Stereochemical studies on the novel monoamine oxidase B substrates (1R,6S)- and (1S,6R)-3-methyl-6-phenyl-3-aza-bicyclo[4.1.0]heptane

Article abstract of DOI:10.1016/j.bmc.2008.02.014

Previous studies have established the unexpected monoamine oxidase-B (MAO-B) substrate properties of racemic 3-methyl-6-phenyl-3-aza-bicyclo[4.1.0]heptane, the 3,4-cyclopropyl analog of the achiral proneurotoxin 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridin

Full text of DOI:10.1016/j.bmc.2008.02.014

Available online at www.sciencedirect.com
Bioorganic & Medicinal Chemistry 16 (2008) 3557–3564
Stereochemical studies on the novel monoamine oxidase
B substrates (1R,6S)- and (1S,6R)-3-methyl-6-phenyl-
3-aza-bicyclo[4.1.0]heptane
Philippe Bissel,a,* Ashraf Khalil,a John M. Rimoldi,b Kazuo Igarashi,a
Dale Edmondson,c Anthony Millera and Neal Castagnoli, Jr.a,d
aDepartment of Chemistry, Virginia Polytechnic Institute and State University, West Campus Dr., Blacksburg, VA 24061-0212, USA
bDepartment of Medicinal Chemistry and Laboratory for Applied Drug Design and Synthesis,
The University of Mississippi, MS 38677, USA
cThe Departments of Biochemistry and Chemistry, Emory University, Atlanta, GA 30322, USA
dThe Edward Via Collage of Osteopathic Medicine, Blacksburg, VA 24061, USA
Received 11 January 2008; revised 31 January 2008; accepted 5 February 2008
Available online 8 February 2008
Abstract—Previous studies have established the unexpected monoamine oxidase-B (MAO-B) substrate properties of racemic
3-methyl-6-phenyl-3-aza-bicyclo[4.1.0]heptane, the 3,4-cyclopropyl analog of the achiral proneurotoxin 1-methyl-4-phenyl-1,2,3,6-
tetrahydropyridine (MPTP). The two stereocenters present in this compound provide an opportunity to examine the enantioselec-
tivity and diastereoselectivity of the MAO-B-catalyzed ring a-carbon oxidation of cyclic tertiary amines to give the corresponding
conjugated iminiumyl metabolites. This paper reports the results of such stereochemical studies using expressed human MAO-B as
the catalyst.
ꢀ 2008 Elsevier Ltd. All rights reserved.
1. Introduction
dihydropyridiniumyl metabolite 2H+.14–17 This bio-
transformation is regiospecific in that oxidation at the
homoallylic, unconjugated position to give the imi-
niumyl isomer 3H+ is not observed.18 The MAO-B sub-
strate properties of MPTP are dependent on the
presence of the 4,5-p bond of the heterocyclic moiety
The mitochondrial flavoenzyme monoamine oxidase A
and B (MAO-A and MAO-B) catalyze the two-electron
a-carbon oxidation of a variety of primary and second-
ary amines including the aminyl neurotransmitters
dopamine, serotonin and norepinephrine.1,2 Reports of
tertiary aminyl substrates are few3,4 and, with the excep-
tion of the class of compounds under consideration in
this study, no cyclic tertiary amine has been described
as an MAO substrate. This exceptional class of com-
pounds includes 5-membered5,6 and 6-membered7 cyclic
tertiary allylamines that undergo ring a-carbon MAO-
catalyzed oxidation to give cyclic iminiumyl metabo-
lites.8–13 An important example of this pathway is the
MAO-B-catalyzed bioactivation of the parkinsonian
inducing proneurotoxin 1-methyl-4-phenyl-1,2,3,6-tetra-
hydropyridine [MPTP (1)] to give the conjugated
since the corresponding piperidinyl analog 4 is not an
19,20
MAO substrate
(Scheme 1).
In view of the p-bond characteristics of the cyclopropyl
group,21–25 studies were undertaken recently to examine
the human placental mitochondrial MAO-A and baboon
liver mitochondrial MAO-B substrate properties of
racemic 3-methyl-6-phenyl-3-aza-bicyclo[4.1.0]heptane
H+
C6H5
C6H5
C6H5
C6H5
FAD FADH2
2
6
MAO-B
N
N
N
N
Keywords: Monoamine oxidase B; Cyclopropyl analog of MPTP; E-
nantioselectivity; Diastereoselectivity.
CH3
4
CH3
3H+
CH3
2H+
CH3
1
*
Corresponding author. Tel.: +1 540 231 8233; fax: +1 540 231
3255; e-mail: pbissel@vt.edu
Scheme 1.
0968-0896/$ - see front matter ꢀ 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2008.02.014

3558
P. Bissel et al. / Bioorg. Med. Chem. 16 (2008) 3557–3564
(5ab),26 a fused cyclopropyl analog of MPTP. These stud-
ies established that 5ab is biotransformed in the presence
of MAO-B to the 3-methyl-6-phenyl-3-azabicy-
clo[4.1.0]hept-2-enyliminiumyl species 6ab+ (Scheme 2).27
the three carbon signals at d 55.1, 45.1 and 51.9 ppm can
be assigned with confidence to the three carbon atoms
attached to nitrogen. Furthermore, the most deshielded
of these three signals (d 55.1 ppm) must be due to C(2)
since it is disubstituted and since the cyclopropyl group
is deshielding.29 The three highest field proton signals in
the spectrum (d 1.42, 0.88 and 1.04 ppm) can be assigned
to the three protons attached to the cyclopropyl group,
C(1)H, C(7)Ha and C7Hb. The most deshielded of these
three signals (d 1.42 ppm) must be due to the methine
proton, that is, C(1)H. The proton signals at d 2.89
(dd, J = 11.8gem and 6.3vic Hz) and 2.72 (dd,
J = 11.5gem and 1.7vic Hz) correlate with C(2) and there-
fore these signals must be C(2)HR and C(2)HS. The cis
and trans relationships between C(1)H and these two
signals should provide an opportunity to assign the
pro-R and pro-S protons. The calculated (Chem 3D)
dihedral angle for H(1)–C(1)–C(2)–HR(2) is ꢀ38ꢁ. The
corresponding dihedral angle for H(1)–C(1)–C(2)–
HS(2) is ꢀ74ꢁ. The Karplus relationship predicts a cis
coupling with J38ꢁ of ꢀ6 Hz and a trans coupling with
J74ꢁ of ꢀ2 Hz. The experimentally determined vicinal
cis and trans coupling constants are 6.3 and 1.7 Hz.
Therefore, the signal at d 2.72 is assigned to C(2)HR,
the proton cis to C(1)H, and the signal at d 2.89 to
C(2)HS, the proton trans to C(1)H. Several 2D correla-
tions support these assignments (see Fig. 1). Particularly
convincing is the NOESY observed between C(1)H and
C(2)HR but not C(2)HS.
Compound 5 bears two stereocenters at C(1) and C(6).
Furthermore, C(2), the carbon atom reported to under-
go MAO-B-catalyzed oxidation, is a pro-stereocenter.
The present studies report the results of efforts to char-
acterize the enantioselectivity (5a vs 5b) and diastereose-
lectivity (2HR vs 2HS) of the expressed human MAO-B-
catalyzed oxidation of 5. Of particular interest is the
opportunity to examine these stereochemical parameters
with a substrate having fixed geometries at the positions
b and c to the nitrogen atom.
C6H5
C6H5
HS
HR
HS
HR
N
N
CH3
CH3
5a
(1R,6S)
5b
(1S,6R)
2. Results and discussion
2.1. Synthesis
The resolution of 5ab was accomplished by recrystalliza-
tion of the diastereomeric salts obtained with (R)- and
(S)-camphorsulfonic acid. The diastereomeric composi-
tions of the intermediate stages of the resolution were
monitored by 13C NMR. The absolute configurations
of the 5a and 5b were established by X-ray crystallogra-
phy of the perchlorate salt of 5b.
The preparation of the required monodeuterated sub-
strates was approached by reduction of the iminiumyl
double bond of 6+. Compounds 6a+ and 6b+ were ob-
tained in pure form from 7a/8a and 7b/8b, respectively,
by the reaction pathway reported previously for the syn-
thesis of 6ab+ (Scheme 3, illustrated with 7a/8a).27 The
deuteration was attempted with a variety of deuterium
enriched reagents [NaCND3, Li-9-BBN-D, LiAl(tO-
Bu)3D] under a variety of conditions (H2O, MeOH,
THF).30 The best diastereoselectivity was obtained with
NaBD4 in methanol at 0 ꢁC (Scheme 3, the reaction is
illustrated with intermediate 6a+). The product obtained
from 6a+ displayed an NMR spectrum with the signal
for 2HR (d 2.72 ppm) integrating for 0.75 protons and
the signal for 2HS (d 2.89 ppm) integrating for 0.25 pro-
tons. Consequently deuteride attack occurred preferen-
tially from the side opposite the cyclopropyl group to
give 3 parts 5a-(2S)-d1 and 1 part 5a-(2R)-d1 (the ‘5a-
2-d1 75:25 isotopoepimeric mixture’). The corresponding
‘5b-2-d1 75:25 isotopoepimeric mixture’ was obtained
starting with N-oxides 7b/8b.
An examination of the potential diastereoselective loss
of 2HR versus 2HS of 5a and 5b required the syntheses
of the isotopoepimers 5a-(2R)-d1 or 5a-(2S)-d1 and 5b-
(2R)-d1 or 5b-(2S)-d1. Characterization of the products
obtained from the synthetic efforts (see below) was made
possible by NMR studies that led to the identification of
the 2HR and 2HS signals of 5.
COSY, NOESY, HSQC and HMBC spectral studies led
to assignments of the carbon and proton signals of 5 as
summarized in Figure 1 and Table 1 with the (1S,6R)-
enantiomer 5b. A molecular model of 5b is included for
clarity.
Analysis of the spectral data that led to the critical
assignments for the signals of 2HR and 2HS is as follows:
The inverse isotopoepimeric mixture, the ‘5a-2-d1 25:75
isotopoepimeric mixture’ was obtained by NaBH4
reduction of 6a+-2-d1. Intermediate 6a+-2-d1 was pre-
pared by the reaction sequence leading to 6a+-d0 only
starting with 5a-2,2-d2 (Scheme 4). The synthesis of
5a-2,2-d2 was achieved by cyclopropylation of the
known MPTP analog31 1-6,6-d2 followed by resolution
of the resulting 5ab-2,2-d2. The conversion of 5a-2,2-d2
to 6a+-2-d1 proceeded by treatment of the N-oxides 7a/
8a-2,2-d2 as described for 7a/8a-d0. NaBH4 converted
6a+-2-d1 to a mixture in which 5a-(2S)-d1 was the minor
H+
C6H5
6
FADH2
C6H5
FAD
7
1
MAO-B
N3
N
CH3
CH3
6ab+
5ab
Scheme 2.

P. Bissel et al. / Bioorg. Med. Chem. 16 (2008) 3557–3564
3559
7.2-7.4
H
0.88/1.04 ppm
C6H5
H
H
H
H
H
N
CH3
HMBC
H
2.17/2.18 ppm
2.22/2.29 ppm
H
H
HS
HR
H
H
HS
HR
H1
HS
HR
1.42 ppm
2.89 ppm
2.72 ppm
C6H5
C6H5
C6H5
H
H
HS
H
H
N
N
CH3
N
CH3
HR
CH3
COSY
NOESY
2.27 ppm
Figure 1. Summary of 2D correlations and chemical shift assignments for the proton signals of 5b.
Table 1. Chemical shifts, proton–proton couplings and coupling constants for 5b
Position
dC (ppm)
H
dH (ppm) (multiplicity)
J (Hz)
1
2
19.5 (CH)
55.1 (CH2)
C(1)H
1.42 (m)
C(2)HR
C(2)HS
N(3)CH3
C(4)H2
2.72 (dd)
2.89 (dd)
2.27 (s)
11.8gem, 6.3vic
11.5gem, 1.7vic
3
4
45.1
51.9 (CH2)
2.22 (ddd)
2.29 (ddd)
2.17 (ddd)
2.18 (ddd)
5
31.8 (CH2)
C(5)H2
6
7
22.4 (q-C)
18.1 (CH2)
28
(7)HaHb
0.88 (dd)
1.04 (dd)
4.3gem
,
4.3gem, 9.1vic
5.3vic
10
148 (q-C)
20,60
3050
40
128.8 (CH)
127.5 (CH)
125.7 (CH)
C(20,30,40,50,60)H5
7.2–7.4 (m)
C6H5
C6H5
C6H5
C6H5
+
2.72 ppm
(1) TFAA
NaBD4
2.89 ppm
H
D
D
H
N
N
N
(2) Alumina
N
CH3
CH3
O
CH3
CH3
5a-(2R)-d1
5a-(2S)-d1
6a+
7a/8a
(75%)
(25%)
The "5a-2-d1 75:25 isotopoepimeric mixture"
Scheme 3.
C6H5
C6H5
C6H5
C6H5
C6H5
C6H5
(1) Zn(Et)2/CH2I2
m-CPBA
(1) TFAA
(2) Alumina
H
D
D
H
NaBH4
+
D
D
D
D
D
D
CH2Cl2/TFA
(2) Resolution
N
N
N
N
N
D
N
CH3
CH3
CH3
CH3
CH3
O
CH3
5a-(2R)-d1
5a-(2S)-d1
7a/8a-2,2-d2
6a+-2-d1
1-6,6-d2
5a-2,2-d2
(25%)
(75%)
The "5a-2-d1 25:75 epimeric mixture"
Scheme 4.
isotopoepimer and 5a-(2R)-d1 the major isotopoepimer
(Scheme 4). NMR analysis confirmed the expected com-
position of 25% 5a-(2S)-d1 and 75% 5a-(2R)-d1. The cor-
responding reaction sequence with 5b-2,2-d2 was not
pursued.
2.2. Enzymology
A few examples of enantioselective MAO-catalyzed con-
versions of chiral substrates32,3,33 and mechanism-based
inactivators34–37 have been reported. The present paper

3560
P. Bissel et al. / Bioorg. Med. Chem. 16 (2008) 3557–3564
describes the results of studies on the enantioselectivity
of the ring a-carbon oxidations of 5a versus 5b.
Support for the proposed diastereoselectivity of this
reaction was sought by examining the metabolic fate
of the ‘5a-2-d1 25:75 isotopoepimeric mixture’. The incu-
bations were conducted with 2.4 lM MAO-B and
500 lM substrate. The increased ratio in the concentra-
tion of enzyme to substrate was prompted by a possible
deuterium isotope effect referred to above since the ma-
jor component of this mixture, 5a-(2R)-d1, bears the
deuterium cis to the cyclopropyl group. LC-ESI/MS ex-
tracted ion chromatographic analyses again indicated
that less than 5% of the starting substrate remained fol-
lowing the 2-h incubation period. The average values of
3 such incubations gave a ratio for 6a+-d0:6a+-2-d1 of
68:32 (Scheme 6). This value, which is close to the in-
verse of the 29:71 ratio obtained with the ‘5a-2-d1
75:25 isotopoepimeric mixture’, is consistent with the
conclusion that the MAO-B-catalyzed oxidation of 5a
is a highly diastereoselective process with 2HR being
preferentially lost.
The substrate properties of the (1R,6S)-enantiomer 5a
and the (1S,6R)-enantiomer 5b were examined spectro-
photometrically (kmax 258 nm) with the aid of a microti-
ter plate reader using expressed human MAO-B
(50 nM). All plots of metabolite concentration versus
time were linear for 30 min. Double reciprocal plots of
the rates of metabolite formation versus the concentra-
tion of 5a (100–1000 lM) gave
a Km value of
221 27 lM and a Vmax value of 135 5 nmol 6a+
formed/min-nmol MAO-B (or minꢁ1). The correspond-
ing values for 5b (250–3000 lM) were 753 22 lM and
127 5 minꢁ1. The ratio of the Vmax/Km values for 5a/5b
is 617/127 = 4.9 (the average of 4 determinations for
each enantiomer). When repeated with 250 nM MAO-
B and a data collection period of 5 min, the average
5a/5b ratio of Vmax/Km was 4.8. This reaction, therefore,
is enantioselective with 5a being the preferred substrate.
A final experiment examined the fate of the ‘5b-2-d1
75:25 isotopoepimeric mixture’. LC-ESI/MS extracted
ion chromatographic analyses of the 2-h incubation
mixtures indicated only 88% consumption of substrate
even though the enzyme:substrate ratio was 2.4:500.
The average percentage (3 determinations) of the inte-
grated ion current at m/z 186 attributable to 6b+-d0
was 68%; the corresponding value for 6b-2-d1 at m/z
187, when corrected for the 13C-satellite contribution
from m/z 186, was 32% (Scheme 7). This result docu-
ments that, as observed with the (1R,6S)-enantiomer
5a, the 2HR proton is selectively lost in the MAO-B-cat-
alyzed oxidation of the (1S,6R)-enantiomer 5b. In the
case of 5a, however, the 2HR proton is cis to the cyclo-
propyl group while in 5b the 2HR proton is trans to the
cyclopropyl group. Consequently, although the configu-
rations at C(1) and C(6) have an impact on the Vmax/Km
values of 5a versus 5b, the configurations of these stereo-
centers do not alter the stereochemical course of the
reaction since 2HR is lost selectively with both 5a and
5b. This behavior may prove useful when attempting
to model the complex between MAO-B and cyclic ter-
tiary aminyl substrates.
A high degree of a-carbon enantioselectivity has also
been reported for prochiral MAO substrates. Examples
include the mitochondrial MAO substrates dopamine
(9),38,39 benzylamine (10),40 and 1-aminoheptane
(11).41 All conversions are reported to proceed with
selective loss of the pro-R proton.
OH
CH3
HO
C6H5
(H2C)4
HS
H2N
HR
HS
HR
HS
HR
H2N
11
H2N
10
9
In the case of 5, the C(2) protons are diastereotopic. The
metabolic fate of the ‘5a-2-d1 75:25 isotopoepimeric mix-
ture’ (1 mM) was examined in the presence of MAO-B
(1 lM). LC-ESI/MS extracted ion chromatographic anal-
yses following work-up of the 2 h incubation mixtures
documented that less than 5% of the starting substrate re-
mained. Consequently, even though studies on related
tetrahydropyridinyl substrates have shown significant
primary kinetic deuterium isotope effects,42,43 under the
conditions used in the present study, any such isotope ef-
fect should not have a significant influence on the distribu-
tion of deuterium in the products of this reaction. The
deuterium content of 6a+ is summarized in Scheme 5.
The percentages (average of 3 determinations) of the total
integrated ion currents (m/z 186 + m/z 187) were 29% for
6a+-d0 (m/z 186) and, when corrected for the 13C-satellite
peak of m/z 186, 71% for 6a+-2-d1 (m/z 187). Although the
use of the mixture of monodeuterated isotopoepimers in
this study may limit the accuracy of this analysis, these
values clearly are consistent with a reaction that proceeds
with the selective loss of the proton and deuteron cis to the
cyclopropyl group in this ‘5a-2-d1 75:25 isotopoepimeric
mixture’.
3. Summary and conclusions
The results of these studies document that the MAO-B-
catalyzed ring a-carbon oxidation of 5 is enantioselec-
tive with an almost 5-fold preference for the (1R,6S)
enantiomer 5a based on Vmax/Km values. A summary
of the results of the diastereoselectivity of this reaction
(2HR vs 2HS) is presented in Scheme 8. In all cases the
predominant pathway in the oxidation of 5 to 6+ in-
volves the loss of 2HR even though 2HR is cis to the
cyclopropyl group in 5a and trans to the cyclopropyl
group in 5b. This outcome is consistent with the behav-
ior of prochiral substrates in that substituents b to the
nitrogen atom do not alter the stereochemical course
of the reactions with selective loss of the a-pro-R proton
observed with all substrates reported. The present re-
sults provide additional insight into the enzyme-medi-
ated stereochemical control of this type of oxidation in

P. Bissel et al. / Bioorg. Med. Chem. 16 (2008) 3557–3564
3561
C6H5
C6H5
C6H5
C6H5
FAD +H+ FADH2
+
+
H
D
D
H
N
N
N
D
N
H
MAO- B
CH3
CH3
5a-(2S)-d1 (75%)
CH3
6a+-2-d1 (71%)
CH3
6a+-d0 (29%)
5a-(2R)-d1 (25%)
The "5a-2-d1 75:25 isotopoepimeric mixture"
Scheme 5.
C6H5
C6H5
C6H5
C6H5
FAD +H+
FADH2
+
D
H
H
D
+
N
N
N
D
N
CH3
H
MAO-B
CH3
5a-(2S)-d1 (25%)
CH3
5a-(2R)-d1 (75%)
CH3
6a+-d0 (68%)
6a+-2-d1 (32%)
The "5a-2-d1 25:75 isotopoepimeric mixture"
Scheme 6.
C6H5
C6H5
C6H5
C6H5
FAD +H+ FADH2
MAO-B
+
+
D
H
H
D
N
N
N
D
N
H
CH3
CH3
CH3
CH3
5b-(2R)-d1 (75%)
5b-(2S)-d1 (25%)
6b+-d0 (68%)
6b+-2-d1 (32%)
The "5b-2-d1 75:25 isotopoepimeric mixture"
Scheme 7.
C6H5
C6H5
C6H5
C6H5
C6H5
C6H5
+
+
D
H
H
D
H
D
D
H
+
D
H
H
D
N
N
N
N
N
N
CH3
5a-(2S)-d1 (75%)
CH3
CH3
CH3
CH3
5b-(2S)-d1 (25%)
CH3
5a-(2R)-d1 (75%)
5a-(2R)-d1 (25%)
5a-(2S)-d1 (25%)
5b-(2R)-d1 (75%)
FAD +H+
FAD +H+
FAD +H+
FADH2
C6H5
FADH2
C6H5
FADH2
C6H5
C6H5
C6H5
C6H5
+
+
+
N
H
N
D
N
D
N
H
N
H
N
D
CH3
CH3
6a+-2-d1 (32%)
CH3
6a+-2-d1 (71%)
CH3
6a+-d0 (68%)
CH3
6b+-d0 (68%)
CH3
6b-2-d1 (32%)
6a+-d0 (29%)
Scheme 8.
that, unlike previous examples, the geometries of the
atoms b and c to the nitrogen atom are fixed. It may
be reasonable to speculate that the stereochemical
course of the MAO-B-catalyzed oxidations of amines
in general is dictated by the geometries of the nitrogen
and a-carbon atoms and a structurally rigid enzyme–
substrate complex. Molecular modeling studies that will
take advantage of these results and the reported X-ray
structure of the expressed human MAO-B used in this
work44–46 are being pursued.
4. Experimental
Important notice: some 1,4-disubstituted-1,2,3,6-tetra-
hydropyridines are known nigrostriatal neurotoxins

3562
P. Bissel et al. / Bioorg. Med. Chem. 16 (2008) 3557–3564
and should be handled using disposable gloves in a
properly ventilated hood following good laboratory
practices. Detailed procedures for the safe handling of
MPTP have been reported.47
J = 15 Hz, 1H), 2.86 (s, 3H), 2.97 (bs, 1H), 3.28 (s,
1H), 3.96 (bs, 1H), 7.21 (m, 1H), 7.31 (m, 2H), 7.38
(m, 2H); 13C NMR (125.8 MHz, CD3OD) d 15.1, 17.6,
18.8, 19.1, 22.5, 24.4, 26.5, 28.3, 42.0, 42.3, 42.7, 46.9,
49.3, 53.6, 58.3, 126.6, 127.4, 128.4, 145.0, 217.1. Anal.
Calcd for C23H33NO4S (419.59): C, 65.84; H, 7.93; N,
3.34. Found: C, 65.78; H, 7.90; N, 3.34. An ethereal
solution of the corresponding free base 5b (1.9 g,
10 mmol) was treated with methanolic HCl (1.2 equiv)
and the resulting solid was recrystallized from MeOH/
ether to give 1.8 g (82%) of 5bÆHCl: mp 212–213 ꢁC;
aD = +66.9 (c 0.69 g/100 mL in MeOH). Anal. Calcd
for C13H18ClN (223.74): C, 69.79; H, 8.11; N, 6.26.
Found: C, 69.63; H, 8.04; N, 6.13.
4.1. General methods
Proton and 13C NMR spectra were recorded on a JEOL
500-MHz spectrometer. X-ray analysis of 5bÆHClO4 was
performed on an Oxford Diffraction Xcalibur Gemini
diffractometer equipped with a Sapphire 3 CCD detec-
tor.48 Enzyme kinetic studies were performed on the
SpectraMax 384 microplate reader and the data were
evaluated with SoftMax PRO v. 4.7.1 software equipped
with a pre-programmed Michaelis–Menten protocol. All
assays were carried out at 37 ꢁC in 50 mM potassium
phosphate buffer (pH 7.0). Recombinant human
MAO-B (50 nM or 250 nM) was used in the experiments
to estimate the values of Vmax and Km for 5a and 5b.
Plots of OD (420 nm) versus time were linear for the first
30 min at 50 nM MAO-B and for 5 min at 250 nM
MAO-B. The diastereoselectivity studies on the mono-
deutero isotopoepimers were performed on 1 mM sub-
strate and 1–2.4 lM MAO-B. Incubations (0.5 mL
final volume) were conducted in 50 mM potassium
phosphate buffer (pH 7.0) at 37 ꢁC for 2 h at which time
acetonitrile (0.5 mL) was added. After vortexing and
centrifugation, the supernatant fraction was evaporated
under vacuum and the residue in 0.5 mL methanol was
filtered through a 1 lm syringe filter and 25 lL of the fil-
trate in 25 mL methanol was analyzed by LC-ESI/MS in
the positive ion mode on an API Sciex 365 triple quad-
rupole tandem mass spectrometry (Applied Biosystems,
Foster City, CA) under the following conditions: The
auxiliary gas (N2) flow-rate was 2 L/min, the curtain
gas flow-rate was 1.3 L/min at 60 p.s.i. and the turbo
ion spray temperature was 400 ꢁC; the HPLC system
consisted of an Agilent 1100 series HPLC pump,
diode-array detector and a Zorbax SB-C18 column
(2.1 · 150 mm, 5 lm). The mobile phase—metha-
nol:10 mM ammonium formate (80:20, v/v), was deliv-
ered in an isocratic mode at a flow-rate of 0.2 mL/min.
The deuterium contents of the iminiumyl metabolites
were estimated by selected ion monitoring of the appro-
priate ions taking into account the 13C contribution of
the d0 species to the d1 species. The results recorded in
Section 2 are the average of three such determinations
for each of the three substrates analyzed.
4.1.2. (1R,6S)-3-Methyl-6-phenyl-3-aza-bicyclo[4.1.0]hep-
tane HCl (5aÆHCl). The solvent from the filtrate ob-
tained after the first crystallization with the (1R)-(ꢁ)-
10-camphorsulfonic acid was removed under reduced
pressure and a saturated aqueous solution of K2CO3
(100 mL) was added. The aqueous phase was extracted
with ether (3· 30 mL) and the combined organic layers
were washed with water (3· 30 mL), brine (2· 30 mL),
dried over Na2SO4 and filtrated. After removal of the
solvent under reduced pressure the residue (7.7 g,
41 mmol) was dissolved in ether (80 mL) and a solution
of (1S)-(+)-10-camphorsulfonic acid (9.5 g, 41 mmol) in
MeOH (20 mL) was added at room temperature. The
collected crystals were recrystallized five times from
ether/MeOH to give (1R,6S)-3-methyl-6-phenyl-3-azabi-
cyclo[4.1.0]heptane (1S)-(+)-10-camphorsulfonate [5aÆ(S)-
C10H16O4S] as white crystals (4.76 g, 28%): mp 177–
178 ꢁC; aD = ꢁ11.5 (c 0.84 g/100 mL in MeOH). Anal.
Calcd for C23H33NO4S (419.59): C, 65.84; H, 7.93; N,
3.34. Found: C, 65.83; H, 7.93; N, 3.35. An ethereal
solution of the corresponding free base 5a (1.8 g,
10 mmol) was treated with methanolic HCl (1.2 equiv)
and the resulting precipitate was recrystallized from
MeOH/ether to give 5aÆHCl (1.3 g, 78%) as white crys-
tals: mp 212–213 ꢁC; aD = ꢁ70.8 (c 0.62 g/100 mL in
MeOH). Anal. Calcd for C13H18ClN (223.74): C,
69.79; H, 8.11; N, 6.26. Found: C, 69.41; H, 7.94; N,
6.04.
4.1.3. (1S,6R)-3-Methyl-6-phenyl-3-azabicyclo[4.1.0]hep-
tane HClO4 (5bÆHClO4). A methanolic solution of
HClO4 (1 mL, 0.64 mmol) was added to an ethereal
solution (2 mL) of the free base 5b (100 mg, 0.53 mmol).
The precipitate was filtrated and recrystallized from
MeOH/ether to give 5bÆHClO4 (137 mg, 90%) as white
crystals: mp 164–165 ꢁC; 1H NMR (500 MHz, CD3OD)
d 1.12 (m, 1H), 1.22 (m, 1H), 1.60 (m, 1H), 2.40 (m, 2H),
2.86 (s, 3H), 2.95 (m, 1H), 3.14 (m, 1H), 3.30 (m, 1H),
3.98 (m, 1H), 7.20 (t, J = 7.5 Hz, 1H), 7.30 (t,
J = 7.5 Hz, 2H), 7.36 (d, J = 7.5 Hz, 2H); 13C NMR
(125.8 MHz, CD3OD) d 15.0, 17.7, 22.5, 28.3, 41.9,
49.3, 53.9, 126.6, 127.4, 128.4, 145.0. Anal. Calcd for
C13H18ClNO4 (287.74): C, 54.26; H, 6.31; N, 4.87.
Found: C, 54.42; H, 6.23; N, 4.90.
4.1.1. (1S,6R)-3-Methyl-6-phenyl-3-azabicyclo[4.1.0]hep-
tane HCl (5bÆHCl). A solution of (1R)-(ꢁ)-10-camphor-
sulfonic acid (8.82 g, 38 mmol) in MeOH (20 mL) was
added to a solution of the free base 5ab27 (14.23 g,
76 mmol) in ether (80 mL) at room temperature. The
crystals (13.7 g, 86%) were collected and recrystallized
five times from ether/MeOH to give (1S,6R)-(ꢁ)-3-
methyl-6-phenyl-3-azabicyclo[4.1.0]heptane
(1R)-10-
camphorsulfonate [5bÆ(R)-C10H16O4S] as white crystals
(4.46 g, 28%): mp 177–178 ꢁC; aD = +13.8 (c 1.05 g/
100 mL in MeOH); 1H NMR (500 MHz, CD3OD) d
0.84 (s, 3H), 1.11 (s, 3H), 1.24 (m, 1H), 1.39 (m, 1H),
1.62 (m, 2H), 1.87 (d, J = 18 Hz, 1H), 2.02 (m, 2H),
2.33 (m, 1H), 2.42 (m, 2H), 2.64 (m, 1H), 2.77 (d,
4.1.4. (1R,6S)-3-Methyl-6-phenyl-3-azabicyclo[4.1.0]hept-
2-enyl perchlorate ð6aþ-d0ꢂClO4ꢁÞ. The procedure as de-
scribed previously27 for the synthesis of racemate 6ab+

P. Bissel et al. / Bioorg. Med. Chem. 16 (2008) 3557–3564
3563
was followed. Recrystallization of the crude product
4.1.10. (1R,6S)-3-Methyl-6-phenyl-3-aza-bicyclo[4.1.0]hep-
tane 5a-(2S)-d1 from 6aþ-2-d1ClO4ꢁ. NaBH4 (12 mg)
was added at 0 ꢁC to a solution of 6a+-d1 (46 mg,
0.16 mmol) in MeOH (1 mL). After 30 min water was
added and the mixture was extracted with ether; 1H
NMR (500 MHz, CD3OD) d 0.81 (m, 1H), 0.97 (m,
1H), 1.37 (m, 1H) 2.22, (m, 4H), 2.24 (s, 3H), 2.62 (m,
0.25H), 2.77 (m, 0.75H), 7.17 (m, 1H), 7.28 (m, 4H).
The free base was treated with methanolic HCl to give
5a-(2S)-d1HCl (22 mg, 60%) as a white powder: mp
214–215 ꢁC.
from MeOH/ether gave 6aþ-d0ꢂClO4 (386 mg, 30%)
ꢁ
as white crystals: mp 126–127 ꢁC; aD = ꢁ255 (c 1.41 g/
100 mL in MeOH). Anal. Calcd for C13H16ClNO4
(285.72): C, 54.65; H, 5.64; N, 4.90. Found: C, 54.48;
H, 5.63; N, 4.91.
4.1.5. (1S,6R)-3-Methyl-6-phenyl-3-azabicyclo[4.1.0]hept-
2-enyl perchlorate ð6bþ-d0ꢂClO4ꢁÞ. The same procedure
gave 6bþ-d0ꢂClO4 (161 mg, 37%) as white crystals:
ꢁ
mp 127–128 ꢁC; aD = +234 (c 0.58 g/100 mL in MeOH).
Anal. Calcd for C13H16ClNO4 (285.72): C, 54.65; H,
5.64; N, 4.90. Found: C, 54.30; H, 5.54; N, 4.84.
Acknowledgments
4.1.6. (1R,6S)-3-Methyl-6-phenyl-3-azabicyclo[4.1.0]hep-
tane HCl [5a-(2S)-d1ÆHCl] from 6aþ-d0ClO4ꢁ. NaBD4
(15 mg) was added at 0 ꢁC to a solution of 6a+-d0
(50 mg, 0.17 mmol) in CD3OD (1 mL). After 30 min
water was added and the mixture was extracted with
This work was supported by the Harvey W. Peters Re-
search Center. The authors thank Dr. Carla Slebodnick
for the X-ray analysis.
1
ether; H NMR (500 MHz, CD3OD) d 0.82 (m, 1 H),
0.99 (m, 1H), 1.39 (m, 1H), 2.21 (m, 4H), 2.24 (m,
3H), 2.62 (m, 0.75), 2.80 (m, 0.25H), 7.15 (m, 1H),
7.28 (m, 4H). The free base was treated with methanolic
HCl to give 5a-(2S)-d1ÆHCl (21 mg, 52%) as a white
powder: mp 215–216 ꢁC.
References and notes
1. Singer, T. P. J. Neural Trans. (Suppl.) 1987, 21, 31.
2. Waldmeier, P. C. J. Neural Transm. 1987, 23, 55.
3. Rochat, B.; Kosel, M.; Boss, G.; Testa, B.; Gillet, M.;
Baumann, P. Biochem. Pharmacol. 1998, 56, 15.
4. Dixon, C. M.; Park, G. R.; Tarbit, M. H. Biochem.
Pharmacol. 1994, 47, 1253.
5. Lee, Y.; Ling, K.-Q.; Lu, X.; Silverman, R. G.; Shepard,
E. M.; Dooley, D. M.; Sayre, L. M. J. Am. Chem. Soc.
2002, 124, 12135.
4.1.7. (1S,6R)-3-Methyl-6-phenyl-3-azabicyclo[4.1.0]hep-
tane HCl [5b-(2S)-d1ÆHCl] from 6bþ-d0 ꢂ ClO4ꢁ. NaBD4
(15 mg) was added at 0 ꢁC to a solution of 6b+-d0
(50 mg, 0.17 mmol) in CD3OD (1 mL). After 30 min
water was added and the mixture was extracted with
6. Wang, Y.-X.; Mabic, S.; Castagnoli, N., Jr. Bioorg. Med.
Chem. 1998, 6, 143.
1
ether; H NMR (500 MHz, CD3OD) d 0.81 (m, 1H),
7. Kalgutkar, A. S.; Dalvie, D. K.; Castagnoli, N., Jr.;
Taylor, T. J. Chem. Res. Toxicol. 2001, 14, 1139.
8. Singer, T. P. In Chemistry and Biochemistry of Flavoen-
0.99 (m, 1H), 1.38 (m, 1H), 2.21 (m, 4H), 2.24 (m,
3H), 2.62 (m, 0.75), 2.79 (m, 0.25H), 7.15 (m, 1H),
7.28 (m, 4H). The free base was treated with HCl meth-
anolic to give 5b-(2S)-d1ÆHCl (13 mg, 53%) as a white
powder: mp 215–216 ꢁC.
zymes (II); Muller, F., Ed.; CRC Press: London, 1991; pp
437–470.
¨
9. Bissel, P.; Bigley, M. C.; Castagnoli, K.; Castagnoli, N.,
Jr. J. Bioorg. Chem. 2002, 10, 3031.
4.1.8. (1R,6S)-2,2-d2-3-Methyl-6-phenyl-3-azabicyclo[4.1.0]-
heptane HCl (5a-2,2-d2ÆHCl). To a solution of racemic
5ab-2,2-d231 (7.11 g, 41 mmol) in ether (45 mL) was added
10. Youngster, S. K.; McKeown, K. A.; Jin, Y. Z.; Ramsey,
R. R.; Heikkila, R. E.; Singer, T. P. J. Neurochem. 1989,
53, 1837.
a
solution
a
(1S)-(+)-10-camphorsulfonate (4.72 g,
11. Singer, T. P.; Castagnoli, N., Jr.; Trevor, A. Neurol.
Neurobiol. 42A (Progr. Catecholamine Res., Pt. A) 1988,
75.
12. Singer, T. P.; Salach, J. I.; Crabtree, D. Biochem. Biophys.
Res. Commun. 1985, 127, 707.
13. Salach, J. I.; Singer, T. P.; Castagnoli, N., Jr.; Trevor, A.
Biochem. Biophys. Res. Commun. 1984, 125, 831.
14. Weisman, J.; Trevor, A.; Chiba, K.; Peterson, L.;
Castagnoli, N., Jr. J. Med. Chem. 1985, 28, 997.
15. Chiba, K.; Trevor, A.; Castagnoli, N., Jr. Biochem.
Biophys. Res. Commun. 1984, 120, 574.
20 mmol) in MeOH (10 mL). The salt was recrystallized
three times from MeOH/ether to give the (S)-(+)-10-cam-
phorsulfonic acid salt as white crystals (1.53 g, 18%): mp
176–177 ꢁC. This salt was treated with an aqueous solu-
tion of K2CO3 and the free base was extracted with ether.
The free base (40 mg, 0.21 mmol) was treated with a meth-
anolic solution of HCl to give 5a-2,2-d2ÆHCl (45 mg, 95%)
as a white solid: mp 216–217 ꢁC; H NMR (500 MHz,
CD3OD) d 1.14 (m, 1H), 1.25 (m, 1H), 1.60 (m, 1H),
2.43 (m, 2H), 2.83 (s, 3H), 2.91 (m, 1H), 3.35 (m, 1H),
7.22 (m, 1H), 7.31 (m, 2H), 7.38 (m, 2H).
1
16. Youdim, M. B. H. Prog. Neuro-Psychopharmacol. Biol.
Psychiatry 1989, 13, 363.
17. Cohen, G. Ann. N. Y. Acad. Sci. 2000, 899, 112.
18. Peterson, L. A.; Caldera, P. S.; Trevor, A.; Chiba, K.;
Castagnoli, N., Jr. J. Med. Chem. 1985, 28, 1432.
19. Heikkila, R. E.; Hess, A.; Duvoisin, R. C. Life Sci. 1985,
36, 231.
20. Langston, J. W.; Irwin, I.; Langston, E. B.; Forno, L. S.
Neurosci. Lett. 1984, 50, 289.
21. Tidwell, T. T. The Chemistry of the Cyclopropyl Group;
John Wiley & Sons Ltd, 1987, pp 565–622.
22. Battiste, M. A.; Couch, M. W.; Rehberg, R. J. Phys.
Chem. 1977, 81, 64.
4.1.9. (1R,6S)-3-Methyl-6-phenyl-3-azabicyclo[4.1.0]hept-
2-enyl perchlorate ð6aþ-2-d1ꢂClO4ꢁÞ. The same proce-
dure as described previously27 for the synthesis of race-
mate 6ab+ was followed. The solid wꢁas recrystallized
from MeOH/ether to give 6aþ-d1ꢂClO4 (166 mg, 22%)
1
as white crystals: mp 127–128 ꢁC; H NMR (500 MHz,
CD3OD) d 2.02 (m, 1H), 2.30 (m, 1H), 2.47 (m, 3H),
3.64 (s, 3H), 3.78 (m, 2H), 7.27 (m, 1H), 7.34 (m, 2H),
7.44 (m, 2H).

3564
P. Bissel et al. / Bioorg. Med. Chem. 16 (2008) 3557–3564
23. Wiberg, K. B.; Laidign, K. E. J. Org. Chem. 1992, 57,
5092.
´
24. Kimura, S.; Ito, S.; Yoshifuji, M.; Veszpremi, T. J. Org.
Chem. 2003, 68, 6820.
25. Handbook of Chemistry and Physics, 73rd ed.; 1992–1993,
pp 9–2.
31. Mabic, S.; Castagnoli, N., Jr. J. Labelled. Compd.
Radiopharm. 1996, 38, 255.
´
32. Nikisch, G.; Mathe, A. A.; Czernik, A.; Eap, C. B.;
´
Jimenez-Vasquez, P.; Brawand-Amey, M.; Baumann, P.
J. Clin. Psychophamacol. 2004, 24, 283.
33. Otton, S. V.; Gillam, E. M. J.; Lennard, M. S.; Tcker, G.
T.; Woods, H. F. Br. J. Clin. Pharmacol. 1990, 30, 751.
34. Silverman, R. B.; Cesarone, J. M.; Lu, X. J. Am. Chem.
Soc. 1993, 115, 4955.
35. Smith, R. A.; White, R. L.; Krantz, A. J. Med. Chem.
1988, 31, 1558.
36. Robinson, J. B. Biochem. Pharmacol. 1985, 34, 4105.
37. Reynolds, G. P.; Rausch, W. D.; Riederer, P. Brit. J. Clin.
Pharmacol. 1980, 9, 521.
38. Yu, P. H.; Bailey, B. A.; Durden, D. A.; Boulton, A. A.
Biochem. Pharmacol. 1986, 35, 1027.
39. Yu, P. H. Biochem. Cell Biol. 1988, 66, 853.
40. Yu, P. H.; Davis, B. A. Intern. J. Biochem. 1988, 20, 1197.
41. Battersby, A. R.; Buckley, D. G.; Staunton, J.; Williams,
P. J. J. Chem. Soc., Perkin Trans. 1 1979, 2550.
42. Ottoboni, S.; Caldera, P.; Trevor, A.; Castagnoli, N., Jr.
J. Biol. Chem. 1989, 264, 13684.
43. Zhao, Z.; Mabic, S.; Kuttab, S.; Franot, C.; Castagnoli,
K., ; Castagnoli, N., Jr. Bioorg. Med. Chem. 1998, 6,
2531.
44. Edmondson, D. E.; Mattevi, A.; Binda, C.; Lim, M.;
Hubalek, F. Curr. Med. Chem. 2004, 11, 1983.
45. Binda, C.; Li, M.; Hubalek, F.; Restelli, N.; Edmondson,
D. E.; Mattevi, A. Proc. Natl. Acad. Sci. U.S.A. 2003, 100,
9750.
46. Binda, C.; Hubalek, F.; Li, M.; Herzig, Y.; Sterling, J.;
Edmondson, D. E.; Mattevi, A. J. Med. Chem. 2004, 47,
1767.
26. Enantiomeric pairs are identified by the lower case Arabic
letters ‘a’ and ‘b’ where ‘a’ refers to the (1R,6S)-configu-
ration and ‘b’ refers to the (1S,6R)-configuration. The
designation ‘a,b’ refers to the (1R,6S),(1S,6R) racemates.
When the discussion refers to both enantiomers and
racemates the Arabic numeral will stand alone. Designa-
tions for diastereomers will be by Arabic numbers except
when referring to the isotopoepimers RR’C(2)HD that
will be identified as (R) and (S). Protonated species that
may be in equilibrium with the corresponding free base
will be designated with an appended ‘H+’ whereas those
charged species that may not equilibrate to a free base will
be designated with ‘+’ only. MS2 fragment ions will be
identified by lower case Roman numerals and neutral
fragments by upper case Arabic letters.
27. Rimoldi, J. M.; Puppai, S. G.; Isin, E.; Bissel, P.; Khalil, A.;
Castagnoli, N., Jr. Bioorg. Med. Chem. 2005, 13, 5080.
28. The small geminal coupling constant cyclopropyl methy-
lene protons has been observed by others see: Ali, Z.;
Khan, S. I.; Pawar, R. S.; Ferreira, D.; Khan, I. A. J. Nat.
Prod. 2007, 70, 107.
29. Sauers, R. R. Tetrahedron 1998, 54, 337.
30. Catalytic reductions of 6+ were less successful. GC/MS
analyses documented that catalytic hydrogenation of 6ab+
in the presence of Pd/C always led to a mixture of products
resulting from reduction of the cyclopropyl ring and/or the
iminiumyl double bond (Gray, A. P.; Kraus, H.; Heitme-
ier, D. H. J. Org. Chem. 1968, 33, 3007; Irwin, W. J.;
McQuillin, F. J. Tetrahedron Lett. 1968, 2195). Mixtures
resulting from hydrogenation of the iminiumyl double
bond and the phenyl group were obtained in the presence
of Pt. Regioselective hydrogenation of the iminiumyl
double bond was achieved in the presence of Raney Ni but
deuteration of 6ab+ gave a product that proved to be a 1:1
mixture of 5ab-(2R)-d1 and 5ab-(2S)-d1..
47. Pitts, S. M.; Markey, S. P.; Murphy, D. L.; Weiss, A. In
Recommended Practices for the Safe Handling of MPTP;
Markey, S. P., Castagnoli, N., Jr., Trevor, A. J., Kopin, I.
J., Eds.; Academic Press: New York, 1995; p 703.
48. CCDC 673555 contains the supplementary crystallo-
graphic data for this paper. These data can be obtained
free of charge from The Cambridge Crystallographic Data
Centre via www.ccdc.cam.ac.uk/data/request/cif.