Probing the activities and mechanisms of leukotriene A4 hydrolase with synthetic inhibitors

Article abstract of DOI:10.1002/(SICI)1521-3765(19980904)4:9<1698::AID-CHEM1698>3.0.CO;2-J

Leukotriene (LT) A₄ hydrolase catalyzes the hydrolysis of leukotriene A₄ to form leukotriene B₄, a potent inflammatory mediator. Recently, the synthesis and evaluation of the highly effective competitive LTA₄ hydrolase inhibitor 2 (K(i) = 1.6 nM) was described. In the present study, we describe the biological activity of 2 against LTB₄ biosynthesis, as well as the design, synthesis, and evaluation of a new series of inhibitors intended to probe the active site of the enzyme. On the basis of these results and of previously reported site-directed mutagenesis and inhibition studies, the mechanisms of peptide and epoxide hydrolysis catalyzed by LTA₄ hydrolase are discussed.

Full text of DOI:10.1002/(SICI)1521-3765(19980904)4:9<1698::AID-CHEM1698>3.0.CO;2-J

FULL PAPER
Probing the Activities and Mechanisms of Leukotriene A4 Hydrolase with
Synthetic Inhibitors
J. Heather Hogg, Ian R. Ollmann, Anders Wetterholm, Martina Blomster Andberg,
Jesper Haeggström, Bengt Samuelsson, and Chi-Huey Wong*
Abstract: Leukotriene (LT) A4 hydrolase catalyzes the hydrolysis of leukotriene A4
to form leukotriene B4, a potent inflammatory mediator. Recently, the synthesis and
evaluation of the highly effective competitive LTA4 hydrolase inhibitor 2 (Ki ˆ
1.6 nm) was described.[25] In the present study, we describe the biological activity of
2 against LTB4 biosynthesis, as well as the design, synthesis, and evaluation of a new
series of inhibitors intended to probe the active site of the enzyme. On the basis of
these results and of previously reported site-directed mutagenesis and inhibition
studies, the mechanisms of peptide and epoxide hydrolysis catalyzed by LTA4
hydrolase are discussed.
Keywords: enzyme inhibitors
hydrolases hydroxamates
metalloenzymes ´ peptides
´
´
´
Introduction
Epoxide Hydrolase
H
O
Leukotriene (LT) A4 hydrolase (EC3.3.2.6) catalyzes the
hydrolysis of LTA4 [(5S,6S)-5,6-oxido-7,9-trans-11,14-cis-eico-
satetraenoic acid] to LTB4 [(5S,12R)-dihydroxy-6,14-cis-8,10-
trans-eicosatetraenoic acid], which acts as a chemokine for
neutrophils.[1±3] As high levels of LTB4 have been detected in
many sites of inflammation,[4, 5] LTB4 has been implicated in a
number of inflammatory diseases including gout,[6, 7] psoria-
sis,[8] inflammatory bowel disease,[9] bronchial asthma,[10, 11]
and rheumatoid arthritis.[12]
The enzyme also exhibits an intrinsic aminopeptidase
activity (Figure 1) which is selective for N-terminal arginine-
containing tripeptides,[13] but accepts dynorphins and enke-
phalins as substrates too.[14] The active site contains a single
zinc atom, and site-directed mutagenesis indicates that each of
the three putative zinc ligands is required for both the peptide
and epoxide hydrolase activities.[15] In addition, Tyr383 plays
critical roles in both activities.[16, 17] These results suggest that
the two activities occur at a common or overlapping
site.[14, 18±20] Indeed, a series of tight-binding peptide isosteres
synthesized as inhibitors of the peptidase activity are also
H
COOH
COOH
HO
LTA4
hydrolase
O
HO
LTA4
LTB4
Aminopeptidase
H
O
O
H
+
OH
H
H3+N
O
O
O-
H3+N
N
OH
LTA4
O-
hydrolase
O
H
N
N
H
+
HN
H3+N
O-
O
O
+
+
HN
HN
H2N NH2
+
HN
H2N NH2 H2N NH2
H2N NH2
Figure 1. The epoxide hydrolase and aminopeptidase activities of LTA4
hydrolase.
competitive inhibitors of the hydrolase activity,[21±25] including
for example mercaptoamine 1[26] and hydroxamate 2[25]
(Figure 2), which have Ki values of 0.35nm and 1.6nm,
respectively, measured against the peptidase activity of the
enzyme. It has been found that there is a strong linear
correlation between the potencies of competitive inhibitors
against the two activities.[27] Thus, if a competitive inhibitor is
strongly active against the peptidase activity, it will also, in
general, be a strongly active competitive inhibitor of the
hydrolase activity.
[*] Prof. C.-H. Wong, Dr. J. H. Hogg, Dr. I. R. Ollmann
Department of Chemistry and Skaggs Institute for Chemical Biology
The Scripps Research Institute, 10550 North Torrey Pines Rd.
La Jolla, California 92037 (USA)
Fax: (‡1)619-784-2409
E-mail: wong@scripps.edu
Dr. M. B. Andberg, Dr. J. Haeggström, Prof. B. Samuelsson
Department of Medical Biochemistry and Biophysics
Division of Chemistry II
While the mechanism of the aminopeptidase activity is
comparatively well understood as LTA4 hydrolase is a
member of the superfamily of zinc metalloproteases, which
The Karolinska Institute, S-17177 Stockholm (Sweden)
1698
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1998
0947-6539/98/0409-1698 $ 17.50+.50/0
Chem. Eur. J. 1998, 4, No. 9

1698±1713
posited to arise from a high-affinity contact between the
potent metal ligands at the polar head of these molecules
accompanied by a substantial contribution arising from the
long hydrophobic benzyloxyphenyl tail, presumably in some
part of the same large hydrophobic pocket normally occupied
by LTA4. It is known that LTA4 methyl ester is not a substrate
for this enzyme[36] (though it is a suicide inhibitor); this
suggests the importance of the acid moiety in molecular
recognition and the likelihood of a highly selective carbox-
ylate recognition site within the enzyme active site.
With derivatives of metallophilic aminopeptidase inhibitor
1, several possible binding arrangements of these inhibitors
relative to the epoxide hydrolysis transition state were
examined.[24] Because the zinc is thought to activate water in
the aminopeptidase activity, this same role was examined in
the epoxide hydrolase activity. A superimposition of a model
of 1 over a model of the transition state with the zinc at the
reacting water molecule (Figure 3A) suggested that (if this
overlapped model is correct) the enzymeꢁs carboxylate-
recognition element would be in the vicinity of the benzyl
group of 1. On the basis of this hypothesis, a series of
derivatives of 1 with an additional carboxylate substituent
were prepared; each derivative resulted in a minimum of a
10000-fold loss of affinity, suggesting that this model is
unlikely to be correct. An alternative binding mode as shown
in Figure 3B may occur.
O
O
SH
H3+N
1
OH
N
H3+N
O
2
OH
O
Figure 2. A competitive inhibitor of LTA4 hydrolase 1 (Ki ˆ 0.35nm) and a
hydroxamate-type inhibitor 2 (Ki ˆ 1.6nm).
includes thermolysin and aminopeptidase M, the mechanism
of the hydrolase activity remains unclear. The enzyme has
very little sequence similarity with other epoxide hydrolases
and also differs from them in several other respects.[28, 29] Thus,
no analogies between its mode of action and that of the other
enzymes can be drawn. Following the discovery that LTA4 can
irreversibly inhibit the enzyme, it was proposed that the
epoxide hydrolase mechanism occurred in two distinct
steps.[30] Tyrosine 378 was found to be the nucleophile
responsible for the irreversible inactivation.[31] Interestingly,
the Y378F mutant has enhanced specific activity (kcat) over
the wild-type enzyme. However, it produces large quantities
of the D6-trans-D8-cis isomer of LTA4, suggesting its role is to
hold the substrate in the correct geometry in the active site
rather than act as a proton donor or reactive intermediate of
some kind.[32, 33] A recent finding that the Y383F mutant
produces substantial quantities of an isomer of LTB4,
characterized by a syn-5,6 diol produced by a formal syn-
facial attack on the epoxide, resulted in the proposal that the
hydrolase activity operated by a two-stage mechanism char-
acterized by an SN1 opening of the epoxide to form a trienyl
cation followed by nucleophilic attack either by water[17] in the
case of LTB4 biosynthesis, or Tyr378 in the case of irreversible
inactivation.[31]
Zn
+
NH3
HO
HS
O
O
O
A
OH
Zn
H3+N
HO
SH
O
The single active-site zinc ion may participate in the
catalysis in one of two ways. It may act as a Lewis acid on
the epoxide, thus facilitating its opening, or it may coordinate
the reacting water molecule, activating it toward general-base
catalysis, as is thought to occur in the aminopeptidase
mechanism.[34] Because of the geometry of the transition state
(as suggested by the stereochemistry of the double bonds in
the substrate and product) it is apparent that the reacting
water molecule and scissile epoxide residue are on opposite
faces of the triene, making it unlikely that the zinc acts at both
oxygens at the same time. Which mode of action is actually
involved has still not been determined, and remains an
important question to be answered.
O
O
OH
B
Figure 3. A and B show the overlay of 1 and LTA4 in the binding site as
possible modes of inhibition. Current data supports model B.
A second model which placed the zinc at the scissile
epoxide was then considered. Modeling the metallophilic
head of the inhibitor to lie at, or near, the position of the
scissile epoxide suggested that the addition of a pentanoic acid
substituent to the metallophilic head group of the inhibitor
would strongly mimic the C1 ± C5 region of the substrate
(Figure 4). Hydroxamate 2 was designed to have a spacer with
four methylene groups between the hydroxamate and carbox-
ylate moieties. This modification resulted in a dramatic
increase in binding potency (Ki now 1.6nm) over the parent
hydroxamate (Table 1).[25] Hydroxamate 4, with a shorter
linker, binds slightly more weakly (Ki ˆ 3.4nm). The corre-
sponding methyl esters of these compounds, 3 (Ki ˆ 28nm)
and 5 (Ki ˆ 11nm), both show an order of magnitude decrease
in binding, demonstrating that this binding pocket prefers the
Here we report on a new series of hydroxamate inhibitors
of LTA4 hydrolase and their implications on the mechanism of
epoxide hydrolysis.[16, 17, 31, 32, 35]
Results and Discussion
Inhibition of the peptidase activity: The extremely tight
interaction between inhibitor 1 and LTA4 hydrolase was
Chem. Eur. J. 1998, 4, No. 9
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1998
0947-6539/98/0409-1699 $ 17.50+.25/0
1699

C.-H. Wong et al.
Incorporation of groups such as an amine 15 (IC50
0.40mm) or a phenyl group 13 (IC50 ˆ 0.11mm) in the place of
the carboxylate again results in a decrease in activity (the
synthesis of 13 and 15 is shown in Scheme 2). While the
methyl esters 3 and 5 are not as potent as their corresponding
acids, they are significantly better inhibitors than the others in
this series. This may suggest that there is a hydrogen-bonding
interaction responsible for the change in the binding free
energy rather than a charge ± charge interaction.
FULL PAPER
ˆ
OH
O
Zn++
O
N
O
O
H3+N
OH
OH
O
Figure 4. A possible binding mode of 2 with the hydroxamate moiety
located at the epoxide binding site and coordinated to the zinc. The two
carboxylates are overlapped. Current data supports this binding mode.
There still remained a third reasonable alternative binding
model for these compounds, which again placed the zinc
adjacent to the reacting water molecule during the epoxide
hydrolase reaction. This model orients the hydrophobic
portion of the inhibitor in the opposite direction from that
in the first model (Figure 5). Here, the hydroxamate lies
proximal to the end of the triene and
free carboxylate. As the chain which links the carboxylate gets
longer (10 and 11, the synthesis of which is shown in
Scheme 1) the binding potencies decrease, with the corre-
sponding esters again about ten times less active than the
acids.
the carboxylate lies near the epoxide,
perhaps interacting with the same
Lewis acid as the opening epoxide. If
(CH2)n
O
O
OCH3
BnO
BnO
HNBoc
NHBoc
NHOH
OH
O
O
O
O
n
N
this is the case, one would predict that
by extending our inhibitor to incorpo-
rate a second carboxylate which would
overlay that of LTA4, we should again
see an increase in potency as the result
of finally achieving the putative carbox-
ylate recognition site (Figure 6).
40% - 70%
(not optimized)
MeO
6 (n = 6)
7 (n = 8)
O
1. LiOH, THF:MeOH:H2O
2. HCl/HOAc
HCl/HOAc
96%
90% - 95%
To this end, we studied a final set of
hydroxamic acids (32 ± 41). Initially
these compounds were synthesized us-
ing an amine as a linker (Schemes 3 and
4). In this way, simply by varying the
amino acid used, a number of com-
pounds may be quickly synthesised with
a defined stereochemistry at the posi-
tion alpha to the amine. It was also
hypothesized that having a positively
charged amine in the vicinity of the
carboxylate might help mimic the
charge distribution of an opening epox-
ide, a feature of this model.
Cl-
Cl-
BnO
BnO
+NH3
+NH3
OH
OH
N
O
n
N
O
n
HO
MeO
10 (n = 6)
11 (n = 8)
8 (n = 6)
9 (n = 8)
O
O
Scheme 1. Synthesis of carboxylate chain links from (2S)-2-N'-Boc-amino-3-(4-benzyloxyphenyl)-N-
hydroxypropylamine.
Table 1. IC50 and Ki values for hydroxamates containing a carboxylate moiety (against the
peptidase activity of LTA4 hydrolase).
X-
O
H3+N
OH
An initial exploration of the prefer-
red stereochemistry using l- (37, IC50 ˆ
0.031mm) and d-alanine (38, IC50 ˆ
0.041mm) derivatives showed very little
preference for a single diastereomer
(Table 2). As l-amino acids are more
readily available and the l-isomer ap-
peared to be slightly more potent, the
structure of 37 was chosen as a template
for further development. Both com-
pounds, however, were less potent than
the original compound 2. Addition of a
second carboxylate on a linker of the
appropriate length, in 40, resulted in a
small increase in potency (Ki ˆ 20nm),
but nowhere near the two orders of
magnitude we saw with the incorpora-
tion of the first carboxylate. The Ki
value for 40 (20nm) compares some-
N
O
R
IC50 (µM)
IC50 (µM)
R
R
O
O
O
O
0.018
(Ki = 35 nM)
0.011
(Ki = 1.6 nM)
10
8
2
OH
OH
0.40
0.055
(Ki = 28 nM)
OMe
O
3
4
OMe
OH
11
9
0.025
0.16
0.11
0.011
(Ki = 3.4 nM)
OH
O
O
O
OMe
NH2
0.025
(Ki = 11 nM)
OMe
5
13
0.40
15
1700
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1998
0947-6539/98/0409-1700 $ 17.50+.50/0
Chem. Eur. J. 1998, 4, No. 9

Leukotriene A4 Hydrolase
1698±1713
in hydrogen-bonding interactions.
Use of a thioether as a linker rather
than an amine led to inhibitor 46
(IC50 ˆ 0.02mm, Ki ˆ 14nm) which
appears to be more potent than its
amine-containing analog 37 (Ki ˆ
78nm), but still not as good as the
parent compound 2 (Ki ˆ 1.6nm).
This may be a better choice for a
linker for further inhibitors.
BnO
BnO
O
O
HNBoc
OH
N
NHBoc
O
O
O
NHOH
60% (not optimized)
12
TFA-
BnO
+N H3
OH
TFA/CH2Cl2
95%
O
N
13
The failure to identify any com-
pelling increases in potency on add-
ing a second carboxylate in this
series of molecules indicates these
modifications contribute only weak
additional interactions, if any, with
the enzyme. The only binding model
to date with good predictive capa-
bility has been model two, on which
2 was based (Figure 4). It is of note
that this model predicts that the
active-site zinc operates at the ep-
oxide, suggesting that its role in
epoxide hydrolysis may be as a
Lewis acid, facilitating the initial
stage of the SN1 conversion of LTA4
to LTB4.
BnO
O
O
NHBoc
BnO
HNBoc
OH
N
NHBoc
O
O
O
NHOH
62% (not optimized)
NHBoc
14
TFA-
BnO
+N H3
OH
N
TFA/CH2Cl2
99%
O
NH2 TFA-
+
15
Scheme 2. Synthesis of amine and phenyl derivatives of (2S)-2-N'-Boc-amino-3-(4-benzyloxyphenyl)-N-
hydroxypropylamine.
H
Zn++
O
O
O3, DMS
BnO
H
HO
OH
BnO
N
O
O
O
O
O
O
O
16
H3+N
OH
R
CO2Me
H2N
CO2Me
1.
BnO
HO
N
R
Figure 5. Another possible binding mode of 2 with the carboxylate moiety
located at the epoxide binding site and the zinc ion coordinated to a water
molecule.
2. NaBH4, MeOH
3. (Boc)2O
Boc
O
O
CO2Me
R
H2, Pd/C
N
H
Zn++
O
Boc
O
H
HO
X
CO2H
O
22 R = L-CH3
23 R = D-CH3
24 R = L-(CH2)2CO2CH3
25 R = L-(CH2)3CO2CH3
26 R = L-(CH2)4NHBoc
17 R = L-CH3
18 R = D-CH3
19 R = L-(CH2)2CO2CH3
20 R = L-(CH2)3CO2CH3
21 R = L-(CH2)4NHBoc
N
O
O
H3+N
CO2H
OH
Figure 6. Design of new inhibitors based on the binding mode shown in
Figure 5.
Scheme 3. Synthesis of hydroxamic acids 17 ± 26.
Though it is far from certain that the zinc atom acts as a
Lewis acid at the epoxide, a growing body of indirect evidence
supports this role. There appears to be a consistent substrate-
design motif which specifies a preferred distance between the
carboxylate termini of the substrates and the sites of zinc-
catalyzed chemistry. The distance between the epoxide and
carboxylic acid in LTA4 is very similar to that between the
scissile amide bond and the C-terminus of a tripeptide, the
preferred substrate size for the enzymeꢁs aminopeptidase
activity,[13] and also the length of a series of known peptidase
inhibitors[37, 38] (Figure 7).
what favorably against the shorter 39 (Ki ˆ 78nm), but the
difference is not great. An amino group rather than a
carboxylate leads to a tenfold reduction in potency (41,
IC50 ˆ 0.20). As observed in the earlier series of carboxylate-
containing inhibitors, the corresponding methyl esters (32 ±
36) were all less potent than their analogous acids.
As the presence of the charged amino linker led to
decreased potency (2 vs. 37), a final compound which uses a
thioether as a linker was synthesized (Scheme 5). The
thioether moiety is perhaps a better mimic of a methylene
group as it has no associated charge and does not participate
Chem. Eur. J. 1998, 4, No. 9
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1998
0947-6539/98/0409-1701 $ 17.50+.25/0
1701

C.-H. Wong et al.
FULL PAPER
required for the aminopeptidase reaction but
not for the epoxide hydrolase activity. On the
basis of homology with thermolysin, it is thought
to be a proton source for the S1 amine during
peptide hydrolysis.[16] The Y383F, Y383Q, and
Y383H mutants of LTA4 hydrolase have severe-
ly diminished aminopeptidase activities. In the
epoxide hydrolysis case, these mutations allow
water to attack at the C6 end of the trienyl
cation in a syn-facial sense at the site of the
broken epoxide C O bond, rather than the
usual C12 position.[17] If this is to be explained
simply as the appearance of a hole in the enzyme
adjacent to residue 383, as the result of the
reduction of size, then it appears that Tyr383, and
by inference the zinc atom, are adjacent to the
epoxide.
Finally, it should be noted that the attack of
water on LTA4 in the active site occurs from the
energetically disfavored face of the molecule as
predicted by molecular orbital theory. If the zinc
atom activates water for a general base catalysis,
the epoxide may open concurrently with a
general acid catalysis as part of an SN2' mech-
anism. However, at least in the case of mutants
at position 383, it seems fairly clear that the
mechanism is more SN1 in character. Further-
more, if the conserved zinc atom is to act directly
(as opposed to allosterically) to accelerate the
reaction as part of an SN1 mechanism, then it
should act to accelerate the rate-determining
step, the opening of the epoxide.
For the sake of exploring all possibilities, one
could argue that even if the epoxide hydrolysis
reaction is SN2 in nature, if the zinc operated as a
water activator and Glu296 as a base in this
reaction, the absence of Glu296 would not be
observed in the overall enzymatic rates as it
would act after the rate-determining step in the
reaction. Similarly, one could argue that the
observed distance similarities between the
lengths of tripeptides and the C1 ± C6 region of
LTA4 is a coincidence, and the observation of the
5,6-isomer of LTB4 as a result of mutations at
position 383 arises from some remote allosteric
effect. It is clear that there still remains much to
be understood about the structure of LTA4
hydrolase and the epoxide hydrolysis mecha-
nism.
CO2Me
BnO
1. isobutylchloroformate,
BocN
OH
N
HO
TEA
N
R
O
CO2Me
R
Boc
O
BnO
NHBoc
NHOH
N
2.
Boc
22 R = L-CH3
23 R = D-CH3
27 R = L-CH3
28 R = D-CH3
29 R = L-(CH2)2CO2CH3
30 R = L-(CH2)3CO2CH3
31 R = L-(CH2)4NHBoc
80% - 90%
24 R = L-(CH2)2CO2CH3
25 R = L-(CH2)3CO2CH3
26 R = L-(CH2)4NHBoc
1. LiOH, THF:MeOH:H2O
2. HCl/HOAc
HCl/HOAc
95%
90%
Cl-
Cl-
BnO
BnO
+NH3
OH
+NH3
OH
N
O
N
O
CO2Me
R
CO2H
R
+
+
N
N
Cl-
H2
Cl-
H2
37 R = L-CH3
38 R = D-CH3
39 R = L-(CH2)2CO2H
40 R = L-(CH2)3CO2H
41 R = L-(CH2)4NH2 • HCl
32 R = L-CH3
33 R = D-CH3
34 R = L-(CH2)2CO2CH3
35 R = L-(CH2)3CO2CH3
36 R = L-(CH2)4NH2 • HCl
Scheme 4. Synthesis of hydroxamic acids 27 ± 41.
Table 2. IC50 and Ki values for extended hydroxamates 6 ± 54 and 6 ± 63 (against the
peptidase activity of LTA4 hydrolase).
Cl-
+ OH
O
NH3
N
O
CO2R
R'
X
IC50 (µM)
0.4
R' (R = H)
IC50 (µM)
0.03
R' (R = Me)
+
(L)
(L)
X =
CH3
CH3
37
38
32
NH2
Cl-
(D)
(D)
CH3
CH3
0.04
33
34
0.2
0.08
(Ki = 0.07)
(L)
(L)
39
40
41
0.1
CO2H
CO2Me
0.02
CO2H
CO2Me
(L)
(L)
(L)
(L)
35
36
0.2
1.4
(Ki = 0.02)
NH2
NH2
0.20
0.03
(Ki = 0.014)
(L)
(L)
CH3
CH3
46
45
0.4
X = S
Site-directed mutagenesis studies published to date further
corroborate this view. Glu296 (sequentially adjacent to His295
The degree to which this model is in agreement with an
earlier model proposed by Orning, Gierse, and Fitzpatrick,[13]
which places LTA4 in the S1' pocket, remains unclear. In that
work, it was originally proposed that the S1' pocket bound
LTA4 on the basis of its observed preference for hydrophobic
substituents as well as some data derived from the inhibitory
behavior of arphamenine. As the S1 pocket prefers arginine,
whose delocalized positive charge might be a good mimic of
the developing delocalized trienyl cation in a SN1-like LTA4
hydrolysis transition state, there remains some basis upon
,
one of the zinc ligands) is required for the aminopeptidase
activity, acting as the general base in the deprotonation of
water. However, it does not play this role during epoxide
hydrolysis; the E296A mutant displays full epoxide hydrolase
activity.[35] This behavior would support the theory that the
zinc atom may act as a Lewis acid for the epoxide opening as
no general base would be expected to be needed in the vicinity
of the zinc ion in such a mechanism. In addition, Tyr383 is also
1702
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1998
0947-6539/98/0409-1702 $ 17.50+.50/0
Chem. Eur. J. 1998, 4, No. 9

Leukotriene A4 Hydrolase
1698±1713
Inhibition of the epoxide hydrolase
activity: The IC50 and Ki values of
several of the hydroxamates have
been determined against the epox-
ide hydrolase activity of purified
enzyme (Table 3). These values are
all consistently an order of magni-
tude higher than those against the
peptidase activity. This reflects the
fact that these assays are run at
much higher enzyme concentra-
tions, necessary for the detection
of the product (see experimental).
In these assays, the enzyme concen-
tration is 360nm; thus the IC50 value
for 2 (0.14mm) is an upper limit
rather than a close approximation
of the Ki. The Ki values for our two
best inhibitors, 2 and 46, were not
determined as their inhibition
curves were too steep under these
high enzyme concentrations to
gather meaningful and reproduci-
ble data.
CO2Me
1. NaOH (aq)
2.
HO
KSAc
HO
HO
SAc
S
Br
O
O
O
42
43
Br
CO2Me
100%
NHBoc
OH
BnO
1. isobutylchloroformate, TEA
BnO
N
O
NHBoc
CO2Me
2.
NHOH
44
S
91%
1. LiOH, THF:MeOH:H2O
2. HCl/HOAc
HCl/HOAc
73%
84%
Cl-
Cl-
+
+
BnO
BnO
NH3
NH3
OH
OH
N
O
N
O
CO2H
CO2Me
S
S
46
45
Scheme 5. Synthesis of compounds 45 and 46, which contain a thioether linker.
Zn++
O O
O
O
H3N+HO
H3N+
OH
OH
N N
H
N
H
O
Table 3. IC50 and Ki values for selected hydroxamates (against the epoxide hydrolase activity
of LTA4 hydrolase).
Compound
IC50 (µM)
0.14
Ki (µM)
O
H3N+TFA-
BnO
BnO
Figure 7. Overlay of 2 and a tripeptide.
OH
N
N.D.[a]
O
O
2
CO2H
which to suspect that the pocket plays host to
LTA4. It appears reasonable that the benzyloxy-
phenyl portion of the inhibitors discussed here
and in previous works binds in either the S1 or
S1' subsites, because of their likely proximity to
the zinc atom. Some evidence, including a
strong requirement for a free amine adjacent
to the zinc ligand[24] and the efficacy of a series
of peptide transition-state isosteres,[21±23] sug-
gests that these inhibitors bind S1 despite its
preference for arginine. The identification of
the subsite which binds the large, rigid, hydro-
phobic benzyloxyphenyl group might help elu-
cidate a likely home for LTA4 in the active site,
therefore O-benzyl-l-tyrosine p-nitroanilide
hydrochloride 47 was prepared by Tesserꢁs
method.[39] Unfortunately, it was found that this
material was highly insoluble in water and could
not be evaluated as a substrate for the enzyme.
O-Benzyl-l-serine p-nitroanilide 48 was also
prepared and found not to be a substrate at
10mm. Larger, more soluble, O-benzyl tyrosine
containing peptidic substrates or chimeric pep-
tides containing LTA4-like structures may help
resolve this issue.
Cl-
+NH3
OH
N
10
0.7
1.1
6.9
CO2H
Cl-
BnO
+NH3
OH
N
O
37
39
N.D.
CO2H
+
N
Cl-
H2
Cl-
BnO
BnO
BnO
+NH3
OH
N
O
1.2
0.8
CO2H
+
N
CO2H
Cl-
H2
Cl-
+NH3
OH
N
O
CO2H
0.5
0.4
0.2
40
46
+
N
CO2H
Cl-
O
H2
Cl-
+
NH3
OH
N.D.
N
CO2H
S
[a] N.D.: not determined.
Chem. Eur. J. 1998, 4, No. 9
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1998
0947-6539/98/0409-1703 $ 17.50+.25/0
1703

C.-H. Wong et al.
FULL PAPER
Whole cell assays: A number of these hydroxamates were also
shown to be effective inhibitors of LTB4 synthesis in
polymorphonuclear leukocytes stimulated with the ionophore
A23187 or incubated with LTA4 (Table 4). Further experi-
ments with 2 also showed it to be selective for LTA4 hydrolase
(Figure 8). At approximately 1mm, the production of the
nonenzymatic hydrolysis products of LTA4 has doubled while
the production of LTB4 has essentially been stopped, indicat-
ing that while LTA4 hydrolase has been inhibited, the enzymes
leading up to the production of LTA4 have remained
unaffected. However, at higher concentrations we also see a
fall in the production of the nonenzymatic hydrolysis prod-
ucts, suggesting that our inhibitor is now interfering with other
processes in this pathway. There is, however, a window of
opportunity at 1mm, where hydroxamate 2 is selective for
LTA4 hydrolase. Mercaptoamine 1 has a similar profile;
however, it shows optimal selectivity at a
concentration of 10mm rather than at 1mm.[40]
Addition of exogenous arachidonic acid to
the assays of hydroxamate 2 fully restores the
production of the nonenzymatic hydrolysis
products; however, the inhibition of LTB4
production by LTA4 hydrolase is still complete
(Table 5). Again, a similar result is seen in the
case of mercaptoamine 1.[40]
Table 4. IC50 values for selected hydroxamates (against the epoxide hydrolase activity of
LTA4 hydrolase in isolated human polymorphonuclear leukocytes).
Incubation
with LTA4
Stimulation with
ionophor A23187
Compound
IC50 (µM)
IC50 (µM)
TFA-
+NH3
OH
BnO
BnO
O
O
0.2
0.2
2
N
CO2H
Conclusions
Cl-
+NH3
OH
With compounds 2 and 4, we have demonstrat-
ed that the inclusion of a free carboxylate
moiety in our hydroxamate class of inhibitors is
crucial for tight binding to LTA4 hydrolase.
When this carboxylate is placed in a position
more distal from the hydroxamate moiety we
see a decrease in activity, and an even greater
decrease is seen when the carboxylate is re-
placed by another moiety such as an amine or a
hydrophobic group. Incorporation of a second
carboxylate, as suggested by one of the binding
modes shown in Figure 5B, did show a small
increase in the potency of the inhibitor when
viewed in context (37 vs. 40), but it was not as
significant as that seen with the incorporation of
the first carboxylate.[25] This behavior, in con-
junction with previously published mutagenesis
results and enzyme-kinetic and inhibitory stud-
ies, suggests the possible modes of action of the
enzyme (Figures 9 and 10). While in the amide-
N
10
0.3
0.2
CO2H
Cl-
BnO
+NH3
OH
N
O
N.I.[a]
N.D.[a]
CO2H
39
+
N
CO2H
Cl-
H2
Cl-
BnO
BnO
+NH3
OH
N
O
CO2H
N.I.
0.5
N.D.
0.8
40
46
+
CO2H
N
Cl-
H2
Cl-
+
NH3
OH
N
O
CO2H
S
[a] N.I.: no inhibition; N.D.: not determined.
Table 5. Effects of hydroxamic acid 2 on LTB4 biosynthesis in intact
polymorphonuclear lymphocytes (PMNL). PMNL (20 Â 106 in 1 mL) were
preincubated with or without inhibitor for 10 min on ice followed by 15 min
at 378C and further incubated with 2mm ionophore A23187 Æ 30mm
arachidonic acid (AA) for 5 min at 378C. Reactions were quenched with
MeOH and samples processed and analyzed as described in the exper-
imental section. The all-trans isomers of LTB4 are D6-trans- and 12-epi-D6-
trans-LTB4. Formation of the respective compounds are expressed in % of
the control samples assayed in the absence of inhibitor.
% of the control
all-trans isomers LTB4
5-HETE
of LTB4
Control
100
60
0
100
0
0
100
0
0
[I] ˆ 100mm
[I] ˆ 500mm
Control ‡ AA
100
132
96
100
0
0
100
135
100
*
Figure 8. Effects of 6 ± 26 on the formation of LTB4
(
), and the
[I] ˆ 100mm ‡ AA
[I] ˆ 500mm ‡ AA
&
nonenzymatic hydrolysis products of LTA4 (all-trans isomers of LTB4,
)
in isolated human polymorphonuclear leukocytes.
1704
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1998
0947-6539/98/0409-1704 $ 17.50+.50/0
Chem. Eur. J. 1998, 4, No. 9

Leukotriene A4 Hydrolase
1698±1713
cleavage reaction, with the involve-
ment of a different general base and a
different general acid.
His299
His299
His295
His295
N
N
Glu296
Glu296
N
N
N
N
N
Glu318
-O
O
N
Glu318
O2N
HO
O
H
O
O
O
O
Experimental Section
O
H
O-
O2N
O
OH
NH2
CH3
NH2
CH3
General: The reagents used were commer-
cially available and used without further
purification. Methylene chloride (CH2Cl2)
was distilled from CaH2, and tetrahydrofuran
(THF) was distilled from sodium and benzo-
phenone prior to use. Anhydrous methanol
was purchased from Aldrich. All reactions
were run under an inert atmosphere of argon
unless otherwise noted. The HCl/acetic acid
solution used for Boc deprotections was made
by bubbling HCl gas through glacial acetic
acid for 5 min.
N
H
N
H
OH
OH
OH
H
O
Tyr378
Tyr378
Tyr383
Tyr383
Figure 9. Proposed mechanism of the amidase activity.
1H NMR chemical shifts (d) are reported
relative to tetramethylsilane (CDCl3), the
solvent peak at d ˆ 2.49 ([D6]DMSO), or the
solvent peak at d ˆ 3.30 ([D4]methanol). 13C
NMR spectra reported in CDCl3 are refer-
enced to the solvent peak at d ˆ 77.0, those in
[D4]methanol are referenced to the solvent
peak at d ˆ 49.0, and those in [D6]DMSO are
referenced to the solvent peak at d ˆ 39.5.
Thin-layer chromatography was performed on
silica-gel plates (0.25 mm, Merck) and flash
chromatography was performed on silica gel
(230 ± 400 mesh, Merck). The known reaction
of hydroxamic acids with ferric chloride to
give a red color was used systematically to
further confirm the presence of the hydrox-
amate moiety. All yields are unoptimized.
His299
His299
His295
His295
N
N
Glu296
Glu296
N
N
N
N
N
Glu318
-O
O
N
Glu318
-O
O
O
O
O
O
O
H2O
O
O-
O
O
+
O-
O-
OH
OH
H
OH
Tyr383
Tyr383
Tyr378
Tyr378
Inhibition studies of the peptidase activity of
LTA4 hydrolase: All assays were performed in
Tris-HCl buffer (50mm, pH 8.0) with l-alanyl-
p-nitroanilide (1.87mm) as substrate. LTA4
hydrolase (1.4 mg) purified from human leu-
kocytes was added for each assay (final
volume ˆ 1.0 mL, [E] ˆ 20nm). The rate of
formation of p-nitroaniline was spectrophoto-
metrically monitored at 405nm. The high
enzyme concentration as compared to inhib-
itor concentration ([E]t ꢀ [I]t) was accounted
for by using the appropriate kinetic equations
for tight-binding inhibitors and the Ki values
were determined using nonlinear regression
methods.[24]
A
His299
His295
N
Glu296
N
N
N
Glu318
-O
O
Tyr378
O
O
H
H
O
HO
O
Enz—B:
OH
HA—Enz
Inhibition studies of the epoxide hydrolase
activity of LTA4: The epoxide hydrolase
activity was determined from short-time
(15 s) incubations of enzyme (2.5 mg ˆ
360nm) and inhibitor (0.01 ± 10mm) dissolved
in DMSO (final conc ˆ 0.5%; v/v), in 2-[4-(2-
hydroxyethyl)-1-piperazinyl]ethanesulfonic
acid (HEPES; 50mm), pH 8, (100 mL) with
LTA4 (67mm) at room temperature. Reactions
were quenched with 2 vols of MeOH. PGB1
was added as internal standard, and samples
Tyr383
O-
O
B
Figure 10. Proposed mechanisms A and B for the epoxide hydrolase activity.
were extracted and analyzed by RP-HPLC, as previously described.[41]
Enzyme and inhibitor were preincubated for 45 min at room temperature
prior to activity determinations.
cleavage reaction, the zinc ion appears to be close to the
scissile bond, two possible modes of action are proposed for
the epoxide hydrolase activity. The active-site zinc atom may
act as a Lewis acid on the epoxide during LTA4 hydrolysis,
though no direct evidence of such a role has yet been found, or
it may activate the reacting water molecule, as in the amide
Preparation and incubation of granulocytes: Human granulocytes were
prepared from buffycoat by dextran sedimentation, centrifugation on
Lymfoprepꢂ, and hypotonic lysis of the remaining erythrocytes as
previously described.[42] The cells were resuspended at a concentration of
Chem. Eur. J. 1998, 4, No. 9
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1998
0947-6539/98/0409-1705 $ 17.50+.25/0
1705

C.-H. Wong et al.
FULL PAPER
1
20 Â 106 cellsmL in Dulbeccoꢁs phosphate-buffered saline, pH 7.4. Ali-
N-Hydroxy-N-[(2S)-N'-Boc-2-amino-3-phenylpropyl]-7-carboxymethyl-
nonanamide (7): Isobutyl chloroformate (73 mL, 0.56 mmol) was added to a
solution of sebacic acid monomethyl ester (121 mg, 0.56 mmol) and
triethylamine (78 mL, 0.56 mmol) in THF (5 mL) at 08C. The resulting
slurry was stirred for 30 min and then (2S)-2-N'-Boc-amino-3-(4-benzylox-
yphenyl)-N-hydroxypropylamine (140 mg, 0.37 mmol) in THF (5 mL) was
added. The reaction mixture was stirred for 30 min and then poured into
CH2Cl2 (50 mL), washed (1n HCl, saturated NaHCO3, saturated NaCl),
and dried (MgSO4). Purification by flash chromatography (1:3 EtOAc/
hexanes) and recrystallization (THF/hexanes) yielded compound 7 as a
white solid (87 mg, 41%). The major by-product was iso-butyl carbamate.
1H NMR (500 MHz, CDCl3, 258C, TMS): d ˆ 8.61 (s, 1H), 7.43 (d, J ˆ 7 Hz,
2H), 7.39 (t, J ˆ 7.5 Hz, 2H), 7.33 (t, J ˆ 7.0 Hz, 1H), 7.10 (d, J ˆ 8.5 Hz,
2H), 6.92 (d, J ˆ 8.5 Hz, 2H), 5.05 (s, 2H), 4.67 (m, 1H), 4.22 (dd, J ˆ 13.5
and 13.5 Hz, 1H), 4.13 (m, 1H), 3.66 (s, 3H), 3.03 (dd, J ˆ 13.5 and 2.5 Hz,
1H), 2.77 (m, 2H), 2.53 (m, 1H), 2.29 (m, 3H), 1.65 ± 1.52 (m, 4H), 1.39 (s,
9H), 1.28 (m, 8H); 13C NMR (125 MHz, CDCl3, 258C): d ˆ 174.8, 174.3,
158.2, 157.8, 136.9, 130.0, 128.6, 127.9, 127.4, 115.1, 80.9, 70.0, 51.4, 50.4, 48.1,
quots (1 mL) were preincubated with or without various concentrations of
inhibitor for 10 min on ice followed by 15 min at 378C prior to the addition
of A23187 (2mm) Æ arachidonic acid (30mM). After 5 min, the incubations
were quenched with 1 vol of MeOH and subjected to solid-phase extraction
and RP-HPLC. The eluate was monitored at 270 nm and 235 nm, for the
detection and quantitation of LTB4 and 5-hydroxyeicosatetraenoic acid (5-
HETE), respectively.
Cell culture: Whole blood was obtained from healthy volunteers by
venepuncture and collected into EDTA-treated Vacutainers (Becton
Dickinson, Rutherford, NJ). After mixing for 45 ± 50 min, the blood was
transferred to LeukoPREP tubes (Becton Dickinson, Lincoln Park, NJ)
and the peripheral blood mononuclear cells (PBMC) were separated
according to the manufacturerꢁs instructions. The isolated PBMC were
washed with two 15 mL volumes of Dulbeccoꢁs phosphate-buffered saline
(D-PBS) and the cell pellet was resuspended at a concentration of 2 Â 106
1
cellsmL in RPMI 1640 media supplemented with 10% fetal bovine
serum (FBS), 50mm betamercaptoethanol, 1mm sodium pyruvate,
50 mgmL gentamycin, and 0.1mm nonessential amino acids. Cells were
added to a 96-well plate precoated with OKT3 (Ortho Biotech, Raritan,
NJ). Plates were coated as previously described by Ritchie et al.[43] 100 mL
(2 Â 105 cells) of the cell suspension were added to each well followed by
the addition of sufficient phorbol myristate acetate (PMA) to achieve a
final concentration of 50 ngmL 1. Drugs or vehicles were then added to
cultures for a final well volume of 200 mL. Cultures were maintained in a
humid atmosphere of 95% air and 5% CO2 at 378C for 72 h. 6 h prior to
harvest, each well was pulsed with 1 mCi of 3H thymidine (6.7 Cimmol
New England Nuclear, Boston, MA). Cells were harvested onto glass-fiber
filters with a Tomtec-96 cell harvester (Tomtec, Orange CT) and the
radioactivity incorporated was measured with a Packard Matrix 9600 direct
beta counter (Packard, Meridian, CT).
‡
37.2, 34.1, 32.5, 29.4, 29.2, 29.1, 28.2, 24.9, 24.6; HRMS [M‡Cs] calcd for
1
C32H46N2O7Cs: 703.2359, found: 703.2344.
N-Hydroxy-N-[(2S)-2-amino-3-(4-benzyloxyphenyl)propyl]-7-carboxyme-
thylheptanamide (8): Hydroxamate 6 (20 mg, 0.037 mmol) was treated with
a solution of HCl/acetic acid (1 mL) for 30 min. The reaction mixture was
diluted with toluene (10 mL) and the solvent removed from the resulting
mixture. The resulting solid was rinsed with ether and then azeotroped with
toluene to give 8 as a white solid (17.0 mg, 96%). 1H NMR (500 MHz,
[D6]DMSO, 258C): d ˆ 10.11 (s, 1H), 8.11 (brs, 3H), 7.43 (d, J ˆ 7.0 Hz,
2H), 7.39 (t, J ˆ 7.0 Hz, 2H), 7.33 (t, J ˆ 7.0 Hz, 1H), 7.21 (d, J ˆ 8.5 Hz,
2H), 6.97 (d, J ˆ 8.5 Hz, 2H), 5.07 (s, 2H), 3.81 (dd, J ˆ 8.0 and 14.5 Hz,
1H), 3.56 (s, 3H), 3.55 (m, 1H), 3.47 (dd, J ˆ 4.0 and 14.0 Hz, 1H), 2.89 (dd,
J ˆ 5.5 and 14.0 Hz, 1H), 2.75 (dd, J ˆ 8.5 and 14.0 Hz, 1H), 2.36 (t, J ˆ
7.0 Hz, 2H), 2.27 (t, J ˆ 7.5 Hz, 2H), 1.47 (m, 4H), 1.24 (m, 4H); 13C NMR
(125 MHz, [D6]DMSO, 258C): d ˆ 174.3, 173.4, 157.4, 137.1, 130.5, 128.5,
128.0, 127.9, 127.7, 114.9, 69.2, 51.2, 50.2, 48.7, 35.1, 33.2, 31.7, 28.5, 28.3, 24.3,
1
,
Relative binding model generation: The models of the binding orientation
of the substrate LTA4 relative to the inhibitors discussed here were
obtained as a result of a combination of model overlay with plastic
molecular models (Maruzen, Japan) augmented by computational molec-
ular modeling and minimization techniques using Insight95 and Discover
‡
23.7; HRMS [M‡Cs] calcd for C25H34N2O5Cs: 575.1522, found: 575.1538.
N-Hydroxy-N-[(2S)-2-amino-3-(4-benzyloxyphenyl)propyl]-9-carboxyme-
thylnonanamide (9): Hydroxamate 7 (14.1 mg, 0.025 mmol) was treated
with a solution of HCl/acetic acid (1 mL) for 30 min. The reaction mixture
was diluted with toluene (10 mL) and the solvent removed from the
resulting mixture. The resulting solid was rinsed with ether and then
azeotroped with toluene to give 9 as a white solid (12.0 mg, 97%). 1H NMR
(500 MHz, [D6]DMSO, 258C): d ˆ 10.08 (s, 1H), 8.08 (brs, 3H), 7.43 (d, J ˆ
7.0 Hz, 2H), 7.39 (t, J ˆ 7.0 Hz, 2H), 7.33 (t, J ˆ 7.0 Hz, 1H), 7.21 (d, J ˆ
8.5 Hz, 2H), 6.97 (d, J ˆ 8.5 Hz, 2H), 5.07 (s, 2H), 3.81 (dd, J ˆ 8.0 and
14.5 Hz, 1H), 3.56 (s, 3H), 3.54 (m, 1H), 3.47 (dd, J ˆ 4.5 and 14.5 Hz, 1H),
2.88 (dd, J ˆ 5.5 and 14.0 Hz, 1H), 2.75 (dd, J ˆ 8.0 and 14.0 Hz, 1H), 2.36
2.98 (MSI) on
computational experiments revealed that the compounds discussed in this
work were found to exhibit large number of conformers which
a Silicon Graphics Octane workstation. Preliminary
a
corresponded to local energetic minima. In each proposed model, it was
verified computationally that conformers similar to those shown in
Schemes 4, 5, 6, and 7 were energetically reasonable. The compounds
were aligned in such a way that portions of the molecule posited to interact
with the zinc atom were in spacially similar areas. When possible,
hydrophobic regions of inhibitors were paired with hydrophobic regions
of the substrate, transition-state complex and hydrogen-bond acceptors and
charged regions on the inhibitor were paired with similar moieties in the
substrate transition-state complex.
(t, J ˆ 7.0 Hz, 2H), 2.26 (t, J ˆ 7.0 Hz, 2H), 1.47 (m, 4H), 1.22 (m, 8H); 13
C
NMR (125 MHz, [D6]DMSO, 258C): d ˆ 174.4, 173.4, 157.4, 137.1, 130.5,
128.5, 128.0, 127.9, 127.7, 114.9, 69.2, 51.2, 50.2, 48.7, 35.1, 33.3, 31.7, 28.8,
(2S)-2-N'-Boc-amino-3-(4-benzyloxyphenyl)-N-hydroxypropylamine: This
‡
28.6, 28.5, 24.4, 23.9; HRMS [M‡Cs] calcd for C27H38N2O5Cs: 603.1835,
compound was synthesized as previously described.[25]
found: 603.1855.
N-Hydroxy-N-(2S)-N'-Boc-2-amino-3-phenylpropyl-7-carboxymethylhep-
tanamide (6): Isobutyl chloroformate (76 mL, 0.59 mmol) was added to a
solution of suberic acid monomethyl ester (128 mL, 0.71 mmol), and
triethylamine (104 mL, 0.59 mmol) in THF (10 mL) at 08C. The resulting
slurry was stirred for 30 min and then (2S)-2-N'-Boc-amino-3-(4-benzyl-
oxyphenyl)-N-hydroxypropylamine (200 mg, 0.54 mmol) was added. The
cooling bath was removed and the reaction mixture stirred for 30 min. The
reaction mixture was then poured into CH2Cl2 (100 mL), washed (1n HCl,
saturated NaHCO3, saturated NaCl), and dried (MgSO4). Purification by
flash chromatography (1:3 EtOAc/hexanes) and recrystallization (THF/
hexanes) yielded compound 6 as a white solid (217 mg, 74%). 1H NMR
(500 MHz, CDCl3, 258C, TMS): d ˆ 8.61 (s, 1H), 7.43 (d, J ˆ 7 Hz, 2H), 7.39
(t, J ˆ 7.5 Hz, 2H), 7.33 (t, J ˆ 7.0 Hz, 1H), 7.10 (d, J ˆ 8.5 Hz, 2H), 6.93 (d,
J ˆ 8.5 Hz, 2H), 5.05 (s, 2H), 4.65 (d, J ˆ 9 Hz, 1H), 4.22 (dd, J ˆ 13.5 and
13.5 Hz, 1H), 4.13 (m, 1H), 3.66 (s, 3H), 3.03 (dd, J ˆ 13.5 and 2.5 Hz, 1H),
2.77 (m, 2H), 2.53 (m, 1H), 2.29 (m, 3H), 1.65 ± 1.52 (m, 4H), 1.39 (s, 9H),
1.32 (m, 4H); 13C NMR (125 MHz, CDCl3, 258C): d ˆ 174.7, 174.2, 158.2,
157.8, 136.9, 130.1, 128.6, 128.0, 127.4, 115.1, 80.9, 70.0, 51.4, 50.4, 48.1, 37.3,
N-Hydroxy-N-[(2S)-2-amino-3-(4-benzyloxyphenyl)propyl]-7-carboxy-
heptanamide (10): LiOH (1.0m, 2.5 mL) was added to hydroxamate 6
(50 mg, 0.092 mmol) in THF/MeOH (2:1, 7.5 mL), and the biphasic mixture
was stirred vigorously for 30 min. The mixture was poured into HCl (1n,
20 mL) and extracted with EtOAc. The organic layer was dried (MgSO4)
and the solvent removed to give a pale yellow oil (46 mg, 94%). The oil
(16 mg, 0.030 mmol) was then treated with a solution of HCl/acetic acid
(1 mL) for 30 min. The reaction mixture was diluted with toluene (10 mL)
and the solvent removed from the resulting mixture. The resulting solid was
rinsed with ether and then azeotroped with toluene to give an oil. CH2Cl2
(5 mL) was added and 10 precipitated out as a white solid (12.3 mg, 87%).
1H NMR (500 MHz, [D6]DMSO, 258C): d ˆ 11.87 (brs, 1H), 10.07 (s, 1H),
8.06 (brs, 3H), 7.43 (d, J ˆ 7.0 Hz, 2H), 7.39 (t, J ˆ 7.0 Hz, 2H), 7.33 (t, J ˆ
7.0 Hz, 1H), 7.21 (d, J ˆ 8.5 Hz, 2H), 6.97 (d, J ˆ 8.5 Hz, 2H), 5.07 (s, 2H),
3.81 (dd, J ˆ 8.0 and 14.0 Hz, 1H), 3.53 (m, 1H), 3.46 (dd, J ˆ 4.0 and
14.0 Hz, 1H), 2.87 (dd, J ˆ 5.5 and 13.5 Hz, 1H), 2.75 (dd, J ˆ 8.0 and
14.0 Hz, 1H), 2.36 (t, J ˆ 7.0 Hz, 2H), 2.17 (t, J ˆ 7.5 Hz, 2H), 1.46 (m, 4H),
1.24 (m, 4H); 13C NMR (125 MHz, [D6]DMSO, 258C): d ˆ 174.5, 174.3,
157.4, 137.1, 130.5, 128.5, 128.0, 127.9, 127.7, 114.9, 69.2, 50.2, 48.7, 35.1, 33.6,
‡
34.0, 32.4, 29.0, 28.9, 28.2, 24.8, 24.4; HRMS [M‡Cs] calcd for
C30H42N2O7Cs: 675.2046, found: 675.2058.
1706
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1998
0947-6539/98/0409-1706 $ 17.50+.50/0
Chem. Eur. J. 1998, 4, No. 9

Leukotriene A4 Hydrolase
1698±1713
‡
‡
31.7, 28.5, 28.4, 24.4, 23.8; HRMS [M‡Cs] calcd for C24H32N2O5Cs:
26.6, 20.8; HRMS [M‡Cs] calcd for C21H29N3O3Cs: 504.1263, found:
561.1366, found: 561.1379.
504.1277.
N-Hydroxy-N-[(2S)-2-amino-3-phenylpropyl]-9-carboxynonanamide (11):
LiOH (1.0m, 2.0 mL) was added to hydroxamate 7 (50 mg, 0.088 mmol) in
THF/MeOH (2:1, 6.0 mL), and the biphasic mixture was stirred vigorously
for 30 min. The mixture was poured into HCl (1n, 30 mL) and extracted
with EtOAc. The organic layer was dried (MgSO4) and the solvent removed
to give a pale yellow oil (49 mg, 99%). The oil (16 mg, 0.029 mmol) was
then treated with a solution of HCl/acetic acid (1 mL) for 30 min. The
reaction mixture was diluted with toluene (10 mL) and the solvent removed
from the resulting mixture. The resulting solid was rinsed with ether and
then azeotroped with toluene to give 11 as a white solid (13.4 mg, 94%). 1H
NMR (500 MHz, [D6]DMSO, 258C): d ˆ 11.96 (brs, 1H), 10.08 (s, 1H), 8.08
(brs, 3H), 7.43 (d, J ˆ 7.0 Hz, 2H), 7.39 (t, J ˆ 7.0 Hz, 2H), 7.33 (t, J ˆ 7.0 Hz,
1H), 7.21 (d, J ˆ 8.5 Hz, 2H), 6.97 (d, J ˆ 8.5 Hz, 2H), 5.07 (s, 2H), 3.81 (dd,
J ˆ 8.0 and 14.0 Hz, 1H), 3.53 (m, 1H), 3.47 (dd, J ˆ 4.5 and 14.5 Hz, 1H),
2.88 (dd, J ˆ 5.5 and 14.0 Hz, 1H), 2.75 (dd, J ˆ 8.0 and 14.0 Hz, 1H), 2.35
Benzyl 4-pentenoate: 4-Pentenoic acid (2.0 g, 20 mmol), benzyl bromide
(2.8 mL, 24 mmol), anhydrous potassium carbonate (13.8 g, 100 mmol),
and tetrabutylammonium iodide (500 mg) were combined in anhydrous
acetone (40 mL) and stirred overnight at room temperature. The reaction
mixture was filtered and the solvent removed. The residue was taken up in
EtOAc, washed (1n HCl, saturated NaHCO3, saturated NaCl) and dried
(MgSO4). Purification by flash chromatography (1:30 EtOAc/hexanes then
1:1 EtOAc/hexanes) gave compound benzyl 4-pentenoate as a yellow liquid
(3.6 g, 96%). 1H NMR (500 MHz, CDCl3, 258C): d ˆ 7.36 (m, 5H), 5.84
(ddt, J ˆ 6.0, 10.5 and 17 Hz, 1H), 5.12 (s, 2H), 5.05 (ddt, J ˆ 17, 1.5, and
1.5 Hz, 2H), 5.00 (ddt, J ˆ 10, 1.5, and 1.5 Hz, 2H), 2.47 (m, 2H), 2.40 (m,
2H); 13C NMR (125 MHz, CDCl3, 258C): d ˆ 172.9, 159.8, 136.5, 135.9,
‡
128.5, 128.2, 115.5, 66.2, 33.5, 28.8; HRMS [M‡H] calcd for C12H15O2:
191.1071, found: 191.1062.
(t, J ˆ 7.0 Hz, 2H), 2.17 (t, J ˆ 7.0 Hz, 2H), 1.46 (m, 4H), 1.23 (m, 8H); 13
C
Benzyl 4-oxobutyrate (16): Ozone was bubbled through a solution of
benzyl 4-pentenoate (3.6 g, 19.0 mmol) in a 3:1 mixture of CH2Cl2 and
MeOH (100 mL) at 788C until a blue color persisted. Oxygen was then
bubbled through until the solution was colorless again and then excess
dimethyl sulfide (4.4 mL) was added. The reaction mixture was stirred for
30 min, before NaHCO3 (5%, 100 mL) was added and the cooling bath
removed. The layers were separated and the aqueous layer extracted with
CH2Cl2. TLC analysis showed two overlapping spots. After leaving to stand
in CH2Cl2 for 2 days, TLC analysis showed only one spot, namely the
aldehyde, suggesting that the other spot was the ozonide. Purification by
flash chromatography (1:3 EtOAc/hexanes) yielded pure aldehyde 16
(3.36 g, 92%). 1H NMR (500 MHz, CDCl3, 258C): d ˆ 9.80 (s, 1H), 7.35 (m,
5H), 5.13 (s, 2H), 2.81 (t, J ˆ 6.5 Hz, 2H), 2.68 (t, J ˆ 6.5 Hz, 2H); 13C
NMR (125 MHz, CDCl3, 258C): d ˆ 199.9, 172.0, 135.6, 128.5, 128.2, 128.2,
NMR (125 MHz, [D6]DMSO, 258C): d ˆ 174.5, 174.3, 157.4, 137.1, 130.5,
128.5, 128.0, 127.9, 127.7, 114.9, 69.2, 50.2, 48.7, 35.1, 33.7, 31.8, 28.8, 28.7,
‡
28.6, 24.5, 23.9; HRMS [M‡Cs] calcd for C26H36N2O5Cs: 589.1679, found:
589.1695.
N-Hydroxy-N-[(2S)-N'-Boc-2-amino-3-(4-benzyloxyphenyl)propyl]-5-
phenylpentanamide (12): Isobutyl chloroformate (76 mL, 0.59 mmol) was
added to a solution of 5-phenylvaleric acid (127 mg, 0.71 mmol) and
triethylamine (104 mL, 0.75 mmol) in THF (10 mL) at 08C. The resulting
slurry was stirred for 30 min and then (2S)-2-N'-Boc-amino-3-(4-benzylox-
yphenyl)-N-hydroxypropylamine (200 mg, 0.54 mmol) was added. The
cooling bath was removed and the reaction mixture was stirred for 30 min.
The mixture was then poured into CH2Cl2 (100 mL), washed (1n HCl,
saturated NaHCO3, saturated NaCl), and dried (MgSO4). Purification by
flash chromatography (1:3 EtOAc/hexanes) and recrystallization (THF/
‡
66.6, 38.4, 26.5; HRMS [M‡H] calcd for C11H13O3: 193.0865, found:
193.0860.
1
hexanes) yielded compound 12 as a white solid (172 mg, 60%). H NMR
(500 MHz, CDCl3, 258C): d ˆ 8.62 (s, 1H), 7.43 (d, J ˆ 7 Hz, 2H), 7.39 (t,
J ˆ 7.5 Hz, 2H), 7.33 (t, J ˆ 7.0 Hz, 1H), 7.25 (dd, J ˆ 6.5 and 7.5 Hz, 2H),
7.16 (d, J ˆ 7 Hz, 3H), 7.10 (d, J ˆ 8.5 Hz, 2H), 6.92 (d, J ˆ 8.5 Hz, 2H), 5.05
(s, 2H), 4.65 (m, 1H), 4.22 (dd, J ˆ 13.5 and 13.5 Hz, 1H), 4.13 (m, 1H),
3.03 (dd, J ˆ 13.5 and 2.5 Hz, 1H), 2.77 (m, 2H), 2.61 (m, 2H), 2.57 (m,
1H), 2.32 (m, 1H), 1.64 (m, 4H), 1.36 (s, 9H); 13C NMR (125 MHz, CDCl3,
258C): d ˆ 174.6, 158.2, 157.8, 142.4, 136.9, 130.1, 128.6, 128.4, 128.2, 128.0,
127.5, 125.6, 115.2, 80.9, 70.0, 50.5, 48.1, 37.3, 35.7, 32.3, 31.3, 28.2, 24.4;
Benzyl N-Boc-N-[(1S)-1-carboxymethylethyl]-4-aminobutyrate (17):
A
solution of l-alanine methyl ester hydrogen chloride (394 mg, 2.8 mmol)
in CH2Cl2 (10 mL) was treated with triethylamine (390 mL, 2.8 mmol) and
the resulting salts were filtered off to give a solution of the free amine. This
was added to a solution of aldehyde 16 (500 mg, 2.6 mmol) in CH2Cl2
(20 mL), and MgSO4 (500 mg) was added. The mixture was stirred at room
temperature for 2 h. The reaction mixture was then filtered and the solvent
removed to give the imine. This was taken up in MeOH (20 mL, 08C) and
sodium borohydride (295 mg, 6.5 mmol) was then added in one portion and
the resulting mixture was stirred for 30 min. Saturated NH4Cl (20 mL) was
added and the methanol removed in vacuo. CH2Cl2 (40 mL) was added and
the layers separated. The aqueous layer was extracted twice more with
CH2Cl2 and the combined organic layers washed (saturated NaHCO3,
saturated NaCl) and dried (MgSO4). Removal of the solvent gave the
secondary amine (0.65 g, 88%) which was used without further purifica-
tion.
‡
HRMS [M‡Cs] calcd for C32H40N2O5Cs: 665.1992, found: 665.1979.
N-Hydroxy-N-[(2S)-2-amino-3-(4-benzyloxyphenyl)propyl]-5-phenylpen-
tanamide (13): Hydroxamate 12 (20 mg, 0.038 mmol) was treated with a
solution of HCl/acetic acid (1 mL) for 30 min. The reaction mixture was
diluted with toluene (10 mL) and the solvent removed from the resulting
mixture. The resulting solid was rinsed with ether and then dissolved in
CH2Cl2. Removal of the solvent gave 13 as a white solid (17.6 mg, 99%). 1H
NMR (500 MHz, CDCl3, 258C): d ˆ 10.09 (s, 1H), 8.07 (brs, 3H), 7.43 (d,
J ˆ 7.0 Hz, 2H), 7.36 (t, J ˆ 7.0 Hz, 2H), 7.32 (t, J ˆ 7.0 Hz, 1H), 7.25 (t, J ˆ
7.0 Hz, 2H), 7.21 ± 7.13 (m, 3H), 7.19 (d, J ˆ 8.5 Hz, 2H), 6.96 (d, J ˆ 8.5 Hz,
2H), 5.06 (s, 2H), 3.80 (dd, J ˆ 7.0 and 14.5 Hz, 1H), 3.52 (br, 1H), 3.45 (dd,
J ˆ 4.5 and 14.5 Hz, 1H), 2.87 (dd, J ˆ 4.5 and 14.0 Hz, 1H), 2.75 (dd, J ˆ
8.0 and 14.0 Hz, 1H), 2.55 (t, J ˆ 7.0 Hz, 2H), 2.40 (brt, J ˆ 7.0 Hz, 2H), 1.51
(m, 4H); 13C NMR (125 MHz, [D6]DMSO, 258C): d ˆ 174.2, 157.4, 142.1,
137.1, 130.5, 128.5, 128.4, 128.3, 128.2, 128.0, 127.9, 127.7, 125.6, 114.9, 69.2,
Di-tert-butyl pyrocarbonate (550 mg, 2.53 mmol) was added to a solution of
the amine (0.65 g, 2.3 mmol) in dioxane (2 mL), THF (2 mL), and water
(2 mL) under argon at 08C and stirring was continued at room temperature.
The reaction was complete after 30 min (TLC, 1:1 EtOAc/hexanes). Water
(10 mL) was added and the product was extracted with EtOAc. The organic
layer was dried (MgSO4) and the solvent removed in vacuo to give
compound 17 as a colorless oil (784 mg, 90%). This product was used
without further purification.
‡
50.2, 48.7, 41.9, 35.11, 35.0, 31.6, 30.7, 23.6; HRMS [M‡Cs] calcd for
C27H32N2O3Cs: 565.1467, found: 565.1478.
N-Hydroxy-N-[(2S)-2-amino-3-(4-benzyloxyphenyl)propyl]-5-aminopen-
tanamide (15): Hydroxamate 14 (19.1 mg, 0.033 mmol) was treated with a
solution of HCl/acetic acid (1 mL) for 30 min. The reaction mixture was
diluted with toluene (10 mL) and the solvent removed from the resulting
mixture. The resulting solid was rinsed with ether and then dissolved in
CH2Cl2. Removal of the solvent gave 15 as a white solid (14.7 mg, 99%). 1H
NMR (500 MHz, [D6]DMSO, 258C): d ˆ 10.25 (s, 1H), 8.19 (brs, 3H), 7.93
(brs, 3H), 7.43 (d, J ˆ 7.5 Hz, 2H), 7.39 (t, J ˆ 7.5 Hz, 2H), 7.33 (t, J ˆ 7.5 Hz,
1H), 7.21 (d, J ˆ 8.5 Hz, 2H), 6.96 (d, J ˆ 8.5 Hz, 2H), 5.07 (s, 2H), 3.81 (dd,
J ˆ 7.0 and 14.5 Hz, 1H), 3.53 (brm, 1H), 3.49 (dd, J ˆ 4.5 and 14.5 Hz, 1H),
2.90 (dd, J ˆ 4.5 and 14.0 Hz, 1H), 2.76 (m, 1H), 2.74 (m, 2H), 2.42 (brt,
2H), 1.53 (m, 4H); 13C NMR (125 MHz, [D6]DMSO, 258C): d ˆ 173.9,
157.4, 137.1, 128.5, 128.1 127.9, 127.7, 114.9, 69.2, 50.2, 48.6, 38.6, 35.1, 31.2,
Benzyl N-Boc-N-[(1R)-1-carboxymethylethyl]-4-aminobutyrate (18): A
solution of d-alanine methyl ester hydrogen chloride (394 mg, 2.8 mmol) in
CH2Cl2 (10 mL) was treated with triethylamine (390 mL, 2.8 mmol) and the
resulting salts were filtered off to give a solution of the free amine. This was
added to a solution of aldehyde 16 (500 mg, 2.6 mmol) in CH2Cl2 (20 mL),
and MgSO4 (500 mg) was added. The mixture was stirred at room
temperature for 2 h and then filtered, and the solvent removed to give
the imine. This was taken directly up in MeOH (20 mL) at 08C, and sodium
borohydride (295 mg, 6.5 mmol) was added in one portion and the mixture
was stirred for 30 min. Saturated NH4Cl (20 mL) was added and the
methanol removed in vacuo. CH2Cl2 (40 mL) was added and the layers
separated. The aqueous layer was extracted twice more with CH2Cl2 and
the combined organic layers washed (saturated NaHCO3, saturated NaCl)
Chem. Eur. J. 1998, 4, No. 9
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1998
0947-6539/98/0409-1707 $ 17.50+.25/0
1707

C.-H. Wong et al.
FULL PAPER
and dried (MgSO4). Removal of the solvent gave the secondary amine
(0.65 g, 88%) which was used without further purification.
Benzyl N-Boc-N-[(1S)-1-carboxymethyl-5-N'-Boc-aminopentyl]-4-amino-
butyrate (21): A solution of e-N-Boc-l-lysine methyl ester hydrogen
chloride (775 mg, 2.6 mmol) in CH2Cl2 (20 mL) was treated with triethyl-
amine (365 mL, 2.6 mmol) and the resulting salts were filtered off to give a
solution of the free amine. This was added to a solution of aldehyde 16
(500 mg, 2.6 mmol) in CH2Cl2 (20 mL), and MgSO4 (500 mg) was added.
The mixture was stirred at room temperature for 2 h. The reaction mixture
was then filtered and the solvent removed to give the imine. This was taken
up directly in MeOH (10 mL) at 08C, sodium borohydride (295 mg,
6.5 mmol) was added in one portion, and the reaction mixture stirred for
30 min. The resulting mixture was worked up as for 18 to yield the
secondary amine which was used immediately without further purification.
Di-tert-butyl pyrocarbonate (550 mg, 2.53 mmol) was added to a solution of
the amine (0.65 g, 2.3 mmol) in CH2Cl2 (10 mL) under argon at 08C and
stirring was continued overnight at room temperature. The solvent was
removed in vacuo and purification by flash chromatography gave
compound 18 as a colorless oil (680 mg, 83%). This compound exists as
two conformers in CDCl3. 1H NMR (500 MHz, CDCl3, 258C): d ˆ 7.32 (m,
5H), 5.12 (s, 2H), 4.39 (m, 0.5H), 3.93 (m, 0.5H), 3.69 (s, 3H), 3.43 (m,
0.5H), 3.34 (m, 0.5H), 3.17 (m, 0.5H), 3.08 (m, 0.5H), 2.39 (m, 2H), 1.91
(m, 2H), 1.44 (d, J ˆ 7.0 Hz, 3H) 1.42 (s, 4H), 1.41 (s, 5H); 13C NMR
(125 MHz, CDCl3, 258C): d ˆ 173.1, 172.9, 172.6, 155.4, 154.8, 135.9, 128.5,
128.2, 80.4, 80.4, 66.2, 55.9, 54.7, 53.4, 52.0, 46.5, 45.4, 31.4, 31.2, 28.3, 24.9,
Di-tert-butyl pyrocarbonate (625 mg, 2.89 mmol) was added to a solution of
the amine in CH2Cl2 (10 mL) under argon at 08C and stirring was continued
overnight at room temperature. Additional di-tert-butyl pyrocarbonate
(200 mg) was added and stirring continued for 24 h. The solvent was
removed in vacuo and purification by flash chromatography gave
compound 21 as a colorless oil (1.28 g, 83%). This compound exists as
two conformers in CDCl3. 1H NMR (500 MHz, CDCl3, 258C): d ˆ 7.35 (m,
5H), 5.12 (s, 2H), 4.58 (br, 1H), 4.33 (m, 0.5H), 3.85 (br, 0.5H), 3.68 (s,
3H), 3.49 (br, 0.5H), 3.29 (br, 0.5H), 3.11 (m, 2H), 3.05 (m, 1H), 2.40 (m,
2H), 2.1 ± 1.7 (m, 4H), 1.49 (m, 2H), 1.45 (s, 5H), 1.44 (s, 9H), 1.40 (s, 4H),
1.35 (m, 2H); 13C NMR (125 MHz, CDCl3, 258C): d ˆ 172.9, 172.1, 155.9,
128.5, 128.2, 80.5, 79.1, 66.2, 60.4, 59.0, 52.0, 47.3, 45.7, 40.3, 31.6, 31.3, 30.0,
‡
24.0, 15.9, 15.3; HRMS [M‡H] calcd for C20H30NO6: 380.2073, found:
380.2079.
Benzyl N-Boc-N-[(1S)-1,3-dicarboxymethylpropyl]-4-aminobutyrate (19):
A solution of l-glutamine dimethyl ester hydrogen chloride (548 mg,
2.6 mmol) in CH2Cl2 (20 mL) was treated with triethylamine (365 mL,
2.6 mmol) and the resulting salts were filtered off to give a solution of the
free amine. This was added to a solution of aldehyde 16 (500 mg, 2.6 mmol)
in CH2Cl2 (20 mL), and MgSO4 (500 mg) was added. The mixture was
stirred at room temperature for 2 h. The reaction mixture was then filtered
and the solvent removed to give the imine. This was taken directly up in
MeOH (10 mL) at 08C, sodium borohydride (295 mg, 6.5 mmol) was added
in one portion, and the reaction mixture stirred for 30 min. The reaction
mixture was worked up as for 18 to yield the secondary amine (0.75 g, 82%)
which was used immediately without further purification.
‡
29.8, 29.5, 29.0, 28.4, 28.3, 24.6, 23.8, 23.5; HRMS [M‡Cs] calcd for
C28H44N2O8Cs: 669.2152, found: 669.2166.
N-Boc-N-[(1S)-1-carboxymethylethyl]-4-aminobutyric acid (22): A solu-
tion of benzyl ester 17 (660 mg, 1.74 mmol) in EtOAc (10 mL) was stirred
vigorously in the presence of Pd/C (10% on carbon, 30 mg) under a
hydrogen atmosphere until no ester remained (2 h). The hydrogen was then
evacuated and replaced with argon. The reaction mixture was filtered
through Celite and the solvent removed to give 22 as a colorless oil (499 mg,
99%) which was used without further purification. This compound exists as
two conformers in CDCl3. 1H NMR (500 MHz, CDCl3, 258C): d ˆ 4.15 (m,
0.5H), 3.96 (m, 0.5H), 3.72 (s, 3H), 3.50 (m, 0.5H), 3.38 (m, 0.5H), 3.20 (m,
0.5H), 3.11 (m, 0.5H), 3.03 (m, 1H), 2.42 ± 2.33 (m, 4H), 2.00 ± 1.78 (m,
4H), 1.64 (m, 2H), 1.44 (s, 4H), 1.40 (s, 5H); 13C NMR (125 MHz, CDCl3,
258C): d ˆ 178.6, 178.2, 172.9, 172.5, 155.1, 80.8, 80.4, 55.9, 54.7, 52.1, 46.3,
Di-tert-butyl pyrocarbonate (545 mg, 2.5 mmol) was added to a solution of
the amine (0.75 g, 2.14 mmol) in CH2Cl2 (5 mL) under argon at 08 C and
stirring was continued overnight at room temperature. Additional di-tert-
butyl pyrocarbonate (150 mg, 0.69 mmol) was added and stirring continued
for 24 h. The solvent was removed in vacuo and purification by flash
chromatography gave compound 19 as a colorless oil (844 mg, 72% from
the aldehyde). This compound exists as two conformers in CDCl3. 1H NMR
(500 MHz, CDCl3, 258C): d ˆ 7.35 (m, 5H), 5.12 (s, 2H), 4.23 (br, 0.5H),
3.92 (br, 0.5H), 3.69 (s, 3H), 3.68 (s, 3H), 3.52 (br, 0.5H), 3.34 (br, 0.5H),
3.03 (m, 1H), 2.42 ± 2.33 (m, 5H), 2.09 (m, 1H), 1.90 (m, 2H), 1.44 (s, 4H),
1.40 (s, 5H); 13C NMR (125 MHz, CDCl3, 258C): d ˆ 173.3, 173.0, 172.6,
171.6, 155.4, 154.7, 136.0, 128.5, 128.2, 80.7, 80.5, 66.2, 59.5, 58.8, 52.1, 51.7,
‡
45.2, 31.1, 28.2, 24.6, 24.0, 15.8, 15.3; HRMS [M‡Cs] calcd for
C13H23NO6Cs: 422.0580, found: 422.0590.
‡
47.8, 46.7, 31.4, 31.3, 30.4, 28.2, 25.5, 24.7, 24.3, 23.6; HRMS [M‡Cs] calcd
N-Boc-N-[(1R)-1-carboxymethylethyl]-4-aminobutyric acid (23): A solu-
tion of benzyl ester 18 (660 mg, 1.74 mmol) in EtOAc (10 mL) was stirred
vigorously in the presence of Pd/C (10% on carbon, 30 mg) under a
hydrogen atmosphere until no ester remained (2 h). The hydrogen was then
evacuated and replaced with argon. The reaction mixture was filtered
through Celite and the solvent removed to give 23 as a colorless oil (499 mg,
99%) which was used without further purification. This compound exists as
two conformers in CDCl3. 1H NMR (500 MHz, CDCl3, 258C): d ˆ 4.15 (m,
0.5H), 3.96 (m, 0.5H), 3.72 (s, 3H), 3.50 (m, 0.5H), 3.38 (m, 0.5H), 3.20 (m,
0.5H), 3.11 (m, 0.5H), 3.03 (m, 1H), 2.42 ± 2.33 (m, 4H), 2.00 ± 1.78 (m,
4H), 1.64 (m, 2H), 1.44 (s, 4H), 1.40 (s, 5H); 13C NMR (125 MHz, CDCl3,
258C): d ˆ 178.6, 178.2, 172.9, 172.5, 155.1, 80.8, 80.4, 55.9, 54.7, 52.1, 46.3,
for C23H33NO8Cs: 584.1261, found: 584.1277.
Benzyl N-Boc-N-[(1S)-1,4-dicarboxymethylbutyl]-4-aminobutyrate (20):
A solution of 2-l-aminoadipic acid dimethyl ester hydrogen chloride
(450 mg, 2.0 mmol) in CH2Cl2 (20 mL) was treated with triethylamine
(300 mL, 2.16 mmol) and the resulting salts were filtered off to give a
solution of the free amine. This was added to a solution of aldehyde 16
(500 mg, 2.6 mmol) in CH2Cl2 (20 mL) and MgSO4 (500 mg) was added.
The mixture was stirred at room temperature for 4 h. The reaction mixture
was then filtered and the solvent removed to give the imine. This was taken
directly up in MeOH (10 mL) at 08C), sodium borohydride (295 mg,
6.5 mmol) was added in one portion, and the reaction mixture stirred for
30 min. The resulting mixture was worked up as for 18 to yield the
secondary amine (0.72 g, 98%) which was used immediately without
further purification.
‡
45.2, 31.1, 28.2, 24.6, 24.0, 15.8, 15.3; HRMS [M‡Cs] calcd for
C13H23NO6Cs: 422.0580, found: 422.0590.
N-Boc-N-[(1S)-1,3-dicarboxymethylpropyl]-4-aminobutyric acid (24): A
solution of benzyl ester 19 (820 mg, 1.8 mmol) in EtOAc (15 mL) was
stirred vigorously in the presence of Pd/C (10% on carbon, 35 mg) under a
hydrogen atmosphere until no ester remained (2 h). The hydrogen was then
evacuated and replaced with argon. The reaction mixture was filtered
through Celite and the solvent removed. Purification by flash chromatog-
raphy (1:1 EtOAc/hexanes then EtOAc) yielded 24 as a colorless oil
(494 mg, 76%). This compound exists as two conformers in CDCl3. 1H
NMR (500 MHz, CDCl3, 258C): d ˆ 4.24 (m, 0.5H), 3.95 (m, 0.5H), 3.73 (s,
3H), 3.69 (s, 3H), 3.57 (m, 0.5H), 3.38 (m, 0.5H), 3.06 (m, 1H), 2.47 ± 2.36
(m, 5H), 2.11 (m, 1H), 1.89 (m, 2H), 1.46 (s, 4H), 1.41 (s, 5H); 13C NMR
(125 MHz, CDCl3, 258C): 178.8, 178.5, 173.4, 171.7, 171.6, 155.4, 155.0, 81.0,
80.6, 59.5, 58.8, 52.2, 51.7, 47.7, 46.6, 31.2, 30.5, 30.4, 28.2, 25.5, 24.7, 24.0,
Di-tert-butyl pyrocarbonate (500 mg, 2.3 mmol) was added to a solution of
the crude amine (0.72 g, 2.0 mmol) in CH2Cl2 (5 mL) under argon at 08C
and stirring was continued overnight at room temperature. Additional di-
tert-butyl pyrocarbonate (200 mg, 0.92 mmol) was added and stirring
continued for 48 h. The solvent was removed in vacuo and purification by
flash chromatography gave compound 20 as a colorless oil (580 mg, 62%
from the amine). This compound exists as two conformers in CDCl3. 1H
NMR (500 MHz, CDCl3, 258C): d ˆ 7.35 (m, 5H), 5.11 (s, 2H), 4.29 (br,
0.5H), 3.86 (br, 0.5H), 3.67 (s, 3H), 3.66 (s, 3H), 3.50 (br, 0.5H), 3.31 (br,
0.5H), 3.03 (m, 1H), 2.42 ± 2.33 (m, 4H), 2.00 ± 1.78 (m, 4H), 1.64 (m, 2H),
1.44 (s, 4H), 1.40 (s, 5H); 13C NMR (125 MHz, CDCl3, 258C): d ˆ 173.5,
173.0, 172.8, 172.1, 171.8, 155.5, 154.8, 135.9, 135.9, 128.5, 128.2, 80.5, 80.5,
66.2, 60.2, 59.0, 52.0, 51.5, 47.4, 46.0, 33.6, 33.5, 31.5, 31.3, 29.7, 28.7, 28.2,
‡
23.5; HRMS [M‡Na] calcd for C16H27NO8Na: 384.1634, found: 384.1645.
‡
24.5, 23.7, 21.9, 21.8; HRMS [M‡Cs] calcd for C24H35NO8Cs: 598.1417,
N-Boc-N-[(1S)-1,4-dicarboxymethylbutyl]-4-aminobutyric acid (25):
A
found: 598.1431.
solution of benzyl ester 20 (560 mg, 1.2 mmol) in EtOAc (10 mL) was
1708
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1998
0947-6539/98/0409-1708 $ 17.50+.50/0
Chem. Eur. J. 1998, 4, No. 9

Leukotriene A4 Hydrolase
1698±1713
stirred vigorously with Pd/C (10% on carbon, 25 mg) under a hydrogen
atmosphere until no ester remained (2 h). The reaction mixture was filtered
through Celite and the solvent removed. Purification by flash chromatog-
raphy (1:1 EtOAc/hexanes, then EtOAc) gave 25 as a colorless oil (395 mg,
90%). This compound exists as two conformers in CDCl3. 1H NMR
(500 MHz, CDCl3, 258C): d ˆ 4.32 (m, 0.5H), 3.90 (m, 0.5H), 3.72 (s, 3H),
3.68 (s, 3H), 3.55 (m, 0.5H), 3.35 (m, 0.5H), 3.08 (m, 1H), 2.48 ± 2.31 (m,
2H), 2.37 (t, J ˆ 7.0 Hz, 2H), 2.02 (m, 1H), 1.96 ± 1.80 (m, 3H), 1.69 (m,
2H), 1.46 (s, 4H), 1.42 (s, 5H); 13C NMR (125 MHz, CDCl3, 258C): d ˆ
178.8, 178.5, 173.7, 173.5, 172.1, 171.8, 155.6, 155.1, 80.9, 80.5, 60.3, 59.0, 52.1,
51.6, 47.2, 45.8, 33.6, 33.5, 31.2, 29.7, 28.7, 28.2, 24.2, 23.6, 21.9, 21.8; HRMS
N-Hydroxy-N-[(2S)-2-N''-Boc-amino-3-(4-benzyloxyphenyl)propyl]-N'-
Boc-N'-[(1S)-1,3-dicarboxymethylpropyl]-4-aminobutanamide (29): Iso-
butyl chloroformate (77 mL, 0.59 mmol) was added to a solution of acid
24 (234 mg, 0.65 mmol) and triethylamine (105 mL, 0.75 mmol) in THF
(10 mL) at 08C. The resulting slurry was stirred for 30 min and then (2S)-2-
N'-Boc-amino-3-(4-benzyloxyphenyl)-N-hydroxypropylamine
(200 mg,
0.54 mmol) was added. The cooling bath was removed and the reaction
stirred for 30 min. The reaction was then quenched by addition of 3-
dimethylaminopropylamine and poured into CH2Cl2 (60 mL), washed (1n
HCl, saturated NaHCO3, saturated NaCl), and dried (MgSO4). Purification
by flash chromatography (1:2 EtOAc/hexanes) yielded compound 29 as a
colorless oil (293 mg, 76%). This compound exists as two conformers in
CDCl3. 1H NMR (500 MHz, CDCl3, 258C): d ˆ 8.65 (brs, 0.5H), 8.61 (brs,
0.5H), 7.43 (d, J ˆ 7.0 Hz, 2H), 7.39 (t, J ˆ 7.0 Hz, 2H), 7.33 (t, J ˆ 7.0 Hz,
1H), 7.10 (d, J ˆ 8.5 Hz, 2H), 6.93 (d, J ˆ 8.5 Hz, 2H), 5.05 (s, 2H), 4.66 (m,
1H), 4.28 (m, 0.5H), 4.18 (m, 1H), 4.15 (m, 1H), 3.96 (m, 0.5H), 3.70 (s,
3H), 3.67 (s, 3H), 3.48 (m, 0.5H), 3.30 (m, 0.5H), 3.09 ± 2.95 (m, 2H), 2.77
(m, 2H), 2.61 (m, 1H), 2.41 (m, 2H), 2.41 ± 2.20 (m, 2H), 2.11 (br, 1H), 1.85
(m, 2H), 1.45 (s, 4H), 1.40 (s, 5H), 1.38 (s, 9H); 13C NMR (125 MHz,
CDCl3, 258C): d ˆ 174.0, 173.9, 173.8, 173.4, 171.8, 171.7, 158.2, 157.9, 157.8,
155.5, 154.8, 136.9, 130.1, 128.7, 128.6, 128.0, 127.4, 115.1, 80.8, 80.7, 80.4,
80.1, 70.0, 59.6, 59.5, 52.1, 51.7, 50.5, 48.3, 48.1, 47.1, 37.3, 37.2, 30.6, 30.6, 30.0,
‡
[M‡Na] calcd for C17H29NO8Na: 398.1791, found: 398.1776.
N-Boc-N-[(1S)-1-carboxymethyl-5-N'-Boc-aminopentyl]-4-aminobutyric
acid (26): A solution of benzyl ester 21 (1.28 g, 2.39 mmol) in EtOAc
(15 mL) was stirred vigorously in the presence of Pd/C (10% on carbon,
45 mg) under a hydrogen atmosphere until no ester remained (2 h). The
hydrogen was then evacuated and replaced with argon. The reaction
mixture was filtered through Celite and the solvent removed. Purification
by flash chromatography (1:1 EtOAc/hexanes then EtOAc) yielded 26 as a
colorless oil (946 mg, 89%). This compound exists as two conformers in
CDCl3. 1H NMR (500 MHz, CDCl3, 258C): d ˆ 5.71 (m, 0.5H), 4.65 (m,
1H), 4.33 (m, 0.5H), 3.87 (m, 0.5H), 3.70 (s, 3H), 3.53 (m, 0.5H), 3.31 (m,
0.5H), 3.11 ± 3.04 (m, 3H), 2.40 ± 2.30 (m, 2H), 2.10 ± 1.70 (m, 4H), 1.53 (m,
2H), 1.46 (s, 4H), 1.43 (s, 9H), 1.40 (s, 5H); 13C NMR (125 MHz, CDCl3,
258C): d ˆ 178.0, 177.9, 172.3, 172.0, 156.0, 155.7, 155.1, 80.7, 80.3, 79.1, 60.3,
58.9, 52.0, 47.0, 45.5, 40.2, 31.1, 29.8, 29.7, 29.5, 28.9, 28.3, 28.2, 24.3, 23.6,
‡
29.9, 29.3, 29.2, 28.2, 28.2, 25.6, 24.7, 24.1, 23.3; HRMS [M‡Cs] calcd for
C37H53N3O11Cs: 848.2734, found: 848.2755.
N-Hydroxy-N-[(2S)-2-N''-Boc-amino-3-(4-benzyloxyphenyl)propyl]-N'-
Boc-N'-[(1S)-1,4-dicarboxymethylbutyl]-4-aminobutanamide (30): Isobu-
tyl chloroformate (75 mL, 0.59 mmol) was added to a solution of acid 25
(215 mg, 0.57 mmol) and triethylamine (110 mL, 0.76 mmol) in THF
(10 mL) at 08C. The resulting slurry was stirred for 30 min and then (2S)-
2-N'-Boc-amino-3-(4-benzyloxyphenyl)-N-hydroxypropylamine (200 mg,
0.54 mmol) was added. The cooling bath was removed and the reaction
mixture stirred for 30 min. The reaction was then quenched by addition of
3-dimethylaminopropylamine and poured into CH2Cl2 (60 mL), washed
(1n HCl, saturated NaHCO3, saturated NaCl), and dried (MgSO4).
Purification by flash chromatography (1:2 EtOAc/hexanes) yielded com-
pound 30 as a colorless oil (292 mg, 75%). This compound exists as two
conformers in CDCl3. 1H NMR (500 MHz, CDCl3, 258C): d ˆ 8.68 (brs,
0.5H), 8.65 (brs, 0.5H), 7.43 (d, J ˆ 7.0 Hz, 2H), 7.39 (t, J ˆ 7.0 Hz, 2H), 7.33
(t, J ˆ 7.0 Hz, 1H), 7.10 (d, J ˆ 8.5 Hz, 2H), 6.93 (d, J ˆ 8.5 Hz, 2H), 5.03 (s,
2H), 4.80 (m, 0.5H), 4.76 (brd, J ˆ 9.0 Hz, 0.5H), 4.31 (m, 0.5H), 4.20 ± 4.05
(m, 2H), 3.89 (m, 0.5H), 3.68 (s, 3H), 3.65 (s, 3H), 3.45 (m, 0.5H), 3.27 (m,
0.5H), 3.09 ± 2.96 (m, 2H), 2.76 (m, 2H), 2.60 (m, 1H), 2.41 ± 2.20 (m, 3H),
1.98 (br, 1H), 1.83 (m, 3H), 1.67 (m, 2H), 1.45 (s, 4H), 1.39 (s, 5H), 1.37 (s,
9H); 13C NMR (125 MHz, CDCl3, 258C): d ˆ 173.9, 173.8, 173.7, 173.4,
172.1, 171.9, 158.1, 157.9, 157.7, 155.6, 154.8, 136.8, 130.0, 128.7, 128.5, 127.9,
127.4, 115.0, 80.8, 80.6, 80.4, 80.1, 70.0, 60.2, 60.1, 58.9, 51.9, 51.5, 50.5, 48.3,
48.1, 47.9, 46.5, 37.2, 37.2, 33.7, 33.5, 30.0, 29.7, 29.6, 29.2, 28.8, 28.2, 28.1, 24.2,
‡
23.4; HRMS [M‡Cs] calcd for C21H38N2O8Cs: 579.1682, found: 579.1664.
N-Hydroxy-N-[(2S)-2-N''-Boc-amino-3-(4-benzyloxyphenyl)propyl]-N'-
Boc-N'-[(1S)-1-carboxymethylethyl]-4-aminobutanamide (27): Isobutyl
chloroformate (39 mL, 0.30 mmol) was added to a solution of acid 22
(87 mg, 0.30 mmol) and triethylamine (42 mL, 0.30 mmol) in THF (5 mL,
08C). The resulting slurry was stirred for 30 min and then hydroxylamine
(2S)-2-N'-Boc-amino-3-(4-benzyloxyphenyl)-N-hydroxypropylamine
(100 mg, 0.27 mmol) was added. The cooling bath was removed and the
reaction mixture stirred for 30 min. The reaction mixture was then poured
into CH2Cl2 (50 mL), washed (1n HCl, saturated NaHCO3, saturated
NaCl), and dried (MgSO4). Purification by flash chromatography (1:2
EtOAc/hexanes) yielded compound 27 as a colorless oil (109 mg, 63%).
This compound exists as two conformers in CDCl3. 1H NMR (500 MHz,
CDCl3, 258C): d ˆ 8.66 (s, 1H), 7.43 (d, J ˆ 7 Hz, 2H), 7.39 (t, J ˆ 7.5 Hz,
2H), 7.33 (t, J ˆ 7.0 Hz, 1H), 7.10 (d, J ˆ 8.5 Hz, 2H), 6.93 (d, J ˆ 8.5 Hz,
2H), 5.05 (s, 2H), 4.62 (m, 1H), 4.45 (m, 0.5H), 4.19 (m, 2H), 3.96 (m,
0.5H), 3.71 (s, 3H), 3.38 ± 3.15 (m, 2H), 3.09 (m, 1H), 2.81 (m, 2H), 2.62 (m,
1H), 2.31 (m, 1H), 1.87 (m, 2H), 1.45 (s, 9H), 1.40 (s, 3H), 1.38 (s, 9H); 13
C
NMR (125 MHz, CDCl3, 258C): d ˆ 174.1, 173.9, 172.9, 172.6, 158.2, 157.8,
155.4, 154.8, 136.9, 130.0, 128.7, 128.5, 127.9, 127.4, 115.1, 80.9, 80.7, 80.3,
80.0, 70.0, 55.9, 54.7, 52.0, 50.51, 48.4, 48.1, 47.1, 45.9, 37.3, 37.2, 34.6, 31.5,
‡
‡
23.3, 21.9, 21.8; HRMS [M‡Cs] calcd for C38H55N3O11Cs: 862.2891, found:
30.0, 29.1, 28.3, 24.7, 23.7, 22.6, 16.0, 15.9, 15.4, 14.1; HRMS [M‡Cs] calcd
862.2868.
for C34H49N3O9Cs: 776.2523, found: 776.2505.
N-Hydroxy-N-[(2S)-2-N''-Boc-amino-3-(4-benzyloxyphenyl)propyl]-N'-
Boc-N'-[(1R)-1-carboxymethylethyl]-4-aminobutanamide (28): Isobutyl
chloroformate (77 mL, 0.59 mmol) was added to a solution of acid 23
(189 mg, 0.65 mmol) and triethylamine (105 mL, 0.75 mmol) in THF
(10 mL) at 08C. The resulting slurry was stirred for 30 min and then (2S)-
2-N'-Boc-amino-3-(4-benzyloxyphenyl)-N-hydroxypropylamine (200 mg,
0.54 mmol) was added. The cooling bath was removed and the reaction
stirred for 30 min. The reaction was then quenched by addition of 3-
dimethylaminopropylamine and poured into CH2Cl2 (60 mL), washed (1n
HCl, saturated NaHCO3, saturated NaCl), and dried (MgSO4). Purification
by flash chromatography (1:2 EtOAc/hexanes) yielded compound 28 as a
colorless oil (285 mg, 82%). This compound exists as two conformers in
N-Hydroxy-N-[(2S)-2-N''-Boc-amino-3-(4-benzyloxyphenyl)propyl]-N'-
Boc-N'-[(1S)-1-carboxymethyl-5-N'-Boc-aminopentyl]-4-aminobutanamide
(31): Isobutyl chloroformate (77 mL, 0.59 mmol) was added to a solution of
acid 26 (290 mg, 0.65 mmol) and triethylamine (105 mL, 0.75 mmol) in THF
(10 mL) at 08C. The resulting slurry was stirred for 30 min and then (2S)-2-
N'-Boc-amino-3-(4-benzyloxyphenyl)-N-hydroxypropylamine
(200 mg,
0.54 mmol) was added. The cooling bath was removed and the reaction
mixture stirred for 30 min. The reaction was then quenched by addition of
3-dimethylaminopropylamine and poured into CH2Cl2 (60 mL), washed
(1n HCl, saturated NaHCO3, saturated NaCl), and dried (MgSO4).
Purification by flash chromatography (1:2 EtOAc/hexanes) yielded com-
pound 31 as a colorless oil (389 mg, 90%). This compound exists as two
conformers in CDCl3. 1H NMR (500 MHz, CDCl3, 258C): d ˆ 8.67 (brd,
J ˆ 14.0 Hz, 1H), 7.43 (d, J ˆ 7.0 Hz, 2H), 7.39 (t, J ˆ 7.0 Hz, 2H), 7.33 (t,
J ˆ 7.0 Hz, 1H), 7.10 (d, J ˆ 8.5 Hz, 2H), 6.93 (d, J ˆ 8.5 Hz, 2H), 5.05 (s,
2H), 4.79 (m, 1H), 4.65 (m, 1H), 4.38 (m, 0.5H), 4.19 (m, 1H), 4.15 (m,
1H), 3.88 (m, 0.5H), 3.69 (s, 3H), 3.42 (m, 0.5H), 3.25 (m, 0.5H), 3.13 (m,
1H), 3.06 (m, 1H), 2.78 (m, 2H), 2.59 (m, 0.5H), 2.31 (m, 1H), 1.95 (m,
1H), 1.85 (br, 2H), 1.51 (m, 2H), 1.45 (s, 3H), 1.43 (s, 6H), 1.40 (s, 3H), 1.38
(s, 6H), 1.38 (s, 9H); 13C NMR (125 MHz, CDCl3, 258C): d ˆ 174.0, 173.8,
172.4, 172.2, 158.2, 157.9, 157.8, 156.0, 155.0, 136.9, 130.1, 128.6, 128.0, 127.4,
115.1, 80.8, 80.7, 80.4, 80.1, 79.0, 70.0, 60.4, 60.3, 59.0, 52.0, 50.5, 48.4, 48.2,
1
CDCl3. H NMR (500 MHz, CDCl3, 258C): d ˆ 8.65 (d, J ˆ 12.0 Hz, 1H),
7.43 (d, J ˆ 7.0 Hz, 2H), 7.39 (t, J ˆ 7.0 Hz, 2H), 7.33 (t, J ˆ 7.0 Hz, 1H), 7.10
(d, J ˆ 8.5 Hz, 2H), 6.93 (d, J ˆ 8.5 Hz, 2H), 5.05 (s, 2H), 4.65 (m, 1H), 4.44
(m, 0.5H), 4.18 (m, 1H), 4.15 (m, 1H), 3.96 (m, 0.5H), 3.69 (s, 3H), 3.35 (m,
0.5H), 3.31 (m, 0.5H), 2.76 (m, 2H), 2.59 (m, 0.5H), 2.31 (m, 1H), 1.85 (m,
2H), 1.45 (s, 4H), 1.40 (s, 5H), 1.38 (s, 9H); 13C NMR (125 MHz, CDCl3,
258C): d ˆ 174.1, 173.9, 172.9, 172.6, 158.2, 157.8, 155.5, 154.8, 136.9, 130.1,
128.6, 128.0, 127.5, 115.1, 81.0, 80.8, 80.4, 80.0, 70.0, 55.9, 54.7, 52.0, 50.5,
48.3, 48.1, 47.2, 45.9, 37.3, 37.2, 30.0, 29.2, 28.3, 28.2, 24.7, 23.7, 16.0, 16.0, 15.4;
‡
HRMS [M‡Cs] calcd for C34H49N3O9Cs: 776.2523, found: 776.2544.
Chem. Eur. J. 1998, 4, No. 9
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1998
0947-6539/98/0409-1709 $ 17.50+.25/0
1709

C.-H. Wong et al.
FULL PAPER
47.8, 46.3, 46.2, 40.3, 37.2, 30.0, 29.8, 29.7, 29.5, 29.2, 29.0, 28.4, 28.3, 28.2,
N-Hydroxy-N-[(2S)-2-amino-3-(4-benzyloxyphenyl)propyl]-N'-[(1S)-1-c-
arboxymethyl-5-aminopentyl]-4-aminobutanamide, HCl salt (36): Hydrox-
amate 31 (50 mg, 0.063 mmol) was treated with a solution of HCl/acetic
acid (1 mL) for 30 min. The reaction mixture was diluted with toluene
(10 mL) and the solvent removed from the resulting mixture. The resulting
solid was rinsed with ether and then azeotroped with toluene to give 36 as a
pale yellow solid (38 mg, 99%). 1H NMR (500 MHz, [D6]DMSO, 258C):
d ˆ 10.25 (s, 1H), 9.75 (brs, 1H), 9.30 (brs, 1H), 8.20 (brs, 3H), 8.04 (brs,
3H), 7.43 (d, J ˆ 7.0 Hz, 2H), 7.36 (t, J ˆ 7.0 Hz, 2H), 7.32 (t, J ˆ 7.0 Hz, 1H),
7.19 (d, J ˆ 8.5 Hz, 2H), 6.96 (d, J ˆ 8.5 Hz, 2H), 5.06 (s, 2H), 4.03 (br, 1H),
3.83 (dd, J ˆ 7.0 and 14.0 Hz, 1H), 3.76 (s, 3H), 3.55 (br, 1H), 3.48 (dd, J ˆ
3.0 and 14.0 Hz, 1H), 2.96 (br, 2H), 2.94 (dd, J ˆ 3.0 and 14.0 Hz, 1H),
2.80 ± 2.69 (m, 3H), 2.53 (t, J ˆ 7.0 Hz, 2H), 1.96 ± 1.79 (m, 4H), 1.55 (m,
2H), 1.42 (m, 1H), 1.29 (m, 1H); 13C NMR (125 MHz, [D6]DMSO, 258C):
d ˆ 173.2, 169.2, 157.2, 130.5, 128.5, 128.1, 127.9, 127.7, 114.9, 89.4, 69.2, 58.5,
‡
24.3, 23.9, 23.6, 23.3; HRMS [M‡Cs] calcd for C42H64N4O11Cs: 933.3626,
found: 933.3649.
N-Hydroxy-N-[(2S)-2-amino-3-(4-benzyloxyphenyl)propyl]-N'-[(1S)-1-
carboxymethylethyl]-4-aminobutanamide, HCl salt (32): Hydroxamate 27
(44 mg, 0.068 mmol) was treated with a solution of HCl/acetic acid (1 mL)
for 30 min. The reaction mixture was diluted with toluene (10 mL) and the
solvent removed from the resulting mixture. The resulting solid was rinsed
with ether and then azeotroped with toluene to give 32 as a white solid
(28.1 mg, 94%). 1H NMR (500 MHz, [D6]DMSO, 258C): d ˆ 10.31 (s, 1H),
9.64 (brs, 1H), 9.27 (brs, 1H), 8.22 (brs, 3H), 7.43 (d, J ˆ 7.0 Hz, 2H), 7.39
(t, J ˆ 7.0 Hz, 2H), 7.33 (t, J ˆ 7.0 Hz, 1H), 7.21 (d, J ˆ 8.5 Hz, 2H), 6.97 (d,
J ˆ 8.5 Hz, 2H), 5.07 (s, 2H), 4.12 (m, 1H), 3.82 (dd, J ˆ 8.0 and 14.0 Hz,
1H), 3.74 (s, 3H), 3.54 (br, 1H), 3.47 (dd, J ˆ 4.0 and 14.5 Hz, 1H), 3.00 ±
2.88 (m, 2H), 2.93 (dd, J ˆ 6.0 and 13.5 Hz, 1H), 2.76 (dd, J ˆ 7.5 and
13.5 Hz, 1H), 2.52 (t, J ˆ 7.0 Hz, 2H), 1.85 (m, 2H), 1.45 (d, J ˆ 7.0 Hz, 3H);
13C NMR (125 MHz, [D6]DMSO, 258C): d ˆ 173.3, 169.9, 157.4, 137.1, 130.5,
128.5, 128.1, 127.9, 127.7, 114.9, 69.2, 54.2, 53.0, 50.2, 48.7, 44.8, 35.1, 29.2,
‡
53.0, 48.7, 46.0, 38.2, 35.0, 29.0, 26.3, 21.2, 20.5; HRMS [M‡H] calcd for
C27H41N4O5: 501.3077, found: 501.3087.
N-Hydroxy-N-[(2S)-2-amino-3-(4-benzyloxyphenyl)propyl]-N'-[(1S)-1-
carboxyethyl]-4-aminobutanamide, HCl salt (37): LiOH (1.0m, 2.5 mL)
was added to hydroxamate 27 (73 mg, 0.11 mmol) in THF/MeOH (2:17,
5 mL) and the biphasic mixture was stirred vigorously for 30 min. The
mixture was poured into HCl (1n, 30 mL) and extracted with EtOAc. The
organic layer was dried (MgSO4) and the solvent removed to give a pale
yellow oil (70 mg, 99%). The oil (26 mg, 0.041 mmol) was then dissolved in
HCl/acetic acid (1 mL) and stirred for 30 min. The reaction mixture was
diluted with toluene (10 mL) and the solvent removed from the resulting
mixture. The resulting solid was rinsed with ether and then azeotroped with
toluene to give 37 as a white solid (18.5 mg, 89%). 1H NMR (500 MHz,
[D6]DMSO, 258C): d ˆ 10.26 (s, 1H), 9.13 (brs, 2H), 8.18 (brs, 3H), 7.43 (d,
J ˆ 7.0 Hz, 2H), 7.39 (t, J ˆ 7.0 Hz, 2H), 7.33 (t, J ˆ 7.0 Hz, 1H), 7.21 (d, J ˆ
8.5 Hz, 2H), 6.97 (d, J ˆ 8.5 Hz, 2H), 5.07 (s, 2H), 3.96 (m, 1H), 3.84 (dd,
J ˆ 8.0 and 15.0 Hz, 1H), 3.55 (br, 1H), 3.46 (dd, J ˆ 4.5 and 14.5 Hz, 1H),
2.92 (m, 3H), 2.76 (dd, J ˆ 8.5 and 14.0 Hz, 1H), 2.52 (t, J ˆ 7.0 Hz, 2H),
1.84 (m, 2H), 1.43 (d, J ˆ 7.0 Hz, 3H); 13C NMR (125 MHz, [D6]DMSO,
258C): d ˆ 173.4, 171.2, 157.6, 137.3, 130.7, 128.7, 128.3, 128.1, 128.0, 115.1,
‡
22.5, 20.6, 14.4; HRMS [M‡H] calcd for C24H34N3O5: 444.2498, found:
444.1339.
N-Hydroxy-N-[(2S)-2-amino-3-(4-benzyloxyphenyl)propyl]-N'-[(1R)-1-
carboxymethylethyl]-4-aminobutanamide, HCl salt (33): Hydroxamate 28
(58 mg, 0.090 mmol) was treated with a solution of HCl/acetic acid (1.5 mL)
for 30 min. The reaction mixture was diluted with toluene (10 mL) and the
solvent removed from the resulting mixture. The resulting solid was rinsed
with ether and then azeotroped with toluene to give 33 as a white solid
(42.8 mg, 92%). 1H NMR (500 MHz, [D6]DMSO, 258C): d ˆ 10.32 (s, 1H),
9.68 (brs, 1H), 9.31 (brs, 1H), 8.23 (brs, 3H), 7.43 (d, J ˆ 7.0 Hz, 2H), 7.39
(t, J ˆ 7.0 Hz, 2H), 7.33 (t, J ˆ 7.0 Hz, 1H), 7.21 (d, J ˆ 8.5 Hz, 2H), 6.97 (d,
J ˆ 8.5 Hz, 2H), 5.07 (s, 2H), 4.13 (m, 1H), 3.83 (dd, J ˆ 8.0 and 14.5 Hz,
1H), 3.74 (s, 3H), 3.55 (br, 1H), 3.48 (dd, J ˆ 4.0 and 14.5 Hz, 1H), 2.95 (m,
3H), 2.77 (dd, J ˆ 8.0 and 14.0 Hz, 1H), 2.52 (t, J ˆ 7.5 Hz, 2H), 1.88 (m,
2H), 1.45 (d, J ˆ 7.0 Hz, 3H); 13C NMR (125 MHz, [D6]DMSO, 258C): d ˆ
173.2, 169.9, 157.3, 137.1, 130.5, 128.5, 128.1, 127.9, 127.7, 114.9, 69.2, 54.2,
‡
53.0, 50.2, 48.7, 44.8, 35.0, 29.0, 20.6, 14.4; HRMS [M‡Cs] calcd for
‡
69.4, 54.6, 50.4, 48.9, 45.0, 35.3, 29.2, 20.8, 14.7; HRMS [M‡Cs] calcd for
C24H33N3O5Cs: 576.1475, found: 576.1489.
C23H31N3O2Cs: 562.1318, found: 562.1335.
N-Hydroxy-N-[(2S)-2-amino-3-(4-benzyloxyphenyl)propyl]-N'-[(1S)-1,3-
dicarboxymethylpropyl]-4-aminobutanamide, HCl salt (34): Hydroxamate
29 (75 mg, 0.091 mmol) was treated with a solution of HCl/acetic acid
(1 mL) for 30 min. The reaction mixture was diluted with toluene (10 mL)
and the solvent removed from the resulting mixture. The resulting solid was
rinsed with ether and then azeotroped with toluene to give 34 as a white
solid (52.7 mg, 99%). 1H NMR (500 MHz, [D6]DMSO, 258C): d ˆ 10.30 (s,
1H), 9.75 (brs, 1H), 9.39 (brs, 1H), 8.22 (brs, 3H), 7.43 (d, J ˆ 7.0 Hz, 2H),
7.39 (t, J ˆ 7.0 Hz, 2H), 7.33 (t, J ˆ 7.0 Hz, 1H), 7.21 (d, J ˆ 8.5 Hz, 2H), 6.97
(d, J ˆ 8.5 Hz, 2H), 5.07 (s, 2H), 4.09 (br, 1H), 3.82 (dd, J ˆ 8.0 and 14.0 Hz,
1H), 3.72 (s, 3H), 3.59 (s, 3H), 3.55 (br, 1H), 3.48 (dd, J ˆ 3.5 and 14.0 Hz,
1H), 2.93 (dd, J ˆ 5.5 and 14.0 Hz, 1H), 2.93 (m, 2H), 2.77 (dd, J ˆ 8.0 and
N-Hydroxy-N-[(2R)-2-amino-3-(4-benzyloxyphenyl)propyl]-N'-[(1R)-1-
carboxyethyl]-4-aminobutanamide, HCl salt (38): LiOH (1.0m, 2.5 mL)
was added to hydroxamate 28 (90 mg, 0.13 mmol) in THF/MeOH (2:1,
7.5 mL) and the biphasic mixture was stirred vigorously for 1 h. The mixture
was poured into HCl (1n, 50 mL) and extracted with EtOAc. The organic
layer was dried (MgSO4) and the solvent removed to give a pale yellow oil
(84 mg, 95%). The oil (30 mg, 0.048 mmol) was then dissolved in HCl/
acetic acid (1 mL) and stirred for 30 min. The reaction mixture was diluted
with toluene (10 mL) and the solvent removed from the resulting mixture.
The resulting solid was rinsed with ether and then azeotroped with toluene
to give 38 as a white solid (22.3 mg, 93%). 1H NMR (500 MHz, [D6]DMSO,
258C): d ˆ 10.28 (s, 1H), 9.11 (brs, 2H), 8.19 (brs, 3H), 7.43 (d, J ˆ 7.0 Hz,
2H), 7.39 (t, J ˆ 7.0 Hz, 2H), 7.33 (t, J ˆ 7.0 Hz, 1H), 7.21 (d, J ˆ 8.5 Hz,
2H), 6.97 (d, J ˆ 8.5 Hz, 2H), 5.07 (s, 2H), 3.92 (m, 1H), 3.84 (m, 1H), 3.55
(br, 1H), 3.46 (dt, J ˆ 3.0 and 14.5 Hz, 1H), 2.92 (m, 3H), 2.76 (dd, J ˆ 8.5
and 14.0 Hz, 1H), 2.52 (t, 2H), 1.84 (m, 2H), 1.42 (d, J ˆ 7.0 Hz, 3H); 13C
NMR (125 MHz, [D6]DMSO, 258C): d ˆ 173.2, 171.0, 157.2, 137.1, 130.2,
128.5, 128.1, 127.9, 127.7, 114.9, 69.2, 54.5, 50.1, 48.7, 44.8, 35.1, 28.9, 20.6,
14.0 Hz, 1H), 2.50 (m, 4H), 2.17 (m, 1H), 2.10 (m, 1H), 1.88 (m, 2H); 13
C
NMR (125 MHz, [D6]DMSO, 258C): d ˆ 173.2, 172.1, 168.8, 157.3, 137.1,
130.5, 128.5, 128.1, 127.9, 127.7, 114.9, 69.2, 57.9, 53.0, 51.6, 50.2, 48.7, 45.5,
‡
35.0, 29.0, 28.8, 23.9, 20.5; HRMS [M‡Cs] calcd for C27H37N3O7Cs:
648.1686, found: 648.1699.
N-Hydroxy-N-[(2S)-2-amino-3-(4-benzyloxyphenyl)propyl]-N'-[(1S)-1,4-
dicarboxymethylbutyl]-4-aminobutanamide, HCl salt (35): Hydroxamate
30 (46 mg, 0.063 mmol) was treated with a solution of HCl/acetic acid
(1 mL) for 30 min. The reaction mixture was diluted with toluene (10 mL)
and the solvent removed from the resulting mixture. The resulting solid was
rinsed with ether and then azeotroped with toluene to give 35 as a white
solid (27.4 mg, 71%). 1H NMR (500 MHz, [D6]DMSO, 258C): d ˆ 10.31 (s,
1H), 9.55 (brs, 1H), 9.32 (brs, 1H), 8.21 (brs, 3H), 7.43 (d, J ˆ 7.0 Hz, 2H),
7.39 (t, J ˆ 7.0 Hz, 2H), 7.33 (t, J ˆ 7.0 Hz, 1H), 7.21 (d, J ˆ 8.5 Hz, 2H), 6.97
(d, J ˆ 8.5 Hz, 2H), 5.07 (s, 2H), 4.09 (br, 1H), 3.80 (dd, J ˆ 8.5 and 14.0 Hz,
1H), 3.75 (s, 3H), 3.58 (s, 3H), 3.55 (br, 1H), 3.48 (dd, J ˆ 4.0 and 14.0 Hz,
1H), 2.92 (dd, J ˆ 5.0 and 14.0 Hz, 1H), 2.90 (m, 2H), 2.78 (dd, J ˆ 8.5 and
14.0 Hz, 1H), 2.51 (t, J ˆ 7.0 Hz, 2H), 2.34 (t, J ˆ 7.0 Hz, 2H), 1.86 (m, 4H),
1.62 (m, 1H), 1.46 (m, 1H); 13C NMR (125 MHz, [D6]DMSO, 258C): d ˆ
173.2, 172.8, 169.2, 157.4, 137.1, 130.5, 128.5, 128.1, 127.9, 127.7, 114.9, 69.2,
58.4, 53.0, 51.4, 50.2, 48.7, 45.5, 35.0, 32.5, 29.0, 28.0, 20.5, 19.8; HRMS
‡
14.5; HRMS [M‡Cs] calcd for C23H31N3O2Cs: 562.1318, found: 562.1328.
N-Hydroxy-N-[(2S)-2-amino-3-(4-benzyloxyphenyl)propyl]-N'-[(1S)-1,3-
dicarboxypropyl]-4-aminobutanamide, HCl salt (39): LiOH (1.0m, 2.5 mL)
was added to hydroxamate 29 (114 mg, 0.16 mmol) in THF/MeOH (2:1,
7.5 mL) and the biphasic mixture stirred vigorously for 1 h. The mixture was
poured into HCl (1n, 30 mL) and extracted with EtOAc. The organic layer
was dried (MgSO4) and the solvent removed to give a pale yellow oil
(106 mg, 96%). The oil (37 mg, 0.054 mmol) was then dissolved in HCl/
acetic acid (1 mL) and stirred for 30 min. The reaction mixture was diluted
with toluene (10 mL) and the solvent removed from the resulting mixture.
The resulting solid was rinsed with ether and then azeotroped with toluene
to give 39 as a white solid (28.1 mg, 94%). 1H NMR (500 MHz, [D6]DMSO,
258C): d ˆ 10.27 (s, 1H), 9.20 (vbrs, 2H), 8.19 (brs, 3H), 7.43 (d, J ˆ 7.0 Hz,
2H), 7.39 (t, J ˆ 7.0 Hz, 2H), 7.33 (t, J ˆ 7.0 Hz, 1H), 7.21 (d, J ˆ 8.5 Hz,
2H), 6.97 (d, J ˆ 8.5 Hz, 2H), 5.07 (s, 2H), 3.92 (m, 1H), 3.84 (m, 1H), 3.55
(br, 1H), 3.46 (m, 1H), 3.00 ± 2.86 (m, 3H), 2.75 (dd, J ˆ 8.5 and 14.0 Hz,
‡
[M‡Cs] calcd for C28H39N3O7Cs: 662.1842, found: 662.1852.
1710
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1998
0947-6539/98/0409-1710 $ 17.50+.50/0
Chem. Eur. J. 1998, 4, No. 9

Leukotriene A4 Hydrolase
1698±1713
1H), 2.52 (m, 2H), 2.46 (m, 1H), 2.32 (m, 1H), 2.11 (m, 1H), 2.02 (m, 1H),
1.85 (m, 2H); 13C NMR (125 MHz, [D6]DMSO, 258C): d ˆ 173.3, 173.1,
170.0, 157.4, 137.1, 130.5, 128.5, 128.1, 127.9, 127.7, 114.9, 69.2, 58.2, 50.1, 48.7,
45 min (TLC 1:1 EtOAc/hexanes). The reaction mixture was poured into
EtOAc (100 mL), and the layers were separated and the organic layer
washed (1n HCl, brine). Purification by flash chromatography (5%
acetone/toluene) gave the desired product 43 as a colorless oil (0.37 g,
100%). 1H NMR (400 MHz, CDCl3, 258C): d ˆ 3.75 (s, 3H), 3.43 (q, J ˆ
7.0 Hz, 1H), 2.71 (m, 1H), 2.67 (m, 1H), 2.50 (t, J ˆ 7.0 Hz, 2H), 1.92 (m,
‡
45.5, 35.1, 29.3, 29.0, 24.1, 22.5, 20.6; HRMS [M‡Cs] calcd for
C25H33N3O7Cs: 620.1373, found: 620.1398.
N-Hydroxy-N-[(2S)-2-amino-3-(4-benzyloxyphenyl)propyl]-N'-[(1S)-1,4-
dicarboxybutyl]-4-aminobutanamide, HCl salt (40): LiOH (1.0m, 2.5 mL)
was added to hydroxamate 30 (86 mg, 0.12 mmol) in THF/MeOH (2:1,
7.5 mL) and the biphasic mixture was stirred vigorously for 30 min. The
mixture was poured into a mixture of HCl (1n, 10 mL) and saturated NaCl
(10 mL) and extracted with EtOAc. The organic layer was dried (MgSO4)
and the solvent removed to give a pale yellow oil (81 mg, 98%). The oil
(29 mg, 0.041 mmol) was then dissolved in HCl/acetic acid (1 mL) and
stirred for 30 min. The reaction mixture was diluted with toluene (10 mL)
and the solvent removed from the resulting mixture. The resulting solid was
rinsed with ether and then azeotroped with toluene to give 40 as a white
solid (22.5 mg, 95%). 1H NMR (500 MHz, [D6]DMSO, 258C): d ˆ 10.25 (s,
1H), 9.10 (brs, 2H), 8.20 (brs, 3H), 7.43 (d, J ˆ 7.0 Hz, 2H), 7.39 (t, J ˆ
7.0 Hz, 2H), 7.33 (t, J ˆ 7.0 Hz, 1H), 7.21 (d, J ˆ 8.5 Hz, 2H), 6.97 (d, J ˆ
8.5 Hz, 2H), 5.07 (s, 2H), 3.88 (m, 1H), 3.83 (m, 1H), 3.54 (br, 1H), 3.45
(dt, J ˆ 4.5 and 14.0 Hz, 1H), 2.92 (m, 2H), 2.92 (dd, J ˆ 4.5 and 13.5 Hz,
1H), 2.76 (dd, J ˆ 8.0 and 13.5 Hz, 1H), 2.52 (t, 2H), 2.25 (t, J ˆ 7.0 Hz,
2H), 1.84 (m, 2H), 1.64 (m, 1H), 1.46 (m, 1H); 13C NMR (125 MHz,
[D6]DMSO, 258C): d ˆ 174.2, 173.5, 170.5, 157.6, 137.3, 130.7, 128.7, 128.3,
128.1, 128.0, 115.1, 69.4, 68.3, 59.1, 50.3, 48.9, 45.7, 35.3, 33.23, 29.2, 28.3,
‡
2H), 1.45 (d, J ˆ 7.0 Hz, 3H). HRMS [M‡H] calcd for C8H15O4S: 207.0691,
found: 207.0695.
N-Hydroxy-N-[(2S)-2-N'-Boc-amino-3-(4-benzyloxyphenyl)propyl]-S-[(1-
S)-1-carboxymethylethyl]-4-mercaptobutanamide (44): Isobutyl chlorofor-
mate (75 mL, 0.59 mmol) was added to a solution of acid 43 (121 mg,
0.59 mmol) and triethylamine (105 mL, 0.75 mmol) in THF (10 mL) at 08C.
The resulting slurry was stirred for 30 min and then (2S)-2-N'-Boc-amino-3-
(4-benzyloxyphenyl)-N-hydroxypropylamine (200 mg, 0.54 mmol) was
added. The cooling bath was removed and the reaction mixture stirred
for 30 min. The reaction was then quenched by addition of 3-dimethyl-
aminopropylamine and poured into CH2Cl2 (60 mL), washed (1n HCl,
saturated NaHCO3, saturated NaCl), and dried (MgSO4). Purification by
flash chromatography (1:2 EtOAc/hexanes, then 1:1 EtOAc/hexanes)
yielded compound 44 as
a
colorless oil (275 mg, 91%). 1H NMR
(400 MHz, CDCl3, 258C): d ˆ 8.65 (s, 1H), 7.45 ± 7.30 (m, 5H), 7.10 (d,
J ˆ 8.5 Hz, 2H), 6.93 (d, J ˆ 8.5 Hz, 2H), 5.05 (s, 2H), 4.65 (d, J ˆ 8.5 Hz,
1H), 4.19 (q, J ˆ 11.5 Hz, 1H), 4.14 (m, 1H), 3.73 (s, 3H), 3.41 (q, J ˆ
7.0 Hz, 1H), 3.04 (dd, J ˆ 1.5 and 11.0 Hz, 1H), 2.77 (m, 2H), 2.72 ± 2.56 (m,
3H), 1.94 ± 1.82 (m, 1H), 1.43 (dd, J ˆ 0.6 and 7.0 Hz, 3H), 1.39 (s, 9H); 13
C
‡
NMR (100 MHz, CDCl3, 258C): d ˆ 173.7, 158.2, 157.8, 136.9, 130.0, 128.6,
20.8, 20.1; HRMS [M‡Cs] calcd for C26H35N3O7Cs: 634.1529, found:
128.0, 127.5, 115.2, 81.0, 70.0, 52.2, 50.5, 48.1, 40.7, 37.2, 31.4, 31.1, 28.2, 24.2,
634.1538.
‡
17.2; HRMS [M‡H] calcd for C29H40N2O7SCs: 693.1611, found: 693.1625.
N-Hydroxy-N-[(2S)-2-amino-3-(4-benzyloxyphenyl)propyl]-N'-[(1S)-1-
carboxy-5-aminopentyl]-4-aminobutanamide, HCl salt (41): LiOH (1.0m,
2.5 mL) was added to hydroxamate 31 (93 mg, 0.12 mmol) in THF/MeOH
(2:1, 7.5 mL) and the biphasic mixture was stirred vigorously for 1 h. The
mixture was poured into HCl (1n, 50 mL) and extracted with EtOAc. The
organic layer was dried (MgSO4) and the solvent removed to give a pale
yellow oil (87 mg, 96%). The oil (31 mg, 0.039 mmol) was then dissolved in
HCl/acetic acid (1 mL) and stirred for 30 min. The reaction mixture was
diluted with toluene (10 mL) and the solvent removed from the resulting
mixture. The resulting solid was rinsed with ether and then azeotroped with
toluene to give 41 as a white solid (20.1 mg, 86%). 1H NMR (500 MHz,
[D6]DMSO, 258C): d ˆ 10.29 (s, 1H), 9.23 (brs, 2H), 8.20 (brs, 3H), 8.04
(brs, 3H), 7.43 (d, J ˆ 7.0 Hz, 2H), 7.39 (t, J ˆ 7.0 Hz, 2H), 7.33 (t, J ˆ 7.0 Hz,
1H), 7.21 (d, J ˆ 8.5 Hz, 2H), 6.97 (d, J ˆ 8.5 Hz, 2H), 5.07 (s, 2H), 3.86 (m,
1H), 3.84 (dd, J ˆ 8.0 and 14.5 Hz, 1H), 3.55 (br, 1H), 3.47 (dd, J ˆ 2.0 and
14.5 Hz, 1H), 2.92 (m, 3H), 2.80 ± 2.71 (m, 3H), 2.52 (t, 2H), 1.89 (m, 2H),
1.58 (m, 2H), 1.47 (m, 1H), 1.32 (m, 1H); 13C NMR (125 MHz, [D6]DMSO,
258C): d ˆ 173.2, 170.2, 157.4, 137.1, 130.5, 128.5, 128.1, 127.9, 127.7, 114.9,
69.2, 58.8, 50.1, 48.7, 45.6, 38.2, 35.1, 29.0, 28.2, 26.4, 21.3, 20.6; HRMS
N-Hydroxy-N-[(2S)-2-amino-3-(4-benzyloxyphenyl)propyl]-S-[(1S)-1-car-
boxymethylethyl]-4-mercaptobutanamide, HCl salt (45): Hydroxamate 44
(80 mg, 0.14 mmol) was treated with a solution of HCl/acetic acid (2 mL)
for 30 min. The reaction mixture was diluted with toluene (10 mL) and the
solvent was removed from the resulting mixture. The resulting solid was
rinsed with ether and then azeotroped with toluene to give 45 as a white
solid (51 mg, 73%). 1H NMR (400 MHz, [D6]DMSO, 258C): d ˆ 10.06 (brs,
1H), 8.03 (brs, 3H), 7.43 (d, J ˆ 7.0 Hz, 2H), 7.39 (t, J ˆ 7.0 Hz, 2H), 7.33 (t,
J ˆ 7.0 Hz, 1H), 7.21 (d, J ˆ 8.5 Hz, 2H), 6.97 (d, J ˆ 8.5 Hz, 2H), 5.07 (s,
2H), 3.81 (dd, J ˆ 8.0 and 14.0 Hz, 1H), 3.62 (s, 3H), 3.58 ± 3.44 (m, 3H),
2.89 (dd, J ˆ 5.5 and 14.0 Hz, 1H), 2.76 (dd, J ˆ 8.0 and 14.0 Hz, 1H), 2.60
(m, 2H), 2.46 (m, 2H), 1.74 (m, 2H), 1.28 (d, J ˆ 7.0 Hz, 3H); 13C NMR
(100 MHz, [D6]DMSO, 258C): d ˆ 173.6, 173.1, 157.4, 137.1, 130.5, 128.4,
127.9, 127.8, 127.7, 115.1, 69.2, 52.0, 50.1, 48.7, 35.1, 30.7, 30.3, 23.8, 17.2;
‡
HRMS [M‡Cs] calcd for C24H32N2O5SCs: 593.1086, found: 593.1074.
N-Hydroxy-N-[(2S)-2-amino-3-(4-benzyloxyphenyl)propyl]-S-[(1S)-1-car-
boxyethyl]-4-mercaptobutanamide, HCl salt (46): LiOH (1.0m, 2.5 mL)
was added to hydroxamate 44 (90 mg, 0.16 mmol) in THF/MeOH (2:1,
7.5 mL) and the biphasic mixture was stirred vigorously for 30 min. The
mixture was poured into HCl(1n 30 mL) and extracted with EtOAc. The
organic layer was dried (MgSO4) and the solvent removed to give a pale
yellow oil (85 mg, 97%). The oil was then treated with a solution of HCl/
acetic acid (2 mL) for 30 min. The reaction mixture was diluted with
toluene (10 mL) and the solvent removed from the resulting mixture. The
resulting solid was rinsed with ether, azeotroped with toluene, then taken
up in CH2Cl2 and allowed to crystallize overnight to give 46 as a white solid
(67 mg, 84%). 1H NMR (500 MHz, [D6]DMSO, 258C): d ˆ 10.13 (brs, 1H),
8.03 (brs, 3H), 7.43 (d, J ˆ 7.0 Hz, 2H), 7.39 (t, J ˆ 7.0 Hz, 2H), 7.33 (t, J ˆ
7.0 Hz, 1H), 7.21 (d, J ˆ 8.5 Hz, 2H), 6.97 (d, J ˆ 8.5 Hz, 2H), 5.07 (s, 2H),
3.81 (dd, J ˆ 8.5 and 14.5 Hz, 1H), 3.54 (m, 1H), 3.46 (dd, J ˆ 4.0 and
14.5 Hz, 1H), 2.88 (q, J ˆ 7.0 Hz, 1H), 2.90 (dd, J ˆ 5.5 and 14.0 Hz, 1H),
2.75 (dd, J ˆ 7.0 and 13.5 Hz, 1H), 2.61 (m, 2H), 2.46 (m, 2H), 1.73 (m, 2H),
1.28 (d, J ˆ 7.0 Hz, 3H); 13C NMR (100 MHz, [D6]DMSO, 258C): d ˆ 173.6,
173.1, 157.4, 137.1, 130.5, 128.4, 128.0, 127.9, 127.7, 115.1, 69.2, 50.1, 48.7, 35.1,
‡
[M‡Cs] calcd for C26H38N4O5Cs: 619.1897, found: 619.1892.
S-Acetyl-4-mercaptobutyric acid (42): Potassium thioacetate (1.0 g,
9.0 mmol) was added in one portion to a solution of 4-bromobutyric acid
(1.0 g, 6.0 mmol) in anhydrous DMF (5 mL) under argon at room
temperature. The reaction was difficult to monitor by TLC as the starting
material was hard to detect (2:1 hexanes/EtOAc). The reaction mixture
was stirred for a total of 1 h then the solvent removed in vacuo. The residue
was taken up in ether and washed (0.1n HCl, brine). Purification by flash
chromatography yielded VI-58 as a colorless oil (0.75 g, 78%). 1H NMR
(400 MHz, CDCl3, 258C): d ˆ 2.94 (t, J ˆ 7.0 Hz, 2H), 2.45 (t, J ˆ 7.0 Hz,
2H), 2.34 (s, 3H), 1.92 (dt, J ˆ 7.0 and 7.0 Hz, 2H); 13C NMR (100 MHz,
CDCl3, 258C): d ˆ 195.7, 178.9, 32.6, 30.6, 28.2, 24.5.
Methyl (R)-(‡)-2-bromopropionate. (R)-(‡)-2-bromopropionic acid
(1.0 g, 590 mL, 6.5 mmol) was treated with ethereal diazomethane to give
the methyl ester. Acetic acid was added to destroy any excess diazo-
methane and the ethereal solution was concentrated by distillation to give
the product in ether (10 mL, approximately 0.65mm). This solution was
used directly in the following reaction (compound 43).
‡
30.8, 30.3, 23.8, 17.4; HRMS [M‡Cs] calcd for C23H30N2O5SCs: 579.0930,
found: 579.0944.
O-Benzyl-l-tyrosine p-nitroanilide, HCl salt (47): N-Boc-O-benzyl-l-
tyrosine (10 mmol, Advanced Chemtech) and p-nitroaniline (10 mmol)
were dissolved in dry pyridine (40 mL). The solution was chilled to between
158C and 208C on a carefully monitored acetone/dry-ice bath. After
15 minutes, phosphorus oxychloride (11 mmol, 1.0 mL) was added drop-
wise over a period of five minutes. In contrast to the reaction reported by
Tesser, from which this synthetic route was derived, no red by-product was
S-[(1S)-1-carboxymethylethyl]-4-mercaptobutyric acid (43): NaOH (1n,
1.2 mL, 1.2 mmol, deoxygenated) was added dropwise to thioacetate 42
(300 mg, 1.8 mmol) in methanol (7 mL, deoxygenated). The resulting
solution was stirred for 15 min. Additional NaOH (1n, 0.6 mL, 0.6 mmol)
was added followed by methyl (R)-(‡)-2-bromopropionate (5 mL of ꢀ
0.65mm solution, see preceding description). Stirring was continued for
Chem. Eur. J. 1998, 4, No. 9
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1998
0947-6539/98/0409-1711 $ 17.50+.25/0
1711

C.-H. Wong et al.
FULL PAPER
observed.[39] After 45 minutes, the color was observed to have lightened
considerably and the reaction was judged complete by TLC (1:1 EtOAc/
hexanes). Crushed ice (50 mL) was then added and the mixture was
transferred to a separating funnel. Upon melting of the ice, the organic
phase was diluted with ethyl acetate (200 mL) and washed 4 Â 100 mL HCl
(0.1n), 1 Â 100 mL saturated CuSO4, 1 Â 100 mL saturated NaHCO3, and
1 Â 50 mL saturated NaCl. The organic phase was dried under reduced
pressure, and the resultant yellow solid recrystallized from hot isopropyl
alcohol (300 mL), to give p-nitroanilide (3.83 g, 78%) as pale yellow
crystals. 1H NMR (500 MHz, [D6]DMSO, 258C): d ˆ 10.66 (s, 1H), 8.03 (d,
J ˆ 7.2 Hz, 2H), 7.84 (d, J ˆ 7.2 Hz, 2H), 7.42 (d, J ˆ 6.8 Hz, 2H), 7.37 (t, J ˆ
6.8 Hz, 2H), 7.31 (t, J ˆ 6.8 Hz, 1H), 7.23 (d, J ˆ 7.2 Hz, 2H), 6.90 (d, J ˆ
7.2 Hz, 2H), 5.05 (s, 2H), 4.3 (m, 1H), 3.35 (m, 1H), 2.7 ± 3.0 (m, 2H ), 1.32
(s, 9H); 13C NMR (100 MHz, [D6]DMSO, 258C): d ˆ 172.1, 157.0, 155.5,
145.1, 142.2, 137.2, 130.2, 129.8, 128.4, 127.8, 127.6, 125.0, 118.9, 114.4, 78.2,
[5] S. D. Brain, T. J. Williams, Pharm. Ther. 1990, 46, 57.
[6] S. A. Rae, E. M. Davidson, M. J. H. Smith, Lancet 1982, 2, 1122.
[7] D. W. Goldman, L. A. Gifford, T. Marotti, C. H. Koo, E. J. Goetzl,
Fed. Proc. 1987, 46, 200.
[8] M. W. Greaves, R. D. R. Camp, Arch. Dermatol. Res. 1988, 208, S33.
[9] D. W. Stenson, J. Gastroenterol. 1990, 25, 13.
[10] S. Lam, H. Chen, J. C. LeRiche, M. Chan-Yeung, H. Salari, J. Allergy
Clin. Immunol. 1988, 81, 711.
[11] A. J. Wardlaw, H. Hay, O. Cromwell, J. V. Collins, A. B. Kay, J. Allergy
Clin. Immunol. 1989, 84, 19.
[12] E. M. Davidson, S. E. Rae, M. J. Smith, Ann. Rheum. Dis. 1983, 43,
677.
[13] L. Orning, J. K. Gierse, F. A. Fitzpatrick, J. Biol. Chem. 1994, 269,
11269 ± 11273.
[14] K. J. Griffin, J. Gierse, G. Krivi, F. A. Fitzpatrick, Prostaglandins 1992,
44, 251 ± 7.
[15] J. F. Medina, A. Wetterholm, O. Raadmark, R. Shapiro, J. Z.
Haeggström, B. L. Vallee, B. Samuelsson, Proc. Natl. Acad. Sci.
USA 1991, 88, 7620 ± 4.
[16] M. Blomster, A. Wetterholm, M. J. Mueller, J. Z. Haeggström, Eur. J.
Biochem. 1995, 231, 528 ± 534.
[17] M. B. Andberg, M. Hamberg, J. Z. Haeggström, J. Biol. Chem. 1997,
272, 23057 ± 23063.
[18] J. Z. Haeggström, A. Wetterholm, R. Shapiro, B. L. Vallee, B.
Samuelsson, Biochem. Biophys. Res. Commun. 1990, 173, 431 ±
437.
[19] L. Orning, G. Krivi, F. A. Fitzpatrick, J. Biol. Chem. 1991, 266, 1375 ±
8.
[20] A. Wetterholm, J. Z. Haeggström, Biochim. Biophys. Acta 1992, 1123,
275 ± 81.
‡
69.1, 57.1, 36.2, 28.2; HRMS [M‡Na] calcd for C27H29N3O6Na: 514.1954,
found: 514.1938.
p-Nitroanilide (492 mg, 1 mmol) was dissolved in conc. HCl/AcOH
(30 mL), and stirred for 45 minutes at room temperature. The acetic acid
was removed under reduced pressure followed by coevaporation with
toluene (3 Â 30 mL). The resultant yellow solid was used without further
purification. Yield: 410 mg (96%). 1H NMR (500 MHz, [D6]DMSO, 258C):
d ˆ 11.9 (brs, 1H), 8.5 (brs, 3H), 8.236 (d, J ˆ 9.0 Hz, 2H), 7.911 (d, J ˆ
9.0 Hz, 2H), 7.409 (d, J ˆ 5.5 Hz, 2H), 7.361 (t, J ˆ 7.5 Hz, 2H), 7.301 (t, J ˆ
7.0 Hz, 1H), 7.269 (d, J ˆ 8.0 Hz, 2H), 6.934 (s, J ˆ 8.5 Hz, 2H), 5.041 (s,
2H), 4.40 (m, 1H), 3.0 ± 3.3 (m, 2H ), 1.32 (s, 9H); 13C NMR (100 MHz,
[D6]DMSO, 258C): d ˆ 168.3, 158.2, 144.9, 143.4, 137.7, 131.4, 129.5, 129.0,
128.8, 128.4, 128.3, 127.3, 125.6, 119.9, 115.4, 69.7, 55.1, 33.6; HRMS
‡
[M‡Na] calcd for C22H22N3O4: 392.1610, found: 392.1621.
O-Benzyl-l-serine p-nitroanilide (48): N-Boc-O-benzyl-l-serine (10 mmol,
Peninsula) and p-nitroaniline (10 mmol) were dissolved in dry pyridine
(40 mL). The solution was chilled to between 158C and 208C on a
carefully monitored acetone/dry-ice bath. After 15 minutes, phosphorus
oxychloride (11 mmol, 1.0 mL) was added dropwise over a period of five
minutes. After 45 minutes, the color was observed to have lightened
considerably and the reaction was judged complete by TLC (1:1 EtOAc/
hexanes). Crushed ice (50 mL) was then added and the mixture transferred
to a separating funnel. Upon melting of the ice, the organic phase was
diluted with ethyl acetate (200 mL) and washed 4 Â 100 mL HCl (0.1n),
1 Â 100 mL saturated CuSO4, 1 Â 100 mL saturated NaHCO3, and 1 Â
50 mL saturated NaCl. The organic phase was dried under reduced
pressure. This material was highly soluble in isopropyl alcohol, so it was
purified by silica-gel chromatography (10% acetone in toluene) instead, to
give p-nitroanilide (3.68 g, 75%) as a yellow oil. 1H NMR (500 MHz,
[D6]DMSO, 258C): d ˆ 9.05 (brs, 1H), 8.122 (d, J ˆ 8.5 Hz, 2H), 7.578 (d,
J ˆ 8 Hz, 2H), 7.3 (m, 5H), 5.555 (d, J ˆ 7.0 Hz, 1H), 4.567 (m 2H), 4.48 (m,
1H), 3.815 (m 2H), 1.45 (s, 9H); 13C NMR (125 MHz, [D6]DMSO, 258C):
d ˆ 169.2, 155.9, 143.4, 143.3, 136.9, 128.6, 128.2, 127.9, 124.9, 119.1, 80.9,
73.6, 69.3, 54.6, 28.2.
[21] W. Yuan, Z. Zhong, C.-H. Wong, Bioorg. Med. Chem. Lett. 1991, 1,
551 ± 556.
[22] W. Yuan, B. Munoz, C.-H. Wong, J. Z. Haeggström, A. Wetterholm, B.
Samuelsson, J. Am. Chem. Soc. 1992, 114, 6552 ± 6553.
[23] W. Yuan, B. Munoz, C.-H. Wong, J. Z. Haeggström, A. Wetterholm, B.
Samuelsson, J. Med. Chem. 1993, 36, 211 ± 220.
[24] I. R. Ollmann, J. H. Hogg, C.-H. Wong, J. Z. Haeggström, B.
Samuelsson, Bioorg. Med. Chem. 1995, 3, 211.
[25] J. H. Hogg, I. R. Ollmann, C.-H. Wong, J. Z. Haeggström, B.
Samuelsson, Bioorg. Med. Chem. 1995, 3, 1405.
[26] We previously reported that the Ki value for this compound was 18nm.
This assay was conducted at an enzyme concentration of 20nm.
Further inhibition studies have recently been carried out which model
the data with rate equations designed to deal with tight-binding
inhibitors, specifically when [I] < [E], and thus gave a Ki value of
0.35nm. The d-enantiomer was found to be a competitive inhibitor
(determined from its Ki value) with Kis ˆ 25nm.
[27] M. Minami, N. Ohishi, H. Mutoh, T. Izumi, H. Bito, H. Wada, Y.
Seyama, H. Toh, T. Shimizu, Biochem. Biophys. Res. Commun. 1990,
173, 620 ± 6.
[28] M. Arand, D. F. Grant, J. K. Beetham, T. Friedberg, F. Oesch, B. D.
Hammock, FEBS Lett. 1994, 338, 251.
[29] J. McGee, F. Fitzpatrick, J. Biol. Chem. 1985, 260, 12832.
[30] L. Orning, D. A. Jones, F. A. Fitzpatrick, J. Biol. Chem. 1990, 265,
14911 ± 14916.
[31] M. J. Mueller, A. Wetterholm, M. Blomster, H. Jörnvall, B. Samuels-
son, J. Z. Haeggström, Proc. Natl. Acad. Sci. USA 1995, 92, 8383 ±
8387.
[32] M. J. Mueller, M. B. Andberg, B. Samuelsson, J. Z. Haeggström, J.
Biol. Chem. 1996, 271, 24345 ± 24348.
[33] M. J. Mueller, M. Blomster, U. C. T. Opperman, H. Jörnvall, B.
Samuelsson, J. Haeggström, Proc. Natl. Acad. Sci. USA 1996, 93,
5931 ± 5935.
[34] B. L. Vallee, D. S. Auld, Acc. Chem. Res. 1993, 26, 543 ± 551.
[35] A. Wetterholm, J. F. Medina, O. Raadmark, R. Shapiro, J. Z.
Haeggström, B. L. Vallee, B. Samuelsson, Proc. Natl. Acad. Sci.
USA 1992, 89, 9141 ± 5.
p-Nitroanilide (421 mg, 1.0 mmol) was dissolved in conc. HCl/AcOH
(30 mL), followed by stirring for 45 minutes at room temperature. The
acetic acid was removed under reduced pressure followed by coevapora-
tion with toluene (3 Â 30 mL). The resultant yellow oil was purified by
silica-gel chromatography (5% MeOH, 1% triethylamine in dichloro-
1
methane). Yield: 266 mg (84%). H NMR (400 MHz, [D6]DMSO, 258C):
d ˆ 8.23 (d, J ˆ 7.5 Hz, 2H), 7.92 (d, J ˆ 7.5 Hz, 2H), 7.3 (m, 5H), 5.0 (vbrs,
2H), 4.50 (s, 2H), 3.62 (m, 2H), 3.59 (m, 1H); 13C NMR (100 MHz,
[D6]DMSO, 258C): d ˆ 173.5, 145.0, 142.2, 138.2, 128.2, 127.4, 125.0, 118.9,
‡
72.5, 72.1, 55.7; HRMS [M‡Na] calcd for C16H17N3O4: 316.1297, found:
316.1307.
Received: February 20, 1998 [F1019]
[36] A. L. Maycock, M. S. Anderson, D. M. DeSousa, F. A. Kuehl, Jr., J.
Biol. Chem. 1982, 257, 13911 ± 13914.
[37] T. D. Penning, L. J. Askonas, S. W. Djuric, R. A. Haack, S. S. Yu, M. L.
Michener, G. G. Krivi, E. Y. Pyla, Bioorg. Med. Chem. Lett. 1995, 5,
2517 ± 2522.
[1] B. Samuelsson, C. D. Funk, J. Biol. Chem. 1989, 264, 19469 ± 19472.
[2] A. W. Ford-Hutchinson, Fed. Proc. 1985, 44, 25 ± 29.
[3] A. W. Ford-Hutchinson, M. Bray, Nature 1980, 286, 264 ± 265.
[4] B. Samuelsson, S.-E. Dahlen, J. A. Lindgren, C. A. Rouzer, C. N.
Serhan, Science 1987, 2378, 1171.
1712
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1998
0947-6539/98/0409-1712 $ 17.50+.50/0
Chem. Eur. J. 1998, 4, No. 9

Leukotriene A4 Hydrolase
1698±1713
Ê
[38] B. L. Parnas, R. C. Durley, E. E. Rhoden, B. F. Kilpatrick, N. Makkar,
K. E. Thomas, W. G. Smith, D. G. Corley, J. Nat. Prod. 1996, 59, 962 ±
964.
[41] A. Wetterholm, J. F. Medina, O. Radmark, R. Shapiro, J. Z. Haegg-
ström, B. L. Vallee, B. Samelsson, Biochim. Biophys. Acta 1991, 1080,
96 ± 102.
[39] D. T. S. Rijkers, H. P. H. M. Adams, C. Hemker, G. I. Tesser, Tetrahe-
dron 1995, 51, 11235 ± 11250.
[42] P.-J. Jakobsson, B. Odlander, H.-E. Claesson, Eur. J. Biochem. 1991,
196, 395 ± 400.
Ä
[40] A. Wetterholm, J. Z. Haeggström, B. Samuelsson, W. Yuan, B. Munoz,
[43] D. M. Ritchie, D. C. Argentieri, B. L. Aparicio, R. K. Plante, C. Y.
Lau, A. G. Barbone, Immunopharmacology 1995, 17, 805.
C.-H. Wong, J. Pharm. Exp. Ther. 1995, 275, 31.
Chem. Eur. J. 1998, 4, No. 9
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1998
0947-6539/98/0409-1713 $ 17.50+.25/0
1713