Inactivation of bovine plasma amine oxidase by haloallylamines

Article abstract of DOI:10.1016/j.bmc.2005.09.065

Various 2- and 3-haloallylamines were synthesized and evaluated as inhibitors of the quinone-dependent bovine plasma amine oxidase (BPAO). 3-Haloallylamines, which were previously found to be good inhibitors of the flavin-dependent mitochondrial monoamine

Full text of DOI:10.1016/j.bmc.2005.09.065

Bioorganic & Medicinal Chemistry 14 (2006) 1444–1453
Inactivation of bovine plasma amine oxidase by haloallylamines
Jisook Kim,ꢀ Yuming Zhang, Chongzhao Ran and Lawrence M. Sayre*
Department of Chemistry, Case Western Reserve University, Cleveland, OH 44106, USA
Received 20 August 2005; revised 26 September 2005; accepted 28 September 2005
Available online 2 November 2005
Abstract—Various 2- and 3-haloallylamines were synthesized and evaluated as inhibitors of the quinone-dependent bovine plasma
amine oxidase (BPAO). 3-Haloallylamines, which were previously found to be good inhibitors of the flavin-dependent mitochon-
drial monoamine oxidase (MAO), exhibited a time-dependent inactivation of BPAO, with the 2-phenyl analogs being more potent
than the 2-methyl analogs. No plateau of enzyme activity loss was observed, suggestive of a lack of competitive partitioning to nor-
mal turnover. The (E)- and (Z)-2-phenyl-3-fluoro analogs were the most potent (low lM IC50s), with the corresponding 3-bromo
and 3-chloro analogs being >10-fold less potent. In each case, the Z-isomers were more potent than the E-isomers, the reverse of the
configurational inhibitory preference observed with MAO. In contrast to the 2-phenyl analogs, 3-phenyl-2(or 3)-chloroallylamines
displayed a partitioning behavior, consistent with these being both substrates and inactivators of BPAO.
ꢀ 2005 Elsevier Ltd. All rights reserved.
1. Introduction
amine to give the aldehyde. The mechanism involves for-
mation of quinone cofactor-substrate Schiff base, gener-
al base-mediated isomerization to a product Schiff base,
and hydrolysis of the latter to generate product aldehyde
and the reductively aminated form of the cofactor (an
aminoresorcinol). The quinone cofactor is recovered
by O2-dependent oxidation of reduced cofactor, releas-
ing ammonia and H2O2 (Scheme 1).6 Driven largely by
the recent discovery that one of the three TPQ-depen-
dent human gene products known as semicarbazide-
sensitive amine oxidase (SSAO) is identical with the hu-
man vascular adhesion protein-1 (HVAP-1),7,8 the
CAOs have become potential therapeutic targets for
intervening in a number of disease states, including
inflammatory diseases, fibrotic diseases (arthritis and
atherosclerosis), diabetes, cardiovascular disease, and
immune dysfunction diseases such as multiple sclerosis.
The copper-containing amine oxidases (CAOs, EC
1.4.3.6) are a ubiquitous family of enzymes, occurring
in bacteria, yeast, plants, and mammals, that catalyze
the oxidative deamination of primary amines to give
the corresponding aldehydes.1–3 In prokaryotic organ-
isms, these enzymes play major roles in nutrient catabo-
lism, whereas in plants and mammals, they play
important roles in metabolizing biogenic amines (includ-
ing mono-, di-, and polyamines and neurotransmitters)
involved in growth, cell division, differentiation, and
the stress response, as well as in metabolism of xenobi-
otic amines, in some cases with cytotoxic consequences.
Unlike the flavin-dependent mitochondrial monoamine
oxidase (MAO), which can oxidize primary, secondary,
and tertiary amines, the CAOs have a nearly strict pref-
erence for oxidation of primary amines. Most enzymes
in this family contain the trihydroxyphenylalanine qui-
none (TPQ) cofactor,4 derived post-translationally from
an active site tyrosine residue.5 This quinone cofactor,
along with active site general acid–base catalysis,
achieves a pyridoxal-like transamination of the primary
O
O
OH
RCH2NH2
-O
-O
-O
O
N
R
NH3
H2O
H2O
OH
O
RCHO
O2
Keywords: Copper amine oxidase; Enzyme inhibition; Haloallylamine.
*
H2O2
+
-O
Corresponding author. Tel.: +1 216 368 3704; fax: +1 216 368
3006; e-mail: lms3@case.edu
-O
NH2
NH
ꢀ
Present address: Department of Chemistry, Tennessee Technological
University, Cooke, TN, USA.
Scheme 1.
0968-0896/$ - see front matter ꢀ 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2005.09.065

J. Kim et al. / Bioorg. Med. Chem. 14 (2006) 1444–1453
1445
Y
X
Y
X
Y
X
H
H
Ph
Cl
CO2Et
BrF2C
Ph
F
F
a)
b)
CO2Et
CO2Bu-t
c)
H2N
H2N
H2N
Ph
H2N
Ph
CH3
Ph
CO2Et
CO2Bu-t
7
1a: X = Cl, Y = H;
1b: X = H, Y = Cl;
1c: X = Br, Y = H;
1d: X = H, Y = Br;
1e: X = F, Y = H;
1f: X = H, Y = F;
1g: X = F, Y = F.
1h: X = Br, Y = H:
1i: X = H, Y =Br.
1j: X = Cl, Y = Ph;
1k: X = Ph, Y = Cl.
1l
9
8
F
H
H
F
F
F
F
F
d)
+
Ph
CH2OH
Ph
CH2OH Ph
11
CH2OH
Ph
CH2NPhth
14
10
d)
Figure 1. Structures of compounds evaluated in this study.
F
H
CH2NPhth
H
F
e)
f)
It was reported more than 20 years ago that propargyl-
amine and 2-chloroallylamine could inactivate bovine
plasma amine oxidase (BPAO).9 These same types of
derivatives were later designed as potential mecha-
nism-based inhibitors of MAO10 and/or SSAO.11
Although mammalian SSAOs include both a mem-
brane-bound and circulating plasma form, the ability
of haloallylamines other than 2-chloroallylamine to
inactivate BPAO was not studied prior to our report
that (E)- and (Z)-3-chloroallylamine are two of the most
potent inactivators of this enzyme.12 On the basis of the
desire to evaluate inhibitory selectivities that can be
achieved among various mammalian amine oxidases,
we were curious about the potential for inactivation of
BPAO of the (E)- and (Z)-3-halo-2-phenylallylamines
that Merrell Dow developed originally for inhibition
of MAO.10 We also considered replacing the 2-phenyl
group with 2-methyl group, which would result in differ-
ent electronic and steric attributes. In this paper, we thus
describe the inhibitory activities of haloallylamines 1a–l
(Fig. 1) toward BPAO.
g)
Ph
Ph
CH2NPhth
13
12
g)
g)
F
H
H
F
F
F
Ph
CH2NH2
HCl
Ph
CH2NH2
HCl
Ph
CH2NH2
HCl
1f
1e
1g
Scheme 3. Reagents: (a) NaH, CBr2F2, THF; (b) reduced pressure
distillation; (c) DIBAL-H; (d) Ph3P, DEAD, phthalimide; (e) Br2; (f)
NaI; (g) NH2NH2 then HCl.
butoxycarbonyl)phenylacetate with CBr2F2 under basic
conditions at room temperature. Heating crude 8 under
reduced pressure afforded ethyl 3,3-difluoro-2-phenylac-
rylate (9). Diisobutylaluminum hydride (DIBAL-H)
reduction of 9 afforded a 65% yield of (E)-2-phenyl-3-
fluoroallyl alcohol (10) and 15% of 3,3-difluoro-2-phe-
nylallyl alcohol (11) that was contaminated with minor
inseparable (Z)-2-phenyl-3-fluoroallyl alcohol. The
Mitsunobu reaction converted alcohols 10 and 11 into
the corresponding phthaloyl-protected amines 12 and
14. Difluorinated 14 was further purified by recrystalli-
zation. Isomerization of 12–13 was accomplished by a
sequence of syn bromination followed by NaI-mediated
anti debromination. Deprotection of 12, 13, and 14 with
hydrazine and HCl follow-up furnished 1e, 1f, and 1g
hydrochloride salts, respectively.
2. Results
2.1. Synthesis
The synthesis of 2-phenylallylamine and compounds 1a–
c was carried out as described previously (Scheme 2).10
Compound 1d was also obtained by careful silica gel
chromatography (eluant petroleum ether–EtOAc 2:1)
during the separation of 1c.
The preparation of isomeric 3-bromo-2-methylallylam-
ines 1h and 1i hydrochloride salts was accomplished as
shown in Scheme 4. Treatment of 3-chloro-2-methylpro-
pene with potassium phthalimide in DMF, followed by
bromination in CCl4 and then 1,8-diazobicy-
clo[5.4.0]undec-7-ene (DBU)-mediated dehydrobromin-
Fluorinated 2-phenylallylamines 1e–f were prepared by
a modification of the literature route10 that was required
in our hands to afford sufficient yields (Scheme 3). First,
ethyl 2-(t-butoxycarbonyl)-2-(bromodifluoromethyl)-
phenylacetate (8) was obtained by treating ethyl 2-(t-
potassium
NBS
O
1) X2
phthalimide
DMF, 90 oC.
NPhth
a) Br2
Ph
CH3
CCl4
Ph
CH2Br
Ph
+
2) DBU
NK
Me
Me
b) DBU
2
NPhth
3
4
Cl
16
Br
Me
O
15
X
H
X
H
H
X
NH2NH2
H
X
Br
Br
Br
a) NH2NH2
b) HCl
+
+
NH2
NPhth
NPhth
NH2
+
+
Ph
Ph
Ph
Ph
Me
Me
Me
HCl
NH2
NPhth
HCl
NH2
NPhth
17b
5a: X = Cl;
b: X = Br.
6a: X = Cl;
b: X = Br.
1a: X = Cl;
c: X = Br.
1b: X = Cl;
d: X = Br.
1h
1i
17a
Scheme 2.
Scheme 4.

1446
J. Kim et al. / Bioorg. Med. Chem. 14 (2006) 1444–1453
ation, afforded a mixture of (E)- and (Z)-1-bromo-2-
methyl-3-phthalimido-1-propenes 17a and 17b. After
hydrazine-mediated deprotection, the mixture of free
bases was chromatographed to remove excess hydrazine,
and the pooled free base mixture was treated with
methanolic HCl to give salts 1h and 1i, which were sep-
arated by fractional crystallization. The major E-isomer
1h was obtained in pure form, though the minor Z-iso-
mer 1i still contained about 10% of 1h after repeated
crystallization.
phenylacetylene with methyl chloromethyl ether in the
presence of ZnCl2 as Lewis acid to give a mixture of
the known 1,3-dichloro-1-phenylpropene isomers 21a
and 21b,14 which were separated by fractional column
chromatography. The individual isomers were treated
with ammonium hydroxide to give the desired amines
lj and 1k, respectively. The configuration at both the
chloride and amine stages was established on the basis
that the CH2 cis to the phenyl ring in the E-isomers is
shielded relative to the trans CH2 in the Z-isomers.14
The assignment of configuration of the E- and Z-iso-
mers 1a–d and 1h/1i was made on the basis of nuclear
Overhauser enhancement (NOE) experiments (Fig. 2).
Thus, for (E)-3-chloro-2-phenylallylamine (1a), presatu-
ration of the methylene singlet at 4.05 ppm enhanced the
vinyl signal at 6.78 ppm, while the (Z)-isomer showed no
such enhancement. For these analogs, all (E)-isomers
exhibited a (+) NOE, while there was no NOE for the
(Z)-isomers. Once the stereochemistry of the final amine
salts was elucidated, the stereochemistry of the phthali-
mide precursors could be assigned, and we note that
elimination of the 1,2-dihalo intermediates always gave
mainly the E-isomer.
2.2. Inactivation of BPAO by (E)- and (Z)-3-halo-2-
phenylallylamines
Data on the consequence of pre-incubation of 2-phenyl-
allylamine and its various 3-halo analogs with BPAO on
its activity are listed in Table 1, alongside data obtained
previously for rat brain MAO (1d was not tested).10 In
contrast to the weak inhibitory activity seen for the par-
ent compound, the 3-halo derivatives exhibited remark-
ably potent, time- and concentration-dependent
inhibition of BPAO. (Z)-3-Fluoro-2-phenylallylamine
(1f) showed the most potent inhibition, effecting 50%
loss of BPAO activity in 40 min at a concentration of
2 lM. The Z-isomers of 3-chloro- (1b) and 3-bromo-2-
phenylallylamine (1d) both exhibited efficient, but about
15-fold weaker, inactivations than 1f. Compared to
these three Z-isomers, all three E-isomers of 3-halo-2-
phenylallylamine exhibited approximately 10-fold less
effective inhibition, respectively, on the basis of IC50 val-
ues. Interestingly, the 3,3-difluoro-analog exhibited
inhibitory activity in between that of the two configura-
tional 3-monofluoro analogs.
The preparation of 1l started with reduction of a-chloro-
cinnamaldehyde to the corresponding known (Z)-alco-
hol 19,13 which subsequently was converted to
phthalimide derivative 20 by the Mitsunobu reaction
(Scheme 5). Deprotection of 20 with hydrazine readily
afforded (Z)-2-chloro-3-phenylallylamine 1l. Lastly, the
E and Z-isomers of 3-chloro-3-phenylallylamine (1j
and 1k) were prepared starting with the reaction of
Although the 3-halo-2-phenylallylamines showed pseu-
do-first order time-dependent inactivation behavior
(see Fig. 3 as an example), Kitz and Wilson replots15
of the data gave lines passing through the origin in every
case, suggestive of a lack of saturation and negating the
possibility for obtained KI and kinact parameters. In this
situation, the reaction kinetics are best described in
terms of the slope of the line passing through the origin,
equivalent to kinact/KI, as if the inactivations represent
bimolecular reactions between the enzyme and the
inhibitors. However, since the inhibitors have little
intrinsic reactivity, this phenomenon is best explained
in terms of kinact being fast relative to the dissociation
of the initial E–I complex. The relative potencies of inac-
tivation for the 3-halo-2-phenylallylamines are thus best
compared kinetically in terms of the kinact/KI values list-
ed in Table 1. On the basis of these values, it can be seen
that the (Z)- and (E)-3-fluoro analogs are 15–25 times
more potent BPAO inhibitors than the corresponding
3-chloro and 3-bromo analogs of the same
configuration.
15.5% NOE
16.1% NOE
H
H
H
H
H
H
H
H
Br
H
H2N
H
H2N
H2N
Cl
H
H
H2N
Cl
Br
1a
1b
1c
1d
2.9% NOE
H
H
H
H
H
H2N
Br
H2N
Br
H3C
H3C
H
1h
1i
Figure 2. Structural assignments of c-haloallylamine analogs.
H
CHO
Cl
H
CH2OH
H
CH2NPhth
H
CH2NH2
Cl
a)
b)
c)
Ph
Ph
Cl
Ph
Cl
Ph
1l
19
20
18
Cl
H
Cl
H
e)
e)
1j
PhC CH
+
21a
Additional experiments confirmed the irreversibility of
BPAO inactivation by 1a–d, as indicated by re-assay
of activity following separation of enzyme from small
molecules by gel-filtration. However, 24 h dialysis of
the inactivated enzyme preparations was associated with
partial recovery of activity (Table 2). Phenylhydrazine
titration16 of the active site after 90% inactivation of
d)
Ph
Cl
CH2Cl
CH2Cl
Ph
Cl
CH2NH2
CH2NH2
MeOCH2Cl
21b
1k
Ph
Ph
H
H
Scheme 5. Reagents: (a) DIBAL-H; (b) Ph3P, phthalimide, DEAD;
(c) NH2NH2; (d) ZnCl2; (e) NH4OH.

J. Kim et al. / Bioorg. Med. Chem. 14 (2006) 1444–1453
Table 1. The inhibitory potencies of haloallylamine analogs on BPAO and MAO
1447
X
Y
BPAO inhibition 40 min
IC50 (lM)
MAO inhibition 15 min
IC50 (lM)a
kinact/KI (BPAO)
(mMÀ1 minÀ1
)
IC50 MAO/IC50 BPAO
X
Y
NH2
Ph
H
Cl
H
Br
H
F
H
H
Cl
H
Br
H
F
2000
311.7
33.9
400
100
500
40
n.d.b
0.054
0.58
n.d.b
0.52
0.84
13.2
2.2
0.20
0.32
1a
1b
1c
1d
1e
1f
15
100 (IC20
)
29.0
21.4
0.018
0.15
10
0.0008
0.075
1.2
H
F
2.0
8.4
1g
F
a Data from Ref. 10.
b Not determined.
50
40
30
20
10
0
0.0
0.2
0.4
0.6
0.8
1.0
1/[Inhibitor],µM-1
Figure 3. Time-dependent inactivation of BPAO by various concentrations of (Z)-3-fluoro-2-phenyl-2-propenamine (1f) (left), and Kitz and Wilson
replot of the data (right).
Table 2. Reversibility of BPAO inhibition by 3-halo-2-phenylallylamines
Compound
Enzyme activity at end of
incubation before gel-filtrationa (%)
After PDX
G. F. 25 (%)
Enzyme activity at end of
incubation before dialysisb (%)
After dialysisc (%)
1a
1b
1c
1d
3.2
5.2
5.8
0
6.7
3.2
9.1
8.7
0.9
2.5
5.4
20
17
31
a 1a at 1 mM, 1b at 100 lM, 1c at 10 mM, and 1d at 0.5 mM.
b 1a at 1 mM, 1b at 200 lM, and 1c at 10 mM.
c Enzyme activity after dialysis for 24 h against 50 mM, pH 7.2, sodium phosphate buffer at room temperature.
BPAO by 1a failed to exhibit the normal development of
absorbance at 440 nm due to derivatization of the TPQ
cofactor. However, the nitroblue tetrazolium (NBT) as-
say17 on the denatured enzyme following inactivation by
1f was indistinguishable from the denatured control
enzyme.
observed in the 2-methyl series of inactivators. Both
(E)- and (Z)-3-bromo-2-methylallylamines exhibited
very weak inactivation compared to the 2-phenyl inhib-
itors. In the case of (Z)-3-bromo-2-methylallylamine
(1i, containing 10% 1h), 50% inactivation was achieved
at 1 mM in 150 min, which is a 100-fold higher IC50
than the corresponding phenyl analog. The E-isomer
1h exerted somewhat weaker inhibition with an IC50
of 1.5 mM at 150 min. The finding that 1h was a weak-
er inhibitor suggested that the greater activity of 1i did
not reflect the contaminating 1h. The small Z > E
potency difference here for BPAO inhibition contrasts
the 10-fold Z > E potency difference seen for the 2-
phenyl series.
2.3. Inactivation of BPAO by (E)- and (Z)-3-bromo-2-
methylallylamines (1h and 1i)
Since inactivation of BPAO by 3-halo-2-phenylallylam-
ines displayed a general dependence on the relative
configuration between the phenyl ring and the halogen,
we investigated whether a similar tendency would be

1448
J. Kim et al. / Bioorg. Med. Chem. 14 (2006) 1444–1453
Following gel-filtration, a sample of enzyme inactivated
by 5 mM 1i by 97.5% exhibited only slight recovery (to
7%). Upon phenylhydrazine titration of the active site
following inactivation by 1i, the normal development
of absorbance at 440 nm was not observed, consistent
with 1i being a mechanism-based inactivator of BPAO.
2.4. Inactivation of BPAO by halogenated 3-phenylallyl-
amines (1j–l)
Unlike the behavior of 2-phenylallylamine as a weak
inhibitor, 3-phenylallylamine (cinnamylamine) was
found to be a mixed substrate-inhibitor of BPAO,18
exhibiting a maximal turnover rate to cinnamaldehyde
(monitored at 285 nm) at 10 lM. Inhibition must in part
be mediated at an ancillary site, since competitive inhibi-
tion due to substrate activity would not explain the de-
creased turnover at higher concentrations. Halogenated
3-phenylallylamines behaved as inactivators of BPAO,
but the time course showed a different pattern compared
to 3-halo-2-phenylallylamines. Thus, after a quick time-
dependent loss of enzyme reactivity, an apparent plateau
was reached, suggestive of a partitioning between inacti-
vation and turnover to a product with low or nil inhibi-
tory activity (the example for 1j is shown in Fig. 5). IC50
values at the plateau points and partition ratios are listed
in Table 3. The large partition ratios suggest that these
amines are good substrates of BPAO, though the extent
of inactivation achieved at each concentration was
shown to be irreversible by observing no increase in en-
zyme activity after gel-filtration.
Despite the weak inhibition for these compounds, time-
dependent loss of BPAO activity showed no obvious
plateau, suggesting that there was little competing pro-
ductive turnover and thus, that the weak inhibitory ef-
fect principally reflected a weak binding affinity of
these compounds. The time-dependent inhibition data
for 1h are shown in Figure 4 (left), and the finding that
the Kitz and Wilson replot (Fig. 4, right) intersects the
0–0 origin, again indicates a lack of saturation, so that
the inactivation kinetics are best described in terms of
Kinact/Km = 0.0055 minÀ1 mMÀ1
.
The corresponding
value for 1i was Kinact/Km = 0.032 minÀ1 mMÀ1. These
values are 2–3 orders of magnitude lower than what
was observed for 1a–g. The apparently weak activity
and selectivity in the methyl series and presumed diffi-
culty in synthesizing the corresponding fluoro analogs
led to a decision not to evaluate additional analogs.
100
200
180
160
140
120
100
80
10
1mM
2mM
3mM
4mM
5mM
60
40
20
0
1
0.0
0.2
0.4
0.6
0.8
1.0
0
50
100
150
200
250
-1
Time(min)
1/[I], (mm )
Figure 4. Time-dependent loss of BPAO activity upon incubation with various concentrations of (E)-3-bromo-2-methylallylamine (1h) (left), and
Kitz and Wilson replot of the data (right).
100
80
60
40
20
0
-200
0
200
400
600
800 1000 1200 1400 1600
[I]/[E]
Figure 5. Time-dependent inactivation of BPAO by various concentrations of (Z)-3-chloro-3-phenylallylamine 1j (left) and partition ratio plot.

J. Kim et al. / Bioorg. Med. Chem. 14 (2006) 1444–1453
1449
Table 3. IC50 values and partition ratios for the inhibition of BPAO by
1j–l
OH
N
O
O
O
H2O
Compound
IC50 at plateau (mM)
Partition ratio
Ph
HO
Ph
-O
X
X
1j
0.46
0.40
0.8
1360
1280
2400
NH2
1k
1l
:Nu
Ph
X
3. Discussion and conclusion
O
OH
N
Nu = active-site residue or
reduced cofactor ArNH2
In this study, 2-phenyl-3-haloallylamines were found to
be potent irreversible inactivators of BPAO. The general
order of potency was F ꢀ Br ꢁ Cl, and in each case, the
Z-isomers were about 10-fold more potent inactivators
than the E-isomers (the 3,3-difluoro isomer exhibited
activity in between the 3-(Z)-fluoro and 3-(E)-fluoro iso-
mers). As shown in Table 1, this configurational prefer-
ence is opposite to that seen for inhibition of MAO,
where the E-isomers are more potent. A priori, there is
no reason for inhibition of these two amine oxidases
to display the same stereoselectivity, since the respective
active sites have no apparent homology, and the mech-
anisms of oxidative deamination are entirely different.
Data on the relative inhibitory activity of this full series
of compounds (1a–g) on SSAO are not available, but for
the case of 2-(3,4-dihydroxyphenyl)-3-fluoroallylamine,
the E configurational isomer was found to be about
six times more potent than the Z-isomer for the SSAO
activity in rat aorta homogenates.11a Thus even within
the quinone-dependent enzyme class, there may be dif-
ferent stereoselectivities between the tissue-bound and
soluble enzymes. Opposite configurational rank orders
have important implications on the design of selective
inhibitors for the different enzyme families.
Ph
Nu
HO
Ph
Nu
Scheme 6.
could occur at either product Schiff base stage or subse-
quent to hydrolytic release of a,b-unsaturated aldehyde.
Alternatively, the latter aldehyde could ꢁbite backꢂ and
alkylate the reduced cofactor prior to release into bulk
solvent, as we have seen in another case.19 Regardless
of the alkylation mode, such mechanism would be con-
sistent with the eventual partial recovery of activity seen
upon dialysis, since the adducts shown could eventually
hydrolyze (e.g., by a second addition–elimination reac-
tion). It is not clear whether the enhanced potency of
the fluoro with respect to bromo and chloro analogs re-
flects a steric preference or electronic activation effect.
Replacement of the 2-phenyl group with a 2-methyl
group in the 3-bromo series resulted in substantially re-
duced inhibitory potency, though the mechanism ap-
pears to be the same, as no evidence of a partitioning
phenomenon (plateauing of enzyme activity) was ob-
served. Although not directly addressed, we interpret
the weaker inhibitory activity to reflect weaker active
site binding of these analogs. Weaker inhibition can also
be rationalized in that the methyl-substituted haloacryl-
aldehyde turnover product is a weaker Michael acceptor
than the corresponding phenyl analogs and would thus
afford a less efficient inactivating covalent binding activ-
ity. In the 2-methyl case, there was also a smaller Z > E
potency difference.
Interestingly, in the case of MAO, the 3-fluoro isomers
are also more potent than the 3-chloro and 3-bromo
analogs, by an even greater extent than in the case of
BPAO. Thus, as shown by the limited IC50 ratio data
that can be extracted (last column in Table 1), although
the 3-fluoro analogs are the most potent inhibitors of
BPAO, it is the 3-chloro analogs (along with the 3,3-
difluoro analog) that are most selective for inhibition
of BPAO over MAO.
Although the phenylhydrazine and nitroblue tetrazoli-
um (NBT) redox cycling assays were not conducted on
the whole series, we assume a common mechanism of
inhibitory action for all 3-halo-2-phenylallylamines.
Thus, the finding of a lack of phenylhydrazine reactivity
with 1a-inactivated BPAO but retention of cofactor re-
dox activity of the 1f-inactivated BPAO suggests that
these analogs inactivate BPAO by a mechanism that in-
volves covalent alkylation of an active site residue by the
electrophilic a,b-unsaturated aldehyde turnover prod-
ucts, in a manner that blocks access of substrate benzyl-
amine and phenylhydrazine to the quinone cofactor.
The great enhancement in inhibitory potency of 2-phe-
nylallylamine by addition of the 3-halo substituent leads
us to suggest an efficient addition–elimination mecha-
nism of inactivation (Scheme 6), as we proposed in the
case of the simple 3-chloroallylamine E- and Z-iso-
mers.12 Covalent adduction of an active-site nucleophile
In contrast to the 2-phenyl and 2-methyl analogs, the 3-
phenyl chloroallylamines 1j–l were moderately potent
inactivators that all showed a plateau phenomenon, sug-
gestive of partitioning between productive turnover and
inactivation. The (E)- and (Z)-3-chloro-3-phenyl analogs
were more efficient inactivators than the (Z)-2-chloro-3-
phenyl analog on the basis of an observed 2-fold lower
IC50 and 2-fold lower partition ratio. Exactly why there
should be a different behavior observed for these deriva-
tives in terms of the mechanism proposed in Scheme 6 is
not clear, but it is evident that both electronic and steric
effects on the postulated addition–elimination process
would be different for 2-phenyl versus 3-phenyl substitu-
tion. Further work will be required to iron out details of
enzyme modification responsible for the various inhibi-
tory activities seen for this series of compounds.

1450
J. Kim et al. / Bioorg. Med. Chem. 14 (2006) 1444–1453
4. Experimental
To a solution of 7 (5.2 g, 20 mmol) in THF (80 mL) was
suspended NaH (1.6 g, 60% by weight in oil, 40 mmol)
at 0 ꢁC. After H2 was evacuated, dibromodifluorome-
thane (6.2 g, 60 mmol) was introduced at room temper-
ature. The reaction mixture was stirred for 5 days and
then quenched carefully with water (100 mL). The solu-
tion was extracted with ethyl acetate (3 · 50 mL), and
the combined organic layer was dried (Na2SO4) and
concentrated in vacuo at room temperature. The residue
was distilled at 110 ꢁC under reduced pressure to give
pure 9 in 77% yield: bp 55 ꢁC/3 mmHg; 1H NMR
(CDCl3) d 1.31 (t, 3H, J = 7.1 Hz), 4.29 (q, 2H,
J = 7.1 Hz), 7.31–7.42 (5H); 13C NMR (CDCl3, partial
spectrum) d 26.4, 73.9, 140.7, 142.26, 142.31, 142.4,
171.9 (dd, J = 309.6 and 296.0 Hz).
4.1. General
1H NMR spectra were obtained on a Varian Gemini 300
instrument (13C NMR at 75 MHz) or a Vrian Gemini 200
instrument (13C NMR at 50 MHz) with chemical shifts
being referenced to TMS or the solvent peak. In the 13C
NMR line listings, attached proton test (APT) designa-
tions are given as (+) or (À) following the chemical shift.
High-resolution mass spectra (HRMS, FAB) were ob-
tained at 20–40 eV on a Kratos MS-25 A instrument.
UV–vis spectra were obtained using a jacketed (tempera-
ture-controlled) cell compartment and Perkin-Elmer
PECSS software. Bovine plasma amine oxidase was pur-
chased from Sigma or Worthington Biochemical Corpo-
ration. The water used for all studies was doubly
distilled and deionized. All solvents, reagents, and organic
fine chemicals were the most pure available from commer-
cial sources. All column chromatography was conducted
using flash grade silica gel using a hexanes–EtOAc gradi-
ent (increasing polarity), unless specified otherwise.
4.2.3. DIBAL-H reduction of 9. To a solution of 9
(2.12 g, 10 mmol) in THF (60 mL) at À78 ꢁC was added
a solution of DIBAL-H in hexanes (1.5 M, 20 mL,
30 mmol). The reaction mixture was stirred for 1 h and
warmed to 0–5 ꢁC for another 2 h. The reaction was
quenched by methanol and then with 10% aqueous
KOH (10 mL). The organic layer was collected and
the aqueous solution was extracted with ether
(3 · 20 mL). The combined organic layer was dried
(Na2SO4) and concentrated in vacuo. The residue was
purified by column chromatography to give pure (E)-
3-fluoro-2-phenyl-2-propen-1-ol (10): yield 65%; 1H
NMR (CDCl3) d 4.38 (d, 2H, J = 4.4 Hz), 6.88 (d, 1H,
J = 82.5 Hz), 7.33–7.44 (m, 3H), 7.54 (d, 2H,
J = 7.4 Hz). A second column fraction, contaminated
by inseparable (Z)-3-fluoro-2-phenyl-2-propen-1-ol,
was 3,3-difluoro-2-phenyl-2-propen-1-ol (11): yield
15%; 1H NMR (CDCl3) d 4.53 (t, 2H, J = 2.5 Hz),
7.32–7.47 (5H).
4.2. Synthesis
4.2.1. Preparation of 3-halo-2-phenylallylamines (1a–d).
Compounds 1a–d were prepared according to McDon-
aldꢂs procedure.10
(E)-3-Chloro-2-phenylallylamine Æ HCl (1a): 1H NMR
(D2O) d 4.05 (d, 2H, J = 1 Hz), 6.78 (t, 1H, J = 1 Hz),
7.46–7.56 (5H); 13C NMR (D2O) d 47.3, 124.5, 131.9,
132.2, 132.5, 136.6, 138.5.
(Z)-3-Chloro-2-phenylallylamine Æ HCl (1b): 1H NMR
(D2O) d 4.31 (s, 2H), 6.76 (s, 1H), 7.47 (br s, 5H); 13C
NMR (D2O) d 40.8, 125.5, 129.7, 132.0, 136.0, 137.7,
138.1.
4.2.4. Preparation of phthaloyl-protected allylamines
(12–14). To a solution of 3-fluorinated-2-phenylallyl
alcohol 10 or 11 (5 mmol), Ph3P (1.78 g, 7.5 mmol),
and phthalimide (1.1 g, 7.5 mmol) in THF (60 mL)
was added diethyl azodicarboxylate (DEAD) (1.3 g,
7.5 mmol) at 0 ꢁC. After addition, the mixture was
warmed to room temperature and reacted overnight.
The solvent was removed and the residue was purified
by column chromatography. From 10 was obtained 12
in 82% yield: 1H NMR (CDCl3) d 4.57 (dd, 2H,
J = 3.5 and 1.1 Hz), 6.82 (0.5H, one peak of 19F-cou-
pled vinyl doublet), 7.23–7.37 (3H + 0.5H of 19F-cou-
pled vinyl doublet), 7.51 (m, 2H), 7.65 (m, 2H), 7.77
(m, 2H); 13C NMR (CDCl3) d 38.0 (d, J = 10 Hz),
118.2, 123.4, 128.1, 128.4, 128.5, 128.6, 131.9, 132.0,
133.0, 134.1, 148.4 (d, J = 266.2 Hz), 167.9. From im-
pure 11 was obtained, after column chromatography,
crude 14, which was further purified by recrystalliza-
tion: yield 62%; 1H NMR (CDCl3) d 4.70 (t, 2H,
J = 0.6 Hz), 7.22–7.38 (5H), 7.64–7.79 (4H); 13C
NMR (CDCl3) d 35.5 (m), 123.4, 128.0, 128.6, 128.7,
128.8, 131.8, 134.1, 154.7, 167.7.
(E)-3-Bromo-2-phenylallylamine Æ HCl (1c): 1H NMR
(D2O) d 3.95 (s, 2H), 6.85 (s, 1H), 7.34–7.46 (5H); 13C
NMR (D2O) d 48.0, 113.6, 131.4, 131.9, 132.1, 137.8,
140.9.
(Z)-3-Bromo-2-phenylallylamine Æ HCl (1d): 1H NMR
(D2O) d 4.32 (s, 2H), 6.95 (s, 1H), 7.44 (br s, 5H); 13C
NMR (D2O) d 43.5 (+), 115.4 (À), 129.7 (À), 132.0
(À), 132.1 (À), 138.9 (+), 140.5 (+).
4.2.2. Preparation of ethyl 3,3-difluoro-2-phenylacrylate
(9). To a solution of tert-butyl phenylacetate (0.58 g,
3 mmol) in THF was added lithium diisopropylamide
in THF (1 M, 6 mL) at À78 ꢁC for 30 min. The result-
ing solution was treated with ethyl chloroformate
(0.43 g, 3 mmol) and stirred for 2 h, then at room tem-
perature for 1 h. The THF was removed, and saturat-
ed aqueous NH4Cl (50 mL) was added, and the
mixture was extracted with ether (3 · 50 mL). The
combined organic layer was dried (Na2SO4) and con-
centrated in vacuo. The residue was distilled to give
ethyl 2-(tert-butoxycarbonyl)phenylacetate (7) in 87%
yield: 1H NMR (CDCl3) d 1.28 (t, 3H, J = 7.1 Hz),
1.46 (s, 9H), 4.21 (q, 1H, J = 7.1 Hz), 4.22 (q, 1H,
J = 7.1 Hz), 4.53 (s, 1H), 7.34–7.38 (5H); 13C NMR
(CDCl3) d 14.1, 27.9, 59.1, 61.7, 82.4, 128.1, 128.6,
129.3, 133.3, 167.3, 168.5.
To a solution of 12 (0.56 g, 2 mmol) in CH2Cl2 (10 mL)
at 0 ꢁC was added bromine (0.35 g, 2.2 mmol) in CH2Cl2
(5 mL). The cooling bath was then removed and the
solution was stirred for 24 h. After removal of organic

J. Kim et al. / Bioorg. Med. Chem. 14 (2006) 1444–1453
1451
1
solvent and excess Br2 in vacuo, NaI (15 g, 100 mmol)
and acetone (100 mL) were added, and the solution
was heated at reflux for 4 h. The reaction mixture was
decolorized by 10% aqueous Na2SO3 (80 mL). Acetone
was removed, and the water layer was extracted with
ether (3 · 80 mL). The combined ether extract was dried
(Na2SO4) and concentrated in vacuo. Column chroma-
tography of the residue gave pure starting material 12
to give 16 in 79% yield: H NMR (CDCl3) d 1.78 (s,
3H), 4.22 (s, 2H), 4.85 (m, 2H), 7.70–7.89 (4H); 13C
NMR (CDCl3) d 20.5, 43.3, 112.0, 123.4, 132.1, 139.4,
168.1.
4.3. Preparation of (E)- and (Z)-1-bromo-2-methyl-3-
phthalimidopropenes (17a and 17b)
1
and its Z-isomer 13 (0.21 g, 38%): H NMR (CDCl3) d
To a solution of 2-methyl-3-phthalimido-1-propene 16
(5 g, 15 mmol) in chloroform (50 mL) at À10 ꢁC was
added 1.1 equiv of Br2 dropwise over 10 min. The reac-
tion mixture was stirred for an additional 15 min, and
the excess Br2 was removed by bubbling air for 2 h. Sat-
urated aqueous NaCl was added at 0 ꢁC along with a
small amount of pentane (or hexane). The separated
organic layer was washed with 2% aqueous NaHCO3
until the pH of the water layer reached ca. 8, dried over
Na2SO4, and concentrated. The solid was recrystallized
from CHCl3 to give 1,2-dibromo-2-methyl-3-phthalimi-
4.83 (dd, 2H, J = 2.6 and 1.3 Hz), 6.86 (d, 1H,
J = 82.8 Hz), 7.24–7.33 (5H), 7.64 (m, 2H), 7.75 (m,
2H); 13C NMR (CDCl3) d 34.2 (d, J = 7.0 Hz), 120.3
(d, J = 5.8 Hz), 123.3, 127.4, 127.5, 128.2, 128.7, 131.9,
133.5, 134.0, 146.0 (d, J = 265.7 Hz), 167.8.
4.2.5. Preparation of 3-fluorinated-2-phenylallylamines
(1e–g). A mixture of phthaloyl-protected fluorinated
amine (1 mmol) and hydrazine hydrate (100 mg,
2 mmol) in CH3OH (10 mL) was refluxed for 3 h. Aque-
ous HCl (6 N, 1 mL) was added and refluxing was con-
tinued for 30 min. The mixture was cooled and filtered
to remove phthalhydrazide. Evaporation of the filtrate
left a solid residue which typically contained traces of
hydrazine hydrochloride even after recrystallization.
Thus, the weighed residue was dissolved in aqueous
THF containing 4 equiv of Na2CO3, to which was added
di-tert-butyl dicarbonate. After 30 min stirring, and
TLC showed complete conversion of the free amine to
the Boc-protected compound, the solution was concen-
trated in vacuo to remove the THF and extracted with
EtOAc. The organic layer was dried (Na2SO4) and evap-
orated, and the residue was purified by column chroma-
tography. The Boc-protected compound was dissolved
in 6 N HCl, and the solution was evaporated to dryness.
Excess HCl was removed by repetitive dissolution in
water and evaporation. The final residue was recrystal-
lized from EtOH/Et2O to afford hydrazine-free 1e–g
hydrochloride salts.
1
dopropane (1.8 g, 33%): H NMR (CDCl3) d 1.89 (s,
3H), 3.86 (s, 2H), 4.16 (s, 2H), 7.75–7.87 (4H); 13C
NMR (CDCl3) d 29.4 (À), 42.3 (+), 47.5 (+), 46.2 (+),
123.7 (À), 131.8 (+), 134.4 (À), 168.1 (+).
To a solution of 1,2-dibromo-2-methyl-3-phthalimido-
propane (1 g, 2.78 mmol) in DMSO (40 mL) was added
1.5 equiv of DBU. The reaction mixture was heated at
90 ꢁC for 10 min followed by further heating at 150 ꢁC
for 2 h. The cooled reaction mixture was diluted with
ice water and partitioned between EtOAc and water.
The organic layer was dried (Na2SO4) and concentrated,
and the residue was subjected to column chromatogra-
phy (eluant petroleum ether–EtOAc 2:1) to give 0.5 g
(64%) a mixture of 17a/17b. Compound 17a (major iso-
mer): 1H NMR (CDCl3) d 1.85 (s, 3H), 4.28 (s, 2H), 6.28
(s, 1H), 7.70–7.85 (4H); 13C NMR (CDCl3) d 17.9, 43.3,
106.7, 123.5, 131.9, 134.2, 136.1, 167.9. Compound 17b
1
(minor isomer): H NMR (CDCl3) d 1.72 (s, 3H), 4.50
(s, 2H), 6.10 (s, 1H), 7.70–7.85 (4H).
Compound 1e Æ HCl: yield, 62%; mp, 193–195 ꢁC; 1H
NMR (CD3OD) d 3.95 (dd, 2H, J = 3.4, 0.8 Hz), 6.99
(t, 0.5H, J = 0.8 Hz), 7.39–7.54 (5.5H); 13C NMR d
40.5 (d, J = 10.6 Hz), 118.5 (d, J = 6.4 Hz), 129.6 (d,
J = 4.0 Hz), 129.9, 130.0, 132.3, 150.6 (d, J = 268.6 Hz).
Compound 1f Æ HCl: yield, 68%, mp, 156–157 ꢁC; 1H
NMR (CD3OD) d 4.14 (d, 2H, J = 2.4 Hz), 7.02 (s,
0.5H), 7.42–7.49 (5.5H); 13C NMR (CD3OD) d 36.0
(d, J = 7.6 Hz), 119.8 (d, J = 6.6 Hz), 128.2 (d,
J = 3.1 Hz), 130.0, 130.4, 133.7 (d, J = 7.0 Hz), 151.5
(d, J = 267.0 Hz).
4.4. Preparation of (E)- and (Z)-3-bromo-2-methylallyl-
amine hydrochloride salts (1h and 1i)
To a solution of the 17a/17b mixture (5 g, 18 mmol) in
EtOH (70 mL) was added 1.1 equiv of 98% hydrazine
under argon. The mixture was heated at reflux for 3 h
at which time the clear solution became almost com-
pletely solidified and then was cooled to room tempera-
ture. The phthalhydrazide was filtered off and washed
with EtOH, and the filtrate was concentrated. The resi-
due was subjected to column chromatography (eluant
EtOAc–MeOH containing 1% NH4OH) to remove
NH2NH2. The pooled mixture of 1h and 1i was concen-
trated, and the residue was added to a solution of 2 mL
of 12 N HCl in MeOH (15 mL) with stirring at room
temperature for 1 h. The mixture was concentrated
and the last traces of water were removed by azeotropic
distillation with toluene. Residual hydrazine hydrochlo-
ride was eliminated by recrystallizing the mixture two
times from EtOH and once in water, with discarding
of the first small amount of solid that crystallized in each
case. The remaining liquid was concentrated to give an
isomeric mixture of HCl salts, which was recrystallized
1
Compound 1g Æ HCl: yield, 42%, mp, 149–150 ꢁC; H
NMR (CD3OD) d 4.02 (t, 2H, J = 1.9 Hz), 7.43–7.45
(5H).
4.2.6. Preparation of 2-methyl-3-phthalimido-1-propene
(16). To a solution of 3-chloro-2-methyl-1-propene 15
(2 g, 22 mmol) in dry DMF (50 mL) was added potassi-
um phthalimide (3.7 g, 20 mmol). After heating at reflux
for 12 h under nitrogen, the reaction mixture was cooled
to room temperature and diluted with ice water, at
which point the desired product precipitated out. The
solid was filtered, washed several times with water and
methanol, and recrystallized from hexane and CHCl3

1452
J. Kim et al. / Bioorg. Med. Chem. 14 (2006) 1444–1453
from acetone–chloroform. The first crop forming within
2 h gave 100% pure (E)-isomer 1h (1.7 g, 49% yield). The
residual mother liquid was concentrated under reduced
pressure, and the residue was recrystallized from EtOH
and diethyl ether. After several times recrystallization,
the (Z)-isomer 1i (0.3 g, 8.6%) was obtained, but was
still contaminated with 10% of the (E)-isomer.
4.4.2. Preparation of (E)- and (Z)-1,3-dichloro-1-phenyl-1-
propenes (21a and 21b). To a solution of phenylacetylene
(2.10 g, 20 mmol) and methyl chloromethyl ether (2.4 g,
30 mmol) in CH2Cl2 (20 mL) was added ZnCl2 (0.68 g,
5 mmol) at room temperature and the resulting suspen-
sion was stirred for 30 min. When TLC showed the disap-
pearance of starting material, the reaction mixture was
poured into ice-water (300 mL) with stirring and extract-
ed with CH2Cl2. The organic layer was collected, washed
with 5% aqueous sodium carbonate, dried (Na2SO4), and
evaporated, and the residue was subjected to fractional
column chromatography (eluanthexanes) to give the indi-
vidual isomeric chlorides. 21a (E-isomer): yield 11%; 1H
NMR (CDCl3) d 4.05 (d, 2H, J = 8.5 Hz), 6.20 (t, 1H,
J = 8.5 Hz), 7.35–7.50 (m, 5H); 13C NMR (CDCl3) d
41.3 (+), 124.8 (À), 128.5 (À), 128.6 (À), 129.6 (À),
135.7 (+), 137.2 (+). 21b (Z-isomer): yield 12%; 1H
NMR (CDCl3) d 4.42 (d, 2H, J = 7.5 Hz), 6.33 (t, 1H,
J = 8.5 Hz), 7.35–7.45 (m, 3H), 7.63 (m, 2H); 13C NMR
(CDCl3) d 40.8 (+), 122.5 (À), 126.7 (À), 128.5 (À),
129.5 (À), 136.9 (+), 137.3 (+).
1
Compound 1h Æ HCl: H NMR (D2O) d 1.82 (s, 3H),
3.59 (s, 2H), 6.45 (s, 1H); 13C NMR (D2O) d 19.9,
47.6, 112.1, 136.8; HRMS (FAB) calcd for C4H8BrN:
(MH+) m/z 149.9918. Found: 149.9912.
Compound 1i Æ HCl: 1H NMR (D2O) d 1.90 (s, 3H), 3.81
(s, 2H), 6.41 (s, 1H); 13C NMR (D2O) d 19.9, 47.5,
111.9, 136.7.
4.4.1. Preparation of (Z)-2-chloro-3-phenylallylamine
(1l). To a solution of a-chlorocinnamaldehyde 18
(4.2 g, 25 mmol) in CH2Cl2 (30 mL) under argon was
introduced DIBAL-H (1.5 M in toluene, 21 mL,
31.5 mmol) at À78 ꢁC. The resulting mixture was stirred
for 1 h and then warmed to room temperature and
quenched by slow addition of water. The precipitate
was filtered off and washed with CH2Cl2. The filtrate
was extracted with CH2Cl2 and the combined organic
phase was dried (Na2SO4) and evaporated to afford
the known (Z)-2-chlorocinnamyl alcohol 1913 as a color-
less oil (4.05 g, 96%), which was sufficiently pure for the
next step: 1H NMR (CDCl3) d 4.34 (s, 2H), 6.80 (s, 1H),
7.31–7.42 (3H), 7.64 (d, 2H, J = 7.5 Hz); 13C NMR
(CDCl3) d 67.8, 124.9, 128.1, 128.3, 129.2, 132.5, 134.2.
4.4.3. Preparation of (E)- and (Z)-3-chloro-3-phenylallyl-
amines (1j and 1k). Compound 21a or21b (0.19 g, 1 mmol)
was dissolved in methanol (5 mL) and added to NH4OH
(100 mL). The resulting mixture was stirred at room tem-
perature overnight, and TLC showed that the reaction
was complete. After removal of water and methanol in
vacuo, the resulting white solid was dissolved in aqueous
Na2CO3, and the solution was evaporated to dryness to
give a white solid, which was suspended in EtOAc. The
NaCl was filtered off, and the filtrate was acidified by addi-
tion of ethanolic HCl to give white precipitate, which was
collected and washed with ethyl acetate/isopropanol (80/
20) to give pure products.
A solution of DEAD (4.2 g, 24 mmol) in THF (25 mL)
was added to a mixture of 19 (4.0 g, 23.8 mmol), phthal-
imide (3.6 g, 24 mmol), and Ph3P (6.3 g, 24 mmol) in
THF (25 mL) at room temperature. The resulting sus-
pension was stirred for 12 h and then poured into ice
water (300 mL) under stirring. The white solid that pre-
cipitated out was collected by filtration, washed with
water, and dried at 80 ꢁC to give pure N-(2-chlorocinn-
amyl)phthalimide 20 (6.52 g, 91%): mp 87–88 ꢁC; 1H
NMR (CDCl3) d 4.63 (d, 2H, J = 0.8 Hz), 6.82 (s, 1H),
7.25–7.40 (3H), 7.63 (m, 2H), 7.75 (m, 2H), 7.90 (m,
2H); 13C NMR (APT, CDCl3) d 45.6 (+), 123.6 (À),
127.3 (+), 127.8 (À), 128.2 (À), 128.3 (À), 129.3 (À),
131.9 (+), 133.9 (+), 134.3 (À), 167.7 (+).
Compound 1j Æ HCl: yield 87%; mp 192–193 ꢁC; 1H NMR
(CD3OD) d 3.62 (d, 2H, J = 7.5 Hz), 6.11 (t, 1H,
J = 7.5 Hz), 7.40–7.50 (5H); 13C NMR (APT, CD3OD)
d 39.4 (+), 122.1 (À), 129.7 (À), 129.9 (À), 131.0 (À),
136.9 (+), 139.4 (+); HRMS (FAB) calcd for C9H11ClN:
(MH+) m/z 168.0580. Found: 168.0575.
Compound 1k Æ HCl: yield 83%; mp 187–188 ꢁC; 1H
NMR (CD3OD) d 3.93 (d, 2H, J = 7.0 Hz), 6.37 (t, 1H,
J = 7.0 Hz), 7.41–7.44 (3H), 7.68 (m, 2H); 13C NMR
(APT, CD3OD) d 39.4 (+), 119.5 (À), 127.7 (À), 129.8
(À), 130.9 (À), 130.9 (À, 137.9 (+), 139.8 (+); HRMS
(FAB) calcd for C9H11ClN: (MH+) m/z 168.0580. Found:
168.0577.
Heating a mixture of 20 (3.0 g, 10 mmol) and hydrazine
(85%, 0.75 g, 15 mmol) in ethanol (75 mL) for 1 h, addi-
tion of a second aliquot of 15 mmol hydrazine, and con-
tinued heating for 1 h, TLC showed that deprotection
was complete. The mixture was cooled to room temper-
ature, the solidified phthalhydrazide was removed, the
filtrate was concentrated, and the residual oil was dis-
solved in ether. Ethanolic HCl was added, and the white
precipitate (1.4 g, 68%) was collected and purified by
recrystallization from EtOH.
4.5. Time-dependent inactivation of BPAO by candidate
inhibitors
A 0.9 mL aliquot of a solution of candidate inhibitor in
100 mM potassium phosphate buffer, pH 7.2, was mixed
with BPAO (0.1 mL, about 8 lM) and incubated at
30 ꢁC aerobically. Aliquots (0.1 mL) were periodically
withdrawn using disposable calibrated Drummond mic-
ropipettes and diluted with 1.0 mL of benzylamine
(5 mM in 50 mM sodium phosphate buffer, pH 7.2) in a
1 cm quartz cuvette (1.5 mL volume). The rate of oxida-
tion of benzylamine to benzylaldehyde was measured by
recording the increase in absorbance at 250 nm for
Compound 1l Æ HCl: mp 237–238 ꢁC; 1H NMR
(CD3OD) d 4.01 (s, 2H), 7.11 (s, 1H), 7.35–7.45 (3H),
7.70 (m, 2H); 13C NMR (CD3OD) d 48.6, 125.6,
129.5, 130.1, 130.5, 132.5, 134.8; HRMS (FAB) calcd
for C9H11ClN: (MH+) m/z 168.0580. Found: 168.0579.

J. Kim et al. / Bioorg. Med. Chem. 14 (2006) 1444–1453
1453
1 min and compared to the rate of benzylamine oxidation
in a companion control solution of enzyme without inhib-
itor. The concentration of active BPAO was estimated
from the rate of benzylamine oxidation as described
previously.20
lacking plateau behavior and partition ratio plots
for inhibitors exhibiting plateau behavior. Supple-
mentary data associated with this article can be
found, in the online version, at doi:10.1016/
j.bmc.2005.09.065.
4.6. Irreversibility of BPAO inhibition
References and notes
(a) By gel-filtration. The irreversibility of the inhibition
was checked by applying enzyme preparations
(0.5 mL) which were 90–95% inhibited to a PDX G. F.
25 column (1 · 7.8 cm), equilibrated with 100 mM sodi-
um phosphate buffer (pH 7.2), to separate non-covalent-
ly bound small molecules. Control runs established that
no enzyme elutes in the first 33 drops, and that in the
subsequent 10 aliquots of 3 drops each, all enzyme elutes
in aliquots 3–7. Percent activity following gel-filtration
was determined comparing the activity between two
gel-filtrated incubations of enzyme in the absence and
presence of inhibitor. The same initial protocol had to
be worked out each time the column was changed. (b)
By dialysis. The irreversibility of the inhibition was also
checked by adding 0.5 mL enzyme preparations which
were 90–95% inhibited to dialysis tubing (6.4 mm Spec-
trum Spectro/Pro membrance, MW cutoff 12,000–
14,000), following by dialysis against 100 mM sodium
phosphate buffer, pH 7.2, at room temperature for peri-
ods up to 24 h. Percent activity was determined by com-
paring the enzyme activity between two dialysis
experiments, following incubation of enzyme in the ab-
sence and presence of inhibitor.
1. Knowles, P. F.; Dooley, D. M. Met. Ions Biol. Syst. 1994,
30, 361.
2. McIntire, W. FASEB J. 1994, 8, 513.
3. Klinman, J. P. Biochim. Biophys. Acta 2003, 1647, 131.
4. Brown, D. E.; McGuirl, M. A.; Dooley, D. M.; Janes, S.
M.; Mu, D.; Klinman, J. P. J. Biol. Chem. 1991, 266,
4049.
5. Janes, S. M.; Mu, D.; Wemmer, D.; Smith, A. J.; Kaur, S.;
Maltby, D.; Burlingame, A. L.; Klinman, J. P. Science
1990, 248, 981.
6. Mure, M.; Millis, S. A.; Klinman, J. P. Biochemistry 2002,
41, 9269.
7. Bono, P.; Salmi, M.; Smith, D. J.; Jalkanen, S. J. Immunol.
1998, 160, 5563.
8. Salmi, M.; Jalkanen, S. Trends Immunol. 2001, 22,
211.
9. (a) Hevey, R. C.; Babson, J.; Maycock, A. L.; Abeles,
R. H. J. Am. Chem. Soc. 1973, 95, 6125; (b) Abeles,
R. H.; Maycock, A. L. Acc. Chem. Res. 1976, 9,
313.
10. Mcdonald, I. A.; Lacoste, J. M.; Bey, P.; Palfreyman, M.
G.; Zreika, M. J. Med. Chem. 1985, 28, 186.
11. (a) Lyles, G. A.; Marshall, C. M. S.; McDonald, I. A.;
Bey, P.; Palfreyman, M. G. Biochem. Pharmacol. 1987, 36,
2847; (b) Palfreyman, M. G.; McDonald, I. A.; Bey, P.;
Dazin, C.; Zreika, M.; Cremer, G. J. Neural Transm.
Suppl. 1994, 41, 407.
4.7. Phenylhydrazine titration of active site after inacti-
vation of BPAO with inactivators
12. Jeon, H. B.; Sayre, L. M. Biochem. Biophys. Res. Commun.
2003, 304, 788.
Incubation of the enzyme with the test inactivator in
100 mM sodium phosphate buffer, pH 7.2, was allowed
to proceed at 30 ꢁC until there was more than 80% loss
of activity after gel-filtration or dialysis. A control was
run at the same time. To the gel filtered or dialyzedenzyme
fractions was added 4 lL of phenylhydrazine hydrochlo-
ride (2 mM) in water and the absorbance at 440 nm was
measured. The percentage decrease in absorbance at
440 nm was determined compared to the control
experiment.
13. Barma, D. K.; Baati, R.; Valleix, A.; Mioskowski, C.;
Falck, J. R. Org. Lett. 2001, 3, 4237.
14. Marcuzzi, F.; Melloni, G.; Zucca, M. V. Gazz. Chim. Ital.
1987, 117, 219.
15. Silverman, R. B. Mechanism-Based Enzyme Inactivation:
Chemistry and Enzymology; CRC Press: Boca Raton,
Florida, 1988; Vol. I.
16. Janes, S. M.; Klinman, J. P. Biochemistry 1991, 30,
4599.
17. Fluckiger, R.; Woodtli, T.; Gallop, P. M. Biochem.
Biophys. Res. Commun. 1988, 153, 353.
18. Cinnamylamine has also been reported to both be a
substrate and time-dependent inhibitor of rat liver
MAO-B: Williams, C. H.; Lawson, J.; Backwell, F. R.
Biochem. J. 1988, 256, 911; Williams, C. H.; Lawson,
J.; Backwell, F. R. Biochim. Biophys. Acta 1992, 1119,
111.
Acknowledgment
This work was supported by National Institutes of
Health Grant GM 48812.
19. OꢂConnell, K. M.; Langley, D. B.; Shepard, E. M.; Duff,
A. P.; Jeon, H.-B.; Sun, G.; Freeman, H. C.; Guss, J. M.;
Sayre, L. M.; Dooley, D. M. Biochemistry 2004, 43,
10965.
Supplementary material
Plots of time-dependent inhibition of BPAO by 1a–
e,g,i,k,l, including Kitz–Wilson replots for inhibitors
20. Jeon, H.-B.; Sun, G.; Sayre, L. M. Biochim. Biophys. Acta
2003, 1647, 343.