Synthesis, molecular modeling studies, and selective inhibitory activity against monoamine oxidase of 1-thiocarbamoyl-3,5-diaryl-4,5-dihydro-(1H)- pyrazole derivatives

Article abstract of DOI:10.1021/jm040903t

A novel series of 1-thiocarbamoyl-3,5-diaryl-4,5-dihydro-(1H)-pyrazole derivatives have been synthesized and investigated for the ability to inhibit selectively the activity of the A and B isoforms of monoamine oxidase (MAO). All the synthesized compounds show high activity against both the MAO-A and the MAO-B isoforms with K_i values between 27 and 4 nM and between 50 and 1.5 nM, respectively, except for a few derivatives whose inhibitory activity against MAO-B was in the micromolar range. Knowing that stereochemistry may be an important modulator of biological activity, we performed the semipreparative Chromatographic enantioseparation of the most potent, selective, and chiral compounds. The separated enantiomers were then submitted to in vitro biological evaluation. The selectivity of the (-)-(S)-1 enantiomer against MAO-B increases twice and a half, while the selectivity of the (-)-(S)-4 enantiomer against MAO-A triples. Both the MAO-A and MAO-B isoforms respectively of the 1O5W and IGOS models deposited in the Protein Data Bank were considered in the computational study. The docking study was carried out using several computational approaches with the aim of proposing possible binding modes of the MAO enantioselective compounds 1 and 4.

Full text of DOI:10.1021/jm040903t

J. Med. Chem. 2005, 48, 7113-7122
7113
Synthesis, Molecular Modeling Studies, and Selective Inhibitory Activity
against Monoamine Oxidase of 1-Thiocarbamoyl-3,5-diaryl-4,5-dihydro-(1H)-
pyrazole Derivatives
Franco Chimenti,† Elias Maccioni,‡ Daniela Secci,*,† Adriana Bolasco,† Paola Chimenti,† Arianna Granese,†
Olivia Befani,§ Paola Turini,§ Stefano Alcaro,# Francesco Ortuso,# Roberto Cirilli,| Francesco La Torre,|
Maria C. Cardia,‡ and Simona Distinto‡
Dipartimento di Studi di Chimica e Tecnologia delle Sostanze Biologicamente Attive, Universita` degli Studi di Roma “La
Sapienza”, P.le A. Moro 5, 00185 Rome, Italy, Dipartimento di Scienze Biochimiche “A. Rossi Fanelli” and Centro di Biologia
Molecolare del CNR Universita` degli Studi di Roma “La Sapienza”, P.le A. Moro 5, 00185 Rome, Italy, Dipartimento di
Scienze Farmaco Biologiche “Complesso Ninı` Barbieri”, Universita` degli Studi di Catanzaro “Magna Graecia”, 88021
Roccelletta di Borgia (CZ), Italy, Laboratorio di Chimica del Farmaco, Istituto Superiore di Sanita`, Viale Regina Elena 299,
00161 Rome, Italy, and Dipartimento Farmaco Chimico Tecnologico, Universita` degli Studi di Cagliari, Via Ospedale 72,
09124 Cagliari, Italy
Received December 14, 2004
A novel series of 1-thiocarbamoyl-3,5-diaryl-4,5-dihydro-(1H)-pyrazole derivatives have been
synthesized and investigated for the ability to inhibit selectively the activity of the A and B
isoforms of monoamine oxidase (MAO). All the synthesized compounds show high activity
against both the MAO-A and the MAO-B isoforms with Ki values between 27 and 4 nM and
between 50 and 1.5 nM, respectively, except for a few derivatives whose inhibitory activity
against MAO-B was in the micromolar range. Knowing that stereochemistry may be an
important modulator of biological activity, we performed the semipreparative chromatographic
enantioseparation of the most potent, selective, and chiral compounds. The separated
enantiomers were then submitted to in vitro biological evaluation. The selectivity of the (-)-
(S)-1 enantiomer against MAO-B increases twice and a half, while the selectivity of the (-)-
(S)-4 enantiomer against MAO-A triples. Both the MAO-A and MAO-B isoforms respectively
of the 1O5W and 1GOS models deposited in the Protein Data Bank were considered in the
computational study. The docking study was carried out using several computational approaches
with the aim of proposing possible binding modes of the MAO enantioselective compounds 1
and 4.
Introduction
common substrates for MAO-A and MAO-B. The two
MAO isoforms are composed of 527 and 520 amino acids,
respectively, with 70% of the amino acid identity.4 Both
isoforms contain the FAD cofactor covalently linked to
a cysteine residue in the active site.5 The catalytic
mechanism of MAOs has been proposed with two
different mechanisms: one radical, with the iminium
cation as intermediate, and the other polar nucleo-
philic.6
MAOs are important in the metabolism of monoamine
neurotransmitters, and as a result, MAO inhibitors
(MAOI) are studied for the treatment of several psy-
chiatric and neurological diseases. MAO-B inhibitors are
coadjuvant in the treatment of Parkinson’s disease7 and
also Alzheimer’s disease.8 MAO-A inhibitors are used
as antidepressant and antianxiety drugs.9
In our previous studies, we focused attention on the
importance of the presence of a 1-acetyl group in the
pyrazoline nucleus for MAO inhibitory activity.10,11
Other authors recently evaluated some 1-N-substi-
tuted thiocarbamoyl-3-phenyl-5-thienyl-2-pyrazolines
and found MAO inhibitory activity in the 450-22 µM
range.12
Our knowledge of the chemistry and the MAO inhibi-
tory activity of pyrazoline derivatives prompted us to
study the influence of the substitution of the 1-acetyl
group with a 1-unsubstituted thioamide one. The dipole
Monoamine oxidases (MAO) are widely distributed
enzymes among mammals, plants, and prokaryotic and
eukaryotic microrganisms that catalyze oxidatively
amines to aldehydes. They are divided into two classes:
aminoxidases containing copper(II) (EC, 1.4.3.6) and
aminoxidases containing flavin (EC, 1.4.3.4).1 The former
aminoxidases contain copper(II) 2,4,5-trihydroxyphen-
ylalanine quinone as cofactor (TPQ-Cu) and are inhib-
ited by semicarbazide.2 The latter aminoxidases contain
flavin adenine dinucleotide (FAD) as cofactor and exist
as two isoforms, MAO-A and MAO-B, which differ in
substrate specificity and sensitivity to inhibitors. MAO-A
preferentially deaminates serotonin and norepinephrine
and is selectively inhibited by clorgyline, whereas
MAO-B preferentially deaminates â-phenylethylamine
and benzylamine and is selectively inhibited by L-
deprenil.3 Dopamine, tyramine, and tryptamine are
* To whom correspondence should be addressed. Phone and fax: +39
06 4991 3763. E-mail: daniela.secci@uniroma1.it.
† Dipartimento di Studi di Chimica e Tecnologia delle Sostanze
Biologicamente Attive, Universita` degli Studi di Roma “La Sapienza”.
‡ Universita` degli Studi di Cagliari.
§ Dipartimento di Scienze Biochimiche “A. Rossi Fanelli” and Centro
di Biologia Molecolare del CNR Universita` degli Studi di Roma “La
Sapienza”.
# Universita` degli Studi di Catanzaro “Magna Graecia”.
| Istituto Superiore di Sanita`.
10.1021/jm040903t CCC: $30.25 © 2005 American Chemical Society
Published on Web 10/20/2005

7114 Journal of Medicinal Chemistry, 2005, Vol. 48, No. 23
Chimenti et al.
Scheme 1a
Direct Analytical and Semipreparative HPLC
Enantioseparations of Samples 1 and 4. The enan-
tiomers of racemic compounds 1 and 4 were separated
on the analytical and the semipreparative scale by
enantioselective HPLC on an amylose phenylcarbamate-
based chiral stationary phase (CSP), the Chiralpak AD
CSP.18 Baseline analytical enantioseparations (resolu-
tion factor Rs > 6) were achieved within 25 min by using
binary mixture n-hexanes-ethanol 20:80 (v/v) or pure
ethanol as mobile phases. The temperature column of
25 °C and flow rates of 1.0 and 0.5 mL min-1 were
utilized. The sign of the optical rotation at a wavelength
of 365 nm of the two enantiomers of 1 and 4 was
determined by on-line polarimetric detection during
HPLC enantioseparations on AD CSP. The second
eluting enantiomers of 1 and 4 were dextrorotatory in
both normal phase and polar organic conditions.
For the semipreparative runs, an amount of 5 mg of
racemic sample was injected onto a 10 mm i.d. Chiral-
pak AD CSP by using pure ethanol as the mobile phase.
After semipreparative separations, the collected frac-
tions were analyzed by an analytical Chiralpak AD
column (250 mm × 4.6 mm i.d.) to determine their
enantiomeric excess (ee). Enantiomeric excess values
greater than 99.0% and yields ranging from 70% to 90%
for both enantiomers of compounds 1 and 4 were
obtained.
a Reagents and conditions: (a) KOH, EtOH, H2O; (b) EtOH,
thiosemicarbazide, KOH.
moments of many compounds indicate that the CdS
group is more polarizable than the CdO group because
of the larger kernel of electrons in the S atom.13
Therefore, the presence of an unsubstituted thiocar-
bamoyl group on the N1 position of the pyrazoline
nucleus appears to be important for the inhibitory
activity against both MAO isoforms and its importance
is comparable with that of the acetyl group previously
attested.10,11 For this reason we prepared and assayed
for the MAO-A and -B inhibitory activities the new
1-thiocarbamoyl-3,5-diaryl-4,5-dihydro-(1H)-pyrazole de-
rivatives 1-10.
Owing to the presence of a chiral center at the C5
position of the pyrazole moiety and knowing that
stereochemistry may be an important modulator of
biological activity,14 we performed the semipreparative
chromatographic enantioseparation of the most potent,
selective, and active chiral compounds. To explain the
selectivity of these enantiomers, a molecular modeling
study was carried out using different docking ap-
proaches with the Protein Data Bank (PDB) enzymatic
models of both MAO isoforms.
The specific rotations of the isolated enantiomers
measured at 589 nm were as follows: (+)-1, [R]23D +241
(c 0.10, CHCl3); (-)-1m [R]23D -228 (c 0.10, CHCl3); (+)-
4, [R]23D +208 (c 0.17, CHCl3); (-)-4, [R]23D -208 (c 0.19,
CHCl3). The (+)-enantiomer of the compounds examined
was the more retained isomer on the polysaccharide-
based AD CSP, and no inversion of elution order
resulted in the scale-up procedure.
Stereochemical Characterization of the Enan-
tiomers of 1 and 4. Taking into account the specific
rotations of enantiomers of the structurally related
chiral compound 1-acetyl-3-(4-hydroxyphenyl)-5-(2-clo-
rophenyl)-4,5-dihydro-(1H)-pyrazole (Figure 1, com-
pound b), we used the sign of optical rotation as an
experimental criterion for the configurational assign-
ment of the stereogenic center C5 located at the pyrazole
moiety.19 The assignment of the absolute configuration
of enantiomers of compounds 1 and 4 could be empiri-
cally established according to their chiroptical and
chromatographic properties. As above-reported, the on-
line polarimetric measurements at 365 nm furnished a
positive sign for the optical rotation for second eluting
enantiomers of 1 and 4 both in ethanol and in n-hex-
anes-ethanol 20:80 (v/v). For the corresponding less
retained enantiomers the sign of the rotations was
inverted, as expected. Off-line polarimetric analysis at
589 nm indicated that the sign of rotation of polarized
light was unmodified in chloroform solution (Figure 1,
compound b).
Chemistry
To verify the effects of structural modifications on
both inhibition and selectivity toward MAO-A and
MAO-B, the substituted 4,5-dihydro-(1H)-pyrazoles 1-10
have been synthesized. In particular, the influence, on
the biological behavior, of the introduction of different
aromatic rings in the 3 and 5 positions of the dihydro-
(1H)-pyrazole nucleus has been investigated.
The starting 1,3-diaryl-2-propen-1-ones (chalcones)
A-L have been synthesized according to literature
methods.15-17
The 1-thiocarbamoyl-3,5-diaryl-4,5-dihydro-(1H)-pyra-
zoles 1-10 were synthesized by reacting 1 equiv of
chalcone, 2 equiv of thiosemicarbazide, and KOH in
ethanol, as depicted in Scheme 1 and listed in Table 1.
By this method products 1-10 where synthesized
with yields ranging from 48% to 58%. Crude products
where crystallized from ethanol/isopropyl alcohol. The
structures of the compounds were fully characterized
by means of both mass spectrometry and 1H NMR
spectroscopy. In particular, the methylene protons and
the methyne proton, in positions 4 and 5, respectively,
of the dihydro-(1H)-pyrazole ring, give rise to a well-
defined system of three double doublets, indicating not
only the formation of the pyrazoline but also the exact
position of the CdN double bond (Figure 1, compound
a).
Single-crystal X-ray diffraction analysis shows that
the (+)-enantiomer of compound b has the absolute (R)-
configuration.19 The (R)-enantiomer of compound b is
dextrorotatory at 365 and 589 nm in both ethanol and
chloroform solutions. The sign of optical rotation of (R)-b
stereoisomer at 365 nm in n-hexanes-ethanol (70:30,
v/v), online-monitored during chiral HPLC on AD CSP,
was also +.

Pyrazole Derivatives
Journal of Medicinal Chemistry, 2005, Vol. 48, No. 23 7115
Table 1. Chemico Physical Properties of 1-Thiocarbamoyl-3,5-diaryl-4,5-dihydro-(1H)-pyrazole Derivatives 1-10
a 0.1 M potassium phosphate buffer (pH 7.2) capable of
restoring 90-100% of the activity of the enzyme. After
a first assessment of their inhibitory ability on the MAO,
the compounds were tested to determine their activity
toward MAO-A and MAO-B selectively in the presence
of the specific substrates, serotonin and benzylamine,
respectively (Table 2).
From Table 2, which shows the MAO inhibition data
along with the MAO inhibitory selectivity (SI B/A and
A/B), it can be seen that all the synthesized compounds
show high activity against both the MAO-A and the
MAO-B isoforms with Ki values between 27 and 4 nM
and between50 and 1.5 nM, respectively, except for
derivatives 3, 4, and 6 whose inhibitory activity was in
the micromolar range against MAO-B.
The 3,5-diphenyl derivatives 1, 2, 4, and 5, which bear
a 4-chlorophenyl substituent in the 5 position, show high
activity against both MAO-A and MAO-B but with
opposite selectivity. In fact, derivative 1 with a methyl
group in the Ar shows a selectivity toward MAO-B (SI-
(A/B) ) 20.7), while derivative 4 with a 4-fluorine in
the same position shows selectivity against MAO-A.
The derivatives 3 and 6, 5-furyl substituted, exhibit
selectivity for the MAO-A.
Figure 1. 1-Thiocarbamoyl-3,5-diaryl-4,5-dihydro-(1H)-pyra-
zole (a) and 1-acetyl-3-(4-hydroxyphenyl)-5-(2-clorophenyl)-4,5-
dihydro-(1H)-pyrazole (b).
Thus, with the assumption of a structural analogy
between the studied compounds, the absolute configu-
ration of enantiomers of 1 and 4 as (-)-(S) and (+)-(R)
is straightforward.20 Furthermore, of all analytes re-
ported, 1, 4, and compound b exhibited the same sense
chiral recognition mechanism and, consequently, the
same enantiomer elution order on AD CSP in the
presence of normal-phase n-hexanes-ethanol eluents,
with preferential retention of the (+)-enantiomer. These
data permit us to advance the hypothesis that the
structurally related analytes investigated undergo same
chiral recognition process on AD CSP and supports the
presumption that the (+)-enantiomers of chiral analytes
1, 4 and compound b possess the same absolute (R)-
configuration.
Compounds 7-10 have high activity against both
isoforms, but for compounds 7 and 8 a little selectivity
against MAO-B can be observed.
Results and Discussion
Furthermore, owing to the presence of a chiral center
at the C5 position of the pyrazole moiety, semiprepara-
tive chromatographic enantioseparation of the most
potent, selective, and active chiral compounds 1 (KiMAO-B
All compounds were first evaluated for their ability
to inhibit MAO in the presence of kynuramine as a
substrate (Table 2). It is interesting to note that all
compounds act through the reversible mode, as shown
by dialysis performed over 24 h in a cold room against
) 1.5 × 10-9; SI ) 20.7) and 4 (KiMAO-A ) 6.0 × 10-9
;
SI ) 166) was performed. The separated enantiomers

7116 Journal of Medicinal Chemistry, 2005, Vol. 48, No. 23
Chimenti et al.
Table 2. Inhibitory Activities of 1-Thiocarbamoyl-3,5-diaryl-4,5-dihydro-(1H)-pyrazole Derivatives 1-10a
a Data represent mean values of at least of three separate experiments. b SI: selectivity index ) Ki(MAO-B)/Ki(MAO-A)
.
c SI: selectivity
index ) Ki(MAO-A)/Ki(MAO-B)
.
Table 3. Inhibitory Activities of 1-Thiocarbamoyl-3,5-diaryl-4,5-dihydro-(1H)-pyrazole Derivatives 1-4a as Racemates and Single
Enantiomersa
a Data represent mean values of at least of three separate experiments. b SI: selectivity index ) Ki(MAO-B)/Ki(MAO-A)
.
c SI: selectivity
index ) Ki(MAO-A)/Ki(MAO-B)
.
were then submitted to in vitro biological evaluation
(Table 3). From the results of these experiments it has
been possible to point out a difference between the
racemic mix and the individual enantiomers in inhibit-
ing the two isoforms selectively. This is particularly
evident in compound 1, for which anti-MAO-B activity
varies slightly, while selectivity increases twice and a
half for the enantiomer (-)-(S), reaching 50.
For compound 4, in the same way but for anti-MAO-A
activity, selectivity triples for the enantiomer (-)-(S),
reaching 480.
The computational work was carried out with the aim
to rationalize the biological results obtained for each
enantiomer of compounds 1 and 4. It is worth noting
that the main difference in the inhibition activity of
these stereoisomers was found when the methyl of the
first aromatic moiety is replaced by a fluorine atom. This
substitution was responsible for the remarkable differ-
ences in the inhibition activity mainly against MAO-B
isoform, with a factor between 2400 and 519, respec-
tively, for the (-)-(S) and (+)-(R) enantiomers of
compounds 1 and 4. The MAO-A effects of such substi-
tutions were much lower with factors between 0 and 1,
respectively. In the entire set of compounds, the highest
substitution effect on the MAO-A inhibition is only 12.5
(data computed with racemate inhibition activities).
The catalytic cleft comparison of bovine MAO-A and
MAO-B isoforms revealed a high level of identity.21
Between them only 1 of 12 residues is not conserved;
Ile335 is replaced with Tyr326. The computational study
started with the conformational analysis of (R)-isomers
of compounds 1 and 4. As expected for the position and
nature of F and Me substitutions, the Monte Carlo (MC)
search gave identical results in terms of conformational

Pyrazole Derivatives
Journal of Medicinal Chemistry, 2005, Vol. 48, No. 23 7117
Table 4. GLUE Binding Modes (BM) and Interaction Energies in kcal/mol
distribution in both compounds: two conformers within
12.5 kcal/mol above the global minimum. The main
differences were due to two thiourea moiety rotamers
attached to the pyrazole N1 atom (data not shown).
Both conformers obtained in the MC search and their
Z mirrored images ((S)-stereoisomers) were used for an
intensive series of docking experiments with the MAO-A
and MAO-B isoforms corresponding to 1O5W and 1GOS
PDB models. The docking experiments were started
following a computational approach reported in our
recent publications.22 The MOLINE method23 was car-
ried out after the enzyme pretreatment (see Experi-
mental Section), fixing the Cartesian origin of the XYZ
axes onto the FAD N5 atom and varying the resolution
and the compression factor. Unlike our previous results
obtained with coumarine derivatives,22 we found binding
modes of compounds 1 and 4 in the enzymatic cleft of
both isoforms only in a few cases. So we decided to use
GLIDE24 as an alternative docking method. Using no
energy filtering control and a box size including all
catalytic residues, with this approach it was possible
to generate 10 configurations of each compound in the
enzymatic clefts. Unfortunately, the unique OPLS-AA
force field implemented in GLIDE, version 1.8, gener-
ated a tetrahedral unrealistic geometry in the sp2 carbon
of the thiourea. Probably as a consequence of this poor
parametrization, the GLIDE scoring functions related
to this ensemble gave a weak correlation with experi-
mental data. However, we used the 80 binding modes
(10 for each ligand complex) as a starting point for
further energetic analysis based on multiminimization
procedures in different conditions with MacroModel,
version 7.2.25 (see Experimental Section), again with no
interesting results. Therefore, similar experiments were
carried out using the GLUE flexible docking approach
included in the GRID suite26 considering all probes
related to the chemical groups of compounds 1 and 4
(see Experimental Section). Multiple binding modes
(BM) varying from 3 to 7 were detected with this
approach. Despite the fact that the conformational
treatment of the thiourea moiety was better, the inter-
action energies did not give an acceptable correlation
with the experimental data, not even in this case (Table
4).
presence of noncovalent ligands complexed to the MAO-B
isoform. To take into account this issue and to consider
the reciprocal induced fit effects adequately, the lowest
interaction energy configurations BM1 (Table 4) ob-
tained by the GLUE docking simulations were submit-
ted to the GROMACS28 molecular dynamic approach.
After standard building procedures, with the aim of
minimizing van der Waals contact with the solvent, the
eight models were submitted to preliminary energy
optimization with no constraints, followed by a dynamic
conformational assessment of the water molecules, the
ligand, and the FAD. After a short molecular dynamics
(MD) run, we observed free binding interaction energies
converted into affinity constants in good agreement with
the experimental data. Moreover, similar to a reported
analysis of MAO-B binding modes,22 the distance be-
tween each docked ligand and FAD N5 was considered
as a geometrical descriptor possibly related to inhibition
data (see Experimental Section). As a matter of fact,
the GLUE-GROMACS docking configurations showed
interesting correlations with experimental data from the
point of view of energy and geometry (Table 5).
Details of the binding modes obtained in the GLUE-
GROMACS, sorted by the experimental inhibition con-
stants of Table 3, are reported in Figures 2 and 3.
The binding mode of the most active MAO-B inhibitor
(S)-1 observed at the end of the MD simulation was
stabilized in the Ar1 moiety portion by stacking interac-
tions with Tyr398 and Tyr435. Electrostatic contribu-
tions were found between the Cl atom and the side chain
OH of Tyr188. The thiourea moiety showed hydrophobic
contact with Tyr60 and a weak hydrogen bond with a
water molecule. The Ar moiety, located far from enzyme
cofactor, established an interaction with the hydropho-
bic core defined by Phe168, Leu171, Cys172, Ile199, and
Tyr326 (Figure 2a).
In the same enzymatic cleft, its enantiomer (R)-1 was
similarly oriented as far as regards the aromatic ring
positions. No Ar1 stacking was observed, but the van
der Waals contacts were evident with the isoalloxazine
FAD ring, Tyr398, and Tyr435. Moreover, with this
residue the thiourea S atom established a first hydrogen
bond and a second one with a water molecule close to
Cys172. Similar to its enantiomer, the Ar moiety (R)-1
interacted with the same cluster of hydrophobic residues
(Figure 2b).
The reason for such unsatisfactory results was at-
tributed mainly to a missing explicit solvent in the above
models. In a recent publication, Binda et al.27 high-
lighted the importance of structural water in the
catalytic region close to the FAD cofactor also in the
The binding mode of (S)-4 in the MAO-A cleft was
characterized by a totally different pattern with respect
to the two docking results reported above. The fluori-

7118 Journal of Medicinal Chemistry, 2005, Vol. 48, No. 23
Table 5. GLUE-GROMACS Binding Mode Analysis
Chimenti et al.
Figure 2. GLUE-GROMACS binding configurations of complexes [MAO-B/(S)-1] (a), [MAO-B/(R)-1] (b), [MAO-A/(S)-4] (c), and
[MAO-A/(R)-4] (d). The inhibitors are displayed in bold polytubes, cofactor in cyan CPK, and residues of the enzymatic clefts in
polytube models. For clarity, all other residues are depicted in structure colored ribbon and hydrogen atoms are omitted. Water
molecules are displayed as balls and stick models.
nated Ar moiety was positioned close to the cofactor in
a hydrophobic cavity delimited by Tyr197, Phe352,
Tyr407, and Tyr444. An electrostatic attraction between
F and a water molecule stabilized this binding mode.
Moreover, two intermolecular hydrogen bonds were
detected: one between the thiourea S atom and a water
molecule close to Phe352 and Ile335 and another
between the amide hydrogen and the Thr336 backbone.
The Ar1 is located in a hydrophobic core delimited by
Leu97, Phe208, Val210, and Leu337 (Figure 2c).

Pyrazole Derivatives
Journal of Medicinal Chemistry, 2005, Vol. 48, No. 23 7119
Figure 3. GLUE-GROMACS binding configurations of complexes [MAO-A/(R)-1] (a), [MAO-A/(S)-1] (b), [MAO-B/(R)-4] (c), and
[MAO-B/(S)-4] (d). The inhibitors are displayed in bold polytubes, cofactor in cyan CPK and residues of the enzymatic clefts in
polytube models. For clarity all other residues are depicted in structure colored ribbon, and hydrogen atoms are omitted. Water
molecules are displayed as balls and stick models.
Conversely, in the MAO-A cleft its enantiomer (R)-4
displayed a binding pattern relatively similar to that
of (R)-1 in the isoform B of the enzyme, with the Ar
moiety far form FAD in a nonpolar region delimited by
Leu97, Phe108, Ala111, Phe208, Val210, Cys323, Ile335,
and Leu337. The Ar1 group established van der Waals
contacts without stacking with Tyr407 and Tyr444,
while its Cl atom gave an attractive electrostatic
interaction with a water molecule close to Tyr197 and
to a second solvent molecule. FAD contacts with the
ligand were basically hindered by these water molecules
(Figure 2d).
bond interaction with a bridge water molecule with
Lys305 (Figure 3a).
Its enantiomer (S)-1 in the same MAO-A enzymatic
cleft displayed the opposite pattern, with its Ar moiety
stacking with Tyr407 and Tyr444 and other contacts
with FAD, Tyr197, and Gly443. On the other side of the
ligand, Ar1 recognized a hydrophobic cleft with Leu97,
Phe108, Phe208, Val210, and Cys323. Its Cl atom
established electrostatic interactions with a water mol-
ecule close to Gly110 and Ala111. Those two residues,
quite far from FAD, seemed to prevent a better ligand-
cofactor interaction (Figure 3b).
The binding mode of (R)-4 within the MAO-B enzyme
was characterized by a repulsion interaction of the
ligand Cl atom and FAD N5, which probably prevents
closer contacts with the cofactor as shown by the
distance analysis in Table 5. Other contacts of the Ar1
moiety were detected with the hydrophobic residues
Tyr60, Phe343, Tyr398, and Tyr435. Similar to the
recognition of (R)-1, a hydrogen bond of S thiourea with
Tyr435 was observed as well as the interaction of the
Similar to (R)-4, (R)-1 also showed the same Ar1
position in the MAO-A enzymatic cleft. The Ar moiety,
and in particular its methyl group, established van der
Waals interaction with nonpolar residues such as
Leu97, Phe108, Gly110, Ala111, and Ile325 and a
partial stacking with Phe208. The thiourea gave a
similar interaction with Tyr69, Leu337, and Phe352.
Notably in this binding mode, according to Binda’s
observations,27 the FAD N5 was hindered by a hydrogen

7120 Journal of Medicinal Chemistry, 2005, Vol. 48, No. 23
Chimenti et al.
Chromatographic (HPLC) Resolution of Racemic
Samples 1 and 4. Chiral HPLC of 1 and 4 was performed by
stainless steel Chiralcel OD columns (250 mm × 4.6 mm i.d.
and 250 mm × 10 mm i.d.) (Daicel Chemical Industries, Tokyo,
Japan). HPLC-grade solvents were supplied by Carlo Erba
(Milan, Italy).
Chiral HPLC was performed using a Waters (Milford, MA)
510 pump equipped with a Rheodyne (Cotati, CA) injector, a
1 mL sample loop, an HPLC Perkin-Elmer (Norwalk, CT) oven,
and a Waters model 996 diode array detector (DAD).
The sign of the optical rotation of the two enantiomers of 1
and 4 was measured on-line at wavelengths of 365 and 589
nm by a Perkin-Elmer polarimeter model 241 equipped with
Hg/Na lamps and a 40 µL flow cell. The system was kept at a
constant temperature of 25 °C. The output signal was acquired
and processed by Millenium 2010 software.
Biochemistry. All chemicals were commercial reagents of
analytical grade and were used without purification. Bovine
brain mitochondria were isolated according to Basford.29 In
all experiments the AO activities of the beef brain mitochon-
dria were determined by a fluorimetric method, according to
Matsumoto et al.30 using kinuramine as a substrate at four
different final concentrations ranging from 5 µM to 0.1 mM.
Briefly, the incubation mixtures contained 0.1 mL of 0.25 M
potassium phosphate buffer (pH 7.4), mitochondria (6 mg/mL),
and drug solutions with final concentration ranging from 0 to
10-3 µM.
The solutions were incubated at 38 °C for 30 min. Addition
of perchloric acid ended the reaction. The samples were
centrifuged at 10000g for 5 min, and the supernatant was
added to 2.7 mL of 0.1 N NaOH. The pyrazole derivatives were
dissolved in dimethyl sulfoxide (DMSO), added to the reaction
mixture from 0 to 10-3 µM. To study the inhibition of pyrazole
derivatives on the activities of both MAO-A and MAO-B
separately, the mitochondrial fractions were preincubated at
38 °C for 30 min before adding the specific inhibitors (0.5 µM
L-deprenyl to estimate MAO A activity and 0.05 µM clorgyline
to assay the isoform B), considering that MAO-A is irreversibly
inhibited by a low concentration of clorgyline but is unaffected
by a low concentration of L-deprenyl, which is used in the
MAO-B form. Fluorimetric measurements were recorded with
a Perkin-Elmer LS 50B spectrofluorimeter. The protein con-
centration was determined according to Bradford.31 The results
are reported in Table 2. Dixon plots were used to estimate the
inhibition constant (Ki) of the inhibitors (data reported in Table
3). The data are the mean values of three or more experiments
performed in duplicate.
Ar ring in a hydrophobic area delimited by Trp119,
Leu164, Phe168, Leu171, Ile199, Ile316, and Tyr326
(Figure 3c).
Finally, its enantiomer (S)-1 interacts with MAO-B
in a region significantly far from FAD compared to
previous docking results. The cofactor was actually
hindered by several water molecules. According to
Binda’s observation,27 one of these molecules showed the
same hydrogen bond network with FAD N5 and Lys296.
Ar1 established mainly hydrophobic interactions with
residues already encountered in other recognition pat-
terns, such as Trp119, Leu164, Ile199, Ile316, and
others relatively far from the catalytic region, such as
Phe103, Pro104, and Phe99. The Ar1 F atom was
hydrogen-bonded to the SH of Cys172, and the rest of
the moiety established van der Waals contacts with
Leu171, Ile198, Gln206, and Tyr435. Other interactions
were also displayed between the thiourea moiety and
Glu84, Thr201, and Thr314 (Figure 3d).
As a result of an analysis of the binding modes found
at the end of the MD simulation carried out with best
scoring GLUE configurations, it was possible to draw
the following conclusions. Two main kinds of binding
modes were detected on the basis of the relative posi-
tions of Ar and Ar1 with respect to FAD. They were
unrelated to the degree of inhibition. In agreement with
previous studies,22 the inhibition activity of the 1 and
4 stereoisomers on both enzyme isoforms was regulated
by their ability to establish close and productive inter-
actions with FAD. In particular, the competition with
water molecules, especially those capable of interacting
with FAD N5 with the support of Lys305 (MAO-A) or
Lys296 (MAO-B),27 seemed to play an additional role
in the inhibition activity. In fact the investigation of this
last issue required the adoption of computational simu-
lation with an explicit model of solvation.
We are focusing this computational work onto a larger
library of (1H)-pyrazole derivatives with the aim of
finding more predictive and general models for revers-
ible and selective MAO inhibitors.
Molecular Modeling. The Monte Carlo search of com-
pounds 1 and 4 was carried out using the (R)-stereoisomer. A
thousand conformations were generated randomizing the
rotatable bonds of the common skeleton with an AMBER*
united atom force field and GB/SA water implicit solvation as
implemented in MacroModel, version 7.2.25
Following the computational work of our recent communica-
tions,11,22 the Protein Data Bank MAO-B crystallographic
model 1GOS32 was considered for the docking experiments. As
regards the MAO-A, a crystallographic model of the rat isoform
was reported recently in the PDB with the code 1O5W.33 Both
PDB models were obtained as adducts with two similar
compounds covalently linked to the FAD N5 nitrogen.
The 1GOS model is related to the MAO-B isoform. The
difference between human and bovine sequence at the catalytic
site is related to one mutation; Ile199 is replaced with Phe199.
The comparison of human and rat MAO-A enzymes is
respectively 90% as a global identity and 100% as a binding
site.33 So 1GOS and 1O5W models were considered in the
computational study.
Experimental Section
Chemistry. Melting points are uncorrected and were
determined on a Reichert Kofler thermopan apparatus. 1H
NMR spectra were recorded on a Bruker AMX (300 MHz)
using tetramethylsilane (TMS) as internal standard and
DMSO or CDCl3 as solvents (chemical shifts in δ values, J in
Hz). Electron ionization (EI) mass spectra were obtained by a
Fisons QMD 1000 mass spectrometer (70 eV, 200 µA, ion
source temperature of 200 °C). The samples were introduced
directly into the ion source. Elemental analyses for C, H, and
N were performed on a Perkin-Elmer 240 B microanalyzer,
and the analytical results were within (0.4% of the theoretical
values.
Preparation of 1-Thiocarbamoyl-3-(4-methylphenyl)-
5-(4-chlorophenyl)-4,5-dihydropyrazole (1). A mixture of
1.80 g (0.007 mol) of 1-(4-methylphenyl)-3-(4-chlorophenyl)-2-
propen-1-one and 1.27 g (0.014 mol) of thiosemicarbazide is
refluxed in ethanol (70 mL) under vigorous stirring. After
complete dissolution of the reactants, a solution of 0.78 g (0.014
mol) of KOH in ethanol (70 mL) is added dropwise.The solution
is refluxed for a further 6 h, allowed to cool, and stirred
overnight. A precipitate is formed, which is filtered off and
crystallized from ethanol/isopropyl alcohol to give 1 as a pure
product.
Pretreatment of the original PDB structures consisted of a
48 kcal/mol constrained energy minimization 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 isoalloxazine FAD
ring and relax the active site amino acids. In the resulting
energy minimum structures, the covalent ligands (pargyline
for 1GOS and clorogyline for 1O5W) were removed and used
According to the same procedure, compounds 2-10 were
synthesized.

Pyrazole Derivatives
Journal of Medicinal Chemistry, 2005, Vol. 48, No. 23 7121
as receptor models. This was performed with the force field
AMBER* united atom notation and GB/SA water implicit
model of solvation as implemented in MacroModel, version
7.2.25
distance between the ligand centers of mass and the FAD N5
atom. A longer simulation up to 20 ps revealed no significant
changes in this geometrical descriptor after 2 ps in all
complexes.
All calculations were done by a Linux cluster of 5 Intel Xeon
dual processors at 3.2 GHz with 2 Gb of RAM. Graphic
manipulations and analysis of the docking experiments were
performed by the Maestro Graphical User Interface, version
4.1.012, for Linux operating systems.25 Jmol, version 10, was
used for creating Figures 2 and 3.39
The MOLINE computational approach required as receptor
and ligand input data, respectively, the pretreated enzyme
models and the two conformations obtained in the MC search
of the inhibitors. The docking experiments were carried out
with the same AMBER* united atom force field considered in
the simulations above. According to a recently reported22
similar case study, both the MOLINE grid resolution GR and
the van der Waals compression factor ø were systematically
varied. Two different dielectric constants respectively fixed at
4 (to mimic conditions similar to an “average” protein medium)
and 80 (water environment) were tried in the docking experi-
ments.
According to the GLIDE24 approach, both pretreated enzyme
models were submitted to map calculations using a box of
about 110 000 Å3 centered on the FAD N5 atom. Flexible
docking of the four ligands was done by generating a maximum
number of 1000 configurations, saving the 10 lowest energy
ones and submitting them to the multiminimization proce-
dures as follows. We considered the MMFF, AMBER* united
atoms and all-atom force fields. OPLS* did not work for
missing parameters related to the FAD phosphate moiety. For
solvation models, we considered two different dielectric con-
stants and the water implicit GB/SA solvation model.25 Three
different constraint conditions were systematically studied:
full relaxing (no constraints), protein backbone fixed with a
200 kJ mol-1 Å-1 force constant on each atom, and a shell of
free atoms up to 15 Å from the FAD N5 with the remaining
atoms kept constrained as above. The interaction energy of
all complexes was computed according to the MOLINE method23
and converted to average state equations. Globally, 8 × 10 ×
3 × 3 × 3 ) 2160 conformations were energy-minimized at
the above conditions.
Molecular interaction fields required by the GLUE program
were precalculated by GRID using the standard probes OH2
and H. Specific probes related to the chemical structures of 1
and 4 were C3, C1d, N2, N:d, F, CL, and O. This last probe
was used to mimic the thiourea S atom, not available in the
GLUE probe list. In the generation of the GRID maps, a cube
box of 10 000 Å3 was centered on the FAD N5 atom using the
original PDB models without their covalent ligands. The
flexibility of the inhibitors was considered, allowing free
rotations of all rotatable bonds of compounds 1 and 4. The
lowest interaction energy configurations of each complex were
submitted to GROMACS calculations.
The force field considered in the GROMACS calculations
was the ffgmx force field with the SPC explicit model of
solvation.34 GROMACS specific modules editconf and genbox
were used to add water molecules within a box edge of
minimum 10 Å from the enzyme complex. The module genion
allowed the neutralization of the models by adding three Cl-
and four Na+ counterions to the MAO-A and MAO-B com-
plexes, respectively. The PRODRG service tool available on
the Internet was used for a quick setup of the non-protein
section of the eight complexes.35 Preliminary minimization was
done using 500 iterations of the steepest descents algorithm
with an energy convergence criterion fixed at 1000 kJ mol-1
nm-1. The particle mesh Ewald (PME) method was used to
treat the electrostatic term.36 Dynamic conformational assess-
ment of non-protein atoms was carried out by Berendsen’s
temperature and pressure coupling method37 for 2 ps at 300
K with a time step of 2 fs. The linear constraints solver
algorithm (LINCS) was adopted for all atoms, preventing bond
distortions as suggested for time steps larger than 1 fs.38 The
g_lie GROMACS utility was used to estimate the free-binding
interaction energy of each enzyme-inhibitor complex (FBIE)
using final conformations of the 2 ps molecular dynamics run.
Theoretical affinity constants (tKa) were derived according to
the equation RT ln tKa ) FBIE, where R is the gas constant
and T the room temperature set to 300 K. The same conforma-
tions were submitted to geometrical analysis considering the
Acknowledgment. This work was supported by
grants from MURST. We also acknowledge Mr. Anton
Gerada, a professional translator and Fellow of the
Institute of Translation and Interpreting of London and
Member of AIIC (Association Internationale des Inter-
preˆtes de Confe´rences, Geneva), for the revision of the
manuscript.
Supporting Information Available: 1H NMR, IR spec-
tral data, and elemental analysis results of derivatives 1-10.
This material is available free of charge via the Internet at
http://pubs.acs.org.
References
(1) Singer, T. P.; Von Korff, R. W.; Murphy, D. L. In Monoamino
Oxidase: Structure, Function and Altered Functions; Aca-
demic: New York, 1979.
(2) Mure, M.; Mills, S. A.; Klinman, J. P. Catalytic mechanism of
the Topa quinone containing copper amine oxidases. Biochem-
istry 2002, 41 (30), 9269-9278.
(3) (a) Johnston, J. P. Some observations upon a new inhibitor of
monoamine oxidase in brain tissue. Biochem. Pharmacol. 1968,
17, 1285-1297. (b) Knoll, J.; Magyar, K. Puzzling pharmacologi-
cal effects of monoamine oxidase [MAO] inhibitors. Adv. Bio-
chem. Psychopharmacol. 1972, 5, 393-408.
(4) Bach, A. W.; Lan, N. C.; Johnson, D. L.; Abell, C. W.; Bembenek,
M. E.; Kwan, S. W.; Seeburg, P. H.; Shih, J. C. cDNA cloning of
human liver monoamine oxidase A and B: molecular basis of
differences in enzymatic properties. Proc. Natl. Acad. Sci. U.S.A.
1988, 85, 4934-4938.
(5) Wouters, J. Structural aspects of monoamine oxidase and its
reversible inhibition. Curr. Med. Chem. 1998, 5, 137-162.
(6) (a) Walker, M. C.; Edmondson, D. E. Structure-activity rela-
tionships in the oxidation of benzylamine analogues by bovine
liver mitochondrial monoamine oxidase B. Biochemistry 1994,
33 (23), 7088-7098. (b) Miller, J. R.; Edmondson, D. E. Struc-
ture-activity relationships in the oxidation of para-substituted
benzylamine analogues by recombinant human liver monoamine
oxidase A. Biochemistry 1999, 38 (41), 13670-13683. (c) Silver-
man, R. B.; Hoffman, S. J.; Catus, W. B. A mechanism for
mitochondrial monoamine oxidase catalyzed amine oxidation.
J. Am. Chem. Soc. 1980, 102 (23), 7126-7128. (d) Kim, J. M.;
Bogdan, M. A.; Mariano, P. S. Mechanistic analysis of the
3-methyllumiflavin-promoted oxidative deamination of benzyl-
amine. A potential model for monoamine oxidase catalysis. J.
Am. Chem. Soc. 1993, 115 (23), 10591-10595. (e) Lu, X.;
Rodr`ıguez, M.; Gu, W.; Silverman, R. B. Inactivation of mito-
chondrial monoamine oxidase B by methylthio-substituted ben-
zylamines. Bioorg. Med. Chem. 2003, 11 (20), 4423-4430.
(7) Cesura, A. M.; Pletscher, A. The new generation of monoamine
oxidase inhibitors. Prog. Drug Res. 1992, 38, 171-297.
(8) Saura, J.; Luque, J. M.; Cesura, A. M.; Da Prada, M.; Chan-
Palay, V.; Huber, G.; Loffler, J.; Richards, J. Increased monoam-
ine oxidase B activity in plaque-associated astrocytes of Alzhe-
imer brains revealed by quantitative enzyme radioautography.
Neuroscience 1994, 62 (1), 15-30.
(9) Amrein, R.; Martin, J. R.; Cameron, A. M. Moclobemide in
patients with dementia and depression. Adv. Neurol. 1999, 80,
509-519.
(10) Chimenti, F.; Bolasco, A.; Manna, F.; Secci, D.; Chimenti, P.;
Befani, O.; Turini, P.; Giovannini, V.; Mondovi’, B.; Cirilli, R.;
La Torre, F. Synthesis and selective inhibitory activity against
MAO of 1-acetyl-3,5-diphenyl-4,5-dihydro-(1H)-pyrazole deriva-
tives. J. Med. Chem. 2004, 47 (8), 2071-2074.
(11) Manna, F.; Chimenti, F.; Bolasco, A.; Secci, D.; Bizzarri, B.;
Befani, O.; Turini, P.; Mondov`ı, B.; Alcaro, S.; Tafi, A. Inhibition
of amine oxidases activity by 1-acetyl-3,5-diphenyl-4,5-dihydro-
(1H)-pyrazole derivatives. Bioorg. Med. Chem. Lett. 2002, 12,
3629-3633.
(12) Gokhan, N.; Yesilada, A.; Uc¸ar, G.; Erol, K.; Bilgin, A. A. 1-N-
substituted thiocarbamoyl-3-phenyl-5-thienyl-2-pyrazolines: Syn-
thesis and evaluation as MAO inhibitors. Arch. Pharm. (Wein-
heim, Ger.) 2003, 336, 362-371.

7122 Journal of Medicinal Chemistry, 2005, Vol. 48, No. 23
Chimenti et al.
(13) Walter, W.; Voss, J. In The Chemistry of Amides; Zabicky, J.,
Ed.; Wiley and Sons Ltd.: London, 1970; Chapter 8, pp 385-
415.
(14) Takasaki, W.; Yamamura, M.; Nozaki, A.; Nitanai, T.; Sasahara,
K.; Itoh, K.; Tanaka, Y. Stereoselective pharmacokinetics of RS-
8359, a selective and reversible MAO-A inhibitor, by species-
dependent drug-metabolizing enzymes. Chirality 2005, 17, 135-
141.
(15) (a) Drake, N. L.; Allen, P. Org. Synth. Collect. J. Wiley and Sons,
Inc.: London, 1948; 1, pp 77-78. (b) Kohler, E. P.; Chadwell,
H. M. Org. Synth. Collect. J. Wiley and Sons, Inc.: London, 1948;
1, pp 78-79.
(16) Nielsen, A. T.; Houlihan, W. J. The Aldol Condensation in
Organic Reactions; Wiley and Sons: New York, London, Sydney,
1968; Vol. 16.
(17) Bonsignore, L.; Cabiddu, S.; Maccioni, A.; Marongiu, E. Synthesis
and separation of geometrical isomers of arylacrylonaphthones.
Gazz. Chim. Ital. 1976, 106, 617-624.
(18) Yashima, E.; Yamamoto, C.; Okamoto, Y. Polysaccharide-based
chiral LC columns. Synlett 1998, 344-360.
(19) Cirilli, R.; Ferretti, R.; Gallinella, B.; Turchetto, L.; Bolasco, A.;
Secci, D.; Chimenti, P.; Pierini, M.; Fares, V.; Befani, O.; La
Torre F. Enantiomers of C-5-chiral 1-acetyl-3,5-diphenyl-4,5-
dihydro-(1H)-pyrazole derivatives: Analytical and semiprepara-
tive HPLC separation, chiroptical properties, absolute configu-
ration, and inhibitory activity against monoamine oxidase.
Chirality 2004, 16, 625-636.
(20) Roussel, C.; Del Rio, A.; Pierrot-Sanders, J.; Piras, P.; Van-
thuyne, N. Chiral liquid chromatography contribution to the
determination of the absolute configuration of enantiomers. J.
Chromatogr., A 2004, 1037 (1-2), 311-328.
(21) Abell, C. W.; Kwan, S. W. Molecular characterization of monoam-
ine oxidases A and B. Prog. Nucleic Acid Res. Mol. Biol. 2000,
65, 129-156.
(22) Chimenti, F.; Secci, D.; Bolasco, A.; Chimenti, P.; Granese, A.;
Befani, O.; Turini, P.; Alcaro, S.; Ortuso, F. Inhibition of
monoamine oxidases by coumarin-3-acyl derivatives: biological
activity and computational study. Bioorg. Med. Chem. Lett. 2004,
14, 3697-3703.
(23) Alcaro, S.; Gasparrini, F.; Incani, O.; Mecucci, S.; Misiti, D.;
Pierini, M.; Villani, C. A. Quasi-flexible automatic docking
processing for studying stereoselective recognition mechanisms.
Part I. Protocol validation. J. Comput. Chem. 2000, 21, 515-
530.
(24) Eldridge, M. D.; Murray, C. W.; Auton, T. R.; Paolini, G. V.; Mee,
R. P. Empirical scoring functions 1. The development of a fast
empirical scoring function to estimate the binding affinity of
ligands in receptor complexes. J. Comput.-Aided Mol. Des. 1997,
11, 425-445.
(25) (a) MacroModel, version 7.2; Schro¨dinger Inc.: Portland, OR,
1998-2001. (b) Mohamadi, F.; Richards, N. G. J.; Guida, W. C.;
Liskamp, R.; Lipton, M.; Caufield, C.; Chang, G.; Hendrickson,
T.; Still, W. C. MacroModel, an integrated software system for
modeling organic and bioorganic molecules using molecular
mechanics. J. Comput. Chem. 1990, 11, 440-467.
(26) (a) GRID, version 22a; Molecular Discovery Ltd.: West Way
House, Elms Parade, Oxford, England. (b) Goodford, P. J. A
computational procedure for determining energetically favorable
binding sites on biologically important macromolecules. J. Med.
Chem. 1985, 28, 849-857.
(27) Binda, C.; Li, M.; Huba`lek, F.; Restelli, N.; Edmondson, D. E.;
Mattevi, A. Insights into the mode of inhibition of human
mitochondrial monoamine oxidase B from high-resolution crystal
structures. Proc. Natl. Acad. Sci. U.S.A. 2003, 100, 9750-9755.
(28) (a) Berendsen, H. J. C.; van der Spoel, D.; van Drunen, R.
GROMACS: A message-passing parallel molecular dynamics
implementation. Comput. Phys. Commun. 1995, 91, 43-56. (b)
Lindahl, E.; Hess, B.; van der Spoel, D. GROMACS 3.0:
A
package for molecular simulation and trajectory analysis. J. Mol.
Model. 2001, 7, 306-317.
(29) Basford, R. E. Preparation and properties of brain mitochondria.
Methods Enzymol. 1967, 10, 96-101.
(30) Matsumoto, T.; Suzuki, O.; Furuta, T.; Asai, M.; Kurokawa, Y.;
Rimura, Y.; Katsumata, Y.; Takahashi, I. A sensitive fluoromet-
ric assay for serum monoamine oxidase with kynuramine as
substrate. Clin. Biochem. 1985, 18, 126-129.
(31) Bradford, M. M. A rapid and sensitive method for the quanti-
tation of microgram quantities of protein utilizing the principle
of protein-dye binding. Anal. Biochem. 1976, 72, 248-254.
(32) Binda, C.; Newton-Vinson, P.; Huba`lek, F.; Edmondson, D. E.;
Mattevi, A. Structure of human monoamine oxidase B, a drug
target for the treatment of neurological disorders. Nat. Struct.
Biol. 2002, 9, 22-26.
(33) Ma, J.; Yoshimura, M.; Yamashita, E.; Nakagawa, A.; Ito, A.;
Tsukihara, T. Structure of rat monoamine oxidase A and its
specific recognitions for substrates and inhibitors. J. Mol. Biol.
2004, 338, 103-114.
(34) Berendsen, H. J.; Postma, J. P. M.; van Gunsteren, W. F.;
Hermans, J. Interaction Models for Water in Relation to Protein
Hydration. In Intermolecular Forces; Pullman, B., Ed.; D. Reidel
Publishing Co.: Dordrecht, The Netherlands, 1981; pp 331-342.
(35) (a) URL: http://davapc1.bioch.dundee.ac.uk/programs/prodrg. (b)
Schuettelkopf, A. W.; van Aalten, D. M. F. PRODRG, a tool for
highthroughput crystallography of protein-ligand complexes.
Acta Crystallogr. 2004, D60, 1355-1363.
(36) (a) Darden, T.; York, D.; Pedersen, L. Particle mesh Ewald: An
N-log(N) method for Ewald sums in large systems. J. Chem.
Phys. 1993, 98, 10089-10092. (b) Essmann, U.; Perera, L.;
Berkowitz, M. L.; Darden, T.; Lee, H.; Pedersen, L. A smooth
particle mesh ewald potential. J. Chem. Phys. 1995, 103, 8577-
8592.
(37) Berendsen, H. J. C.; Postma, J. P. M.; vanGunsteren, W. F.; Di
Nola, A.; Haak, J. R. Molecular dynamics with coupling to an
external bath. J. Chem. Phys. 1984, 81, 3584-3590.
(38) Hess, B.; Bekker, H.; Berendsen, H.; Fraaije, J. LINCS: A linear
constraint solver for molecular simulations. J. Comput. Chem.
1997, 18, 1463-1472.
(39) Jmol, version 10; http://jmol.sourceforge.net.
JM040903T