1-Methyl-3-pyrrolines and 2-methylisoindolines: New classes of cyclic tertiary amine monoamine oxidase B substrates

Article abstract of DOI:10.1016/S0968-0896(97)10020-7

Both 1-methyl-3-pyrrolines and 2-methylisoindolines are substrates for MAO-B with V(max)/K(m) values ranging from 200 to 2000 min^-1 mM^-1 at 37°C. These compounds represent new classes of cyclic tertiary amine substrates for this flavoenzyme. The only other known cyclic amines that are MAO-B substrates are 1,4-disubstituted 1,2,3,6-tetrahydropyridinyl derivatives. The presence of an allylic (benzylic) amino functionality in all of these compounds may be linked to their substrate properties since related piperidinyl and pyrrolidinyl analogs are stable in the presence of MAO-B. This paper discusses energetic and geometric features of these compounds in relationship to their substrate properties and in anticipation of their utility to probe the active site of this flavoenzyme.

Full text of DOI:10.1016/S0968-0896(97)10020-7

BMC 859p
BIOORGANIC &
MEDICINAL
CHEMISTRY
Bioorganic & Medicinal Chemistry 6 (1998) 143±149
1-Methyl-3-pyrrolines and 2-methylisoindolines:
new classes of cyclic tertiary amine monoamine
oxidase B substrates
You-Xiong Wang, Stephane Mabic, Neal Castagnoli Jr*
Department of Chemistry, Virginia Tech, Blacksburg, VA 24061-0212, U.S.A.
Received 3 July 1997; accepted 22 August 1997
Abstract
Both 1-methyl-3-pyrrolines and 2-methylisoindolines are substrates for MAO-B with Vmax/Km values ranging from 200
to 2000 min 1 mM at 37ꢀC. These compounds represent new classes of cyclic tertiary amine substrates for this
1
¯avoenzyme. The only other known cyclic amines that are MAO-B substrates are 1,4-disubstituted 1,2,3,6-tetra-
hydropyridinyl derivatives. The presence of an allylic (benzylic) amino functionality in all of these compounds may be
linked to their substrate properties since related piperidinyl and pyrrolidinyl analogs are stable in the presence of
MAO-B. This paper discusses energetic and geometric features of these compounds in relationship to their substrate
properties and in anticipation of their utility to probe the active site of this ¯avoenzyme. # 1998 Elsevier Science Ltd.
All rights reserved.
1. Introduction
examine the catalytic mechanism [8±11] and to investi-
gate the topology of the active sites of both forms of the
enzyme [12±14]. To the best of our knowledge, 1,4-
disubstituted 1,2,3,6-tetrahydropyridines are the only
reported cyclic tertiary amines which display good
MAO-B substrate properties. The corresponding piper-
idinyl analog 4 of MPTP is not a substrate [15] suggest-
ing that the allylamine functionality is important for the
catalytic process. Consistent with this view, the MAO-B
catalyzed oxidation of MPTP occurs regiospeci®cally at
the C-6 allylic position [16].
The ¯avoenzymes monoamine oxidase MAO A and B
(MAO-A and MAO-B) catalyze the oxidative deamina-
tion of brain neurotransmitters such as dopamine and
serotonin as well as a variety of xenobiotic amines [1].
The primary structures of these enzymes have been
established from gene sequences [2,3] but little is known
about the three dimensional features of their active sites.
The parkinsonian inducing nigrostriatal neurotoxin
1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine
[MPTP
(1, R=Ph)] and some of its structurally related analogs
are excellent MAO-B and/or MAO-A substrates [4,5].
MAO-B catalyzes the conversion of MPTP to the
In an attempt to evaluate further the potential
importance of the allylic moiety in the MAO-B cata-
lyzed oxidations of cyclic tertiary amines, the series of
ꢀ,ꢁ-unsaturated ®ve membered cyclic tertiary amines
(pyrrolines and isoindolines) shown below was synthe-
sized. The substrate properties of these compounds are
discussed in terms of their geometric and stereoelec-
tronic properties and are compared to the known
structural requirements for the MAO-B catalyzed
oxidation of tetrahydropyridinyl derivatives and related
compounds.
dihydropyridinium intermediate
2 which undergoes
spontaneous oxidation to the pyridinium species 3, the
ultimate neurotoxin (Scheme 1) [6,7]. These partially
rigid tetrahydropyridinyl derivatives have been used to
*Corresponding author.
e-mail: ncastagnoli@chemserver.chem.vt.edu
0968-0896/98/$19.00 # 1998 Elsevier Science Ltd. All rights reserved
PII: S0968-0896(97)10020-7

144
Y.-X. Wang et al./Bioorg. Med. Chem. 6 (1998) 143±149
Ph
R
Ph
features did not allow unambiguous assignment of the
geometry about the double bond. Arguments based on
steric control of enolate attack on the phenylselenyl
chloride from the least hindered side and AM1 semi-
empirical calculations (the trans-diastereomer 12 is more
stable than the cis-diastereomer 13 by 7 kcal/mol) sup-
port the assignment of this product as the trans-isomer
12. Borane reduction of this lactam gave the pyrrolidine
14 (or its geometric isomer 15). The corresponding
selenoxide (presumably 16) gave the desired pyrroline 6,
a reaction that should proceed by a syn elimination
which is consistent with the proposed trans geometry for
this series. The yield in this reaction was only 30% sug-
gesting that elimination also may have occurred to give
the isomeric enamine 17 which may have degraded dur-
ing work-up of the reaction mixture [17]. The formation
of 17 is not unexpected since calculations show that 6 is
favored over 17 by only 0.3 kcal/mol.
MAO-B
+
N
+
N
6
2
N
1
2
3
MPTP: R = Ph
R
R
R1
(CH2)n
N
R2
N
N
4: R = Ph; n = 2
32: R = H; n = 1
5: R1 = H, R2 = H
6: R1 = Ph, R2 = H
7: R1 = H, R2 = 3-Py
8: R = H
9: R = CH3
10: R = CF3
Scheme 1.
Syntheses of the 2-methylisoindolines 8±10 were
achieved according to the reaction sequence summar-
ized in Scheme 3. The starting materials 4-methylph-
thalic acid (19) and 1,2-bis-hydroxymethylbenzene (21)
were commercially available. The 4-tri¯uoromethyl
analog 20 was obtained by oxidation of 1,2-dimethyl-4-
tri¯uoromethylbenzene (18) [19]. LiAlH4 reduction of
the phthalic acids 19 and 20 [20] provided the corre-
sponding carbinols 22 and 23. The bismesylates 24±26
then were converted to the desired isoindolines by
treatment with methylamine [21].
2. Results and discussion
2.1 Synthesis
Synthesis of 1-methyl-3-phenyl-3-pyrroline (6) was
accomplished via a strategy developed by Chavdarian
for the preparation of the corresponding 2-(3-pyridinyl)
analog 7 (Scheme 2) [17]. The racemic form of 1-methyl-
4-phenyl-2-pyrrolidinone (11) was treated with LDA
followed by reaction of the resulting lithium enolate
with phenylselenyl chloride [18]. The 1H NMR spectrum
of the resulting phenylselenyl product (Scheme 2) was
consistent with a single diastereoisomer but the spectral
2.2 Enzymology
Repeated UV scans (450 to 250 nm) of incubation
mixtures containing 500 ꢂM of each of the ®ve test
Ph
Ph
SeC6H5
or
Ph
SePh
O
Ph
SePh
Ph
SePh
a
b
or
O
O
N
N
N
N
N
12
13
14
15
11
c
O
Ph
Ph
Ph
SePh
N
N
N
17
6
16
.
Scheme 2. Synthetic pathway to 1-methyl-4-phenyl-3-pyrroline (6). (a) i. LDA, THF, ii. PhSeCl; (b) BH3 THF, THF; (c) H2O2 30%.

Y.-X. Wang et al./Bioorg. Med. Chem. 6 (1998) 143±149
145
R
R
R
R
a
c
b
MsOH2C
CH2OMs
'R
R'
HOH2C
CH2OH
N
18: R = CF3; R' = CH3
19: R = CH3; R' = COOH
20: R = CF3; R' = COOH
21: R = H
22: R = CH3
23: R = CF3
24: R = H
25: R = CH3
26: R = CF3
8: R = H
9: R = CH3
10: R = CF3
Scheme 3. Synthetic pathway to the isoindolines 8±10. (a) LiAlH4, THF; (b) MsCl, NEt3; (c) MeNH2.
compounds (5, 6, and 8±10) and 0.16 ꢂM MAO-B were
examined to evaluate qualitatively their MAO-B sub-
strate properties. The time dependent formation of new
chromophores established that all compounds were
substrates for MAO-B. Consequently, 3-pyrrolines and
isoindolines join 1,2,3,6-tetrahydropyridines as cyclic
tertiary amine MAO-B substrates. As with the 6-mem-
bered series, the fully reduced azacycle, 1-methylpyrro-
lidine (32), was stable in the presence of MAO-B.
Kinetic studies on the MAO-B catalyzed oxidations of
these compounds led in all cases to linear initial rate
plots at substrate concentrations that bracketed the Km
values. The Vmax and Km values (Table 1) were obtained
from the corresponding Lineweaver±Burke double recip-
rocal plots using the reported molar extinction coe-
cients for 27, 28 and 29. Since we have assumed that the
molar extinction coecients for the methyl (30) and tri-
¯uoromethyl (31) analogs are the same as the value
reported for 29, kinetic estimates are only approximate.
transfer [22] versus hydrogen atom transfer [23]
mechanisms which have been proposed for the MAO
catalytic pathway.
As expected, formation of the C-6 allylic radical
(ÁÁHf ˆ 6:628 kcal=mol) of MPTP is favored over the
corresponding
C-2
homoallylic
radical
(ÁÁHf ˆ 18:192 kcal=mol). The relatively weak energy
(58.731 kcal/mol) of the allylic C±H bond is due to the
favorable geometry for extended orbital overlap
throughout the tetrahydropyridinyl six membered ring
of the carbon radical. Similarly, the ÁÁHf value
(21.644 kcal/mol) for formation of the ꢃ-carbon radical
of 1-methylpyrrolidine (32) is considerably greater than
the corresponding values for the 3-pyrrolines and iso-
indolines which range from 10.635 to 16.863 kcal/mol.
The substrate properties (Vmax/Km=862 min 1 mM
1
)
of 1-methyl-3-pyrroline (5), however, were not antici-
pated since 1,2,3,6-tetrahydropyridine (1, R=H) is not
an MAO-B substrate [24,25]. The similar ÁÁHf values
for the formation of 33 from 5 (15.857 kcal/mol) and 36
from 8 (16.835 kcal/mol) suggest that the double bond
and the fused phenyl ring are equally ecient in stabi-
lizing ꢃ-carbon radicals.
Summarized in Scheme
4 are possible reaction
sequences to account for product formation with these
two series. Calculations (Table 2) were performed at the
semi-empirical level (AM1) to estimate the change in
energy involved in the conversion of the substrate
molecules to the proposed allylic radicals (33±40) that
are likely to be formed as intermediates leading to the
pyrrolyl and isoindolyl metabolites. In this analysis no
e€ort was made to distinguish between the single electron
Introduction of a phenyl group at C-3 (compound 6)
resulted in an increase in Vmax and a decrease in Km
1
values which makes 6 (Vmax/Km=2054 min 1 mM
)
a
better substrate than MPTP (Vmax/Km=
1431 min 1 mM 1) [26]. Two allylic radicals may form
from 6, and therefore the pathway leading to 28 is not
obvious. Calculations indicate that the trans ꢃ-carbon
radical 34 is favored over the cis isomer 35 by 3.94 kcal/
mol. A comparison of the HOMO energy surfaces
(Fig. 1) of the two ꢃ-carbon radical species provides
evidence of better orbital overlap of the unpaired elec-
tron with the aromatic ring through the double bond for
the trans radical 34. Additional support for the stabili-
zation of 34 by the aromatic group comes from the
coplanarity of the phenyl and pyrrolinyl rings (ꢄ1=0ꢀ,
Fig. 2). In the case of the cis isomer 35, the double bond
is more engaged in resonance with the radical and less
with the phenyl ring resulting in a greater dihedral angle
Table 1
MAO-B catalyzed oxidation of pyrroline and isoindoline
derivatives
Substrate
Vmax
Km
(mM)
Vmax/Km
(min mM)
1
1
(min
)
MPTP
273
214
397
62
0.191
0.248
0.193
0.267
0.126
0.351
1431
862
5
6
2054
231
8
9
10
80
132
639
376

146
Y.-X. Wang et al./Bioorg. Med. Chem. 6 (1998) 143±149
R
R
R
R
and/or
N
N
N
N
R = H:
R = Ph:
33
5
6
27
28
35
34
R
R
R
R
and/or
N
N
N
N
R = H:
R = CH3:
8
9
36
29
30
31
37
39
38
40
R = CF3: 10
Scheme 4. Proposed pathways for the MAO-B catalyzed oxidations of 3-pyrrolines and isoindolines.
(ꢄ1=9ꢀ, Fig. 3) and reducing the steric interactions
between the two rings. Radicals in both the cis and trans
positions increase the N-methyl/ring angle ꢄ2 as a con-
sequence of the better overlap of the nitrogen lone pair
with the radical.
chosen for their similar steric and lipophilic properties
(LogP 9=1.94, LogP 10=2.05) but inverse electronic
e€ects (ꢅp 9= 0.14, ꢅp 10=+0.53). Both substituents
led to a moderate enhancement in activity with the
electron donating methyl group being more in¯uencial
than the electron withdrawing tri¯uoromethyl group
(Table 1). The essentially identical ÁÁHf values for the
formation of the lowest energy species [the trans radical
in both cases (37 from 9 and 39 from 10)], however, do
not o€er a good opportunity to evaluate electronic
e€ects in this series as anticipated.
The improved substrate properties of 6 relative to 5
may be a consequence of the increased stability of the
allylic radical of 34 (ÁÁHf ˆ 10:635 kcal=mol) derived
from 6 versus 33 (ÁÁHf ˆ 15:875 kcal=mol) derived
from 5. The geometry of 6, however, also may con-
tribute to its good substrate properties. The minimum
energy conformer of 6 (Fig. 3) shows that the phenyl
group occupies space that corresponds to that of the
phenyl group of 1-methyl-4-benzyl-1,2,3,6-tetrahydro-
pyridine (1, R=CH2Ph) which is an excellent MAO-B
substrate (Vmax/Km=3580 min 1 mM 1) [26]. It should
be noted, however, that the minimum energy conformer
of 6 shows a dihedral angle of only 6ꢀ between the two
rings as opposed to 30±35ꢀ for MPTP [27].
In summary, the good substrate properties of these
1,4-disubstituted
compounds
and
1,2,3,6-tetra-
hydropyridinyl derivatives, particularly when compared
to the lack of substrate properties of the corresponding
reduced azacyclic systems, suggest that the allylic func-
tionality plays an important role in the MAO-B cataly-
tic process. A general trend is observed between ÁÁHf
values for conversion of the parent amines to the pro-
posed ꢃ-carbon radical intermediates, suggesting that
the C±H bond energy at the carbon center undergoing
The methyl substituted (9) and tri¯uoromethyl sub-
stituted (10) isoindolinyl derivatives initially were
Table 2
Di€erences in the heats of formation (ÁÁHf) of allylic or benzylic radicals 32±39 that may be derived from the substrates MPTP, 4±6,
8±10, and 32a
Compound
MPTP
4
5
6
8
9
10
32
ÁÁHfC2
ÁÁHfC5ꢁC6†
18.195
6.628
19.048
Ð
15.875
Ð
14.575
10.635
16.835
Ð
16.731
16.511
16.757
16.421
21.644
Ð
aValues (kcal/mol) calculated using AM1. NB: ÁHfH ˆ 52:103 kcal=mol.

Y.-X. Wang et al./Bioorg. Med. Chem. 6 (1998) 143±149
147
oxidation may predict substrate properties within this
series. It is clear, however, that the enzyme active site
also is sensitive to the geometry of these types of com-
pounds. Therefore it may be possible to exploit struc-
tural analogs of these azacycles to investigate regions of
the active site of MAO-B that are not readily accessed
with tetrahydropyridinyl derivatives.
3. Experimental
3.1 General
All chemicals were reagent or HPLC grade. Proton
NMR spectra were recorded on a Bruker WP 270 MHz
spectrometer. Chemical shifts are reported in ppm rela-
tive to tetramethylsilane (TMS, ꢆ ˆ 0) and spin multi-
plicities are given as s (singlet), d (doublet), t (triplet), or
m (multiplet). Gas chromatography±electron ionization
mass spectrometry (GC±EIMS) was performed on a
Hewlett-Packard 5890 GC ®tted with an HP-1 capillary
column (20 mÂ200 mmÂ0.33 mm ®lm thickness) which
was coupled to a Hewlett-Packard 5870 mass-selective
detector. Data were acquired using an HP 5970 Chem-
Station. Normalized peak heights are reported as a per-
centage of the base peak. Compound 5 was synthesized
according to the literature [28] and was converted to its
oxalate salt in dry Et2O, which was recrystallized from
MeOH/Et2O: mp 137±138ꢀC; 1H NMR (DMSO-d6) ꢆ
5.91 (s, 2H, vinyl of C3 and C4), 4.02 (s, 4H, CH2 of C2
and C5), 2.87 (s, 3H, CH3). The free base of 2-methyliso-
indoline (8) was prepared as reported previousꢀly [29]
and was converted to its oxalate salt: mp 166±167 C; 1H
NMR (DMSO-d6) ꢆ 7.32±7.41 (m, 4H, phenyl), 4.53 (s,
4H, CH2 of C1 and C3), 2.94 (s, 3H, CH3); GC-EIMS
Fig. 1. HOMO energy surfaces of the two possible ꢃ-carbon
radicals 34 and 35 derived from 1-methyl-3-phenyl-3-
pyrroline (6).
(free base) tR 4.00 min, m/z (M +) 133. Enzyme kinetic
.
studies were performed on a Beckman Model DU-7400
spectrophotometer. Melting points were determined
using a Thomas±Hoover melting point apparatus and
are uncorrected.
Fig. 2. Geometry of 1-methyl-3-phenyl-3-pyrroline (6) and the
trans (34) and cis (35) ꢃ-carbon radicals.
3.1.1 Trans-1-methyl-4-phenyl-3-phenylselenyl-2-
pyrrolidinone (12)
To a solution of diisopropylamine (2.39 g, 0.0236 mol)
in THF (50 ml) was added n-BuLi (2.5 M, 8.80 ml,
0.022 mol) at 20ꢀC with stirring under N2. After the
addition the mixture was kept at 20ꢀC for 30 min and
then was cooled to 78ꢀC. A solution of 1-methyl-4-
phenyl-2-pyrrolidinone [18] (11, 1.80 g, 0.01027 mol) in
THF (10 ml) was then added dropwise to the LDA
solution at 78ꢀC over 15 min. The resulting solution
was stirred at 78ꢀC for 40 min and 0ꢀC for 1 h. The
solution was quenched with water and extracted with
Et2O. The resulting crude product was chromato-
graphed (silica-gel 35 g, eluent CH2Cl2) to give 1.98 g
(58%) of pure 12 as an oil: 1H NMR (CDCl3) ꢆ 7.10±7.67
Fig. 3. Overlay of the minimum energy conformers of 1-
methyl-3-phenyl-3-pyrroline (6) and 1-methyl-4-benzyl-1,2,3,6-
tetrahydropyridine (1, R = CH2Ph).

148
Y.-X. Wang et al./Bioorg. Med. Chem. 6 (1998) 143±149
(m, 10H, phenyls), 3.85 (d, 1H, C3, J3 ˆ 5:4 Hz), 3.47
benzene (24) 1H NMR (CDCl3) ꢆ 7.41±7.53 (m, 4H,
phenyl), 5.36 (s, 4H, CH2), 2.99 (s, 6H, CH3); 3,4-bis-
mesyloxytoluene (25) 1H NMR (CDCl3) ꢆ 7.21±7.39 (m,
4H, phenyl), 5.32 (s, 4H, CH2), 2.99 (s, 3H, CH3), 2.97
(s, 3H, CH3), 2.39 (s, 3H, CH3; 1,2-bis-mesyloxy-4-tri-
¯uoromethylbenzene (26) 1H NMR (CDCl3) ꢆ 7.57±7.75
(m, 4H, phenyl), 5.39 (s, 2H, CH2), 5.38 (s, 2H, CH2),
3.07 (s, 3H, CH3), 3.06 (s, 3H, CH3).
4
(m, 1H, C4), 3.22±3.38 (m, 2H, C5), 2.84 (s, 3H, CH3);
GC-EIMS tR 10.11 min, m/z (M.+) 331; HRMS
(CI). Calcd for C17H18NO76Se: 328.0581. Found:
328.0586.
3.1.2 Trans-1-methyl-3-phenylselenyl-4-phenylpyrrolidine
(14)
To a solution of 12 (1.32 g, 0.004 mol) in THF (15 ml)
.
was added BH3 THF (1.0 M, 24 ml, 0.024 mol) at room
3.1.5 Oxalate salts of isoindolines 9 and 10
temperature with stirring under N2. The mixture was
kept at room temperature for 30 min and then was
heated under re¯ux for 22 h. After cooling in an ice-
water bath, the reaction was quenched with 6 N aqueous
HCl and the resulting mixture was heated under re¯ux
for 3 h and then was cooled, made basic with aqueous
NaOH and extracted with Et2O. The residue obtained
after rotary evaporation was puri®ed by column chro-
matography (silica-gel 50 g, eluent CH2Cl2 to 10%
MeOH in CH2Cl2) to give 1.20 g (95%) of pure 14 as an
oil: 1H NMR (CDCl3) ꢆ 7.23±7.42 (m, 10H, phenyls),
3.91 (m, 1H, C4), 3.66 (m, 2H, C2), 3.46 (m, 2H, C5),
To a solution of each of the bismesylates 25 and 26
(0.084 mol) was added dropwise a solution of methyla-
mine in THF (2M, 12.6 ml, 0.0252 mol) at 0ꢀC with
stirring. The mixtures were kept at 0ꢀC for an additional
4 h and then at room temperature for 15 h. The resulting
mixtures were washed successively with water and brine.
The oxalate salts were prepared in dry Et2O and recrys-
tallized from MeOH/Et2O. The oxalate salt of 2,5-
dimethylisoindoline (9) was obtained in 36% yield:
mp 177±178ꢀC; 1H NMR (DMSO-d6) ꢆ 7.14±7.27
(m, 3H, phenyl), 4.49 (s, 4H, CH2 of C1 and C3), 2.93
(s, 3H, N-CH3), 2.31 (s, 3H, CH3 of C5); GC-EIMS
(free base) tR 4.60 min, m/z (M.+) 147. Anal. calcd for
C12H15NO4: C, 60.75; H, 6.37; N, 5.90%. Found C,
60.50; H, 6.28; N, 5.85%. The oxalate salt of 2-methyl-
5-tri¯uoromethylisoindoline (10) was obtained in 45%
2.87 (s, 3H, CH3), 1.68 (m, 1H, C3); GC±EIMS tR 8.99
+
.
min, m/z (M
)
317; HRMS (CI). Calcd for
C17H20N76Se: 314.0787. Found: 314.0779.
yield: mp 174±175ꢀC; H NMR (DMSO-d6) ꢆ 7.59±7.79
1
3.1.3 Oxalate salt of 1-methyl-3-phenyl-3-pyrroline (6)
To a solution of 14 (0.60 g, 0.0019 mol) in 15 ml
THF was added 30% H2O2 (0.30 g, 0.0266 mol) drop-
wise at 0ꢀC with stirring. The mixture was kept at 0ꢀC
for 30 min and then at room temperature for 1.5 h.
Following the addition of aqueous 10% Na2SO3 (5 ml)
and 10% aqueous Na2CO3 (10 ml), the mixture was
extracted with Et2O. Column chromatography of the
organic isolate (silica-gel 25 g, eluents EtOAc to 10%
MeOH in EtOAc) gave the free base 6 which was further
puri®ed as its oxalate salt (0.14 g, 30%): 1H NMR
(CD3OD) ꢆ 7.25±7.37 (m, 5H, phenyl), 6.18 (m, 1H,
C4), 4.44 (s, 2H, C2), 4.21 (unresolved, 1H, C5), 2.98
(s, 3H, CH3); GC±EIMS (free base) tR 6.38 min, m/z
(m, 3H, phenyl), 4.51 (s, 4H, CH2 of C1 and C3), 2.93
(s, 3H, N-CH3); GC-EIMS (free base) tR 4.09 min, m/z
(M.+) 201. Anal. calcd for C12H12F3NO4: C, 49.49; H,
4.15; N, 4.81%. Found C, 49.26; H, 4.18; N, 4.72%.
3.2 Enzyme substrate studies
The isolation and puri®cation of MAO-B from beef
liver were carried out using the procedures reported by
Salach [30] with the following modi®cations. We did not
subject the MAO-B preparation to the glucose gradient
puri®cation step. The speci®c activity of MAO-B
(9 nmol/ml) was established with MPTP as substrate at
30ꢀC (Vmax=204 min 1) as reported earlier [31]. The
MAO-B preparation was found to be stable when stored
at 15ꢀC over the period of this study.
(M +) 159; Anal. calcd for C13H15NO4: C, 62.64;
.
H, 6.07; N, 5.62%. Found C, 62.48; H, 6.08; N,
5.53%.
Solutions of the oxalate salts of the test compounds in
phosphate bu€er (pH 7.4, 0.5 mM, ®nal volume 500 ꢂl)
in a 1 ml quartz cuvette were treated with 10 ꢂl of the
MAO-B preparation (®nal concentration 0.16 ꢂM) and
the cuvette was placed in a Beckman Model DU-7400
spectrophotometer maintained at 37ꢀC. The substrate
properties were evaluated qualitatively by obtaining a
series of scans (450±250 nm) versus time over a 1 h per-
iod for each compound.
Kinetic studies with MAO-B were carried out using a
Beckman DU-7400 spectrophotometer. Solutions of the
test compounds (®nal volume 510 ꢂl, ®nal substrate
concentrations 125 to 4000 ꢂM) in 100 mM sodium
3.1.4 Bismesylates 24±26
To a solution of methanesulfonyl chloride (5.50 g,
0.048 mol) in dry CH2Cl2 (40 ml) was added dropwise a
solution of diol 21, 22, or 23 [21] (0.012 mol) and trie-
thylamine (5.59 g, 0.055 mol) in CH2Cl2 (15 ml) at 0ꢀC
with stirring under nitrogen. After an additional 30 min
at 0ꢀC the reaction mixture was washed successively
with 30 ml each of ice-cold water, 10% HCl, saturated
NaHCO3 and brine. The organic layer was separated
and dried over MgSO4 to give quantitative yields of the
oily bismesylates 24, 25, and 26 which were used in the
next step without further puri®cation: 1,2-bismesyloxy-

Y.-X. Wang et al./Bioorg. Med. Chem. 6 (1998) 143±149
149
phosphate (pH 7.4) were incubated in the presence of
0.16ꢂM MAO-B. The rates of oxidation were obtained
by monitoring the increment in absorbance of the
metabolic aromatic products 27±31 over a 30±120 s time
period. The Vmax and Km values were calculated from
double reciprocal plots.
[7] Salach JI, Singer TP, Castagnoli N Jr, Trevor A. Biochem
Biophys Res Commun 1984;125:831.
[8] Kalgutkar AS, Castagnoli
1992;35:4165.
N
Jr.
J
Med Chem
[9] Kuttab S, Kalgutkar A, Castagnoli N Jr. Chem Res Tox-
icol 1994;7:740.
[10] Nimkar SK, Anderson A, Rimoldi JM, Stanton M, Cas-
tagnoli KP, Mabic S, Wang X-Y, Castagnoli N Jr. Chem
Res Toxicol 1996;9:1013.
[11] Rimoldi JM, Wang Y-X, Nimkar SK, Kuttab SH,
Anderson AH, Burch H, Castagnoli N Jr. Chem Res
Toxicol 1995;8:703.
3.3 Calculations
Molecular properties were determined using the semi-
empirical method AM1 [32] within MacSpartan soft-
ware (1996, Wavefunction, Inc.). Calculations used the
default restricted Hartree±Fock (RHF) method for the
self-consistent ®eld (SCF) except for calculations on the
radical species that were performed using the unrest-
ricted Hartree±Fock (UHF) model. The ACD/LogP 1.0
software from ACD/Labs was used to calculate the
LogP values.
[12] Efange SMN, Boudreau RJ.
Design 1991;5 :405.
J Computer-aided Mol
[13] Altomare C, Carrupt PA, Gaillard P, El Tayar N, Testa B,
Carotti A. Chem Res Toxicol 1992;5:366.
[14] Palmer S, Mabic S, Castagnoli N Jr. J Med Chem 1997.
[15] Heikkila RE, Manzino L, Cabbat FS, Duvoisin RC.
Neurosci Lett 1985;58:133.
[16] Ottoboni S, Caldera P, Trevor A, Castagnoli N Jr. J Biol
Chem 1989;264:13684.
[17] Chavdarian CG. J Org Chem 1983;48:1529.
[18] Rasmussen CR, Gardocki JF, Plampin JN, Twardzik
BL, Reynolds BE, Molinari AJ, Schwartz N, Bennetts
WW, Price BE, Marakowski J.
1978;21:1044.
J
Med Chem
Acknowledgements
[19] Pawlowski G, Hanack M. Synth Commun 1981;11:351.
[20] Farrooq O. Synthesis 1994:1035.
[21] Ding Z, Tufariello J. J Synth Commun 1990;20:227.
[22] Silverman RB. Account Chem Res 1995;28:335.
[23] Anderson A, Kuttab S, Castagnoli N Jr. Biochemistry
1996;35:335.
This study was supported by the National Institute of
Health (NS 28792) and the Harvey W. Peters Center for
the Study of Parkinson's Disease.
[24] Maret G, Testa B, Jenner P, El Tayar N, Carrupt PA.
Drug Metabolism Rev 1990;22:291.
[25] Heikkila RE, Manzino L, Cabbat FS, Duvoisin RC.
Neurosci Lett 1985;58:133.
References
[1] Kuhn DM, Murphy DL, Youdim MBH. In Kuhn DM,
Murphy DL, Youdim MBH, editors. Physiological and
clinical aspects of monoamine oxidase. Structure and
[26] Wang Y-X, Castagnoli N Jr. J Med Chem 1995;38:1904.
[27] Harris D. In Harris D, editor. Quantitative structure-
activity relationships of monoamine oxidase catalyzed
oxidation of 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine
analogs; Blacksburg:Virginia Tech, 1996:137.
function of amine oxidases,
1985;22:231.
Boca Raton:CRC,
[28] Anderson WK, Milowsky AS.
1985;50:5423.
[29] Hernandez JE, Fernandez S, Arias G. nth Commun
1988;18:2055.
[30] Salach JI, Weyler W. In Salach JI, Weyler W, editors.
Preparation of the ¯avin-containing aromatic amine oxi-
dases of human placenta and beef liver; London:Aca-
demic, 1987;142:627.
[31] Kalgutkar AS, Castagnoli K, Hall A, Castagnoli N Jr. J
Med Chem 1994;37:944.
J
Org Chem
[2] Back AJW, Lan NC, Johnson DL, Abell CW, Bembenek
ME, Kwan SW, Seeburg P, Shih JC. Proc Natl Acad Sci
USA 1988;85:4934.
[3] Hsu Y-PP, Weyler W, Chen S, Sims KB, Rinehart WB,
Utterback M, Powell JF, Break®eld XO. J Neurochem
1988;51:1321.
[4] Youngster SK, Sonsalla PK, Sieber B-A, Heikkila RE. J
Pharmacol Exp Therapeutics 1989;249:820.
[5] Krueger MJ, Efange SMN, Michelson RH, Singer TP.
Biochemistry 1992;31:5611.
[6] Chiba K, Trevor A, Castagnoli N Jr. Biochem Biophys
Res Commun 1984;120:574.
[32] Dewar MJS, Zoebish EG, Healy EF, Stewart JJP. J Amer
Chem Soc 1985;107:8646.