Inhibition of bovine plasma amine oxidase by 1,4-diamino-2-butenes and -2-butynes

Article abstract of DOI:10.1016/S0968-0896(03)00521-2

Bovine plasma amine oxidase (BPAO) was previously shown to be irreversibly inhibited by propargylamine and 2-chloroallylamine. 1,4-Diamine versions of these two compounds are here shown to be highly potent inactivators, with IC₅₀ values near 20 μM. Mono-N-alkylation or N,N-dialkylation greatly lowered the inactivation potency in every case, whereas the mono-N-acyl derivatives were also weaker inhibitors and enzyme activity was recoverable. The finding that the bis-primary amines 1,4-diamino-2-butyne (a known potent inhibitor of diamine oxidases) and Z-2-chloro-1,4-diamino-2-butene are potent inactivators of BPAO is suggestive of unexpected similarities between plasma amine oxidase and the diamine oxidases and implies that it may be unwise to attempt to develop selective inhibitors of diamine oxidase using a diamine construct.

Full text of DOI:10.1016/S0968-0896(03)00521-2

Bioorganic & Medicinal Chemistry 11 (2003) 4631–4641
Inhibition of Bovine Plasma Amine Oxidase by
1,4-Diamino-2-butenes and -2-butynes
Heung-Bae Jeon, Younghee Lee, Chunhua Qiao, He Huang and Lawrence M. Sayre*
Department of Chemistry, Case Western Reserve University, Cleveland, OH 44106, USA
Received 8 March 2003; accepted 21 July 2003
Abstract—Bovine plasma amine oxidase (BPAO) was previously shown to be irreversibly inhibited by propargylamine and 2-
chloroallylamine. 1,4-Diamine versions of these two compounds are here shown to be highly potent inactivators, with IC50 values
near 20 mM. Mono-N-alkylation or N,N-dialkylation greatly lowered the inactivation potency in every case, whereas the mono-N-
acyl derivatives were also weaker inhibitors and enzyme activity was recoverable. The finding that the bis-primary amines 1,4-dia-
mino-2-butyne (a known potent inhibitor of diamine oxidases) and Z-2-chloro-1,4-diamino-2-butene are potent inactivators of
BPAO is suggestive of unexpected similarities between plasma amine oxidase and the diamine oxidases and implies that it may be
unwise to attempt to develop selective inhibitors of diamine oxidase using a diamine construct.
# 2003Elsevier Ltd. All rights reserved.
Introduction
none at the expense of reduction of O2 to H2O2, with
release of ammonia. Across the plant and animal king-
doms, there are several sub-classes of copper amine
oxidases with different tissue distribution and substrate
preferences. The mammalian plasma amine oxidases,
the human form of which is often known as the soluble
semicarbazide-sensitive amine oxidase (SSAO),5 prefer
to metabolize arylalkylamines including benzylamine,
the phenethylamine neurotransmitters, and exogenous
amine drugs. On the other hand, a family of plant
enzymes and mammalian enzyme found in kidney,
intestines, and placenta, referred to as diamine oxidases
(DAOs), prefers to metabolize aliphatic 1,4- and 1,5-
diamines and the neurotransmitter histamine. It is
believed that these latter enzymes contain a cation
binding site located at an appropriate distance (ꢀ10 A)
from the reactive cofactor carbonyl,6 such that electro-
The copper-containing amine oxidases convert unbran-
ched primary amines to the corresponding aldehydes by
use of an active-site quinone-mediated transamination
reaction. Based on the fairly recent elucidation of the
structures of the quinone cofactors, there has been
renewed interest in the mechanistic aspects of oxidative
deamination and in the development of useful inhibi-
tors. Most of the enzymes in this class utilize a 2,4,6-
trihydroxyphenylalanine quinone (TPQ) cofactor
derived from an active site tyrosine.1,2 A different
cofactor, lysine tyrosylquinone (LTQ), that represents
an active site lysine–tyrosine cross-link,3 is used by a
sub-class of these enzymes, lysyl oxidases, responsible
for oxidative deamination of the lysine side chains of
the connective tissue proteins collagen and elastin.
The mechanism of the deamination half-reaction4
involves formation of a Schiff base between quinone
cofactor and substrate amine, isomerization of the
‘substrate Schiff base’ to a ‘product Schiff base’ with
reduction of the quinone (the key transamination step),
and hydrolysis of the product Schiff base (see Scheme 1
for abbreviated mechanism using TPQ). Subsequently,
the reduced cofactor is converted to the starting qui-
*Corresponding author. Tel.: +1-216-368-3704; fax: +1-216-368-
3006; e-mail: lms3@cwru.edu; URL: http://www.cwru.edu/artsci/
chem/faculty/sayre/index.html
Scheme 1.
0968-0896/$ - see front matter # 2003Elsevier Ltd. All rights reserved.
doi:10.1016/S0968-0896(03)00521-2

4632
H.-B. Jeon et al. / Bioorg. Med. Chem. 11 (2003) 4631–4641
static binding (and possibly hydrogen-bonding) of one
protonated terminus of the diamine (or imidazolium
ring in histamine) positions the other primary amine
terminus properly for Schiff base formation with the
cofactor.
cases suggests the absence of an active-site nucleophile
suitably positioned to be irreversibly alkylated by the
immediate aminoenal turnover products.
The preliminary reports that 48,11 and 511 are mechan-
ism-based inactivators of BPAO raises the question as
to whether this enzyme manifests some selectivity for
diamines over monoamines, at least in terms of
substrate action leading to inactivation. Thus, to clarify
if diamine-based inactivators could be relied upon to
exert selective inactivation of DAO, we investigated the
structural dependence of diamine-based inactivation on
BPAO by varying the nature of the second amino group
in compounds 4–7.
Based on work mainly directed at inactivating pyridoxal
phosphate (PLP)-dependent transamination enzymes,
Abeles and co-workers reported nearly 30 years ago
that amines bearing an unsaturated C–C bond at the b-
position could inactivate bovine plasma amine oxidase
(BPAO).7,8 Propargylamine (1) and 2-chloroallylamine
(2) were found to be inactivators, though allylamine (3)
was found to be only a substrate with no inactivation
tendency. Also, the propargylic diamine analogue 1,4-
diamino-2-butyne (4) was reported to exhibit 50%
8
inactivation of BPAO at 2.5 mM in 0.23min despite
Results
little knowledge that diamines are substrates for this
enzyme. However, few details were revealed by these
early studies. More recently, Frebort and co-workers
reported that 4 is a potent mechanism-based inactivator
Design and synthesis
The chosen structural modifications were to convert the
second primary amine group to a secondary or tertiary
amino group or to acylate it. Alkylation would prevent
metabolism at that end (N-methyl secondary amines are
not substrates for the copper amine oxidases) but not
affect the charge properties. On the other hand, acyla-
tion would abolish metabolism at that end as well as
prevent protonation. In the case of the unsymmetrical
compound 5, there would be two different regioisomers
possible for each type of modification. According to
these guidelines, the compounds in addition to 4–7 that
were prepared here for study are given in Chart 1.
of pea DAO9 and Aspergillus niger amine oxidase.10
A
preliminary report from our lab claimed inactivation of
BPAO by 4 as well as by a diamine version 5 of 2-
chloroallyl amine.11
trans- (6) and cis-1,4-Diamino-2-butene (7) were pre-
pared as previously reported.13 Though 4 has been
made by the Gabriel synthesis from 1,4-dichloro-2-
butyne, we found it more convenient to convert the
dichloride directly to 4 with NH4OH.15 The secondary
and tertiary amine analogues of 4 were synthesized
according to the routes shown in Scheme 3. Mono-
substitution of 1,4-dichloro-2-butyne with potassium
phthalimide to give 1-chloro-4-phthalimido-2-butyne
(19)16 took advantage of the use of excess dichloride to
prevent bis-derivatization (unreacted dichloride was
readily separated). When the primary amines methyla-
mine, benzylamine, and phenethylamine were added in
excess to 19, the secondary amine analogues 8–10 were
afforded directly by a one-pot reaction, with removal of
the phthaloyl protecting group occurring under the
basic reaction conditions. However, when the secondary
An early study reported that 1,4-diamino-2-butene
(without specification of geometrical isomerism), was
converted to pyrrole by pig kidney diamine oxidase
(PKDAO).12 As a follow-up to this report, we evaluated
the individual trans- (6) and cis-1,4-diamino-2-butene
(7) isomers with PKDAO.13 As shown in Scheme 2,
both isomers were found to be substrates and irrever-
sible mechanism-based inactivators of DAO. The trans
isomer 6 was the more potent inactivator, exerting
almost complete inhibition at 1 mM in 1 h, whereas the
cis isomer 7 exhibited only 54% inhibition under the
same conditions, owing to competitive ring closure of
the initial cis-aminoenal turnover product. The inde-
pendently synthesized trans-aminoenal was found to be
a potent enzyme inactivator.13 In addition it was shown
that pyrrole formation starting with the trans isomer 6
arises from addition–elimination of substrate or product
amine to the trans-aminoenal turnover product, con-
verting it to the cis-aminoenal.13 Interestingly, both 6
and 7 appear to be pure substrates for the grass pea
diamine oxidase.14 The lack of inactivation in these
Chart 1. The compounds prepared for evaluation of inhibition of
bovine plasma amine oxidase.
Scheme 2.

H.-B. Jeon et al. / Bioorg. Med. Chem. 11 (2003) 4631–4641
4633
amines dimethylamine and bis[2-(2-pyridyl)ethyl]amine
were used, the intermediates 20 and 21 were first
obtained, requiring deprotection with hydrazine to give
desired tertiary amines 11 and 12 (Scheme 3). In later
studies, it was found that superior yields of the N-
methyl and N,N-dimethyl analogues 8 and 11 could be
obtained from the reaction of Boc-protected 4-chloro-2-
butynamine17 with methanolic methylamine or dime-
thylamine, respectively, followed by deprotection
(Scheme 3). Monoacylation of 4 with di-tert-butyl
dicarbonate at pH 10 gave N-Boc-1,4-diamino-2-butyne
(13). Acetylation of the latter compound (Ac2O, Et3N)
followed by acid hydrolysis gave N-acetyl-1,4-diamino-
2-butyne (14).
directly, we added di-tert-butyl dicarbonate to the mix-
ture to give Boc protected compounds 24 and 25, which
could be separated by flash chromatography and
deprotected by acid hydrolysis to give the individual
isomers 17 and 18 (Scheme 5).
Structural distinction between 17 and 18 was aided by
comparison of their 1H and 13C NMR spectra to that of
1
5. As shown in Chart 2, the H NMR (3.89 ppm) and
13C NMR (37.2 ppm) signals of C4 in 5 match those of
1
C1 in 17 but with no position in 18. Also, the H NMR
(4.01 ppm) and 13C NMR (45.4 ppm) signals of C1 in 5
are consistent with those of C1 in 18 but with no posi-
tion of 17. The stereochemistry in 17 and 18 was
assigned as Z, as in the case of 5, based on the knowl-
edge that HCl addition to alkynes in HOAc occurs anti
rather than syn.19 In the case of 17, we confirmed the
stereochemistry through an NOE difference experiment.
Enhancements of greater than 0.3% that were observed
for irradiation of the two methylene and vinyl positions
are shown in Chart 3. The enhancements observed
between the two methylene positions were only 0.1 and
0.2%, as opposed to large values that would be expected
in the case of the E isomer.
trans-N,N-Dimethyl-1,4-diamino-2-butene (15) was
prepared starting with monosubstitution of trans-1,4-
dichloro-2-butene
with
hexamethylenetetramine
(HMTA) on one side, followed by substitution with
dimethylamine on the other side, and finally acid
hydrolysis as shown in Scheme 4. trans-N-Acetyl-1,4-
diamino-2-butene (16) was prepared from trans-1,4-
dichloro-2-butene by its reaction with potassium
phthalimide to give 22, which was in turn reacted with
HMTA followed by acid hydrolysis to give 23, followed
by acetylation and finally hydrazinolysis of the phthali-
mide protecting group (Scheme 4).
Enzyme studies
All compounds were evaluated for their ability to inac-
tivate the benzylamine oxidase activity of commercially
Z-2-Chloro-1,4-diamino-2-butene (5) was prepared by
addition of HCl to 1,4-diamino-2-butyne (4) as descri-
bed.18 Z-1-Amino-3-chloro-4-(dimethylamino)-2-butene
(17) and Z-1-amino-2-chloro-4-(dimethylamino)-2-
butene (18) were prepared as a mixture by the same
procedure18 from 1-amino-4-(dimethylamino)-2-butyne
(11). Since it was very difficult to separate 17 and 18
Scheme 5.
Scheme 3.
Chart 2. Chemical shift correlation of compounds 5, 17, and 18.
Chart 3. Nuclear Overhauser effect enhancements for compound 17 in
CD3OD.
Scheme 4.

4634
H.-B. Jeon et al. / Bioorg. Med. Chem. 11 (2003) 4631–4641
available BPAO (Sigma) at 30 ꢁC, pH 7.2, determined
by assaying the rate of benzaldehyde production, and
the data are listed in Table 1. Irreversibility of inhibition
was demonstrated by comparing the activity before and
after gel filtration to remove non-covalently bound
small molecules and additionally in the cases of 8 and 11
by 24 h dialysis against 100 mM pH 7.2 phosphate buf-
fer. For most inhibitors, the time-dependent loss of
enzyme activity was incomplete and plateaued at an
activity level that monotonically decreased as the initial
concentration of inhibitor increased. This plateau beha-
vior is indicative of classical partitioning between turn-
over and inactivation and permits determination of
partition ratios.20
1,4-Diamino-2-butyne-based inactivators
Initially the inactivation of BPAO by 1,4-diamino-2-
butyne (4)8,11 was reinvestigated, since kinetic para-
meters and the partition ratio were not reported. In the
concentration range of 40–400 mM, loss of enzyme
activity followed a simple exponential as shown in Fig-
ure 1. KI and kinact values (Table 1) determined by the
method of Kitz and Wilson21 (Fig. 2) indicate that 4 is a
very efficient inactivator of BPAO (kinact/KI=1.4 mMꢂ1
minꢂ1). Although complete inactivation by 4 was
observed at 0.1 mM and higher, loss of activity at lower
concentrations reached a plateau, indicative of a parti-
tioning between inactivation and turnover. The concen-
tration of 4 needed to reach a plateau with 50% loss of
activity was 20 mM. From the plateau data (not shown),
the graphically-determined partition ratio was found to
be 20. BPAO inactivated by 4 did not recover activity
either by gel filtration or by overnight dialysis. Evidence
for modification of the active site was that treatment of
the inactivated enzyme with phenylhydrazine did not
result in the characteristic absorption at 450 nm resulting
from derivatization of the quinone cofactor.22 In addition,
samples of the enzyme that were 95–100% inactivated
by 4 exhibited only 10–20% of the nitroblue tetrazolium
(NBT) quinone redox cycling activity23 of the unmodi-
fied enzyme, following denaturation.
Table 1. Inhibitory data of compounds
Compd IC50 (mM)a Partition
ratio
Kinetic data
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
0.020 (0.012)d
0.024
2
Reversible
0.4d
1.0
20
29
KI=0.92 mM, kinact=1.3min ꢂ1
KI=0.19 mM, kinact=1.7 minꢂ1
KI=11.8 mM, kinact=0.14 minꢂ1
73% inhibition in 1 min, 2.5 mMc
60% inhibition in 26 min, 2 mM
Nob
1800d
555
330
4000d
138
e
0.5
1.0 (1.4)d
0.1
50% inhibition in 50 min, 2.5 mM
60% inhibition in 25 min, 0.15 mM
1.2e
0.35e
3.5
e
N.D.
e
N.D.
4e
2.0
53% inhibition in 30 min, 2 mM
20% inhibition at 2 mMf
ND, not determined.
aThe number listed represents the concentration needed to achieve
50% inactivation at the point of activity plateau (5–120 min) for those
inhibitors that exhibited partitioning (partition ratio listed), or the
concentration needed to achieve 50% inactivation in 30 min, for those
inhibitors that did not exhibit plateau behavior. Unless indicated
otherwise, the loss of activity is irreversible even after gel filtration
and/or 22-h dialysis.
The mono N-methyl- (Fig. 3), N-benzyl-, and N-phe-
nethyl- secondary amine analogues 8–10 of 1,4-dia-
mino-2-butyne exhibited a markedly reduced inhibitory
potency compared to 4. Nonetheless, all three exhibited
a concentration- and time-dependent irreversible inacti-
vation characterized by plateau behavior indicative of
partitioning (plateau-determined IC50 values and parti-
tion ratio data is given in Table 1). The N,N-dimethyl
derivative 11 exhibited an even weaker inhibitory
potency than the mono N-methyl derivative 8 but was
bNo evidence of plateau behavior was seen for the concentration range
used.
cThe activity of enzyme inhibited by 7 recovered to 75% over 9 min.
dThis measurement used BPAO obtained from Worthington.
eThe enzyme inhibited by these compounds lost activity within 30 min
and then began to recover. The activity recovered to ꢀ90% in ꢃ12 h.
fThe inhibition was not time-dependent.
Figure 1. Time dependent inactivation of BPAO by various con-
centrations of 1,4-diamino-2-butyne (4).
Figure 2. Kitz and Wilson plot for inactivation of BPAO by 4.

H.-B. Jeon et al. / Bioorg. Med. Chem. 11 (2003) 4631–4641
4635
again an irreversible inactivator. For both 8 and 11, fur-
ther loss of enzyme activity after about 2 h was very slow
but still apparent (see Fig. 3 for 8). The lack of a clear
plateau at the point of consumption of inhibitor has been
seen before, and was interpreted to reflect a slow inhibi-
tion by the accumulated aldehyde turnover product.24
The listed plateau IC50 values and partition ratios (Table
1) were determined in these two cases from the psuedo-
plateau points of the observed biphasic behavior.
Whereas BPAO inactivated by 8 and 11 failed to react
with phenylhydrazine, preparations of BPAO inacti-
vated by 8 (95%) or 11 (94%) still gave 40–50% of the
NBT redox cycling response of the unmodified enzyme.
Interestingly, another tertiary amine analogue, N,N-
[bis(2-pyridyl)ethyl]-1,4-diamino-2-butyne (12) exhib-
ited a reasonably potent irreversible inactivation (pla-
teau IC50 value of 0.1 mM). The relatively slow onset of
inhibition even though [I]ꢄ[E] suggests that this com-
pound is not inhibiting by chelating the Cu(II) out of
the enzyme. The mono-acetyl and mono-Boc derivatives
14 and 13 exhibited an 18- and 60-fold drop, respec-
tively, in BPAO inhibitory potency relative to 4 (Table
1), and in these cases, the partially inhibited enzyme
recovered slowly over time (see Fig. 4 for 14). The lack
of permanent inhibition in these cases may reflect an
instability (e.g., toward hydrolysis) of the covalent
adduct generated.
The huge drop in inactivation potency seen upon addi-
tion of even a single N-methyl group to diamine 4 is
remarkable, and the large partition ratio seen for this
compound (8) suggests that it is still being efficiently pro-
cessed by BPAO, but mainly as a substrate rather than
inactivator. To confirm the substrate activity of both 8
and 11, O2 uptake studies were performed. Dioxygen
consumption was linear with concentration at low con-
centrations of inhibitor, but attempts to observe satura-
tion behavior at higher concentration (to permit
calculation of Vmax and Km) were thwarted by enzyme
inactivation occurring within the (short) time frame of
determining the O2 consumption rate. In fact, the mea-
sured O2 consumption actually began to decrease with
increasing concentration or 8 and 11 above about 2–3
mM. Thus, for comparative purposes, the second-order
rate constant was computed from the linear portion of the
plot of O2 consumption rate versus concentration at low
concentration. Under the conditions of the assay (pH 7.2,
30 ꢁC), the k values for 8, 11, and benzylamine were 8.6,
9.3, and 22 mMꢂ1 minꢂ1, respectively, indicating that 8
and 11 are excellent substrates for BPAO, though not as
good as benzylamine. Although a determination of sub-
strate behavior for the parent diamine 4 was precluded by
the potent inactivation that occurs even at relatively low
concentrations, one can compare the above numbers to
the calculated pseudo-second-order processing of 4 as an
inactivator (kinact/KI=1.4 mMꢂ1 minꢂ1).
Allylamine-based inhibitors
Although it has been reported that allylamine is a sub-
strate rather than inactivator of BPAO,7 the fact that
the allylic diamines 6 and 7 inactivate PKDAO13 sug-
gested that they may also inactivate BPAO. trans-1,4-
Diamino-2-butene (6) was found to be a time- and con-
centration-dependent but relatively weak irreversible
inactivator of BPAO, inducing a 50% enzyme activity
loss in 30 min at 2 mM. The graphically determined KI
and kinact values (Table 1) indicate an inactivation effi-
ciency (kinact/KI) for 6 that is 119 times lower than its
acetylenic analogue 4. The N,N-dimethyl derivative 15
of 6 was an even less potent inactivator, but was still
irreversible (Table 1). On the other hand, whereas trans-
N-acetyl-1,4-diamino-2-butene (16) also exhibited
weaker inhibition than 6 (Table 1), enzyme activity
recovered slowly in this case, as was observed for the
mono-acyl derivatives 13 and 14 of 1,4-diamino-2-
butyne. On account of the weak inhibitory properties of
15 and 16, we did not pursue the determination of inhi-
bitory kinetic parameters.
cis-1,4-Diamino-2-butene (7) exerted 73% inhibition at
1 min at 2.5 mM, with no clear time-dependent activity
loss, but the enzyme activity recovered very rapidly
(Fig. 4). The transient inhibition of benzylamine
Figure 4. Effect of incubation time on inactivation of BPAO by cis-
1,4-diamino-2-butene (7, &) and N-acetyl-1,4-diamino-2-butyne (14,
&).
Figure 3. Time dependent inactivation of BPAO by various con-
centrations of 8.

4636
H.-B. Jeon et al. / Bioorg. Med. Chem. 11 (2003) 4631–4641
Discussion
1,4-Diamino-2-butyne (4) has been confirmed to be a
potent mechanism-based inactivator of BPAO, as it is of
diamine oxidases. In fact, although inactivation of plant
DAO by 4 is not so pronounced so as to preclude
measurement of its substrate kinetic parameters,26
4
appears to be a more pure inactivator of BPAO. Its
increased potency relative to propargylamine (1) as well
as the ability of trans-1,4-diamino-2-butene (6) but not
allylamine (3) to inactivate BPAO, suggests that the
second amino group in the diamine inhibitors provides
for some type of special recognition, either in terms of
binding or in terms of the inactivation chemistry. Due
to structural similarity of diamine oxidases with the
other copper amine oxidases, the improved inactivating
potency of 4 and 6 relative to 1 and 3 for BPAO may
reflect a common recognition feature for both enzymes,
even though diamines are not known to be especially
good substrates of BPAO. This latter point probably
deserves careful scrutiny.
Figure 5. Effect of incubation time on inactivation of BPAO by var-
ious concentrations of 5.
oxidation by 7 is most likely simply the result of sub-
strate activity of 7 (it must bind more tightly than
benzylamine) until it is consumed. The rapid formation
of pyrrole from the cis-aminoenal turnover product
(Scheme 2) would protect the enzyme from being alky-
lated, as was deduced for DAO,12,13 thus rationalizing
the lack of irreversible enzyme inactivation in this case.
If the improved inhibitory efficacy of 4 and 6 is due to a
facilitation of binding by the second positive charge,
then one would expect that at least small N-alkyl and
N,N-dialkyl secondary and tertiary amine derivatives
would preserve inactivating potential, whereas the
mono N-acyl derivatives might not. The N-acyl deriva-
tives 13 and 14 of 4 were found to be weak and at least
partially reversible inhibitors of BPAO, consistent with
the requirement of at least a second basic center on the
installation of a permanent covalent mode of inactiva-
tion. However, even mono-N-methylation of 4 resulted
in a marked weakening of inhibitory potency. The find-
ing that inactivation of BPAO by 4 but not by 8 results
in loss of redox cycling integrity of the quinone cofac-
tor, suggests that 4 induces a very special type of cova-
lent inactivation that incorporates a modified quinone
cofactor alone or together with an active site residue.
The distinctive inactivation by 4 must reflect a mechan-
ism either that requires both primary amino termini or
that is exquisitely sensitive to even the subtle steric and/
or pKa change accompanying introduction of a single
methyl group.
Chloroallylamine-based inactivators
Z-2-Chloro-1,4-diamino-2-butene (5) was found to be a
potent time- and concentration-dependent irreversible
inactivator of BPAO (Fig. 5), consistent with pre-
liminary reports.11,25 Strict pseudo-first-order behavior
permitted Kitz and Wilson determination of kinetic
parameters (Table 1) that indicate this compound is the
most efficient (kinact/KI) inactivator of BPAO studied
here, about 6 times more so than 1,4-diamino-2-butyne
(4). Nonetheless, at low concentration, the activity ver-
sus time plots reached plateaus, indicative of partition-
ing, permitting graphical determination of a partition
ratio of 29. It is noteworthy that the behavior of 5 is so
similar to that of 4, suggesting that these two closely
related compounds, differing by the elements of HCl,
may inactivate BPAO via a common intermediate.
Consistent with this possibility, a preparation of BPAO
that was 99% inhibited by 5 exhibited only 5–10% of
the NBT redox cycling capacity of the control prepara-
tion, similar to what was observed for 4.
Because plasma amine oxidase is usually considered to
have a high affinity for arylalkylamines, we investigated
whether the addition of a benzyl (9) or phenethyl group
(10) to one amino terminus of 1,4-diamino-2-butyne
might improve inactivation efficiency for BPAO. Both
compounds had reduced inactivating potency relative to
4, and they were not improved inhibitors relative to the
N-methyl compound 8.
The two N,N-dimethyl regioisomers of 5 were designed
to evaluate how the positional dependence of the chlor-
ine atom with respect to the primary amine moiety
influences inactivation potency. Z-1-Amino-3-chloro-4-
(dimethylamino)-2-butene (17), which contains a 3-
Frebort and coworkers26 recently proposed a mechan-
ism for the inactivation of plant amine oxidases by 4
that involves its metabolism to 4-amino-2-butynal (or
its tautomer 4-amino-1,2-butadienal), which is a potent
Michael acceptor and alkylates the enzyme to give a
residue-bound 4-amino-2-butenal moiety that cyclizes
to a pyrrole (Scheme 6, R=H). Pyrrole formation
would not be possible with either the N,N-dimethyl or
N-acyl analogues, though the mono-N-alkyl analogues
chloroallylamine part-structure, exhibited
a
time-
dependent irreversible inactivation of BPAO, but with a
weak inhibitory potency. Curiously, the 2-chloro
regioisomer 18, that contains the 2-chloroallylamine
part-structure, did not exhibit time-dependent inactiva-
tion, and exerted only 20% inhibition at 2 mM.

H.-B. Jeon et al. / Bioorg. Med. Chem. 11 (2003) 4631–4641
4637
Scheme 6.
Scheme 7.
would potentially be able to follow this same
mechanism, giving N-alkylpyrroles (Scheme 6,
R=alkyl). Although the fates of N-methyl compound
8 and N,N-dimethyl compound 11 with the plant
enzyme have not been investigated, the data obtained
with BPAO are inconsistent with operation of a common
mechanism. First, inactivation of BPAO by 4 was
accompanied by all but 10–20% loss of the NBT redox
cycling activity, suggesting that the inactivated enzyme
in this case contains a modified cofactor, though side-
chain modification in addition cannot be excluded.
Second, both the N-methyl and N,N-dimethyl compounds
were considerably weaker inhibitors, each exhibiting
only about 50% loss of NBT signal. The implications of
partial loss of redox-cycling are unclear, possibly
reflecting a modified cofactor with reduced but still active
redox properties, but more likely reflecting a hetero-
geneous population of inactivated enzyme, with and
without modified redox-inactive cofactor. If the
mechanism of Scheme 6 were operative for BPAO, the
monomethyl inhibitor 8 should have behaved more like 4
than like 11. Further studies will be needed to elucidate
the basis of the uniquely potent inhibition of BPAO
effected by 4.
tautomerization of 26 and elimination of HCl from 27
would generate the identical allenylimine, which could
be the alkylating moiety.
Importantly, in our efforts to determine the effect of the
two possible N,N-dimethylation regiochemistries on the
chloroallyldiamine 5, the isomer that contains the 2-
chloroallylamine part-structure (18) was found to be a
weaker inactivator than the isomer (17) that contains
the 3-chloroallylamine part-structure (though both were
relatively weak). The greater potency of 17 is consistent
with the finding that the E and Z isomers of 3-chloro-
allylamine were much more potent inhibitors of BPAO
than 2-chloroallylamine,27 consistent with the addition-
elimination mechanism depicted in Scheme 7. Overall,
however, although possibly the first step, simple con-
jugate addition would not explain why, for example,
inactivation by the N-acyl inhibitors 13 and 14 is rever-
sible. The maturation of the initial alkylated enzyme
nucleophiles must vary as a function of the status of the
second amino group according to multistep mechanisms
such as shown in Scheme 6.
Conclusions
Despite trans-1,4-diamino-2-butene (6) being a rela-
tively weak inactivator relative to 4, N,N-dimethyla-
tion at one end of 6 significantly diminished its
inactivation potency, as in the case of 4, again
pointing to some special role of the second primary
The results described here extend previous reports on
the inactivation of BPAO by amines bearing b-unsa-
turation.7,8,11 It is evident that propargylic and chloro-
allylic diamines are highly potent inhibitors, more so
than the simple allylic diamines, and that the inactiva-
tors bearing two primary amine termini (4–6) are more
potent inhibitors than the corresponding derivatives
where one of the primary amino termini is converted to
a secondary or tertiary amine, or to an amide group.
Though of variable potency, all of the diamines were
irreversible inactivators, distinguishing them from the
N-acyl analogues, which displayed weak inhibition, and
in each case the partially inhibited enzyme recovered
activity over time. Though the mechanisms responsible
for enzyme inactivation have not been elucidated,
multiple mechanisms must be at play for the different
families of compounds, and cofactor modification
appears to occur alternatively or additionally to alkyl-
ation of an active site residue in the case of the
uniquely potent inhibitors 4 and 5. The finding that 4,
previously known to be a potent inhibitor of diamine
oxidases, is also a potent inactivator of BPAO, is sug-
gestive of unexpected similarities between these two
enzymes, and thus it is probably unwise to try to inac-
tivate diamine oxidase selectively with diamine-based
inhibitors.
amine group. The cis-allyldiamine analogue
7
behaved as a substrate rather than as an inactivator,
apparently because the immediate cis aminoenal pro-
duct cyclizes to pyrrole as turnover product before
alkylation occurs (Scheme 2).
Abeles had shown many years ago that 2-chloro-
allylamine has similar potency to propargylamine as
an inactivator of BPAO.7 It is thus important to note
that upon extension of these two compounds to 1,4-
diamines, Z-2-chloro-1,4-diamino-2-butene (5) and
1,4-diamino-2-butyne (4) are now seen to have very
similar and more potent inactivating potential. As
mentioned before, this parallel behavior may indicate
a common reactive intermediate mechanism of inacti-
vation since the two compounds differ only by the
elements of HCl. One possibility would be that after
processing of 4 and 5 to the product Schiff base
stages 26 and 27 (Scheme 7), addition of an active site
nucleophile to 26 and substitution of the same active
site nucleophile for Clꢂ (addition-elimination) in 27
would give the same initial product. Alternatively,

4638
H.-B. Jeon et al. / Bioorg. Med. Chem. 11 (2003) 4631–4641
Experimental
N-Methyl-1,4,-diamino-2-butyne diperchlorate (8) from
19. Methylamine gas was slowly bubbled through a
solution of 2.3g (10 mmol) of 19 in 50 mL of acetoni-
trile at room temperature for 40 h. The solvent and
excess methylamine was removed by evaporation to
afford 1.9 g (87%) of 1-phthalimido-4-(methylamino)-2-
butyne as a white solid. A solution of the latter (1.3g, 6
mmol) and hydrazine (85%, 15 mL) in 50 mL of ethanol
was heated at reflux for 2.5 h. After cooling the mixture
to 0 ꢁC, phthalhydrazide was removed by filtration.
Evaporation of the filtrate left the crude amine free base
which was subjected to flash column chromatography
(CH2Cl2–MeOH–NH4OH 10:10:1 as eluent) to afford
0.44 g (75%) of 8. The free base form is highly hygro-
scopic and was converted to the di-HClO4 salt immedi-
ately: 1H NMR (CD3OD) d 2.82 (s, 3H), 3.95 (t, J=1.82
Hz, 2H), 4.04 (t, J=1.84 Hz, 2H); 13C NMR (CD3OD)
30.3 (+), 33.1 (ꢂ), 38.9 (+), 78.2 (+), 81.8 (+).
General methods
1H NMR spectra were obtained on a Varian Gemini
300 (13C NMR at 75 MHz) or 200 (13C NMR at
50 MHz) instruments, with chemical shifts being refer-
enced to TMS or the solvent peak. In the 13C NMR line
listings, attached proton test (APT) designations are
given as (+) or (ꢂ) following the chemical shift. The
NOE difference spectra were recorded using a Varian
Inova 600. High-resolution mass spectra (HRMS, elec-
tron impact) were obtained at 20–40 eV on a Kratos
MS-25A instrument. UV–vis spectra were obtained
using a jacketed (temperature-controlled) cell compart-
ment and Perkin-Elmer PECSS software. Doubly dis-
tilled water was used for all enzyme experiments and
model studies. Melting points are uncorrected. Thin-
layer and preparative thin-layer chromatography were
run on Merck silica gel 60 plates with 254 nm indicator.
All column chromatography was run using flash-grade
silica gel. All solvents, reagents, and organic fine che-
micals were the most pure available from commercial
sources. Bovine plasma amine oxidase (BPAO) (100
units/gram of protein) and PDX G. F. 25 were purchased
from Sigma. In later studies, BPAO was purchased from
Worthington Biochemical. All evaporations were con-
ducted at reduced pressure using a rotary evaporator.
N-Benzyl-1,4-diamino-2-butyne dihydrochloride (9). A
solution of 3.0 g (12.8 mmol) of 19 and 8.3g (76.8
mmol) of benzylamine in dry CH3CN (50 mL) was
heated at reflux for 6 h under Ar. The solid generated
during the reaction was filtered, and the filtrate was
concentrated in vacuo. The residue obtained upon sol-
vent evaporation was subjected to column chromato-
graphy using MeOH as eluent to give the free base 9 as
1
an oil in 39% yield: H NMR (CDCl3) d 2.74 (s, 3H),
1,4-Diamino-2-butyne (4). A solution of 2.6 g (20 mmol)
of 1,4-dichloro-2-butyne and 50 mL of concentrated
NH4OH was heated to 40 ꢁC for 3h in a 350 mL pres-
sure bottle. The mixture rapidly became homogenous.
Excess ammonia was evaporated carefully at room
temperature, the remaining solution was acidified to pH
5, and the acidic solution was evaporated almost to
dryness. The residue was adjusted to pH 11 and extrac-
ted with diethyl ether using a continuous extractor. The
organic layer was dried (Na2SO4) and concentrated to
afford a brown oil which was distilled to give 0.7 g
(40%) of the main fraction of 415 (16 mm Hg, 96 ꢁC):
mp 42–46 ꢁC.
3.30 (s, 2H), 3.37 (s, 2H), 3.74 (s, 2H), 7.14–7.26 (5H);
13C NMR (CDCl3) d 31.18, 37.57, 52.37, 80.77, 83.24,
127.13, 128.38 (appears to be a composite of two 2C
signals), 139.30. The free base (0.8 g) was dissolved in
CH2Cl2 (10 mL) and ethyl ether (10 mL), and HCl gas
was bubbled through the solution for 10 min. The gen-
erated solid was filtered and recrystallized from EtOH
.
containing a trace of H2O to give 9 2HCl in 92% yield:
mp 215–217 ꢁC. H NMR (D2O) d 3.96 (s, 2H), 4.01 (s,
1
2H), 4.35 (s, 2H), 7.51 (s, 5H); 13C NMR (D2O) d 29.13,
35.69, 50.21, 76.85, 80.26, 129.46 (2C), 130.03 (appears
to be a composite of two (C+2C) signals), 130.32. Anal.
.
calcd for C11H16Cl2N2 0.8H2O: C, 50.51; H, 6.78; N,
10.71. Found: C, 50.23; H, 6.25; N, 10.88. HRMS calcd
for C11H14N2 (M+ꢂH) m/z (rel intensity) 173.1079,
found 173.1075 (9%).
Z-2-Chloro-1,4-diamino-2-butene dihydrochloride (5). A
solution of 1,4-diamino-2-butyne dihydrochloride (350
mg, 2.23mmol) was heated at reflux in a 1:1 mixture of
concd HCl and glacial acetic acid (30 mL) for 1 week.
The reaction mixture was concentrated and the solid
was recrystallized from glacial AcOH to give the pro-
duct 5 2HCl in 72% yield: H NMR (D2O) d 3.89 (d,
J=6.6 Hz, 2H), 4.01 (s, 2H), 6.28 (t, J=6.6 Hz, 1H);
13C NMR (D2O) d 37.2, 45.4, 124.5, 131.7.
N-Phenethyl-1,4-diamino-2-butyne dihydrochloride (10).
A solution of 2.0 g (8.6 mmol) of 19 and 8.3g (68.5
mmol) of phenethylamine in dry CH3CN (40 mL) was
heated at reflux for 12 h under Ar. The solid which was
generated during the reaction was filtered, and the fil-
trate was concentrated in vacuo to remove CH3CN and
excess phenethylamine. The residue was subjected to
column chromatography using MeOH as eluent to give
1
.
1-Chloro-4-phthalimido-2-butyne (19). 1,4-Dichloro-2-
butyne (20 mL, 0.204 mol) was added to the stirred
suspension of potassium phthalimide (4.7 g, 25.5 mmol)
in DMF (30 mL). The reaction mixture was heated to
100 ꢁC for 2 h. After cooling, 50 mL of water was added
and the precipitated solid was filtered and dried at high
vacuum to give 1-chloro-4-phthalimido-2-butyne as a
1
the free base 10 in 41% yield: H NMR (CDCl3) d 1.50
(br s, 3H), 2.83 (t, 2H), 2.93 (t, 2H), 3.43 (s, 4H), 7.22–
7.31 (5H). 13C NMR (CDCl3) d 31.42, 36.07, 38.37,
49.91, 80.30, 83.82, 126.17, 128.44 (2C), 128.66 (2C),
139.78. The free base (0.6 g) was dissolved in CH2Cl2
(10 mL) and ether (10 mL), and HCl gas was bubbled
through the solution for 10 min. The generated solid
was filtered and recrystallized from EtOH to give
1
white solid in 70% yield: H NMR (CDCl3) d 4.09 (t,
J=2.13Hz, 2H), 4.48 (t, J=1.71 Hz, 2H), 7.72–7.81
(4H). 13C NMR (CDCl3) d 27.3, 30.2, 77.9, 80.0, 123.6
(2C), 132.0 (2C), 134.3 (2C), 167.0 (2C).
ꢁ
10 2HCl in 90% yield: mp 221.0–222.5 C. 1H NMR
.
(D2O) d 3.07 (t, 2H), 3.46 (t, 2H), 3.93 (s, 2H), 4.01 (s,

H.-B. Jeon et al. / Bioorg. Med. Chem. 11 (2003) 4631–4641
4639
1
2H), 7.35–7.44 (5H). 13C NMR (D2O) d 29.12, 31.67,
36.60, 47.96, 76.77, 80.12, 127.63, 128.97 (2C), 129.27
(2C), 136.26. Anal. calcd for C12H18Cl2N2: C, 55.18; H,
6.95; N, 10.73. Found: C, 54.90; H, 6.87; N, 11.18.
in 60% yield. 11 2HCl: H NMR (CD3OD) d 3.00 (s,
6H), 3.98 (t, J=2.01 Hz, 2H), 4.23(t, J=2.01 Hz, 2H);
13C NMR (CD3OD) d 30.19 (+), 42.94 (ꢂ), 47.59 (+),
76.99 (+), 83.57 (+).
.
N,N-Dimethyl-1,4-diamino-2-butyne dihydrochloride (11)
from 19. Triethylamine (7.2 mL, 51.6 mmol) was added
to a mixture of 19 (4 g, 17.2 mmol) and dimethylamine
hydrochloride (2.8 g, 34.4 mmol) in 60 mL of CH2Cl2.
After stirring overnight at room temperature, the reac-
tion mixture was extracted with cold water. The organic
layer was dried over Na2SO4, concentrated, and sub-
jected to flash column chromatography using EtOAc as
eluent to give 1-(N,N-dimethyl)amino-4-phthalimido-2-
butyne (20) as a solid in 85% yield: 1H NMR (CDCl3) d
2.15 (s, 6H), 3.11 (t, J=1.95 Hz, 2H), 4.39 (t, J=1.92
Hz, 2H), 7.63–7.77 (4H). 13C NMR (CDCl3) d 27.23,
44.11 (2C), 47.86, 78.38, 78.44, 123.41 (2C), 132.00 (2C),
134.10 (2C), 166.98 (2C). A mixture of 20 (2.5 g, 10.3
mmol) and hydrazine monohydrate (0.5 mL, 10.3
mmol) in ethanol (50 mL) was heated at reflux for 1 h.
After cooling to 0 ꢁC, the reaction mixture was filtered
and the filtrate was carefully distilled to obtain the free
amine 11 (1.2 mL), which was treated with 1 N HCl to
N,N-Bis[2-(2-pyridyl)ethyl]-1,4-diamino-2-butyne tetra-
hydrochloride (12). A solution of 1.55 g (6.6 mmol) of
19 and 3.04
g
(13.4 mmol) of bis[2-(2-pyr-
idyl)ethyl]amine in dry CH3CN (40 mL) was heated at
reflux for 8 h. The reaction mixture was concentrated,
and the residue was partitioned between CH2Cl2 and
water. The organic layer was separated, dried over
Na2SO4, and concentrated. The residue was subjected to
column chromatography using CH2Cl2–MeOH (5:1) as
eluent to give 4-bis[2-(2-pyridyl)ethyl]amino-1-phthali-
1
mido-2-butyne (21) in 84% yield: H NMR (CDCl3) d
2.90 (br s, 8H), 3.47 (t, 2H), 4.45 (t, 2H), 7.04–7.10 (4H),
7.49 (2H), 7.68–7.86 (4H), 8.44 (m, 2H). 13C NMR
(CDCl3) d 27.39, 36.27 (2C), 42.45, 53.53 (2C), 78.39,
78.48, 121.10 (2C), 123.28 (2C), 123.50 (2C), 132.10
(2C), 134.13 (2C), 136.21 (2C), 149.08 (2C), 160 30 (2C),
167.08 (2C). A solution of 1.8 g (4.5 mmol) of 21 and
anhydrous hydrazine 0.2 mL in MeOH (40 mL) was
heated at reflux for 2 h. The reaction mixture was fil-
tered and the filtrate was concentrated in vacuo. The
residue was dissolved in CH2Cl2, and HCl gas was
bubbled through the solution for 15 min. The generated
give the salt form, 11 2HCl in 73% yield: 1H NMR
.
(D2O) d 2.99 (s, 6H), 3.99 (t, 2H), 4.16 (t, J=1.95 Hz,
2H). 13C NMR (D2O) d 29.3, 42.4 (2C), 47.0, 75.7, 81.9.
HRMS calcd for C6H12N2 (M+) m/z (rel. intensity)
112.1001, found 112.0997 (13%).
salt was filtered and recrystallized ꢁfrom EtOH to give
1
.
12 4HCl in 88% yield: mp 156–159 C. H NMR (D2O)
d 3.61 (t, 4H), 3.85 (t, 4H), 4.01 (s, 2H), 4.35 (br s, 2H),
7.97–8.08 (4H), 8.58 (t, 2H), 8.73(d, 2H). 13C NMR
(D2O) d 28.23(2C), 29.03, 42.89, 51.55 (2C), 74.90,
82.82, 126.25 (2C), 127.80 (2C), 141.84 (2C), 147.61
(2C), 150.56 (2C). Anal. calcd for C18H26Cl4N4 0.5H2O:
C, 48.13; H, 6.01; N, 12.47. Found: C, 48.13; H, 6.13; N,
12.17.
N-Methyl-1,4,-diamino-2-butyne (8) and N,N-dimethyl-
1,4-diamino-2-butyne (11) from 1-tert-butyloxycarbony-
lamino-4-chloro-2-butyne. 1-Amino-4-chloro-2-butyne
.
.
HCl was prepared by HCl-mediated hydrolysis of the
product of reaction of hexamethylenetetramine with
1,4-dichloro-2-butyne:16 1H NMR (D2O) d 3.91 (t,
J=1.39 Hz, 2H), 4.30 (t, J=1.39 Hz, 2H). The amine
salt (10.5 g, 75 mmol) was dissolved in 250 mL of water-
methanol (5:1), and 1.1 equivalents of di-tert-butyl
dicarbonate was added fractionally at room tempera-
ture while maintaining the pH at 10.0 by dropwise
addition of 3N NaOH. At the end of reaction, 200 mL
of water was added and the mixture was extracted with
CH2Cl2 (3ꢅ100 mL). The combined organic layer was
dried (Na2SO4) and evaporated, and the residue was
subjected to flash column chromatography with hex-
anes–CH2Cl2 (2:1) as eluent to give 1-tert-butyloxy-
carbonylamino-4-chloro-2-butyne17 in 90% yield: 1H
NMR (CDCl3) d 1.46 (s, 9H), 3.97 (dt, J=2.08, 5.4,
2H), 4.14 (t, J=2.08 Hz, 2H), 4.70 (br, 1H). To a 50-mL
pressure bottle was added 102 mg (50 mmol) of the lat-
ter Boc-protected compound and 20 mL of a 2 M solu-
tion of either methylamine or dimethylamine in
methanol containing a catalytic amount of KI, and the
mixture was heated to 50 ꢁC for 2 h. After cooling to
room temperature, the mixture was adjusted to pH 3
with 3N HCl and extracted with diethyl ether (dis-
carded). The aqueous layer was adjusted to 11 and
extracted with diethyl ether (3ꢅ30 mL). The combined
organic layer was dried (Na2SO4) and evaporated to
afford the Boc-protected derivatives of 8 and 11. The
Boc group was removed by 3N HCl in ethanol at room
temperature to afford the dihydrochlorides of 8 and 11
N-tert-Butyloxycarbonyl-1,4-diamino-2-butyne (13). To
a solution of 1,4-diamino-2-butyne (1 g, 11.9 mmol) in
20 mL of methanol–water (9:1) was added a solution of
di-tert-butyl dicarbonate in MeOH, while maintaining
the solution at pH 10-10.5 by adding aqueous NaOH,
until about 70% of the 1,4-diamino-2-butyne was con-
sumed according to TLC analysis. The reaction mixture
was concentrated and partitioned between CH2Cl2 and
brine, and the organic layer was dried over Na2SO4,
concentrated, and subjected to flash column chromato-
graphy using EtOAc and then MeOH as eluents to
afford 13 as a white solid in 37% yield: 1H NMR
(CDCl3) d 1.36 (s, 9H), 3.33 (t, J=1.97 Hz, 2H), 3.82 (d,
J=4.95 Hz, 2H), 5.12 (br s, NH). 13C NMR (CDCl3) d
28.3 (3C), 30.5, 31.4, 78.6, 79.7, 83.6, 155.4. HRMS
calcd for C9H17N2O2 (MH+) m/z (rel. intensity)
185.1290, found 185.1288 (2%).
N-Acetyl-1,4-diamino-2-butyne hydrochloride (14). To a
stirred solution of 13 (650 mg, 3.53 mmol) in dichloro-
methane (15 mL) was added triethylamine (0.98 mL,
7.06 mmol) and acetic anhydride (0.66 mL, 7.06 mmol).
After stirring overnight, the reaction mixture was con-
centrated, and the residue was partitioned between CHCl3
and brine. The organic layer was dried, concentrated, and
subjected to flash column chromatography using EtOAc

4640
H.-B. Jeon et al. / Bioorg. Med. Chem. 11 (2003) 4631–4641
as eluent to give 670 mg (83%) of N-acetyl-N0 -tert-
butyloxycarbonyl - 1,4 - diamino - 2 - butyne. 1H NMR
(CDCl3) d 1.34 (s, 9H), 1.90 (s, 3H), 3.79 (m, 2H), 3.90
(m, 2H), 5.30 (br s, NH, 1H), 6.93 (br t, NH, 1H). 13C
NMR (CDCl3) d 22.8, 28.3 (3C), 29.2, 30.4, 78.7, 79.3,
79.8, 155.6, 170.4. The latter intermediate (350 mg, 1.55
mmol) was treated with 1 mL of concd HCl. After 5
min, the reddish homogeneous solution was con-
centrated, and the residue was recrystallized from
as a white solid (4.0 g): H NMR (CD3OD) d 3.58 (d,
1
J=8 Hz, 2H), 4.4 (d, J=6 Hz, 2H), 4.57 (d, J=13Hz,
6H), 4.7 (d, J=13Hz, 6H), 5.9 (m, 1H), 6.2 (m, 1H), 7.8
(m, 4H). 13C NMR (CD3OD) d 40.0, 58.4, 72.0, 79.9,
118.7, 124.2, 133.3, 135.5, 139.8, 169.4. A solution of the
latter salt (3.9 g, 10.3 mmol) in a mixture of ethanol
(100 mL) and con. HCl (5 mL) was heated at reflux for
2 h. On cooling of the solution in ice-water, the pre-
cipitate was filtered off. The filtrate was concentrated to
EtOH–Et2O to give 14 HCl in 61% yield: 1H NMR
give 2.5 g of trans-4-phthalimido-2-butenamine HCl
1
.
.
(D2O) d 2.01 (s, 3H), 3.83 (s, 2H), 3.98 (s, 2H). 13C
NMR (D2O) d 21.9, 29.0, 29.3, 73.7, 83.1, 174.2. HRMS
calcd for C6H9N2O (M+ꢂH), C6H10N2O (M+), and
C6H11N2O (MH+) m/z (rel. intensity) 125.0715,
126.0793, and 127.0871, found 125.0713 (1%), 126.0777
(<1%), and 127.0859 (1%), respectively.
(23 HCl): H NMR (CD3OD) d 3.6 (d, J=6 Hz, 2H),
.
4.35 (d, J=4 Hz, 2H), 5.8 (dt, J1=6 Hz, J2=15 Hz,
1H), 6.0 (dt, J1=5 Hz, J2=15 Hz, 1H), 7.8–7.9 (4H).
Without further purification, triethylamine (1.8 g, 18
.
mmol) was added slowly to a solution of 23 HCl (2.5 g,
10 mmol) in 50 mL of ice-cold methanol, followed by
dropwise addition of a solution of acetic anhydride (1.8
g, 24 mmol) in 10 mL Et2O. The mixture was stirred at
room temperature for 4 h. After the solvent was
removed, the residue was extracted with EtOAc and
washed with water. The organic layer was dried over
Na2SO4 and concentrated. The residue was recrys-
tallized from i-PrOH–Et2O to give 0.2 g of the N-acetyl
derivative of 23. 1H NMR (CDCl3) d 1.9 (s, 3H), 3.85 (t,
2H), 4.3(d, 2H), 5.7 (m, 2H), 6.15 (br s, 1H), 7.73(m,
2H), 7.86 (m, 2H). A solution of the latter acetamide
(0.2 g, 0.8 mmol) in 10 mL of ethanol was heated at
reflux with hydrazine (0.025 g, 0.8 mmol) for 3h. The
reaction mixture was cooled to room temperature, the
precipitate was filtered off, and the filtrate was con-
centrated. The residue was dissolved in methanol and
subjected to flash chromatography using EtOAc–
MeOH (1:1) as eluent to give 0.1 g of 16. 16 HCl salt:
trans-N,N-Dimethyl-1,4-diamino-2-butene dihydrochlor-
ide (15). Hexamethylenetetramine (4.0 g, 28 mmol) in
ethanol (80 mL) was added dropwise to trans-1,4-
dichloro-2-butene (15 mL, 0.14 mol). The mixture was
allowed to stir at room temperature overnight, and then
evaporated to dryness. The residue was triturated with
CHCl3 (10 mLꢅ2), and the solid was filtered and dried
under vacuum to give the light yellow N-(trans-4-
chloro-2-butenyl)hexamethylenetetraminium chloride
(7.0 g) as the intermediate. To a solution of dimethyla-
mine hydrochloride (8.2 g, 0.1 mol) and KOH (7.1 g,
0.12 mol) in 30 mL of water cooled in an ice bath was
added dropwise a solution of the intermediate salt (6.6
g, 25 mmol) in 60 mL water. The mixture was stirred at
room temperature for 4 h and concentrated. The residue
was dissolved in cold ethanol, and KCl was filtered off.
The filtrate was concentrated, and the residue was par-
titioned between water and CHCl3. The aqueous layer
was concentrated to dryness to give a brown viscous oil,
which was heated at reflux in a mixture of ethanol (75
mL) and con. HCl (10 mL) for 2 h. On cooling of the
reaction mixture, the precipitate was filtered off, and the
filtrate was concentrated to give an oily residue, which
was crystallized by treating with i-PrOH–CH3CN to
mp 182–184 ꢁC. H NMR (CD3OD) d 2.2 (s, 3H), 3.58
1
(d, J=6 Hz, 2H), 3.95 (d, J=5 Hz, 2H), 5.8 (m, 1H),
5.95 (m, 1H). 13C NMR (CD3OD) d 21.4, 41.7, 42.5,
125.1, 133.2, 175.0. HRMS calcd for C6H10NO
([MꢂNH2]+) m/z 112.0725, found 112.0762 (94%).
Z-4-Amino-2-chloro-1-(dimethylamino)-2-butene dihy-
drochloride (17) and Z-4-amino-3-chloro-1-(dimethyl-
amino)-2-butene dihydrochloride (18). Compound 11
(450 mg, 2.4 mmol) was heated at reflux in a mixed
solution of concd HCl (15 mL) and glacial acetic acid
(15 mL) for 1 week. After evaporation of volatile mate-
ꢁ
1
.
give 200 mg of the product 15 2HCl: mp 158–160 C. H
NMR (CD3OD) d 2.9 (s, 6H), 3.7 (d, 2H), 3.9 (d, 2H),
6.1 (m, 2H). 13C NMR (CD3OD) d 41.5, 42.9, 59.2,
126.0, 134.2. HRMS calcd for C6H14N2 m/z 114.1157,
found 114.1159 (M+, 69%).
1
rials, the H NMR spectrum of the remaining residue
showed 17 and 18 as the only products in a ratio of 7:10.
To the mixture of 17 and 18 dissolved in 30 mL of
THF–water (2:1), was added sodium hydroxide (960
mg, 24 mmol) and di-tert-butyl dicarbonate (780 mg, 36
mmol), and the solution was stirred overnight at room
temperature. The reaction mixture was extracted with
EtOAc (30 mLꢅ2), and the organic layer was dried over
Na2SO4, concentrated, and subjected to column chro-
matography using EtOAc–MeOH (9:1) as eluent to
obtain faster moving t-Boc derivative 24 (120 mg, 0.48
mmol) and slower moving t-Boc derivative 25 (160 mg,
0.65 mmol). Compound 24: 1H NMR (CDCl3) d 1.37 (s,
9H), 2.17 (s, 6H), 2.98 (s, 2H), 3.83 (t, J=5.99 Hz, 2H),
4.99 (br s, NH, 1H), 5.72 (t, J=5.99 Hz, 1H). 13C NMR
(CDCl3) d 28.4 (3C), 38.0, 44.8 (2C), 66.3, 79.4, 126.0,
trans-N-Acetyl-1,4-diamino-2-butene (16). To a suspen-
sion of potassium phthalimide (8.9 g, 48 mmol) in 50
mL of DMF was added trans-1,4-dichloro-2-butene
(15 mL, 144 mmol) at once. The mixture was heated
to 100 ꢁC for 1 h, cooled to room temperature, and
poured into 200 mL of ice-water. The precipitate was
filtered, washed with water, and dried in air to give
6.0 g of trans-1-chloro-4-phthalimido-2-butene as a
colorless solid: 1H NMR (CDCl3) d 4.0 (d, 2H), 4.3
(d, 2H), 5.9 (m, 2H), 7.7 (q, 2H), 7.8 (q, 2H). A
mixture of the previous intermediate (5.3g, 22.4
mmol) and hexamethylenetetramine (4.7 g, 33.6
mmol) in 60 mL of chloroform was allowed to stir at
room temperature overnight. On cooling of the reac-
tion mixture in ice-water, N-(trans-4 - phthalimido - 2 -
butenyl)hexamethylenetetraminium chloride separated
1
133.4, 155.9. Compound 25: H NMR (CDCl3) d 1.42
(s, 9H), 2.22 (s, 6H), 3.10 (d, J=6.48 Hz, 2H), 3.88 (d,

H.-B. Jeon et al. / Bioorg. Med. Chem. 11 (2003) 4631–4641
4641
J=5.61 Hz, 2H), 5.17 (br. s, NH, 1H), 5.80 (t, J=6.39
References and Notes
Hz, 1H). 13C NMR (CDCl3) d 28.4 (3C), 45.0 (2C), 47.2,
56.7, 79.8, 123.3, 133.6, 155.5. Compounds 24 and 25
were individually treated with 1 mL of concd HCl and
1. Janes, S. M.; Mu, D.; Wemmer, D.; Smith, A. J.; Kaur, S.;
Maltby, D.; Burlingame, A. L.; Klinman, J. P. Science 1990,
248, 981.
2. Janes, S. M.; Palcic, M. M.; Scaman, C. H.; Smith, A. J.;
Brown, D. E.; Dooley, D. M.; Mure, M.; Klinman, J. P. Bio-
chemistry 1992, 31, 12147.
3. Wang, S. X.; Mure, M.; Medzihradszky, K. F.; Burlin-
game, A. L.; Brown, D. E.; Dooley, D. M.; Smith, A. J.;
Kagan, H. M.; Klinman, J. P. Science 1996, 273, 1078.
4. Mure, M.; Mills, S. A.; Klinman, J. P. Biochemistry 2002,
41, 9269.
.
.
evaporated to give 17 HCl and 18 HCl, respectively.
1
.
Compound 17 HCl: H NMR (D2O) d 2.95 (s, 6H), 3.92
(t, J=6.95 Hz, 2H), 4.18 (s, 2H), 6.41 (t, J=6.59 Hz,
1H). 13C NMR (D2O) d 37.3 (ꢂ), 42.5 (2C, +), 62.6 (ꢂ),
128.6 (+), 128.8 (ꢂ). HRMS calcd for C6H1335ClN2
(M+) and C6H1035ClN (M+ꢂNH3) m/z (rel. intensity)
148.0767 and 131.0502, found 148.0776 (<1%) and
131.0504 (45%), respectively. Compound 18 HCl: 1H
.
NMR (D2O) d 2.93(s, 6H), 4.02 (s, 2H), 4.04 (t, J=6.95
Hz, 2H), 6.32 (t, J=7.08 Hz, 1H). 13C NMR (D2O) d
42.9 (2C, +), 45.6 (ꢂ), 54.5 (ꢂ), 121.5 (+), 134.9 (ꢂ).
HRMS calcd for C6H1335ClN2 (M+) and C6H1035ClN
(M+ꢂNH3) m/z (rel. intensity) 148.0767 and 131.0502,
found 148.0758 (6%) and 131.0528 (19%), respectively.
5. Lyles, G. A.; Int, J. Biochem. Cell Biol. 1996, 28, 259.
6. Bardsley, W. G.; Hill, C. M. Biochem. Biophys. Res. Com-
mun. 1970, 41, 1068.
7. Hevey, R. C.; Babson, J.; Maycock, A. L.; Abeles, R. H. J.
Am. Chem. Soc. 1973, 95, 6125.
8. Abeles, R. H.; Maycock, A. L. Acc. Chem. Res. 1976, 9,
313.
9. Pec, P.; Frebort, I. Eur. J. Biochem. 1992, 209, 661.
10. Frebort, I.; Pec, P.; Luhova, L.; Matsushita, K.; Toyama,
H.; Adachi, O. Eur. J. Biochem. 1994, 225, 959.
11. Sayre, L. M.; Wang, F.; Lee, Y.; Huang, H.; Greenaway,
F. T.; He, Z.; Lightning, A.; Kagan, H. M. In Biochemistry of
Vitamin B6 and PQQ; Marino, G., Sannia, G., Bossa, F., Eds.;
Birkhauser: Basel, Switzerland, 1994; p 259.
12. Macholan, L.; Hubalek, F.; Subova, H. Coll. Czech.
Chem. Commun. 1975, 40, 1247.
13. He, Z.; Nadkarni, D. V.; Sayre, L. M.; Greenaway, F. T.
Biochim. Biophys. Acta 1995, 1253, 117.
14. Sebela, M.; Frebort, I.; Lemr, K.; Brauner, F.; Pec, P.
Arch. Biochem. Biophys. 2000, 384, 88.
15. Johnson, A. W. J. Chem. Soc 1946, 1009.
16. Dobson, N. A.; Raphael, R. A. J. Chem. Soc 1958, 3642.
17. Nassal, M.; Buc, P.; Wieland, T. Liebigs Ann. Chem. 1983,
1524.
18. Davtyar, N. M.; Nonezyan, N. G.; Oganesyan, S. A.;
Martirosyan, G. T. Arm. Khim. Zh. 1976, 29, 356; Chem.
Abstr. 1976, 85, 160043z.
Enzymology
Time-dependent inactivation of BPAO by candidate
inhibitors was determined as in our recent studies,24,28
by taking aliquots over time of primary incubations of
BPAO with inhibitors and assaying for the conversion
of benzylamine (10 mM) to benzaldehyde at 250 nm.
Inactivation time courses were determined twice for
most of the concentrations used, though the time points
were non-identical, and thus activity loss profiles were
not statistically evaluated. The figures shown are repre-
sentative plots where the plateau activity was found to
vary byꢆ5% between runs. In the determination of
partition ratios, the points at high [I]/[E] were neglected
in selecting the best straight line.20 Irreversibility of
BPAO inhibition was checked by gel filtration24 or after
24 h dialysis, with the before and after activity assays
varying no more than 5% in the cases of irreversible
inhibitors. Phenylhydrazine titration of the active site22
after incubation of BPAO with irreversible inactivator
was performed as described24,28 on dialyzed or gel-fil-
tered preparations that showed less than 8% activity.
The nitroblue tetrazolium (NBT) redox cycling assays23
were performed as described24,28 on denatured prepara-
tions of enzyme that showed less than 7% activity. No
alteration of the electrophoretic properties of the
enzyme or the intensity of Coomassie staining was
observed for the gel run in parallel.
19. Fahey, R. C.; Lee, D. J. J. Am. Chem. Soc. 1968, 90,
2124.
20. Silverman, R. B. Mechanism-Based Enzyme Inactivation:
Chemistry and Enzymology; CRC: Boca Raton, FL, 1988.
21. Kitz, R.; Wilson, I. B. J. Biol. Chem. 1962, 237, 3245.
22. Janes, S. M.; Klinman, J. P. Biochemistry 1991, 30, 4599.
23. Fluckiger, R.; Paz, M. A.; Gallop, P. M. Methods Enzy-
mol. 1995, 258, 140.
24. Jeon, H.-B.; Sayre, L. M. Biochim. Biophys. Acta 2003,
1647, 343.
25. Shepard, E. M.; Smith, J.; Elmore, B. O.; Kuchar, J. A.;
Sayre, L. M.; Dooley, D. M. Eur. J. Biochem. 2002, 269, 3645.
26. Frebort, I.; Sebela, M.; Svendsen, I.; Hirota, S.; Endo, M.;
Yamauchi, O.; Bellelli, A.; Lemr, K.; Pec, P. Eur. J. Biochem.
2000, 267, 1423.
Acknowledgements
27. Jeon, H.-B.; Sayre, L. M. Biochem. Biophys. Res. Com-
mun. 2003, 304, 788.
28. Lee, Y.; Ling, K.-Q.; Lu, X.; Silverman, R. B.; Shepard,
E. M.; Dooley, D. M.; Sayre, L. M. J. Am. Chem. Soc. 2002,
124, 12135.
We are grateful to the National Institutes of Health for
support of this work through grant GM 48812. We also
thank Dr. Gang Sun for performing the NBT assay on 5
and Dr. Dale Ray for the NOE difference spectra on 17.