Interactions of 1-methyl-3-phenylpyrrolidine and 3-methyl-1-phenyl-3-azabicyclo[3.1.0]hexane with monoamine oxidase B

Article abstract of DOI:10.1016/j.bmc.2010.03.079

The parkinsonian inducing agent 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP) and its corresponding five-membered ring analogue 1-methyl-3-phenyl-3-pyrroline are cyclic tertiary allylamines and good substrates of monoamine oxidase B (MAO-B). The MAO-B catalyzed 2-electron α-carbon oxidation of this class of substrates appears to be dependent on the presence of the allylic π-bond since the corresponding saturated piperidinyl analogue of MPTP is reported not to be an MAO-B substrate. The only saturated cyclic tertiary amine known to act as an MAO-B substrate is the 3,4-cyclopropyl analogue of MPTP, 3-methyl-6-phenyl-3-azabicyclo[4.1.0]heptane. As part of our ongoing studies we have examined the MAO-B substrate properties of the corresponding pyrrolidinyl analogue, 1-methyl-3-phenylpyrrolidine, and the 3,4-cyclopropyl analogue, 3-methyl-1-phenyl-3-azabicyclo[3.1.0]hexane. The results document that both the pyrrolidinyl analogue [K_m = 234 μM; V_max = 8.37 nmol/(min-mg mitochondrial protein)] and the 3,4-cyclopropyl analogue [K_m = 148 μM; V_max = 16.9 nmol/(min-mg mitochondrial protein)] are substrates of baboon liver mitochondrial MAO-B. We also have compared the neurotoxic potential of these compounds in the C57BL/6 mouse. The results led us to conclude that these compounds are not MPTP-type neurotoxins.

Full text of DOI:10.1016/j.bmc.2010.03.079

Bioorganic & Medicinal Chemistry 18 (2010) 4111–4118
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
Interactions of 1-methyl-3-phenylpyrrolidine and
3-methyl-1-phenyl-3-azabicyclo[3.1.0]hexane with monoamine oxidase B
a
a
a
a
b
Anél Pretorius , Modupe O. Ogunrombi , Hendrik Fourie , Gisella Terre’Blanche , Neal Castagnoli Jr. ,
a
a,
*
Jacobus J. Bergh , Jacobus P. Petzer
Pharmaceutical Chemistry, School of Pharmacy, North-West University, Private Bag X6001, Potchefstroom 2520, South Africa
Department of Chemistry, Virginia Tech and Edward Via College of Osteopathic Medicine, Blacksburg, VA 24061, USA
a
b
a r t i c l e i n f o
a b s t r a c t
Article history:
The parkinsonian inducing agent 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP) and its
corresponding five-membered ring analogue 1-methyl-3-phenyl-3-pyrroline are cyclic tertiary allyl-
Received 17 February 2010
Revised 29 March 2010
Accepted 31 March 2010
Available online 3 April 2010
amines and good substrates of monoamine oxidase B (MAO-B). The MAO-B catalyzed 2-electron
a-carbon
oxidation of this class of substrates appears to be dependent on the presence of the allylic -bond since
p
the corresponding saturated piperidinyl analogue of MPTP is reported not to be an MAO-B substrate. The
only saturated cyclic tertiary amine known to act as an MAO-B substrate is the 3,4-cyclopropyl analogue
of MPTP, 3-methyl-6-phenyl-3-azabicyclo[4.1.0]heptane. As part of our ongoing studies we have
examined the MAO-B substrate properties of the corresponding pyrrolidinyl analogue, 1-methyl-3-
phenylpyrrolidine, and the 3,4-cyclopropyl analogue, 3-methyl-1-phenyl-3-azabicyclo[3.1.0]hexane.
Keywords:
Monoamine oxidase B
Saturated substrate
1
3
-Methyl-3-phenylpyrrolidine
-Methyl-1-phenyl-3-
The results document that both the pyrrolidinyl analogue [K
mitochondrial protein)] and the 3,4-cyclopropyl analogue [K
m
= 234 M; Vmax = 8.37 nmol/(min-mg
= 148 lM; Vmax = 16.9 nmol/(min-mg
l
azabicyclo[3.1.0]hexane
-Methyl-3-phenyl-3-pyrroline
m
1
mitochondrial protein)] are substrates of baboon liver mitochondrial MAO-B. We also have compared
the neurotoxic potential of these compounds in the C57BL/6 mouse. The results led us to conclude that
these compounds are not MPTP-type neurotoxins.
Ó 2010 Elsevier Ltd. All rights reserved.
1
. Introduction
C H
C H
6
5
6
5
The flavoenzyme monoamine oxidase B (MAO-B) catalyzes the
2
-electron a-carbon oxidation of a variety of endogenous and exog-
-H+
enous arylalkylamines such as dopamine, epinephrine and benzyl-
amine. Also among its substrates is the parkinsonian inducing
neurotoxin 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine [MPTP
-
+
-
2e , H
1
MAO-B
N
N
2
(1)] (Scheme 1). MPTP is a member of the cyclic tertiary allylami-
CH3
CH3
2H+
nyl class of MAO-B substrates and is oxidized at C-6 of the tetrahyd-
ropyridinyl ring to yield the conjugated eniminiumyl metabolite,
the 1-methyl-4-phenyl-2,3-dihydropyridiniumyl species [MPDP
1
+
+
3
(2H )]. Presumably the corresponding conjugate base 2 undergoes
C H
C H
6 5
6
5
a
second 2-electron oxidation to generate the 1-methyl-4-
+
+
phenylpyridiniumyl metabolite MPP (3 ) that is thought to act as
-
+
-
2e , H
the ultimate neurotoxin.2
,4,5
Inhibitors of MAO-B that block the
+
conversion of MPTP to MPP protect experimental animals against
the neurotoxic action of MPTP.6 The MAO-B substrate properties
of this cyclic tertiary aminyl class of substrates are thought to be
N
N
CH3
2
CH3
3
+
dependent on the presence of the allylic
p-bond since the corre-
sponding saturated piperidinyl analogue 4 of MPTP is reported
Scheme 1. The MAO-B-catalyzed oxidation of MPTP (1) at C-6 of the tetrahydro-
pyridinyl ring to yield the corresponding conjugated eniminiumyl metabolite
+
+
+
*
Corresponding author. Tel.: +27 18 2992206; fax: +27 18 2994243.
E-mail address: jacques.petzer@nwu.ac.za (J.P. Petzer).
MPDP (2H ). MPDP , via the conjugate base 2, is further oxidized to yield the
+
+
pyridiniumyl product MPP (3 ).
0
968-0896/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2010.03.079

4
112
A. Pretorius et al. / Bioorg. Med. Chem. 18 (2010) 4111–4118
C H
H
C H
C H
C H
C H
C H
6 5
6
5
6
5
6
5
6
5
6
5
i
ii
-
+
-
2e, H
N
N
N
N
N
N
MAO-B
CH3
CH3
CH3
CH3
CH3
CH3
6+
10
7
11
4
5
Scheme 5. Synthetic pathway to compounds racemic 10 and racemic 11. Key: (i) H
(atm), PtO , rt; (ii) (C Zn, CH
2
Scheme 2. The structures of 1-methyl-4-phenylpiperidine (4) and 3-methyl-6-
phenyl-3-azabicyclo[4.1.0]heptane (5). The MAO-B catalyzed oxidation of 5 yields
the iminiumyl metabolite 6 .
2
2
H
)
5 2
2 2
I .
+
We also have examined the possibility that the saturated ana-
logues 10 and 11 may act as neurotoxins in the C57BL/6 mouse.14
The neurotoxic action of MPTP is dependent upon its metabolic
conversion by MAO-B to yield the charged pyridinium species
C H
C H
6 5
6
5
+
+ 2
MPP (3 ). After accumulating in the inner mitochondrial mem-
-
-
2e , H+
+
brane of the nigrostriatal nerve terminals, MPP is thought to act
N
N
by inhibiting complex I of the mitochondrial respiratory chain.6
The mitochondrial accumulation is driven by the transmembrane
electrochemical gradient which appears to rely upon the presence
,15
MAO-B
CH3
CH3
8H+
7
+
16,17
of the permanent charge of MPP .
Since the MAO-B catalyzed
C H
6
5
2-electron oxidations of substrates 10 and 11 may yield charged
intermediates and end products, the possibility exists that these
substrates could display MPTP-type neurotoxicity. In this study
we have tested this hypothesis by measuring ex vivo striatal dopa-
mine concentrations in C57BL/6 mice 7 days after administration
-
H+
N
1
3
CH3
of 10 and 11. In addition, the possibility that 10 and 11 may be
converted to charged intermediates by MAO-B was further investi-
gated using mass spectrometry.
9
Scheme 3. The MAO-B-catalyzed oxidation of 1-methyl-3-phenyl-3-pyrroline (7)
+
to yield the iminiumyl 8H that, following deprotonation, gives the stable product
1
-methyl-3-phenylpyrrole (9).
2. Results
2
.1. Chemistry
C H
C H
6 5
6
5
Both 10 and 11 were synthesized from 1-methyl-3-phenyl-3-
1
3
pyrroline (7) as shown in Scheme 5. Atmospheric pressure
hydrogenation of 7 in the presence of PtO gave 10 in high yield
71%). In a reaction similar to that previously described for the
2
N
N
18
(
9
CH3
CH3
cyclopropylation of MPTP, treatment of 7 with an excess of
CH and diethylzinc yielded 11 (24%). Both substrates were con-
verted to their respective oxalate salts. Compounds 10 and 11
2 2
I
1
0
11
1
9
20
Scheme 4. The structures of (±)-1-methyl-3-phenylpyrrolidine (10) and (±)-3-
methyl-1-phenyl-3-azabicyclo[3.1.0]hexane (11).
have been prepared previously by alternative synthetic pathways.
N-Methylation of 4-phenylpiperidine with formic acid/formalde-
hyde (Eschweiler–Clarke reaction) gave compound 4.
1
3,21
not to be an MAO-B substrate (Scheme 2).7,8 The only known satu-
rated cyclic tertiary aminyl MAO-B substrate is the 3,4-cyclopropyl
analogue of MPTP, racemic 3-methyl-6-phenyl-3-azabicyclo[4.1.0]-
2.2. General enzymology
9
heptane (5), that is converted to the corresponding iminiumyl
In the present study we have examined the MAO-B substrate
properties of the saturated cyclic tertiary amines 10 and 11. As en-
zyme source we have employed the mitochondrial fraction ob-
tained from baboon liver tissue since it exhibits a high degree of
+
metabolite 6 . The
p
-bond characteristics of the cyclopropyl group
9
may be responsible for the good substrate properties of 5.
Another member of the cyclic tertiary allylaminyl class of MAO-B
2
2
substrates is the five-membered ring MPTP analogue 1-methyl-3-
MAO-B catalytic activity. Also, based on the similarity of the stea-
dy-state kinetic parameters (K and Vmax) between baboon and
phenyl-3-pyrroline (7) (Scheme 3).1
0–12
MAO-B catalyzes the ring
m
a
-carbon oxidation of 7 to yield 1-methyl-3-phenylpyrrole (9) as
human MAO-B, the interactions of substrates with MAO-B ob-
tained from baboon liver tissue appears to be similar to those of
the human form of the enzyme. The rates of oxidation of 10
the final product. This 2-electron oxidation process most likely pro-
ceeds via 8H , the short lived conjugate acid of the pyrrolyl product
+
22
1
3
9.
As part of our ongoing investigations of five-membered azacy-
and 11 by baboon liver mitochondrial MAO-B were measured by
clic systems, we have examined the interactions of the correspond-
ing saturated pyrrolidinyl [(±)-1-methyl-3-phenylpyrrolidine (10)]
and 3,4-cyclopropyl [(±)-3-methyl-1-phenyl-3-azabicyclo[3.1.0]-
hexane (11)] analogues of 1-methyl-3-phenyl-3-pyrroline with
monitoring the amount of H
enzyme. In the catalytic cycle of MAO-B, one mole of O
to H for each mole of substrate oxidized. The concentration of
in the enzyme reactions can be conveniently measured by
the peroxidase-coupled spectrophotometric assay system de-
scribed previously. For the purpose of this study we have chosen
a discontinuous assay protocol with an incubation time of 20 min
(for substrate 10) or 15 min (for substrate 11). As illustrated in
2
O
2
generated in the presence of the
2
is reduced
2 2
O
H O
2 2
MAO-B (Scheme 4). In view of the
p-bond characteristics of the
2
3
cyclopropyl group, we expected that 11 would be a good MAO-B
substrate. On the other hand, the pyrrolidinyl analogue 10, being a
homolog of 4, should be devoid of MAO-B substrate properties.

A. Pretorius et al. / Bioorg. Med. Chem. 18 (2010) 4111–4118
4113
3
2
2
1
1
0
5
0
5
0
5
0
5
1
1
6
2
8
4
0
11
1
0
-
0
500
1000
S], µM
1500
2000
0
4
8
12
16
20
24
[
Incubation time (min)
Figure 1. The linear plots of the baboon liver mitochondrial MAO-B (0.15 mg
protein/mL) catalyzed oxidations of substrates 10 (500 M; open circles) and 11
50
13
l
2
3
(
2 2
250 lM; filled circles) as measured by H O production.
4
3
2
1
0
0
0
0
0
1
2
4
Figure 1, the H
1 in the presence of baboon liver mitochondrial MAO-B remained
linear for at least 20 min at substrate concentrations of 500 and
50 M, respectively. Control incubations carried out in the ab-
sence of the test substrates documented that the background
2 2
O generated as a result of the oxidation of 10 and
1
7
1
2
l
2
H O
2
generation by the mitochondrial fractions was less than
M. This is approximately 6–11% of the H measured in
M) and in 15 min
M) at Vmax concentrations.
generated (89–94%) by the mito-
1
4
2
2
.3
l
2 2
O
0 min incubations of substrate 10 (21–24
l
incubations of substrate 11 (33–39
This result verifies that the H
l
2
O
2
0
500
1000
1500
2000
chondrial fractions are dependent upon the presence of the test
substrates.
[S], µM
Figure 2. Plots of initial rate versus substrate concentration for the steady-state
baboon liver mitochondrial MAO-B (0.15 mg protein/mL) catalyzed oxidation of test
substrates 10 (open circles, 20 min incubation period) and 11 (filled circles, 15 min
incubation period) at 37 °C (Panel A). The corresponding plots of initial rate vs.
substrate concentration for kynuramine (13, open circles), benzylamine (12, filled
circles), pyrroline 7 (filled triangles), MMTP (14, open triangles), MPTP (1, filled
squares) and 4 (open squares) (Panel B). Incubation periods varied from 6 to 15 min
2 2
To estimate the extent of the H O produced by non-MAO-B
mediated oxidation of substrates 10 and 11, the baboon liver mito-
chondrial MAO-B were pre-inactivated with the MAO-B selective
inactivator (R)-deprenyl.22 Following incubation of substrate 11
(
250
low concentrations of H
concentration (1.79 ± 0.11
lM) with the (R)-deprenyl pre-inactivated mitochondria, only
2
O
2
were observed. For example, the H
2 2
O
(
see Section 4) and the initial rates are expressed as nmoles H
2 2
O formed/min-mg
lM) with the inactivated mitochondrial
protein.
preparation was only 6.6% of that formed with the fully active
preparation (27.3 ± 0.53 M). Similarly, when substrate 10
500 M) was incubated with the (R)-deprenyl pre-inactivated
mitochondria only 1.92 ± 0.06 M of H was observed which is
.3% of the amount formed (20.6 ± 0.08 M) with the fully active
preparation. These results confirm that the oxidations of 10 and
1 in the presence of the baboon liver mitochondrial fraction used
l
(
l
baboon liver mitochondria are known to be devoid of MAO-A activ-
ity. Using the same experimental protocol, analogous results
were obtained with the known MAO-B substrates benzylamine
l
2 2
O
l
2
2
9
(
12), kynuramine (13), MPTP (1) and 1-methyl-4-(1-methylpyr-
rol-2-yl)-1,2,3,6-tetrahydropyridine [MMTP (14)] (Scheme 6). For
example, following incubations of 12 (500 M), 13 (500 M), MPTP
500 M) and MMTP (250 M) with the (R)-deprenyl pre-inacti-
vated mitochondria, the concentrations of H measured were
–13% of those measured with fully active preparations.
1
here as enzyme source are MAO-B dependent. The oxidation activ-
ity not suppressed by (R)-deprenyl has been observed previously
where tetrahydropyridinyl analogues served as the MAO-B sub-
l
l
(
l
l
2 2
O
2
2
strates. At least for baboon liver mitochondria the residual activ-
ity is not related to the action of the MAO-A isoform since the
5
2
.3. Steady-state MAO-B substrate properties of 10 and 11
NH2
The values of the steady-state kinetic parameters (K and Vmax
for the baboon liver mitochondrial MAO-B catalyzed oxidation of
m
)
N
O
the test substrates 10 and 11 were determined by measuring the
2 2
initial rates of H O production at eight substrate concentrations
2
3
spanning at least two orders of a magnitude. Plots of initial rates
versus substrate concentration for 10 and 11 (Fig. 2A) and for 1, 7,
NH2
N
1
m
2, 13 and 14 (Fig. 2B) gave the K and Vmax values reported in Ta-
NH2
ble 1. While the steady-state baboon liver mitochondrial MAO-B
CH3
substrate properties of MPTP (1) and 7 have been previously re-
1
2
13
14
13
ported, the kinetic parameters for these substrates were reevalu-
ated in the present study since different mitochondrial
preparations may exhibit differing activities. The kinetic values
Scheme 6. The structures of benzylamine (12), kynuramine (13) and 1-methyl-4-
1-methylpyrrol-2-yl)-1,2,3,6-tetrahydropyridine [MMTP (14)].
(

4
114
A. Pretorius et al. / Bioorg. Med. Chem. 18 (2010) 4111–4118
Table 1
C H
C H
C H
6
5
6
5
6
5
Values of the steady-state kinetic parameters for the MAO-B catalyzed oxidation of
selected aminyl substrates
2e , H+
-
-
a
b
or
V
max
K
m
(
l
M)
Vmax/K
m
N
MAO-B
N
N
1
1
0
1
8.37 ± 0.25
234 ± 23.5
148 ± 7.36
0.04
0.11
0.14
0.28
0.51
0.17
0.85
16.9 ± 0.60
CH3
CH3
CH3
MPTP (1)
7
MMTP (14)
Benzylamine
Kynuramine
4
11.7 ± 0.15; 6.7c
27.9 ± 0.13; 9.2c
22.2 ± 0.37
81.8 ± 7.01; 173c
100 ± 8.40; 54.6c
43.4 ± 3.20
15a+
15b+
1
0
42.4 ± 0.41
50.0 ± 0.41
not a substrate
249 ± 0.30
58.9 ± 4.89
C H
C H
C H
6
5
6
5
6
5
-
2e, H+
-
or
a
b
c
Values are expressed in nmol/(min-mg mitochondrial protein).
Values are expressed in (min-mg mitochondrial protein)
Values obtained from Ref.13.
À1
N
MAO-B
N
N
.
CH3
CH3
CH3
16a+
16b+
1
1
Scheme 7. The proposed MAO-B-catalyzed oxidation of 10 and 11 to yield the
iminiumyl metabolites 15a /15b and 16a /16b , respectively.
that were recorded for substrate 11 (K
nmol/min-mg protein) documents that 11 acts as a good substrate
m
= 148
l
M; Vmax = 16.9
+
+
+
+
of baboon liver MAO-B. Based on the Vmax/K
m
values, it is apparent
that the cyclopropyl analogue 11 [Vmax/K
m
= 0.11 (min-mg pro-
2.4. Mass spectrometry
À1
À1
tein) ] and MPTP (1) [Vmax/K
m
= 0.14 (min-mg protein) ] have
comparable substrate properties and that both 1 and 11 are poorer
To investigate the structures of the metabolites that form upon
substrates than the pyrrolinyl analogue 7 [Vmax/K
protein) ].
m
= 0.28 (min-mg
the MAO-B catalyzed ring
a-carbon oxidation of 10 and 11,
À1
reactions containing 100 M of the substrates, respectively, and
l
Although poorer than the corresponding pyrrolinyl (7) and
fused cyclopropyl (11) analogues, the saturated pyrrolidinyl ana-
baboon liver mitochondrial MAO-B were incubated for 14 h at
37 °C. LC-ESI-MS analysis of the incubation mixtures showed two
major peaks in the total ion tracings of both incubations. For
substrate 10, the parent compound eluted at 6.77 min with the
logue 10 also proved to be an MAO-B substrate [Vmax/K
m
= 0.04
À1
(
min-mg protein) ]. This was an unexpected outcome in view of
+
the report that the piperidinyl homolog 4, the saturated analogue
expected MH ion at m/z 162. The metabolite derived from 10
7
+
of MPTP (1), is stable in the presence of mouse MAO-B. Compound
eluted at 6.48 min and displayed M at m/z 160 (Fig. 3A). This mass
4
also proved to be stable in the presence of baboon liver mito-
of this metabolite is consistent with the isomeric 2-electron oxida-
+
+
chondrial MAO-B (Fig. 2B and Table 1). These results should prove
useful for the further characterization of cyclic tertiary aminyl
structural features required for MAO-B substrates.
tion products, 15a and 15b (Scheme 7), that may form as a result
of the action of MAO-B on 10. Similarly, for the incubations con-
taining 11, the parent substrate eluted at 6.76 min and displayed
6.77
2
2
1
1
8
4
.4e7
.0e7
.6e7
.2e7
.0e6
.0e6
6.48
0.0
1
2
3
4
5
6
6
7
8
9
10
11
12
13
14
15
16
17
Time, min
.31
3
2
2
2
1
1
8
4
.2e7
.8e7
.4e7
.0e7
.6e7
.2e7
.0e6
.0e6
6.76
0.0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
Time, min
Figure 3. LC-ESI-MS total ion chromatograms of incubations containing 100
protein/mL). The incubations were carried out for 14 h at 37 °C.
lM of 10 (Panel A) and 11 (Panel B), respectively, and baboon liver mitochondrial MAO-B (0.6 mg

A. Pretorius et al. / Bioorg. Med. Chem. 18 (2010) 4111–4118
4115
the expected MH+ ion at m/z 174 while the metabolite eluted at
dopamine concentrations in the dissected striata were determined
by HPLC-ECD analysis as described in Section 4.
+
26
6
.31 min and had M at m/z 172 (Fig. 3B). Again the mass of the
metabolite is consistent with the isomeric iminiumyl species,
As expected MPTP treatment significantly (p <0.01) depleted
the dopamine levels to 36% (Fig. 4A) and 27% (Fig. 4B) of the con-
trol values. In contrast, treatment with neither 10 nor 11 depleted
striatal dopamine levels with the measured dopamine concentra-
tions at 99% (Fig. 4A) and 111% (Fig. 4B) of the control values,
respectively.
+
+
1
6a and 16b (Scheme 7), that are expected to form upon the ac-
tion of MAO-B on 11. The detection of only one metabolite peak for
each of 10 and 11, however, suggests that ring -carbon oxidation
a
of these substrates is highly regioselective. These metabolite peaks
were absent from the total ion chromatograms of incubations of 10
and 11 in the absence of enzyme (data not shown).
2
.5. Neurotoxicity studies
3. Discussion
It has been documented that the administration of MPTP to
The results document that the fused 3,4-cyclopropylpyrrolidi-
nyl analogue 11 has substrate properties comparable to those of
the known MAO-B pyrrolinyl substrate 7 when examined with ba-
boon liver mitochondrial MAO-B. This result is analogous to that
observed with the corresponding six-membered series, that is,
1
4
C57BL/6 mice causes degeneration of nigrostriatal neurons. The
extent of neuronal damage can be estimated by measuring the de-
gree of dopamine depletion in the striatum 7–10 days following
treatment. In a frequently used protocol, aged mice are treated with
2
4,25
9
single intraperitoneal injections of MPTP (95–238
l
mol/kg).
In
MPTP (1) and its fused 3,4-cyclopropylpiperidinyl analogue 5. As
this study we investigated whether similar treatment of C57BL/6
mice with 10 or 11 also results in nigrostriatal injury as indicated
by a loss of dopamine in the striatum. For each of the test com-
pounds a group of control mice (n = 10) was treated with saline
mentioned in the introduction, the p-bond characteristics of the
cyclopropyl group may be responsible for the observation that
the saturated substrates 5 and 11 are approximately as good
MAO-B substrates as their corresponding allylaminyl analogues
9
(
0.20 mL/30 g mouse). Again for each test compound, a positive
control group (n = 10 mice) was treated with MPTP hydrochloride
167 mol/kg). A third group of mice (n = 10 for each test com-
pound) was treated with the oxalate salts of 10 or 11 (238 mol/
kg). The animals were sacrificed 7 days after treatment and the
MPTP (1) and 7, respectively. The saturated pyrrolidinyl analogue
10 also was shown to be a relatively good substrate for baboon li-
ver mitochondrial MAO-B, with a turnover comparable to that of
MPTP. This result is in contrast to the behavior of the saturated
piperidinyl analogue 4 of MPTP that is not an MAO-B substrate,7
behavior that was confirmed in the present study. This result,
(
l
l
,8
therefore, suggests that the presence of an allylic
p-bond or 3,4-
cyclopropyl-fused system is not a prerequisite for cyclic tertiary
amines to display MAO-B substrate properties. This finding should
provide additional opportunities to map the nature of the interac-
tions of MAO-B with tertiary aminyl substrates and should alert
investigators to the possibility of MAO-B substrate potential of
compounds under development as future therapeutic agents.
8
6
4
2
0
0
0
0
0
Also noteworthy is the observation that the a-carbon oxidation
of MPTP and its 3,4-cyclopropyl analogue 5 occurs regioselectively
9
at C-6. This suggests that the electronic properties of the allylic
*
p-bond and cyclopropyl group may contribute to the stabilization
of the corresponding -carbon radicals that may be intermediary
a
9
in the catalytic mechanism of MAO-B. This analysis could explain
why the piperidinyl derivative 4 is stable in the presence of MAO-
B. The finding of the present study that the pyrrolidinyl analogue
1
0 acts as a relatively good MAO-B substrate, however, does not
Saline
MPTP
10
support this hypothesis. In addition to electronic properties, the
three dimensional structures and binding orientations within the
active site are likely to contribute to the substrate properties and
the regioselectivity of oxidation of this class of compounds. The
factors that determine the substrate properties of cyclic tertiary
amines remain to be identified. In the present study, mass spectral
analysis suggests that the MAO-B catalyzed two-electron oxida-
tions of 10 and 11 also occur regioselectively with the formation
of only a single product for each substrate. Evidence suggests that
8
6
4
2
0
0
0
0
0
+
the products are the corresponding iminiumyl metabolites 15a /
+
+
+
1
5b and 16a /16b for the metabolism of 10 and 11, respectively.
The neurotoxicity of MPTP is thought to be the consequence of
*
its MAO-B-catalyzed metabolism in the brain that ultimately yields
+
2,6
the pyridiniumyl mitochondrial toxin MPP . Evidence suggests
that the mitochondrial toxicity and subsequent neurotoxic action
+
16,17
of MPP rely on it being permanently charged.
Consistent with
Saline
MPTP
11
this view, 1-methyl-3-phenyl-3-pyrroline (7), which is metabo-
1
3
lized to a neutral pyrrolyl species, is not neurotoxic. In this inves-
tigation we also have evaluated 10 and 11 as potential MPTP-type
nigrostriatal neurotoxins. The results document that, unlike MPTP,
these cyclic tertiary amines are not neurotoxic. While the reasons
for this lack of neurotoxicity are not known, an explanation may be
Figure 4. Striatal dopamine levels of C57BL/6 male mice (n = 10 mice/group)
measured seven days after treatment (ip) with saline, MPTPÁHCl (167 mol/kg),
mol/kg) (Panel B). Dopa-
l
1
0Áoxalate (238 lmol/kg) (Panel A) or 11Áoxalate (238 l
mine levels are expressed as pmol/mg tissue. *Significantly different (p <0.01) from
the saline treated group.

4
116
A. Pretorius et al. / Bioorg. Med. Chem. 18 (2010) 4111–4118
that the pharmacokinetic properties and in vivo molecular interac-
Column chromatography was carried out with Fluka aluminum
oxide (Brockmann Activity I). To determine the purity of the oxalic
acid salts of 10, 11 and 4, HPLC analyses were carried out with an
Agilent 1100 HPLC system equipped with a quaternary pump and
an Agilent 1100 series diode array detector (see Supplementary
data). HPLC grade acetonitrile (Merck) and Milli-Q water (Milli-
pore) was used for the chromatography.
tions of 10 and 11 and their respective MAO-B-generated products
+
are different from those of MPTP and MPP . Among the pharmaco-
kinetic factors important for the neurotoxic action of MPTP are a
high degree of blood–brain barrier permeability of MPTP and slow
+
27
clearance of MPP from the brain. Among the molecular interac-
+
tions that lead to neuronal death are the accumulation of MPP in
the nigrostriatal nerve terminals via the plasma membrane dopa-
2
8,29
mine transporter (DAT),
the energy-dependent accumulation
4.2. Synthesis of the oxalic acid salt of 1-methyl-3-
phenylpyrrolidine (10)
+
16
of MPP within the inner mitochondrial membrane and inhibi-
15
tion of complex I of the mitochondrial respiratory chain. The pos-
sibility also exists that the proposed iminiumyl metabolites that
arise as a consequence of the MAO-B catalyzed oxidation of 10
and 11 may be converted to neutral species in vivo. Such neutral
species are not expected to be mitochondrial toxins. For example,
deprotonation of 15a/b would yield the corresponding 2-pyrroli-
nyl analogue which may undergo a second
and deprotonation to generate the pyrrolyl species. Iminiumyl
6a/b in turn may rearrange to form the neutral six-membered
1-Methyl-3-phenyl-3-pyrroline (7) (12.04 mmol, 1.92 g) and
2
PtO (1.32 mmol, 0.3 g) in 60 mL methanol was stirred at room
1
8
temperature in an atmosphere of hydrogen. After 14 h of stirring
the catalyst was removed via filtration and the solvent was evapo-
rated to yield the free base 10 as a light yellow oil (71%). Com-
pound 10 was converted into its oxalic acid salt in diethyl ether
and recrystallized from boiling methanol to give white powdery
+
a-carbon oxidation
+
1
1
crystals: yield 36%; mp 148–149 °C; H NMR (Varian Gemini 300,
dihydropyridine species. These proposed transformations require
further investigation.
DMSO-d
3.41 (m, 2H), 3.63 (m, 2H), 7.23–7.34 (m, 5H); C NMR (Varian
Gemini 300, DMSO-d ) d 32.0, 40.6, 42.2, 54.8, 60.0, 127.0, 127.3,
6
) d 2.05 (m, 1H), 2.37 (m, 1H), 2.85 (s, 3H), 3.23 (m, 1H),
1
3
In conclusion, the present study shows that the structural
requirements for cyclic tertiary amines to act as MAO-B substrates
are less stringent than previously thought. Additionally, the data
presented in this paper demonstrate that the MAO-B-catalyzed
oxidations of cyclic tertiary amines leading to charged metabolites
do not necessarily lead to neurotoxic outcomes. The complex se-
quence of events required for the parkinsonian inducing properties
observed with MPTP does extend to the five-membered azacyclic
systems reported here and in earlier publications. The findings re-
ported here may alert investigators to the possibility that drugs
under development which incorporate pyrrolidinyl and azabicy-
clo[3.1.0]hexane moieties are potential substrates for the MAO en-
6
128.63, 140.2, 165.0; EI-HRMS m/z: calcd 161.1205, found
+
161.1201 (M ).
4.3. Synthesis of the oxalic acid salt of 3-methyl-1-phenyl-3-
azabicyclo[3.1.0]hexane (11)
Compound 11 was synthesized from 1-methyl-3-phenyl-3-pyr-
1
3
roline (7) according to the procedure previously reported for the
9
cyclopropylation of MPTP. The crude product (24%) was purified
by aluminum oxide chromatography (petroleum ether/ethyl ace-
tate 50:50) and converted into its oxalic acid salt in diethyl ether.
Compound 11 has been previously reported as the hydrochloric
zymes. It is however unlikely that the
a-carbon oxidations of such
2
0
1
drugs would lead to MPTP-type neurotoxic events.
acid salt: yield 8.4%; mp 141–143 °C; H NMR (Varian Gemini
3
00, CD
3
OD) d 1.22 (m, 1H), 1.37 (m, 1H), 2.16 (m, 1H), 2.96 (s,
3
H), 3.59 (m, 2H), 3.82 (m, 1H), 4.01 (m, 1H), 7.23–7.37 (m, 5H);
4
. Experimental
13
3
C NMR (Varian Gemini 300, CD OD) d 24.4, 32.4, 41.1, 49.5,
5
8.6, 61.5, 128.1, 128.3, 129.8, 139.8, 166.5; FAB-MS m/z: 174
Caution. MPTP is a nigrostriatal neurotoxin and should be han-
+
+
(
MH ); EI-HRMS calcd 173.1205, found 173.1200 (M ).
dled using disposable gloves and protective eyewear. Procedures
3
0
for the safe handling of MPTP have been described previously.
4
.4. Synthesis of the oxalic acid salt of 1-methyl-4-
phenylpiperidine (4)
4
.1. Chemicals and instrumentation
4
-Phenylpiperidine (1.55 mmol; 0.25 g) was cooled to 0 °C and
98% formic acid (9.30 mmol; 358 L) followed by 36.5% aqueous
formaldehyde (9.30 mmol; 702 L) were carefully added. The reac-
All starting materials, unless otherwise stated, were obtained
from Sigma–Aldrich and were used without purification. MPTPÁHCl
1) and kynuramineÁ2HBr (13) were purchased from Sigma–Al-
drich while the oxalate salts of 1-methyl-3-phenyl-3-pyrroline
l
l
(
tion mixture was stirred at 80 °C for 3 h and water (14 mL) fol-
lowed by an aqueous solution (14 mL) of sodium carbonate
(33 mmol) was added to the reaction. The reaction was extracted
to diethylether (3 Â 20 mL) and dried over anhydrous magnesium
sulfate (4 g). Upon removal of the solvent a light yellow oil
remained which was converted into the oxalic acid salt in diethyl
ether. The oxalate salt was recrystallized from boiling methanol
to give colorless crystals: yield 62%; mp 159–160 °C; H NMR (Bru-
ker Avance III 600, DMSO-d ) d 1.93 (m, 4H), 2.74 (m, 4H), 3.00 (m,
2H), 3.43 (d, 2H, J = 10.8 Hz), 7.22 (m, 3H), 7.30 (m, 2H); C NMR
(Bruker Avance III 600, DMSO-d ) d 29.7, 38.5, 42.5, 53.5, 126.6,
1
3
31
(7) and MMTP (14) were prepared as described previously.
Benzylamine (12) (Merck) was converted to the HCl salt in ethanol.
Petroleum ether used in this study had a distillation range of 40–
1
13
6
0 °C. Proton ( H) and carbon ( C) NMR spectra were recorded
on a Varian Gemini 300 spectrometer at frequencies of 300 MHz
and 75 MHz, respectively, and on a Bruker Avance III 600 spec-
trometer at frequencies of 600 and 150 MHz, respectively. Chemi-
cal shifts are reported in parts per million (d) downfield from the
signal of tetramethylsilane added to the deuterated methanol
1
6
13
6
(
CD
d (doublet) or m (multiplet). Fast atom bombardment mass spectra
FAB-MS) were recorded with a VG 7070E mass spectrometer
while direct insertion electron impact ionization mass spectra
EI-HRMS) were obtained with an AutoSpec ETOF (Micromass).
3
OD) or DMSO-d
6
. Spin multiplicities are given as s (singlet),
126.6, 128.6, 144.3, 165.2; EI-HRMS m/z: calcd 175.1361, found
175.1360 (M ).
+
(
4.5. Steady-state MAO-B activity measurements
(
Melting points (mp) were determined on a Stuart SMP10 melting
point apparatus and are uncorrected. UV–vis spectra were re-
corded on a Shimadzu UV-2100 double-beam spectrophotometer.
Baboon liver mitochondria were isolated according to the previ-
3
2
ously reported procedure and stored at À70 °C. To the mitochon-
drial isolates were added one volume sodium phosphate buffer

A. Pretorius et al. / Bioorg. Med. Chem. 18 (2010) 4111–4118
4117
(
100 mM, pH 7.4) containing 50% glycerol (w/v) and the protein
triplicate and the concentrations of the pyrrolyl products were ex-
pressed as means ± SEM. The extent of the H produced by non-
MAO-B mediated oxidation of benzylamine (500 M), kynuramine
(500 M), MPTP (500 M), MMTP (250 M) and 7 (500 M) were
determined following the same protocol.
3
3
concentrations were determined by the method of Bradford.
2 2
O
The K
oxalate salts of the test substrates 10 and 11 were determined by
measuring the initial rates of H generation at eight substrate
concentrations spanning at least two orders of magnitude (12.5–
0003lM). The reactions contained 85 L of the chromogenic solu-
tion and were carried out in a final volume of 500 L (in 100 mM
m
and Vmax values for the MAO-B catalyzed oxidation of the
l
l
l
l
l
2 2
O
2
l
4
.8. Mass spectrometry
2
l
sodium phosphate buffer, pH 7.4). The chromogenic solution was
prepared in 100 mM sodium phosphate buffer (pH 7.4). The en-
zyme concentration used was 0.15 mg mitochondrial protein/mL.
Following incubation at 37 °C for a period of 20 min (substrate
Reactions containing baboon liver mitochondrial MAO-B
0.6 mg protein/mL) and 100 M of the substrates, 10 and 11,
(
l
respectively, were incubated for 14 h at 37 °C in sodium phosphate
buffer (100 mM, pH 7.4). The samples were subsequently centri-
fuged at 16,000g for 10 min and the supernatant fractions were
subjected to LC-ESI-MS analysis. Control reactions containing only
mitochondrial MAO-B (0.6 mg protein/mL) or substrate (10 or 11)
were included to facilitate the identification of the metabolite
peaks.
The LC-ESI-MS system consisted of an Agilent 1100 series bin-
ary gradient pump, an autosampler and a vacuum solvent degasser
coupled to an API 2000 triple quadrupole mass spectrometer (Ap-
plied Biosystems). Analyst 1.4 data acquisition and analysis soft-
ware was employed and mass spectra were acquired over the
scan range m/z 158–177 with the instrument operating in positive
ion mode. Nitrogen (purity 99.995%) was used as both nebulizer
1
0) or 15 min (substrate 11), the reactions were terminated by
the addition of 10 L (R)-deprenyl (5 mM) and placing the reac-
tions on ice. The samples were centrifuged at 16,000g for 10 min
and the H concentrations were measured by the peroxidase-
coupled spectrophotometric assay system described previously.
Quantitative estimations were made by means of a linear calibra-
tion curve that ranged from 11 to 88 M of H . The initial rates
l
2 2
O
2
3
l
2 2
O
as a function of substrate concentration were fitted to the Michae-
lis–Menten equation using the one site binding model incorpo-
rated into the GraphPad Prism software package (GraphPad
Software Inc.). All measurements were carried out in triplicate
and the K
m
and Vmax values were expressed as means ± standard
error of the mean (SEM). The steady-state MAO-B catalyzed oxida-
tion properties of benzylamine (12), kynuramine (13), MPTP (1),
MMTP (14), 7 and 4 were determined following the same protocol
with the exception that the experiments were carried out in dupli-
cate and the incubation times (the time intervals for which oxida-
tion remains linear) were chosen as follows: benzylamine (8 min),
kynuramine (6 min), MPTP (12.5 min), MMTP (10 min) and 7
(
20 psi) and the drying gas (15 L/min, 350 °C). The capillary voltage
and the vaporizer temperature were set at 5500 V and 300 °C,
respectively. The declustering, focusing and entrance potentials
were set to 50, 230 and 8.0 V, respectively. Separation was
achieved with a Gemini C18 column (2 Â 150 mm, 5
lm) (Phe-
nomenex, Torrance, CA) and a mobile phase which consisted ini-
tially of 95% Milli-Q water (containing 0.1% ammonium formate,
pH 5.0) and 5% acetonitrile. The solvent was delivered at a flow rate
(
10 min). For 4 the incubation time was 15 min.
of 250 lL/min and the injection volume was 10 lL. One minute
4
.6. Linearity of oxidation of the aminyl substrates
after injection of the sample, a time-gradient program was initi-
ated and the composition of acetonitrile in the mobile phase was
increased linearly to 90% over a period 7 min and maintained at
this percentage for 4 min. A period of 5 min was allowed for the
instrument to re-equilibrate at the initial conditions before injec-
tion of the next sample.
To determine the time interval for which MAO-B catalyzed pro-
duction of H
1
2
O
2
remains linear, the oxalate salts of test substrates
M) and 85 L of the chromogenic solu-
0 (500 M) and 11 (250
l
l
l
2
3
tion were incubated at 37 °C with 0.15 mg protein/mL of the ba-
boon liver mitochondrial fraction. The reactions were carried out in
1
5
00 mM sodium phosphate buffer (pH 7.4) to a final volume of
00 L. At the required time points (2.5–20 min), the reactions
L (R)-deprenyl (5 mM)
were measured as described pre-
viously. All measurements were conducted in triplicate and the
concentrations of H were expressed as means ± SEM.
4
.9. Animal studies and striatal dopamine measurements
l
were terminated with the addition of 10
and the MAO-B-generated H
l
Male C57BL/6 mice (30–35 g, 9–11 months of age) were pro-
2 2
O
2
3
vided by the Laboratory animal center of the Potchefstroom cam-
pus. The protocols for all animal trials were reviewed and
approved by the Research Ethics Committee of the North-West
University. Five animals were housed per cage in a temperature
2 2
O
4
.7. (R)-Deprenyl studies
To estimate the extent of the H
(
21 ± 0.5 °C) and humidity (50 ± 5% relative humidity) controlled
room on a 12 h light–12 h dark cycle with free access to food and
water. All injections were intraperitoneal (ip) in a volume of
2 2
O produced by non-MAO-B
mediated oxidation of substrates 10 and 11, baboon liver mitochon-
dria (0.3 mg protein/mL) were preincubated with the hydrochloric
acid salt of the MAO-B selective inactivator (R)-deprenyl
0
.2 ml per 30 g mouse. Sterile saline was used as vehicle for all of
the test compounds. Mice were sacrificed by rapid cervical disloca-
tion. The dopamine concentrations in the dissected mouse striata
were determined as described previously and were expressed
as means ± SEM.
À6
(
3
3 Â 10 M). These preincubations were carried out at 37 °C for
2
6
0 min in sodium phosphate buffer (100 mM, pH 7.4).22 The pre-
inactivated mitochondrial fractions were then added to the test sub-
strates 10 (500 M) and 11 (250 M) to yield final concentrations of
.15 mg mitochondrial protein/mL. The volume of these incubation
mixtures were 500 L and contained 85 L of the chromogenic solu-
tion. The reactions were incubated for 15 min at 37 °C and termi-
nated with the addition of 10 L (R)-deprenyl (5 mM). Control
l
l
0
Acknowledgments
l
l
2
3
We are grateful to Jan du Preez and the staff of the Analytical
Technology Laboratory, North-West University for their support.
The NMR and MS spectra were recorded by André Joubert, Johan
Jordaan and Louis Fourie of the SASOL Centre for Chemistry,
North-West University. This work was supported by grants from
the National Research Foundation and the Medical Research Coun-
cil, South Africa.
l
incubation reactions with the fully active enzyme preparation were
carried out following the same procedure with the exception that
for the preincubations the addition of (R)-deprenyl was omitted.
The concentrations of the MAO-B-generated H
2 2
O were measured
as described previously.23 All measurements were conducted in

4
118
A. Pretorius et al. / Bioorg. Med. Chem. 18 (2010) 4111–4118
1
5. Singer, T. P.; Ramsay, R. R.; McKeown, K.; Trevor, A.; Castagnoli, N., Jr.
Supplementary data
Toxicology 1988, 49, 17.
1
6. Sayre, L. M.; Singh, M. P.; Arora, P. K.; Wang, F.; McPeak, R. J.; Hoppel, C. L. Arch.
Biochem. Biophys. 1990, 280, 274.
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.bmc.2010.03.079.
17. Rollema, H.; Johnson, E. A.; Booth, R. G.; Caldera, P.; Lampen, P.; Youngster, S.
K.; Trevor, A. J.; Naiman, N.; Castagnoli, N., Jr. J. Med. Chem. 1990, 33, 2221.
18. Terau, Y.; Kotaki, H.; Imai, N.; Achiwa, K. Chem. Pharm. Bull. 1985, 33, 2762.
19. Blicke, F. F.; Zambito, A. J.; Stenseth, R. E. J. Org. Chem. 1961, 26, 1826.
20. American Cyanamid Co. German Patent, 1978, PN/DE2740562.
References and notes
1
2
3
4
5
.
.
.
.
.
Youdim, M. B. H.; Edmondson, D.; Tipton, K. F. Nat. Rev. Neurosci. 2006, 7,
95.
Chiba, K.; Trevor, A.; Castagnoli, N., Jr. Biochem. Biophys. Res. Commun. 1984,
20, 574.
Weisman, J.; Trevor, A.; Chiba, K.; Peterson, L.; Castagnoli, N., Jr. J. Med. Chem.
985, 28, 997.
Ramsay, R. R.; Krueger, M. J.; Youngster, S. K.; Gluck, M. R.; Casida, J. E.; Singer,
T. P. J. Neurochem. 1991, 56, 1184.
Markey, S. P.; Johannessen, J. N.; Chiueh, C. C.; Burns, R. S.; Herkenham, M. A.
Nature 1984, 311, 464.
Heikkila, R. E.; Manzino, L.; Cabbat, F. S.; Duvoisin, R. C. Nature 1984, 311, 467.
Heikkila, R. E.; Hess, A.; Duvoisin, R. C. Life Sci. 1985, 36, 231.
Langston, J. W.; Irwin, I.; Langston, E. B.; Forno, L. S. Neurosci. Lett. 1984, 50, 289.
Rimoldi, J. M.; Puppali, S. G.; Isin, E.; Bissel, P.; Khalil, A.; Castagnoli, N., Jr.
Bioorg. Med. Chem. 2005, 13, 5808.
21. Pine, S. H.; Sanchez, B. L. J. Org. Chem. 1971, 36, 829.
22. Inoue, H.; Castagnoli, K.; Van der Schyf, C.; Mabic, S.; Igarashi, K.; Castagnoli, N.,
Jr. J. Pharmacol. Exp. Ther. 1999, 291, 856.
23. Holt, A.; Sharman, D. F.; Baker, G. B.; Palcic, M. M. Anal. Biochem. 1997, 244, 384.
24. Di Monte, D. A.; Royland, J. E.; Anderson, A.; Castagnoli, K.; Castagnoli, N., Jr.;
Langston, J. W. J. Neurochem. 1997, 69, 1771.
25. Castagnoli, K. P.; Steyn, S. J.; Petzer, J. P.; Van der Schyf, C. J.; Castagnoli, N., Jr.
Chem. Res. Toxicol. 2001, 14, 523.
26. Harvey, B. H.; Brand, L.; Jeeva, Z.; Stein, D. J. Physiol. Behav. 2006, 87, 881.
27. Castagnoli, K.; Palmer, S.; Anderson, A.; Bueters, T.; Castagnoli, N., Jr. Chem. Res.
Toxicol. 1997, 10, 364.
2
1
1
6.
7.
8.
9.
28. Chiba, K.; Trevor, A. J.; Castagnoli, N., Jr. Biochem. Biophys. Res. Commun. 1985,
128, 1228.
29. Javitch, J. A.; Snyder, S. H. Eur. J. Pharmacol. 1984, 106, 455.
30. Pitts, S. M.; Markey, S. P.; Murphy, D. L.; Weisz, A. In MPTP: A Neurotoxin
Producing A Parkinsonian Syndrome; Markey, S. P., Castagnoli, N., Jr., Trevor, A. J.,
Kopin, I. J., Eds.; Academic Press: New York, 1986; pp 703–716.
31. Bissel, P.; Bigley, M. C.; Castagnoli, K.; Castagnoli, N., Jr. Bioorg. Med. Chem.
2002, 10, 3031.
1
1
1
1
0. Wang, Y. X.; Mabic, S.; Castagnoli, N., Jr. Bioorg. Med. Chem. 1998, 6, 143.
1. Williams, C. H.; Lawson, J. Biochem. J. 1998, 336, 63.
2. Williams, C. H.; Lawson, J. Neurobiology (Bp) 1999, 7, 225.
3. Ogunrombi, M. O.; Malan, S. F.; Terre’Blanche, G.; Castagnoli, K.; Castagnoli, N.,
Jr.; Bergh, J. J.; Petzer, J. P. Life Sci. 2007, 81, 458.
32. Salach, J. I.; Weyler, W. Methods Enzymol. 1987, 142, 627.
33. Bradford, M. M. Anal. Biochem. 1976, 72, 248.
1
4. Smeyne, R. J.; Jackson-Lewis, V. Brain Res. Mol. Brain Res. 2004, 134, 57.