 (the supernatant S2 which is dis-
carded). The fraction R2 was dissolved in Medium F (6 mL/g
of original tissue), gently homogenized and centrifuged at
12,000g for additional 30 min. The resulting mitochondria frac-
tion R3 was washed using 4 mL of Medium A/g of original tis-
sue and again centrifuged at 12,000g for 15 min. The final
mitochondria fraction R4, was homogenized in potassium phos-
phate buffer pH 7.4, 0.25 M. The yield of mitochondria protein
obtained was between 100 and 140 mg per 50 g wet weight
of the original tissue.
4.3.3. Docking simulations
4.2.2. Activity assay
Monoamine oxidase activity was determined using kinur-
amine as a substrate, at four different final concentrations rang-
Docking of 22, 24, 44, and 46 to both MAO-A and MAO-B was
performed with GOLD 3.1,19,20 which uses a genetic algorithm
for determining the docking modes of ligands and proteins. An
advantage of GOLD over other docking methods is the program’s
ability to account for some rotational protein flexibility, as well
as full ligand flexibility. Specifically, OH groups of Ser, Thr, and
Tyr, and amino groups of Lys are allowed to rotate during dock-
ing to optimize H-bonding to the ligand. GOLD requires a user-
defined binding site. It searches for a cavity within the defined
area, and considers all the solvent-accessible atoms in that area
as active site atoms. On the basis of the GOLD score, for each
molecule a bound conformation with high score was considered
as the best bound conformation. The score function that was
implemented in GOLD consisted basically of H-bonding, complex
energy, and ligand internal energy terms. A population of possi-
ble docked orientations of the ligand is set up at random. Each
member of the population is encoded as a ‘chromosome’, which
contains information about the mapping of ligand H-bond atoms
onto (complementary) protein H-bond atoms, mapping of hydro-
phobic points of the ligand onto protein hydrophobic points, and
the conformation around flexible ligand bonds and protein OH
groups. A number of parameters control the precise operation
ing from 5 lM to 0.1 mM, by a sensitive fluorometric assay
according to Matsumoto et al.14 In all assays the incubation
mixtures contained: potassium phosphate buffer, pH 7.4, mito-
chondria (6 mg/mL), drug solutions in DMSO, added to the reac-
tion mixture at a final concentration ranging from 0 to 10À9
.
Solutions were preincubated for 30 min before adding the sub-
strate and then incubated for others 30 min. The inhibitory
activities against MAO-A and MAO-B were determined at
38 °C, after incubation of the mitochondrial fractions for
30 min in the presence of the specific inhibitor
(L-deprenyl
(1 M) or clorgyline (1 M) to estimate the MAO-A and MAO-
l
l
B activity, respectively). The addition of percloric acid ended
the reaction. Then the samples were centrifuged at 10,000g
for 5 min and the supernatant was added to 2.7 mL NaOH 1N.
Fluorometric measurements were recorded at kexc. 317 nm and
kem 393 nm using a Perkin-Elmer LS 50B spectrofluorometer.
Dixon plot were used to estimate the inhibition constant (Ki)
of the inhibitors. Data are the means of three or more experi-
ments each performed in duplicate.

G. La Regina et al. / Bioorg. Med. Chem. 16 (2008) 9729–9740
9739
of the genetic algorithm. Fifty independent docking runs were
Acknowledgments
performed for each docking experiment, using standard default
settings with a population size of 100, a maximum number of
100,000 operations, and a mutation and crossover rate of 95.
The protein input file may be the entire protein structure or a
part of it comprising only the residues that are in the region
of the ligand binding site. In the present study, GOLD was
allowed to calculate interaction energies within a sphere of a
13 Å radius centered on the phenolic oxygen atom of Tyr444
and Tyr435 in MAO-A and MAO-B, respectively. After docking,
the 20 individual binding poses of each ligand were re-ranked
according to the GOLDscore (details in the text). The top-ranked
conformation of each ligand was selected and the corresponding
GOLDscore was then correlated with the inhibition constant
pKi(ÀlogKi).
This work was partially supported by the Italian MIUR (Ministe-
ro dell’Istruzione, Università e Ricerca) and by funds MIUR-PRIN
(Cofin) (EA).
Supplementary data
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.bmc.2008.09.072.
References and notes
1. Mondovì, B. Structure and Function of Amine Oxidases; CRC Press: Boca Raton, FL,
1985.
2. (a) Abell, C. W.; Kwan, S.-W. Prog. Nucl. Acid Res. Mol. Biol. 2001, 65, 129–156;
(b) Grimsby, J.; Lan, N. C.; Neve, R.; Chen, K.; Shih, J. C. J. Neurochem. 1990, 55,
1166–1169; (c) Tipton, K. F.; Boyce, S.; O’Sullivan, J.; Davey, G. P.; Healey, J.
Curr. Med. Chem. 2004, 11, 1965–1982.
3. (a) Cesura, A. M.; Pletscher, A. Prog. Drug Res. 2002, 38, 171–298; (b) Yamada,
M.; Yasuhara, H. Neurotoxicology 2004, 25, 215–221.
4. (a) Drukarch, B.; van Muiswinkel, F. L. Biochem. Pharm. 2000, 59, 1023–1031;
(b) Schapira, A. H. V. Arch. Neurol. 2007, 64, 1083–1088.
5. Boyer, E. W.; Shannon, M. N. Engl. J. Med. 2005, 352, 1112–1120.
6. (a) Sagi, Y.; Drigues, N.; Youdim, M. B. H. Br. J. Pharm. 2005, 146, 553–560; (b)
Youdim, M. Y. B.; Bakhle, Y. S. Br. J. Pharm. 2006, 147, S287–S296.
7. Avramovich-Tirosh, Y.; Amit, T.; Bar-Am, O.; Zheng, H.; Fridkin, M.; Youdim, M.
B. H. J. Neurochem. 2007, 100, 490–502.
8. Silvestri, R.; La Regina, G.; De Martino, G.; Artico, M.; Befani, O.; Palumbo, M.;
Agostinelli, E.; Turini, M. J. Med. Chem. 2003, 46, 917–920.
9. La Regina, G.; Silvestri, R.; Artico, M.; Lavecchia, A.; Novellino, E.; Befani, O.;
Turini, P.; Agostinelli, E. J. Med. Chem. 2007, 50, 922–931.
10. Perez, V.; Unzeta, M. Neochem. Int. 2003, 42, 221–229.
11. Sanz, E.; Romera, M.; Belyk, L.; Marco, J. I.; Unzeta, M. Med. Sci. Monitor 2004,
10, BR477–BR484.
12. Morón, J. A.; Campillo, M.; Perez, V.; Unzeta, M.; Pardo, M. J. Med. Chem. 2000,
43, 1684–1691.
13. Basford, R. E. Methods Enzymol. 1967, 10, 96–101.
14. Matsumoto, T.; Suzuki, O.; Furuta, T.; Asai, M.; Kurokawa, Y.; Rimura, Y.;
Katsumata, Y.; Takahashi, I. Clin. Biochem. 1985, 18, 126–129.
15. Ma, J.; Yoshimura, M.; Yamashita, E.; Nakagawa, A.; Ito, A.; Tsukihara, T. J. Mol.
Biol. 2004, 338, 103–114.
16. Binda, C.; Newton-Vinson, P.; Hubalek, F.; Edmondson, D. E.; Mattevi, A. Nat.
Struct. Biol. 2002, 9, 22–26.
17. Binda, C.; Li, M.; Hubalek, F.; Restelli, N.; Edmondson, D. E.; Mattevi, A. Proc.
Natl. Acad. Sci. U.S.A. 2003, 100, 9750–9755.
18. Database searching (SWISS-PROT), sequence alignment, and analysis of rat,
bovine and human MAO sequences were carried out using FASTA Pearson, W.
R. Proc. Natl. Acad. Sci. U.S.A. 1988, 85, 2444–2448; and BLAST programs Wang,
S.; Pak, Y. J. Phys. Chem. B 2000, 104, 354–359.
19. GOLD 3.1; CCDC Software Limited: Cambridge, UK, 2004.
20. Jones, G.; Willett, P.; Glen, R. C.; Leach, A. R.; Taylor, R. J. Mol. Biol. 1997, 267,
727–748.
21. Schulz-Gasch, T.; Stahl, M. J. Mol. Model. 2003, 9, 47–57.
22. Wang, R.; Lu, Y.; Wang, S. J. Med. Chem. 2003, 46, 2287–2303.
23. Kellenberger, E.; Rodrigo, J.; Muller, P.; Rognan, D. Proteins: Struct. Funct. Bioinf.
2004, 57, 225–242.
24. Warren, G. L.; Andrews, C. W.; Capelli, A. M.; Clarke, B.; LaLonde, J.; Lambert, M.
H.; Lindvall, M.; Nevins, N.; Semus, S. F.; Senger, S.; Tedesco, G.; Wall, I. D.;
Woolven, J. M.; Peishoff, C. E.; Head, M. S. J. Med. Chem. 2006, 49, 5912–5931.
25. Verdonk, M. L.; Cole, J. C.; Hartshorn, M. J.; Murray, C. W.; Taylor, R. D. Proteins
2003, 52, 609–623.
26. Mahboobi, S.; Teller, S.; Pongratz, H.; Hufsky, H.; Sellmer, A.; Botzki, A.; Uecker,
A.; Beckers, T.; Baasner, S.; Schaechtele, C.; Ueberall, F.; Kassack, M. U.; Dove, S.;
Boehmer, F. D. J. Med. Chem. 2002, 45, 1002–1018.
27. Caubere, C.; Caubere, P.; Renard, P.; Bizot-Espiart, J. G.; Ianelli, S.; Nardelli, M.;
Jamart-Gregoire, J. Tetrahedron 1994, 50, 13433–13448.
28. Pigulla, J.; Roeder, E. Arch. Pharm. 1979, 312, 12–18.
29. Ellames, G. J.; Gibson, J. S.; Herbert, J. M.; McNeill, A. H. Tetrahedron 2001, 57,
9487–9497.
30. Inesi, A.; Mucciante, V.; Rossi, L. J. Org. Chem. 1998, 63, 1337–1338.
31. McGahren, W. J.; Kunstmann, M. P. J. Org. Chem. 1972, 37, 902–906.
32. Kuehne, M. E.; Shannon, P. J. J. Org. Chem. 1977, 42, 2082–2087.
33. Ding, C. Z.; Lu, X.; Nishimura, K.; Silverman, R. B. J. Med. Chem. 1993, 36, 1711–
1715.
34. Willoughby, C. A.; Buchwald, S. L. J. Am. Chem. Soc. 1994, 116, 8952–8965.
35. Verdaguer, X.; Lange, U. E. W.; Reding, M. T.; Buchwald, S. L. J. Am. Chem. Soc.
1996, 118, 6784–6785.
All the parameters were used as GOLD default values, and the
ligands were submitted to 20 genetic algorithm runs. The program
was set to allow early termination if the top 3 solutions are within
1.0 Å rmsd of each other, which we feel would indicate that the
poses have reached an optimum based on the fitness function.
The four water molecules detected in the MAO-B binding site were
used in the docking studies. GOLD automatically determined
whether they should be bound or displaced by toggling it on and
off during the docking run, and, furthermore, the orientation of
the water hydrogen atoms were optimized by GOLD during dock-
ing. After docking, the predicted binding poses of each ligand were
re-ranked according to the GOLDscore (details in the text). Finally,
the top-ranked conformation of each ligand was used for further
studies.
4.3.4. Molecular dynamics simulations
All MD simulations were performed using the SANDER mod-
ule in the AMBER suite of programs, employing the Cornell
et al. force field39 to assign parameters for the standard amino
acids. General AMBER force field (GAFF) parameters were as-
signed to ligands, while the partial charges were calculated
using the AM1-BCC method as implemented in the ANTECHAM-
BER suite of AMBER. The complexes were soaked in a box of
TIP3P48 water molecules with a margin of 10 Å along each
dimension. An appropriate number of counterions were added
to neutralize the system. The entire system was then minimized
for 5000 steps using combined steepest descent and conjugate
gradient methods until
a convergence value of 0.001 kcal/
mol Å. Upon minimization, harmonic constraints of 2 kcal/Å2/
mol on the protein backbone atoms of the complexes were ap-
plied. Such a energy minimization was designed to resolve the
clashes between the complex and solvent molecules gradually.
The complexes were then subjected to a two stage equilibration
to further relax the protein and the surrounding solvent. In the
first stage, the systems were heated from 0 to 300 K (using the
Langevin dynamics method) over 50 ps of simulation time. We
performed this stage of equilibration with the volume held con-
stant. In the second stage, the systems were equilibrated for
50 ps using pressure and temperature control to adjust the den-
sity of water to experimental values. A subsequent production
run was performed at a constant temperature of 300 K and a
constant volume, giving a total simulation time of 500 ps. The
time step of the simulations was 2.0 fs with a cut-off of 8 Å
for the no bonded interaction, and SHAKE49 was employed to
keep all bonds involving hydrogen atoms rigid. Coordinates
were saved every 1 ps and used to calculate the averaged struc-
tures from the simulations. The averaged structures over the
last 300 ps of the simulations were energy-minimized as previ-
ously described and store as the final conformation of the li-
gand–enzyme complexes.
36. Molecular Operating Environment (MOE), version 2005.06; Chemical
Computing Group, Inc.: Montreal, Canada, 2005.

9740
G. La Regina et al. / Bioorg. Med. Chem. 16 (2008) 9729–9740
37. Huang, C. C.; Couch, G. S.; Pettersen, E. F.; Ferrin, T. E. Pac. Symp. Biocomput.
1996, 1, 724
45. Bernstein, F. C.; Koetzle, T. F.; Williams, G. J. B.; Meyer, E. F., Jr.; Brice, M. D.;
Rodgers, J. R.; Kennard, O.; Shimanouchi, T.; Tasumi, T. J. Mol. Biol. 1977, 112,
535–542.
46. Hubálek, F.; Binda, C.; Khalil, A.; Li, M.; Mattevi, A.; Castagnoli, N.; Edmondson,
D. E. J. Biol. Chem. 2005, 16, 15761–15766.
47. Authors demonstrated that the binding of two inhibitors, rasagiline and isatin,
to human MAO B I199F mutant enzyme is identical to that of WT enzyme. In
both cases, the side chain of Phe199 extends into the entrance cavity with
essentially identical conformations rather than the ‘open/closed’
conformations seen with Ile199. In the same paper, the authors pointed out
that the presence of this bulky residue in the entrance cavity does not appear
to alter the kinetic properties of the enzyme or the binding affinity of a small
competitive inhibitor.
38. Case, D. A.; Darden, T. A.; Cheatham, T. E., III; Simmerling, C. L.; Wang, J.;
Duke, R. E.; Luo, R.; Merz, K. M.; Pearlman, D. A.; Crowley, M.; Walker, R.
C.; Zhang, W.; Wang, B.; Hayik, S.; Roitberg, A.; Seabra, G.; Wong, K. F.;
Paesani, F.; Wu, X.; Brozell, S.; Tsui, V.; Gohlke, H.; Yang, L.; Tan, C.;
Mongan, J.; Hornak, V.; Cui, G.; Beroza, P.; Mathews, D. H.; Schafmeister,
C.; Ross, W. S.; Kollman, P. AAMBER 9; University of California: San
Francisco, 2006.
39. Cornell, W. D.; Cieplak, P.; Bayly, C. I.; Gould, I. R.; Merz, K. M., Jr.; Ferguson, D.
M.; Spellmeyer, D. C.; Fox, T.; Caldwell, J. W.; Kollman, P. A. J. Am. Chem. Soc.
1995, 117, 5179–5197.
40. Halgren, T. A. J. Comput. Chem. 1996, 17, 490–512.
41. Halgren, T. A. J. Comput. Chem. 1996, 17, 520–552.
42. Halgren, T. A. J. Comput. Chem. 1996, 17, 553–586.
48. Jorgensen, W. L.; Chandrasekhar, J.; Madurs, J.; Impey, R. W.; Klein, M. L. J.
Chem. Phys. 1983, 79, 926–935.
43. Halgren, T. A. J. Comput. Chem. 1996, 17, 587–615.
44. Halgren, T. A. J. Comput. Chem. 1996, 17, 616–641.
49. Ryckaert, J. P.; Ciccotti, G.; Berendsen, H. J. C. J. Comput. Phys. 1977, 23, 327–
333.