Novel 3-amino-2-hydroxy acids containing protease inhibitors. Part 1: Synthesis and kinetic characterization as aminopeptidase P inhibitors

Article abstract of DOI:10.1016/j.bmc.2005.05.040

Novel, potent inhibitors of aminopeptidase P, containing a 3-amino-2-hydroxy acid and a proline or a proline analogues, have been prepared. One part of the bestatin-derived inhibitors was found to inhibit APP from Escherichia coli and from rat intestine a

Full text of DOI:10.1016/j.bmc.2005.05.040

Bioorganic & Medicinal Chemistry 13 (2005) 4806–4820
Novel 3-amino-2-hydroxy acids containing protease inhibitors.
Part 1: Synthesis and kinetic characterization
as aminopeptidase P inhibitors
Angela Stockel-Maschek,a,* Beate Stiebitz,a Regine Koelschb and Klaus Neuberta
¨
aMartin-Luther-University Halle-Wittenberg, Department of Biochemistry, Institute of Biochemistry/Biotechnology,
Kurt-Mothes-Str.3, 06120 Halle/Saale, Germany
bMartin-Luther-University Halle-Wittenberg, Medical Faculty, Institute of Physiological Chemistry,
Hollystr. 1, 06114 Halle/Saale, Germany
Received 1 December 2004; accepted 24 May 2005
Abstract—Novel, potent inhibitors of aminopeptidase P, containing a 3-amino-2-hydroxy acid and a proline or a proline analogues,
have been prepared. One part of the bestatin-derived inhibitors was found to inhibit APP from Escherichia coli and from rat intes-
tine according to a mixed-type mechanism, with Ki values up to 1.26 lM. The other compounds, 3-amino-2-hydroxy acyl prolines of
a different configuration, inhibit APP competitively, according to a slow-binding mechanism, with Ki values in the nanomolar up to
the micromolar range.
ꢀ 2005 Elsevier Ltd. All rights reserved.
1. Introduction
bradykinin.6 Kitamura et al.7 and Prechel et al.8 have
shown that APP together with angiotensin-converting
enzyme inactivates bradykinin.
Aminopeptidase P (APP, EC 3.4.11.9) is a high specific
exopeptidase cleaving N-terminal Xaa-Pro peptide
bonds. Originally, APP was isolated from Escherichia
coli.1 Later it was found in other bacteria and mam-
mals.2 The enzyme is multimer. In mammals, APP oc-
curs membrane bound by a glycosylinositol anchor or
in soluble form.2c,3 It was shown that recombinant
APP from E. coli contains a dinuclear manganese cen-
ter.4 The enzyme is a member of the methionyl amino-
peptidase family (M24). This peptidase family
represents a set of aminopeptidases (APP, methionine
aminopeptidase (MetAP), and prolidase). All these en-
zymes contain cocatalytic metal atoms and are so-called
ꢀpita-breadꢁ-fold enzymes.5
3-Amino-2-hydroxy acids are the characteristic structural
element of several important natural products and thera-
peutics, such as taxol, bestatin, and HIV-protease inhibi-
tors.9 The natural occurring aminopeptidase inhibitors
bestatin, amastatin, probestin, and leuhistin contain
the 3-amino-2-hydroxy acid in the configuration
2S,3R.9c,10 The 3-amino-2-hydroxy acyl residue takes
part in the chelation of the Zn2+ in the active site of
metal-dependent enzymes.11 Bestatin and amastatin are
described as very potent inhibitors of leucine aminopepti-
dase (LAP, EC 3.4.11.1) and aminopeptidase M (APM,
EC 3.4.11.2).10a,11,12 However, only millimolar concen-
trations of both compounds are able to inhibit APP.2b,c
The reason seems to be the different substrate specificities
of these aminopeptidases. Bestatin and amastatin
contain Leu and Val, respectively, at the P01-position,
while APP has an absolute requirement for Pro or proline
homologous structures in this position.13 Therefore, we
synthesized different stereoisomers of 3-amino-2-
hydroxyacyl-prolines, 3-amino-2-hydroxy acid pyrrolid-
ides, and thiazolidides as potential inhibitors of APP
On account of its unique substrate specificity, APP is in-
volved in protein turnover of collagen and regulation
of biological active peptides, such as substance P and
Keywords: Aminopeptidase P; Mixed-type inhibition; Slow-binding
inhibition; Bestatin.
*
Corresponding author. Tel.: +49 345 5524902; fax: +49 345
5527011; e-mail: stoeckel@biochemtech.uni-halle.de
0968-0896/$ - see front matter ꢀ 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2005.05.040

A. Sto¨ckel-Maschek et al. / Bioorg. Med. Chem. 13 (2005) 4806–4820
4807
Scheme 1. General structure of APP inhibitors.
(Scheme 1). Our investigations show that the compounds
exhibit different inhibition mechanisms. For the first time,
slow-binding inhibitors of APP are described.
Fehrentz and Castro14, and Herranz et al.15 (Scheme
2). Thus, Z-protected (R)- or (S)-amino acid N-meth-
yl-O-methylcarboxamides were reduced to the corre-
sponding a-amino aldehydes by LiAlH4.15 These
aldehydes were converted to a mixture of two diaste-
reomers of Z-protected 3-amino-2-hydroxy acid methyl
esters via O-(trimethylsilyl)cyanhydrin. As described by
Herranz et al.15 this synthesis yielded favored the threo
isomer of the Z-protected 3-amino-2-hydroxy acid
methyl ester with the configuration (2S,3R) or
(2R,3S) starting from the (R)- or (S)-amino acid,
respectively. The two diastereomers were separated
by flash chromatography and saponified, subsequently.
Pure diastereomers of Z-protected 3-amino-2-hydroxy
acids were coupled with (S)-Pro-OMe, (S)-Pro-(S)-
Phe-OMe, pyrrolidine, or thiazolidine using the
carbodiimide method. Acidolytic or hydrogenolytic
deprotection of Z-protecting group provided the
expected final compounds in the following cases:
(2S,3R)-AHPB-(S)-Pro-(S)-Phe-OMe (16), (2S,3R)-
AHPB-Pyrr (17), (2S,3R)-AHPB-Thia (18), (2R,3R)-
AHPB-Thia (20), (2R,3S)-AHPB-Thia (23), and
(2S,3R)-AHMH-Thia (26).
2. Results and discussion
2.1. Chemistry
3-Amino-2-hydroxy acids containing peptides and
amides were designed as potential APP inhibitors based
on the fact that bestatin and amastatin are very potent
inhibitors of APM, LAP, and aminopeptidase B.11–13
Peptides (AHPB/AHMH-Pro; AHPB, 3-amino-2-hy-
droxy-4-phenylbutanoic acid; AHMH, 3-amino-2-hy-
droxy-5-methylhexanoic acid), as well as amino acid
pyrrolidides (AHPB-Pyrr; Pyrr, pyrrolidide) and amino
acid thiazolidides (AHPB/AHMH-Thia; Thia, thiazolid-
ides) were synthesized and investigated for their inhibito-
ry efficiency toward APP of a different origin (Scheme 1).
Z-protected (Z, benzyloxycarbonyl) 3-amino-2-hydroxy
acids were synthesized using methods described by
Scheme 2. Synthesis of bestatin-derived inhibitors of APP. (i) 1. TMSCN, 2. HCl/Ether/MeOH, and 3. H2O; (ii) NaOH/dioxane; (iii) EDC/HOBT/
amino compound; and (iv) HBr/acetic acid or H2/Pd/C.

4808
A. Sto¨ckel-Maschek et al. / Bioorg. Med. Chem. 13 (2005) 4806–4820
In contrast, the deprotection of the AHPB- and
AHMH-containing dipeptide methyl esters (1, 5, 7, 9,
and 10) resulted in nonhomogeneous products. Depend-
ing on the configuration of the 3-amino-2-hydroxy acid,
up to three reaction products were obtained. As an
example, the deprotection of compound 1 is shown in
Scheme 3, yielding the methyl ester 13, the acid 14,
and the diketodiazepine 15. After hydrogenation of Z-
(2S,3R)-AHPB-(S)-Pro-OMe (1), the mass spectrum
showed three peaks representing three reaction prod-
ucts. The expected (2S,3R)-AHPB-(S)-Pro-OMe (13)
was detected only by mass spectrometry (m/z [M+H]+
307.21). It was possible to isolate the unexpected di-
ketodiazepine 15 (Scheme 3), which was identified by
diketodiazepine 22, which was also identified by mass
spectrometry (m/z [M+H]+ 275.13) and NMR spectros-
copy. The remaining residue was saponified and purifi-
cation by HPLC yielded the N- and C-terminal
unprotected dipeptide (2S,3R)-AHPB-(S)-Pro-OH (21).
In the case of deprotection of the dipeptide Z-(2S,3S)-
AHPB-(S)-Pro-OMe (9), the reaction was started with
saponification. Subsequent hydrogenation in methanol
yielded the dipeptide methyl ester (2S,3S)-AHPB-(S)-
Pro-OMe (24), characterized by mass spectrometry
1
(m/z [M+H]+ 307.21) and NMR spectroscopy. The H
NMR spectrum exhibits a typical singlet for the protons
of the methyl ester group at 3.67 ppm.
1
mass spectrometry (m/z [M+H]+ 275.13), and H and
1
13C NMR spectroscopy. In the H NMR spectrum of
The dipeptides, (2R,3R)-AHPB-(S)-Pro-OH (19) and
(2S,3R)-AHMH-(S)-Pro-OH (25), were isolated as
homologous products after hydrogenation, immediately
following saponification and purification by HPLC of
the protected dipeptides Z-(2R,3R)-AHPB-(S)-Pro-
OMe (5) and Z-(2S,3R)-AHMH-(S)-Pro-OMe (10),
respectively.
15, a singlet peak for the protons of a methyl ester
group, which was expected for (2S,3R)-AHPB-(S)-Pro-
OMe (13), was not found. A doublet for the proton of
the hydroxyl group was detected at 6.33 ppm
(3J(H,H) = 4.3 Hz). Furthermore, the intensity of the
signal of the nitrogen-bound hydrogen (NH) represent-
ed only one proton. The second isolated product was
(2S,3R)-AHPB-(S)-Pro-OH (14), characterized by mass
At first, formation of seven-membered rings of the
diketodiazepines (3S,4R,7S)-4-benzyl-3-hydroxy-1,5-di-
azabicyclo-[5.3.0]-decan-2,6-dion (15) and (3R,4S,7S)-
4-benzyl-3-hydroxy-1,5-diazabicyclo-[5.3.0]-decan-2,6-di-
on (22), during the deprotection of the Z-protected
dipeptide methyl esters, was a surprising result, but such
ring formations are not unknown in peptide chemistry.
Xaa-Pro dipeptides form 2,5-diketopiperazines, as they
1
spectrometry (m/z [M+H]+ 293.15) and H NMR spec-
troscopy. The acidolytic deprotection (HBr/AcOH) of
the dipeptide Z-(2S,3R)-AHPB-(S)-Pro-OMe (1) also
yielded a mixture of products and crystallization pro-
vided the desired dipeptide methylester 13.
Crystallization, after hydrogenolytic deprotection of the
dipeptide Z-(2R,3S)-AHPB-(S)-Pro-OMe (7), gave the
contain a higher amount of cis peptide bonds.16
A
Scheme 3. Deprotection of Z-(2S,3R)-AHPB-(S)-Pro-OMe (1) as an example for the synthesis of 3-amino-2-hydroxy acyl-(S)-proline, 3-amino-2-
hydroxy acyl-(S)-proline methyl ester, and 4-benzyl-3-hydroxy-1,5-diazabicyclo-[5.3.0]-decan-2,6-dion (diketodiazepin 15 and 22). (i) HBr/acetic acid
or H2/Pd/C; (ii) aqueous NaOH (0.1 N) in dioxane; and (iii) HBr/acetic acid or H2/Pd/C.

A. Sto¨ckel-Maschek et al. / Bioorg. Med. Chem. 13 (2005) 4806–4820
4809
similar reaction was observed by Sakurai et al.17 during
the synthesis of HIV-protease inhibitors. The tripeptide,
(2S,3S)-3-(N-benzyloxycarbonyl-L-asparaginyl)amino-2-
hydroxy-4-phenylbutyryl-L-proline tert-butyl ester, was
converted into the diketooxazine (3S,6S,10S)-3-{10-(N-
benzyloxycarbonyl-L-asparaginyl)-amino-20-phenylethyl}-
1-aza-4-oxabicyclo[4.3.0]-nonan-2.5-dion by intramolec-
ular cyclization.
known that the IC50 values differ around 180 times for
the inhibition of rat APP by the apstatin analogous
compounds (2S,3R)- and (2S,3S)-AHMH-L-Pro-L-Ala-
NH2.19 In contrast to the dipeptides, the Ki values for
the inhibition of bacterial APP by the thiazolidides 18,
20, and 23 are in the same range, irrespective of config-
uration. In the case of rat APP, there are some differen-
2.2. Mixed-type inhibition of APP
One part of the investigated compounds inhibited rat
APP and recombinant E. coli APP, according to a linear
mixed-type mechanism, with inhibition constants in the
micromolar range (Table 1, Scheme 1). In proof of con-
cept, the Dixon plot of the inhibition of rat APP by
(2S,3R)-AHPB-Thia 18 is shown in Figure 1. All lines
are linear and have a common point of intersection in
the second quadrant (Fig. 1A). Linearity of the lines
means the EIS-complex is catalytically inactive. Gener-
ally, APP is inhibited according to a competitive or a
linear mixed-type mechanism. The replot of slopes ver-
sus 1/[S] allows us to distinguish between pure competi-
tive and linear mixed-type inhibition (a 5 1). The
inhibition of APP is a linear mixed-type one, because
the line in the replot of the slopes does not pass through
the origin, as shown in Fig. 1B.18 Furthermore, replots
of slopes versus 1/[S], as well as intercepts versus 1/[S],
were used to calculate the kinetic constants Ki and a
(Eqs. 3 and 4).
Like thiazolidide 18, dipeptide methyl esters (13, 24), tri-
peptide methyl ester (16), pyrrolidide (17), and other
thiazolidides (20, 23, and 26) inhibit APP from both
sources according to the linear mixed-type mechanism
(Scheme 4). In all cases, the inhibition constants Ki are
in the micromolar range (Table 1). Nevertheless, config-
uration of a 3-amino-2-hydroxy acyl residue exerts an
influence on the Ki value. A comparison of the Ki values
of the dipeptide methyl esters (2S,3R)-AHPB-(S)-Pro-
OMe 13 and (2S,3S)-AHPB-(S)-Pro-OMe 24 demon-
strates a great influence of the configuration of the car-
bon atom C-3. The Ki values differ around 60 times and
40 times for E. coli APP and rat APP, respectively. Both
enzymes prefer the 3R configuration. Furthermore, it is
Figure 1. Inhibition of rat APP by (2S,3R)-AHPB-Thia 18. The data
were obtained by the hydrolysis of Lys(Abz)-Pro-Pro-4NA (2–10 lM)
in 40 mM Tricine buffer, pH 7.4, containing 2.4 mM MnCl2 and
different concentrations of 18 at 30 ꢁC. Final enzyme concentration
was 0.24 nM. (A) Dixon plot 1/v vs. [I]. (B) Replot of slopes (s) and
intercepts ( ) of the Dixon plot vs. 1/[S].
•
Table 1. Linear mixed-type inhibition of APP
Inhibitor
E. coli APPa
Kic (lM)
Rat APPb
a
Kic (lM)
a
13
16
17
18
20
23
24
26
(2S,3R)-AHPB-Pro-OMe
(2S,3R)-AHPB-Pro-Phe-OMe
(2S,3R)-AHPB-Pyrr
2.60
1.26
1.5
1.7
24.0
57.0
37.6
27.4
269
0.83
0.84
1.1
5.2
3.8
2.9
2.4
6.6
19.8
2.2
(2S,3R)-AHPB-Thia
(2R,3R)-AHPB-Thia
30.7
60.3
28.2
0.44
2.3
(2R,3S)-AHPB-Thia
(2S,3S)-AHPB-Pro-OMe
(2S,3R)-AHMH-Thia
0.38
0.62
2.1
78.9
854
158
22.9
4.32
a Inhibition of E. coli APP-catalyzed hydrolysis of Lys(Abz)-Pro-Pro-4NA (4–15 lM) was monitored in 40 mM Tris buffer, pH 7.4, containing
0.75 mM MnCl2 and 24 nM APP at 30 ꢁC, as well as different concentrations of inhibitor.
b lRat APP was assayed using Lys(Abz)-Pro-Pro-4NA (2–20 lM) in 40 mM Tricine buffer, pH 7.4, containing 2.4 mM MnCl2, different concen-
trations of inhibitor, and 0.24 nM APP at 30 ꢁC.
c Calculations were performed, as indicated in the experimental section.

4810
A. Sto¨ckel-Maschek et al. / Bioorg. Med. Chem. 13 (2005) 4806–4820
Scheme 4. Kinetic model of linear mixed-type inhibition.
ces between the Ki values of diastereomeric thiazolidides
(18, 20, and 23). In addition, the IC50 values of (2S,3R)-
and (2R,3S)-AHMH-L-Pro-L-Ala-NH2 are also in the
same range for the inhibition of different mammalian
APPs.19
Figure 2. Progress curves of the inhibition of rat APP by (2S,3R)-
AHPB-(S)-Pro-OH 14. APP (0.12 nM) was incubated with Lys(Abz)-
Pro-Pro-4NA (7 lM) in the presence and absence of 14 (20 lM) in
Tricine buffer, pH 7.4, with 2.4 mM MnCl2 at 30 ꢁC. (1) Without
inhibitor, (2) start of the reaction by the addition of APP, (3)
preincubation of APP and 14 and start of the reaction by addition of S.
In general, our results and data from the literature indi-
cate that amides and peptides containing a 3-amino-2-
hydroxy acid are able to inhibit APP from different
sources according to the linear mixed-type mecha-
nism.8,19 Thus, the mixed-type inhibitor apstatin
((2S,3R)-AHPB-Pro-Pro-Ala-NH2), described by Pre-
chel et al.8, has Ki values of 2.6 and 0.64 lM for APP
from rat lung and human lung, respectively. That
means, apstatin is a slightly more potent inhibitor of
mammalian APP than even the best mixed-type inhibi-
tor (26) described here. This could have been caused
by substrate specificity of the enzyme. The new mixed-
type inhibitors correspond only partially to the substrate
specificity of rat APP. Previous studies from Yoshimoto
et al.13 showed that mammalian APP does not hydrolyze
dipeptides. Furthermore, an aromatic residue at the
P02-position is disadvantageous. Apstatin corresponds
to these requirements better than our compounds. E.
coli APP does not have such restrictions with regard
to substrate specificity. Therefore, the Ki values of the
investigated compounds are in the same range as apsta-
tin (APPE. coli Ki = 14 lM) or even up to 10 times better,
for example, the tripeptide 16 (APPE. coli Ki = 1.26 lM).8
the enzyme-catalyzed reaction reaches the same steady-
state velocity vs as in the case when the reaction is trig-
gered by addition of the enzyme.
For calculation of the kinetic constants of all inhibitors
exhibiting a slow-binding behavior, each progress curve
was fitted to Eq. 5 to yield values for the first-order rate
constant kobs, the initial velocity vi, and the steady-state
velocity vs. The velocities vi and vs were fitted into Eq. 6
to calculate the inhibition constant Ki and the overall
inhibition constants Kiꢀ, respectively. The constants kon
and koff representing the forward and reverse rate con-
stants, respectively, for a slow conversion of the initial
EI-complex into a tight-binding complex EI* were
determined according to Eqs. 7 and 8 (Scheme 5). A plot
of kobs versus [I] representative for all the investigated
slow-binding inhibitors is shown in Figure 3. The shape
Like Maggiora et al.,19 we found a more effective inhibi-
tion of rat APP by compounds containing AHMH,
instead of AHPB, suggesting a more preferential interac-
tion of the isobutyl side chain with the mammalian en-
zyme. These results have been corroborated by
substrate specificity studies.13
Scheme 5. Kinetic model of slow-binding inhibition.
2.3. Slow-binding inhibition of APP
The mechanism of inhibition of APP by 3-amino-2-hy-
droxy acid-derived dipeptides with a free carboxyl group
at the C-terminus ((2S,3R)-AHPB-(S)-Pro-OH 14,
(2R,3R)-AHPB-(S)-Pro-OH 19, (2R,3S)-AHPB-(S)-
Pro-OH 21, (2S,3R)-AHMH-(S)-Pro-OH 25), and the
diketodiazepine 15 differs from that discussed above.
The single progress curves of APP inhibition have a
characteristic shape (Fig. 2). The nonlinear nature of
these progress curves demonstrates that these com-
pounds are time-dependent inhibitors of APP. This
time-dependent reaction is reversible. Figure 2 also
shows that after preincubation of enzyme and inhibitor
Figure 3. Replot of kobs vs. [I] for the inhibition of APP by (2S,3R)-
AHPB-(S)-Pro-OH 14. The solid line is the fit to Eq. 7.

A. Sto¨ckel-Maschek et al. / Bioorg. Med. Chem. 13 (2005) 4806–4820
4811
of this curve corresponds to the slow-binding mecha-
nism B, according to Morrison and Walsh (Scheme 5).20
A comparison of the above-mentioned N- and C-termi-
nal unprotected dipeptides (14, 19, 21, and 25) with
mixed-type inhibitors discussed above shows that the
slow-binding inhibitors are more potent for E. coli, as
well as for rat APP. It is presumed that free carboxyl
group of the C-terminal unprotected dipeptides (14,
19, 21, and 25) is relevant to the occurrence of an initial
EI-complex entailing efficient inhibition.
In all cases, the Dixon plots 1/vi versus [I] and 1/vs versus
[I] show that all lines have a common point of intersection
in the second quadrant (for example, Fig. 4A). That
means, slow-binding inhibition takes place, according to
a competitive or a linear mixed-type mechanism. Distinc-
tion between competitive and linear mixed-type inhibi-
tion was made by replotting the slopes versus 1/[S] (Fig.
4B). In all cases, the straight line in the replot of slopes
from Dixon plot versus 1/[S] passes through the origin.
That means, N- and C-terminal unprotected dipeptides
14, 19, 21, and 25, as well as the diketodiazepine 15, inhibit
APP competitively, according to a slow-binding mecha-
nism. The kinetic constants are given in Table 2.
(2S,3R)-AHPB-(S)-Pro-OH (14) is the most efficient
inhibitor with an overall inhibition constant Kꢀi of
170 nM and a second-order rate constant kon=Kꢀi of
588,000 sꢁ1 Mꢁ1, the most potent of all known inhibitors
of bacterial APP and is 100-fold more potent than apsta-
tin (Ki = 14 lM).8 (2S,3R)-AHMH-(S)-Pro-OH (25) is
the most efficient inhibitor for rat APP with an overall
inhibition constant Kꢀi of 1.1 lM and a second-order rate
Surprisingly, the diketodiazepine 15 inhibits bacterial
APP and rat APP with overall inhibition constants Kiꢀ
of 970 nM and 5.1 lM, respectively, nearly to the same
extent as the corresponding aliphatic dipeptide (2S,3R)-
AHPB-(S)-Pro-OH 14 (APPE. coli Ki = 170 nM, APPrat
Ki = 2.1 lM).
The slow-binding inhibitors confirm the preference of
both APPs for compounds containing a 3-amino-2-hy-
droxy acid in the configuration (2S,3R). (2S,3R)-
AHPB-(S)-Pro-OH (14) has nearly 5-fold and 100-fold
lower Kꢀi values for both enzymes than (2R,3S)-
AHPB-(S)-Pro-OH (21) and (2R,3R)-AHPB-(S)-Pro-
OH (19), respectively. Nevertheless, the influence of ste-
reochemistry of the AHPB residue in bestatin ((2S,3R)-
AHPB-(S)-Leu-OH) on the inhibition of LAP and ami-
nopeptidase B is higher than in case of the inhibition of
APP by bestatin-derived compounds. Aoyagi et al.21
found differences up to 1000-fold in the IC50 values for
the inhibition of LAP and aminopeptidase B by the four
stereoisomers of bestatin. Both enzymes have an abso-
lute requirement for the 2S-configuration of the 3-ami-
no-2-hydroxy acid, which was confirmed by Rich et al.13
constant kon=Kꢀ of 58,200 sꢁ1 Mꢁ1. Thus, compound 25 is
i
a slightly better inhibitor than apstatin (Ki = 2.6 lM).8
Both APPs do not show a great preference for the
AHPB (14) or AHMH (25) containing dipeptide in re-
gard of the Kꢀi value but the second-order rate constant
kon=Kꢀi differs for both compounds. E. coli APP prefers
the AHPB-containing dipeptide (14), while the rat
APP is inhibited more efficiently by an AHMH-contain-
ing dipeptide (25). These preferences were observed in
the case of mixed-type inhibitors discussed above, too.
2.4. A comparison of the inhibition of APP with prolidase
and methionine aminopeptidase
In general, 3-amino-2-hydroxy acid pyrrolidides and
thiazolidides are potent inhibitors of APP. On the other
hand, common amino acid pyrrolidides and thiazolid-
ides, for example, Phe-Thia and Ile-Thia, are not able
to inhibit APP efficiently.22 For this reason, the question
arose whether 2-amino-3-hydroxy acid pyrrolidides and
thiazolidides are also inhibitors of APP or not. Inhibi-
tion studies with Ser-Thia and Thr-Thia as 2-amino-3-
hydroxy acid thiazolidides revealed that both com-
pounds (5 mM) inhibit APP from both sources only
up to around 25%. These results demonstrate that the
3-amino-2-hydroxy acid represents an essential structur-
al element of the inhibitors described in this paper.
Figure 4. (A) Estimation of the inhibition mechanism and the overall
inhibition constant Kꢀi by the Dixon plot (1/v vs. [I]) for the inhibition
of rat APP by (2S,3R)-AHPB-(S)-Pro-OH 14. vs was determined from
the progress curves according to Eq. 5. (B) Replot of the slopes from
panel A vs. 1/[S].
This result corresponds to X-ray crystallographic investi-
gations of APP and the related enzymes MetAP and proli-
dase.23 All these enzymes contain a dinuclear metal center
attheactivesite.Theyarecharacterizedbyarelativelynar-

4812
A. Sto¨ckel-Maschek et al. / Bioorg. Med. Chem. 13 (2005) 4806–4820
Table 2. Slow-binding inhibition of APP
Inhibitor
Kia (lM)
14
Ki* (lM)
kon (sꢁ1
)
k )
on=Kꢀi (sꢁ1 Mꢁ1
E. coli APPb
14
15
19
21
25
(2S,3R)-AHPB-Pro
Diketodiazepine
4
0.17 0.04
0.97 0.15
0.10
588,000
38,100
1250
130 10
190 30
0.037
0.015
0.023
0.060
(2R,3R)-AHPB-Pro
(2R,3S)-AHPB-Pro
(2S,3R)-AHMH-Pro
12
2
16
14
3
5
0.74 0.08
0.37 0.07
31,100
162,000
Rat APPc
14
15
19
21
25
(2S,3R)-AHPB-Pro
Diketodiazepine
32 5
109 34
2100 400
2.1 0.4
5.1 0.8
170 20
0.026
0.0081
0.014
0.017
0.064
12,400
1590
82
(2R,3R)-AHPB-Pro
(2R,3S)-AHPB-Pro
(2S,3R)-AHMH-Pro
120
110
1
3
11
1.1 0.1
1
1550
58,200
Ki and Kꢀi are the dissociation constant of the initial complex and the overall inhibition constant, respectively. kon is the forward constant for the
conversion to the EI*-complex.
a Progress curves were calculated, as described in the experimental section.
b Inhibition of E. coli APP-catalyzed hydrolysis of Lys(Abz)-Pro-Pro-4NA (4–15 lM) was determined in 40 mM Tris buffer, pH 7.4, containing
0.75 mM MnCl2, different concentrations of inhibitor, and 5 nM APP at 30 ꢁC.
c Rat APP was assayed by use of Lys(Abz)-Pro-Pro-4NA (4–10 lM) in Tricine buffer containing 2.4 mM MnCl2, different concentrations of inhibitor,
and 0.12 nM APP.
row substrate specificity in comparison to other metal-de-
pendent aminopeptidases.23b All three enzymes show a
close agreement of the active site on the coordination of
the two metal atoms and their bridging water molecule,
as well as the distance between the metal atoms.24 All the
above-mentioned aminopeptidases, as well as LAP, con-
tain a metal-bridging water molecule or hydroxide ion,
which should act as a nucleophile during catalysis.23b It
was shown for the inhibition of MetAP by (2S,3R)-3-ami-
no-2-hydroxy heptanoic acid, for the inhibition of proli-
dase by (2S,3R)-AHMH-(S)-Pro-OH, as well as for the
inhibition of LAP by bestatin that the 2-hydroxy group
of the inhibitors substitutes this important water molecule
or hydroxide ion and bridges the two metal atoms.24,25
led to compounds inhibiting APP according to a classi-
cal linear mixed-type mechanism or a competitive slow-
binding mechanism with inhibition constants in the
nanomolar up to the low micromolar range. We have
shown that the 3-amino-2-hydroxy acid is the decisive
structural element of this new inhibitor. A further paper
will describe the inhibition of other metallopeptidases
and serine proteases by these compounds.
4. Experimental section
4.1. General procedures
Commercially available reagents were purchased from
Fluka, Merck, Sigma, Aldrich, and BACHEM (Germa-
ny). Solvents were dried by standard procedures and dis-
tilled before use. Purity of compounds was checked by
thin-layer chromatography (TLC), reversed-phase HPLC
(Merck, Germany) and by capillary electrophoresis (Bio-
In the meantime, Graham et al.26 solved the crystal
structure of E. coli APP, which was in complex with
the inhibitor apstatin. As postulated from MetAP and
prolidase, they found the replacement of the bridging
hydroxyl group between the two metal atoms by the
hydroxy group at the carbon atom C2 of the inhibitor.
These studies explain the competitive binding of an
inhibitor to APP because the mixed-type they show only
an EI-complex. For confirmation of the mixed-type
inhibition mechanism, addition of a substrate is neces-
sary. Unfortunately, the authors did not mention the
mixed-type inhibition pattern of apstatin.
FocusTM, Bio-Rad, Germany), and 1H NMR and/or 13
C
NMR spectroscopy. TLC was performed with Kieselgel
60 F254 plates (Merck, Germany) using chloroform/
MeOH (9:1) or n-BuOH/EtOAc/AcOH/water (1:1:1:1)
mixtures as eluent. Melting points were determined with
a BOETIUS melting point apparatus and are uncorrect-
ed. Optical rotations were measured on a Polamat A (Carl
Zeiss Jena, Germany) or a chiral detector (Knauer,
Germany) at 25 ꢁC. Electrospray mass spectra were
performed in positive mode on a VG-BIO-Q triple–qua-
druple electrospray mass spectrometer (Fisons Instru-
ments, England). Elemental analyses were performed at
the Department of Chemistry, Martin-Luther-University
Nevertheless, (2S,3R)-3-amino-2-hydroxy heptanoic acid
showed a competitive inhibition pattern of the MetAP,
whereas our investigated compounds inhibited APP,
according to the mixed-type mechanism or competitive
slow-binding mechanism.27 Slow-binding inhibition was
also described for the inhibition of LAP by bestatin.12
1
Halle-Wittenberg, Germany. H and 13C NMR spectra
were recorded on Bruker ARX500 and Varian UNITY
500 spectrometers, respectively. Standard methods were
used to perform one- and two-dimensional experiments,
pulse programs having been taken from the Bruker and
Varian software libraries. Resonance assignments were
carried out by a combined analysis of H,H-COSY,
APT, and HETCOR spectra.
3. Conclusion
In conclusion, we have developed very potent inhibitors
of APP starting from the structure of the well-known
aminopeptidase inhibitor bestatin. This approach has

A. Sto¨ckel-Maschek et al. / Bioorg. Med. Chem. 13 (2005) 4806–4820
4813
3-Benzyloxycarbonylamino-2-hydroxy acids were pre-
pared according to the procedure described by Fehrentz
and Castro14 and by Herranz et al.15 starting from ben-
zyloxycarbonyl-(R)-phenylalanine N-methyl-O-methyl-
7.04. Found: C, 66.47; H, 6.65; N, 6.75; MS ESI: m/
z = 588.17 [M+H+].
4.2.4. (2S,3R)-3-Benzyloxycarbonylamino-2-hydroxy-4-
phenylbutanoic acid pyrrolidide (3). Compound 3 was
synthesized by coupling of (2S,3R)-3-benzyloxycar-
bonylamino-2-hydroxy-4-phenylbutanoic acid (290 mg,
0.88 mmol) and pyrrolidine (80 ll, 0.97 mmol). Yield:
carboxamide,
benzyloxycarbonyl-(S)-phenylalanine
N-methyl-O-methylcarboxamide, or benzyloxycarbon-
yl-(R)-leucine N-methyl-O-methylcarboxamide.
4.2. Synthesis of 3-amino-2-hydroxy acid pyrrolidide, 3-
amino-2-hydroxy acid thiazolidides, 3-amino-2-hydroxyacyl-
(S)-proline methyl ester, 3-amino-2-hydroxyacyl-(S)-pro-
line, or 3-amino-2-hydroxyacyl-(S)-prolyl-(S)-phenylal-
anine methyl ester
270 mg (80%); mp 137–143 ꢁC; TLC: Rf = 0.60
20
(CHCl3/MeOH 9:1); ½aꢂ +1.6 (c 1, CHCl3); elemental
D
analysis Calcd (%) for C22H26N2O4 (382.46): C, 69.09;
H, 6.85; N, 7.32; Found: C, 68.86; H, 6.99; N, 6.86;
1
MS ESI: m/z = 383.30 [M+H]+; H NMR (500 MHz,
CDCl3, 30 ꢁC, TMS) d = 1.76 (A, 1H, C3H2-Pyrr; m,
2H, C4H2-Pyrr), 1.91 (B, 1H, C3H2-Pyrr), 2.96 (A, 1H,
C2H2-Pyrr; m, 2H, CcH2-AHPB), 3.22 (B, 1H, C2H2-
Pyrr), 3.31 (A, 1H, C5H2-Pyrr), 3.44 (B, 1H, C5H2-Pyrr),
4.07 (br s, 1H, CaH-AHPB), 4.13 (br s, 1H, OH), 4.20
4.2.1. General procedure. The corresponding Z-protected
3-amino-2-hydroxy acid (1 mmol) and an amino compo-
nent [pyrrolidine, thiazolidine, (S)-proline methyl ester
hydrochloride, or (S)-prolyl-(S)-phenylalanine methyl
ester hydrochloride (1.1 mmol)] were dissolved in
dichloromethane (10 ml). In the case of the amino
components (S)-proline methyl ester hydrochloride or
(S)-prolyl-(S)-phenylalanine methyl ester hydrochloride,
triethylamine (154 ll, 1.1 mmol) was added. The
solution, cooled to 4 ꢁC, was treated with hydroxybenzo-
triazole (170 mg, 1.1 mmol) and N-(3-dimethylaminopro-
2
(m, 1H, CbH-AHPB), 5.01 (A, 1H, J(H,H) = 12.4 Hz,
2
CH2-Z), 5.06 (B, 1H, J(H,H) = 12.4 Hz, CH2-Z), 5.14
(d, 1H, 3J(H,H) = 9.8 Hz, NH), 7.22–7.36 (m, 10H,
phenyl); 13C NMR (500 MHz, CDCl3, 30 ꢁC, TMS)
d = 23.6 (C4H2-Pyrr), 26.0 (C3H2-Pyrr), 38.9 (CcH2-
AHPB), 45.5 (C2H2-Pyrr), 46.5 (C5H2-Pyrr), 53.2
(CbH-AHPB), 66.7 (CH2-Z), 68.4 (CaH-AHPB), 126.7
(CH-Ph), 127.8 (CH-Ph), 128.0 (2 CH-Ph), 128.5 (2
CH-Ph), 128.7 (2 CH-Ph), 129.5 (2 CH-Ph), 136.6
(CH-Ph), 137.8 (CH-Ph), 156.0 (CO-Z), 170.0 (CO-
AHPB).
pyl)-N0-ethylcarbodiimide
hydrochloride
(210 mg,
1.1 mmol). The mixture was allowed to stir for 1 h at
4 ꢁC and overnight at room temperature. The organic
layer was subsequently washed with water, 0.2 M HCl,
brine, saturated NaHCO3, and brine. The reaction
mixture was dried over MgSO4. The solvent was removed
under reduced pressure and the amide was obtained as an
oil or was crystallized from EtOAc/n-hexane.
4.2.5. (2S,3R)-3-Benzyloxycarbonylamino-2-hydroxy-4-
phenylbutanoic acid thiazolidide (4). Compound 4 was
synthesized by coupling of (2S,3R)-3-benzyloxycarbon-
ylamino-2-hydroxy-4-phenylbutanoic acid (290 mg,
0.88 mmol) and thiazolidine (91 mg, 0.97 mmol). Yield:
4.2.2. (2S,3R)-3-Benzyloxycarbonylamino-2-hydroxy-4-
phenylbutanoyl-(S)-proline methyl ester (1). Compound 1
was synthesized by coupling of (2S,3R)-3-benzyloxycar-
bonylamino-2-hydroxy-4-phenylbutanoic acid (300 mg,
0.91 mmol) and (S)-proline methyl ester hydrochloride
360 mg (90%, E/Z-isomers); decomposition >126 ꢁC;
20
TLC: Rf = 0.66 (CHCl3/MeOH 9:1); ½aꢂ ꢁ5.7 (c 1,
D
CHCl3); elemental analysis Calcd (%) for C21H24N2O4S
(400.50): C, 62.98; H, 6.04; N, 6.99; S, 8.01. Found: C,
63.04; H, 6.28; N, 6.57; S, 8.12; MS ESI: m/z = 401.25
[M+H+]; 1H NMR (500 MHz, CDCl3, 27 ꢁC, TMS)
major isomer: d = 2.95 (m, 4H, C4H2-Thia, CcH2-
AHPB), 3.25 (A, 1H, C5H2-AHPB), 3.48 (B, 1H,
C5H2-Thia), 4.00 (br s, 1H, CaH-AHPB), 4.09 (br s,
1H, OH), 4.17 (m, 1H, CbH-AHPB), 4.24 (A, 1H,
C2H2-Thia), 4.67 (B, 1H, C2H2-Thia), 5.04 (m, 2H,
CH2-Z), 5.14 (br s, 1H, NH), 7.23–7.35 (m, 10H, phen-
yl); minor isomer: d = 2.95 (m, 4H, C4H2-Thia, CcH2-
AHPB), 3.48 (A, 1H, C5H2-Thia), 3.90 (A, 1H, C2H2-
Thia, B, 1H, C5H2-Thia), 4.00 (br s, 1H, CaH-AHPB),
4.09 (br s, 1H, OH), 4.17 (m, 1H, CbH-AHPB), 4.24
(B, 1H, C2H2-Thia), 5.04 (m, 2H, CH2-Z), 5.14 (br s,
1H, NH), 7.23–7.35 (m, 10H, phenyl); 13C NMR
(500 MHz, CDCl3, 30 ꢁC, TMS) major isomer:
d = 30.9 (C5H2-Thia), 38.7 (CcH2-AHPB), 47.7 (C4H2-
Thia), 48.6 (C2H2-Thia), 53.6 (CbH-AHPB), 66.8
(CH2-Z), 68.7 (CaH-AHPB), 126.9 (CH-Ph), 127.9
(CH-Ph), 128.1 (CH-Ph), 128.5 (CH-Ph), 128.8 (CH-
Ph), 129.4 (CH-Ph), 136.5 (CH-Ph), 137.5 (CH-Ph),
137.6 (CH-Ph), 156.0 (CO-Z), 170.1 (CO-AHPB); minor
isomer: d = 29.0 (C5H2-Thia), 38.7 (CcH2-AHPB), 47.2
(C2H2-Thia), 49.1 (C4H2-Thia), 53.6 (CbH-AHPB),
66.8 (CH2-Z), 68.4 (CaH-AHPB), 126.9 (CH-Ph),
(165 mg, 1 mmol). Yield: 290 mg (72%); decomposition
20
>136 ꢁC; TLC: Rf = 0.61 (CHCl3/MeOH 9:1); ½aꢂ
D
ꢁ12.3 (c 1, CHCl3); elemental analysis Calcd (%) for
1
2
C24H28N2O6 ꢃ H2O (449.50): C, 64.13; H, 6.50; N 6.23.
Found: C, 64.66; H, 6.48; N, 5.57; MS ESI: m/z =
441.45 [M+H+]; 13C NMR (500 MHz, CDCl3, 30 ꢁC,
TMS) d = 25.3 (CcH-Pro), 28.6 (CbH-Pro), 38.5 (CdH-
AHPB), 46.5 (CdH-Pro), 52.2 (OCH3), 54.0 (CbH-
AHPB), 59.6 (CaH-Pro), 66.8 (CH2-Z), 68.5
(CaH-AHPB), 126.7 (CH-Ph), 128.1 (CH-Ph), 128.5 (2
CH-Ph),128.7 (2 CH-Ph), 129.5 (2 CH-Ph), 129.6 (2
CH-Ph), 136.8 (CH-Ph), 138.0 (CH-Ph), 155.8 (CO-Z),
171.0 (CO-AHPB), 171.6 (CO-Pro).
4.2.3. (2S,3R)-3-Benzyloxycarbonylamino-2-hydroxy-4-
phenylbutanoyl-(S)-prolyl-(S)-phenylalanine methyl ester
(2). Compound 2 was synthesized by coupling of
(2S,3R)-3-benzyloxycarbonylamino-2-hydroxy-4-phenyl-
butanoic acid (290 mg, 0.88 mmol) and (S)-prolyl-
(S)-phenylalanine
(303 mg, 0.97 mmol). Yield: 336 mg (65%); mp 57–
methyl
ester
hydrochloride
20
62 ꢁC; TLC: Rf = 0.59 (CHCl3/MeOH 9:1); ½aꢂD
ꢁ28.8 (c 1, CHCl3); elemental analysis Calcd (%) for
1
2
C33H37N3O7 ꢃ H2O (596.68): C, 66.43; H 6.42; N

4814
A. Sto¨ckel-Maschek et al. / Bioorg. Med. Chem. 13 (2005) 4806–4820
127.9 (CH-Ph), 128.1 (CH-Ph), 128.5 (CH-Ph), 128.8
(CH-Ph), 129.4 (CH-Ph), 136.5 (CH-Ph), 137.5 (CH-
Ph), 137.6 (CH-Ph), 156.0 (CO-Z), 170.1 (CO-AHPB).
4.2.8. (2R,3S)-3-Benzyloxycarbonylamino-2-hydroxy-4-
phenylbutanoyl-(S)-proline methyl ester (7). Compound
7 was synthesized by coupling of (2R,3S)-3-benzyl-
oxycarbonylamino-2-hydroxy-4-phenylbutanoic
(330 mg, 1 mmol) and (S)-proline methyl ester hydro-
acid
4.2.6. (2R,3R)-3-Benzyloxycarbonylamino-2-hydroxy-4-
phenylbutanoyl-(S)-proline methyl ester (5). Compound
5 was synthesized by coupling of (2R,3R)-3-benzyl-
chloride (182 mg, 1.1 mmol). Yield: 330 mg (75%); mp
20
158–164 ꢁC; TLC: Rf = 0.65 (CHCl3/MeOH 9:1); ½aꢂD
oxycarbonylamino-2-hydroxy-4-phenylbutanoic acid
(100 mg, 0.3 mmol) and (S)-proline methyl ester hydro-
chloride (58 mg, 0.35 mmol). Yield: 100 mg (76 %);
ꢁ56.0 (c 1, CHCl3); elemental analysis Calcd (%) for
1
C24H28N2O6 ꢃ H2O (449.50): C, 64.13; H, 6.50; N,
6.23. Found: 2C, 64.76; H. 6.53; N, 5.82; ESI MS:
m/z = 441.31 [M+H]+; 13C NMR (500 MHz, CDCl3,
27 ꢁC, TMS) d = 24.6 (CcH2-Pro), 28.6 (CbH2-Pro),
38.6 (CcH2-AHPB), 46.0 (CdH2-Pro), 52.3 (OCH3),
52.9 (CbHNH), 59.5 (CaH-Pro), 66.7 (CH2-Z), 68.4
(CaHOH), 126.8 (CH-Ph), 127.8 (2 CH-Ph), 128.0
(CH-Ph), 128.5 (2 CH-Ph), 128.7 (2 CH-Ph), 129.4 (2
CH-Ph), 136.5 (CH-Ph), 137.5 (CH-Ph), 156.0 (CO-Z),
170.7 (CO-AHPB), 172.3 (CO-Pro).
20
oil; TLC: Rf = 0.61 (CHCl3/MeOH 9:1); ½aꢂ ꢁ22.0
D
(c 2.9, CHCl3); elemental analysis Calcd (%) for
1
C24H28N2O6 ꢃ H2O (449.50): C, 64.13; H, 6.50; N,
6.23. Found: 2 C, 64.37; H, 6.45; N, 5.89; MS ESI:
m/z = 441.45 [M+H+]; 13C NMR (500 MHz, CDCl3,
27 ꢁC, CHCl3) d = 24.5 (CcH2-Pro), 28.6 (CbH2-Pro),
34.9 (CcH2-AHPB), 46.3 (CdH2-Pro), 52.3 (OCH3),
53.6 (CbH-AHPB), 59.3 (CaH-Pro), 66.8 (CH2-Z), 71.0
(CaH-AHPB), 126.6 (2 CH-Ph), 127.9 (4 CH-Ph),
129.3 (2 CH-Ph), 129.4 (2 CH-Ph), 136.4 (CH-Ph),
137.7 (CH-Ph), 156.1 (CO-Z), 169.9 (CO-AHPB),
172.2 (CO-Pro).
4.2.9. (2R,3S)-3-Benzyloxycarbonylamino-2-hydroxy-4-
phenylbutanoic acid thiazolidide (8). Compound 8 was
synthesized by coupling of (2R,3S)-3-benzyloxycarbon-
ylamino-2-hydroxy-4-phenylbutanoic acid (330 mg,
1 mmol) and thiazolidine (103 mg, 1.1 mmol). Yield:
4.2.7. (2R,3R)-3-Benzyloxycarbonylamino-2-hydroxy-4-
phenylbutanoic acid thiazolidide (6). Compound 6 was
synthesized by coupling of (2R,3R)-3-benzyloxycarbon-
ylamino-2-hydroxy-4-phenylbutanoic acid (220 mg,
0.67 mmol) and thiazolidine (70 mg, 0.74 mmol) Yield:
330 mg (71%, E/Z-isomers); mp 133–137 ꢁC; TLC:
20
Rf = 0.64 (CHCl3/MeOH 9:1); ½aꢂ +4.5 (c 0.7, CHCl3);
D
elemental analysis Calcd (%) for C21H24N2O4S ꢃ 1 12 H2O
(427.52): C, 59.00; H, 6.36; N, 6.55; S, 7.50. Found: C,
58.53; H, 5.76; N, 6.05; S, 7.59; MS ESI: m/z = 401.16
230 mg (86%, E/Z-isomers); mp 116–122 ꢁC; TLC:
20
1
Rf = 0.65 (CHCl3/MeOH 9:1); ½aꢂ +6.5 (c 1, CHCl3);
[M+H]+; H NMR (500 MHz, CDCl3, 30 ꢁC, TMS) iso-
D
elemental analysis Calcd (%) for C21H24N2O4S
(400.50): C, 62.98; H, 6.04; N, 6.99; S, 8.01. Found:
C, 62.60; H, 6.21; N, 6.64; S, 8.03; MS ESI:
m/z = 401.16 [M+H]+; 1H NMR (500 MHz, CDCl3,
30 ꢁC, TMS) isomer 1: d = 2.76 (A, 1H, C4H2-Thia),
2.86 (m, 2H, CcH2-AHPB), 3.02 (B, 1H, C4H2-Thia),
3.78 (A, 1H, C5H2-Thia), 3.90 (B, 1H, C5H2-Thia),
3.97 (br s, 1H, OH), 4.17 (m, 1H, CbH-AHPB),
mer 1: d = 2.97 (m, 4H, C4H2-Thia, CcH2-AHPB), 3.26
(A, 1H, C5H2-Thia), 3.48 (B, 1H, C5H2-Thia), 3.97 (br
s, 1H, OH), 4.10 (m, 1H, CaH-AHPB), 4.16 (m, 1H,
CbH-AHPB), 4.24 (A, 1H, C2H2-Thia), 4.67 (B, 1H,
C2H2-Thia), 5.03 (A, 1H, 2J(H,H) = 16.2 Hz, CH2-Z),
2
5.06 (B, 1H, J(H,H) = 15.9 Hz, CH2-Z), 5.10 (d, 1H,
3J(H,H) = 8.7 Hz, NH), 7.23–7.36 (m, 10H, phenyl);
isomer 2: d = 2.97 (m, 4H, C4H2-Thia, CcH2-AHPB),
3.48 (A, 1H, C5H2-Thia), 3.91 (A, 1H, C2H2-Thia; B,
1H, C5H2-Thia), 3.97 (br s, 1H, OH), 4.16 (m,
2H, CaH-AHPB, CbH-AHPB), 4.24 (B, 1H, C2H2-
Thia), 5.03 (A, 1H, 2J(H,H) = 16.2 Hz, CH2-Z), 5.06
(B, 1H, 2J(H,H) = 15.9 Hz, CH2-Z), 5.10 (d, 1H,
3J(H,H) = 8.7 Hz, NH), 7.23–7.36 (m, 10H, phenyl);
13C NMR (500 MHz, CDCl3, 30 ꢁC, TMS) isomer 1:
d = 30.9 (C5H2-Thia), 38.7 (CcH2-AHPB), 47.4 (C4H2-
Thia), 48.7 (C2H2-Thia), 53.6 (CbH-AHPB), 66.8
(CH2-Z), 68.7 (CaH-AHPB), 126.9 (CH-Ph), 127.9
(CH-Ph), 128.1 (2 CH-Ph), 128.5 (2 CH-Ph), 128.8 (2
CH-Ph), 129.4 (2 CH-Ph), 136.5 (CH-Ph), 137.5 (CH-
Ph), 156.0 (CO-Z), 170.1 (CO-AHPB); isomer 2:
d = 29.0 (C5H2-Thia), 38.7 (CcH2-AHPB), 47.2 (C2H2-
Thia), 49.1 (C5H2-Thia), 53.6 (CbH-AHPB), 66.8
(CH2-Z), 68.4 (CaH-AHPB), 126.9 (CH-Ph), 127.9
(CH-Ph), 128.1 (2 CH-Ph), 128.5 (2 CH-Ph), 128.8 (2
CH-Ph), 129.4 (2 CH-Ph), 136.5 (CH-Ph), 137.5 (CH-
Ph), 156.0 (CO-Z), 170.1 (CO-AHPB).
2
4.20 (A, 1H, J(H,H) = 10.2 Hz, C2H2-Thia), 4.50 (m,
1H, CaH-AHPB), 4.65 (B, 1H, 2J(H,H) = 10.0 Hz,
C2H2-Thia), 5.06 (s, 2H, CH2-Z), 5.33 (d, 1H,
3J(H,H) = 3.9 Hz, NH), 7.15–7.36 (m, 10H, phenyl);
isomer 2: d = 2.86 (A, 1H, C4H2-Thia, m, 2H, CcH2-
AHPB), 2.95 (B, 1H, C4H2-Thia), 3.46 (A, 1H, C5H2-
Thia), 3.90 (B, 1H, C5H2-Thia), 3.97 (br s, 1H, OH),
4.17 (m, 1H, CbH-AHPB), 4.52 (A, 1H,
2J(H,H) = 9.4 Hz, C2H2-Thia), 4.57 (m, 1H, CaH-
AHPB), 4.61 (B, 1H, 2J(H,H) = 9.4 Hz, C2H2-Thia),
5.06 (s, 2H, CH2-Z), 5.33 (d, 1H, 3J(H,H) = 3.9 Hz,
NH), 7.15–7.36 (m, 10H, phenyl); 13C NMR
(500 MHz, CDCl3, 27 ꢁC, TMS); isomer 1: d = 31.1
(C5H2-Thia), 34.3 (CcH2-AHPB), 48.0 (C4H2-Thia),
48.9 (C2H2-Thia), 54.0 (CbH-AHPB), 66.9 (CH2-Z),
71.4 (CaH-AHPB), 126.7 (CH-Ph), 127.9 (CH-Ph),
128.2 (CH-Ph), 128.5 (CH-Ph), 129.1 (CH-Ph), (CH-
Ph), 136.1 (CH-Ph), 137.1 (CH-Ph), 137.2 (CH-Ph),
156.1 (CO-Z), 169.5 (CO-AHPB); isomer 2: d = 29.0
(C5H2-Thia), 34.3 (CcH2-AHPB), 48.0 (C4H2-Thia),
48.5 (C2H2-Thia), 54.0 (CbH-AHPB), 66.9 (CH2-Z),
71.0 (CaH-AHPB), 126.7 (CH-Ph), 127.9 (CH-Ph),
128.2 (CH-Ph), 128.5 (CH-Ph), 129.1 (CH-Ph), 136.1
(CH-Ph), 137.1 (CH-Ph), 137.2 (CH-Ph), 156.1 (CO-
Z), 169.2 (CO-AHPB).
4.2.10. (2S,3S)-3-Benzyloxycarbonylamino-2-hydroxy-4-
phenylbutanoyl-(S)-proline methyl ester (9). Compound 9
was synthesized by coupling of (2S,3S)-3-benzyloxycar-
bonylamino-2-hydroxy-4-phenylbutanoic acid (180 mg,
0.54 mmol) and (S)-proline methyl ester hydrochloride

A. Sto¨ckel-Maschek et al. / Bioorg. Med. Chem. 13 (2005) 4806–4820
4815
(92 mg, 0.56 mmol). Yield: 150 mg (63%); mp 60–62 ꢁC;
TLC: Rf = 0.60 (CHCl3/MeOH 9:1); elemental analysis
chloride (93 mg, 0.56 mmol). Yield: 110 mg (58%); oil;
20
D
TLC: Rf = 0.60 (CHCl3/MeOH 9:1); ½aꢂ ꢁ46 (c 0.6,
1
Calcd (%) for C24H28N2O6 ꢃ H2O (444.78): C, 64.78;
CHCl3);
elemental
analysis
Calcd
(%)
for
H, 6.45; N, 6.29. Found: C,4 64.78; H, 6.14; N, 6.07;
C21H30N2O6 ꢃ H2O (415.48): C, 60.71; H, 7.52; N, 6.74.
1
2
Found: C, 60.82; H, 7.48; N, 6.53; MS ESI:
MS ESI: m/z = 441.31 [M+H]+.
m/z = 407.24 [M+H]+.
4.2.11. (2S,3R)-3-Benzyloxycarbonylamino-2-hydroxy-5-
methylhexanoyl-(S)-proline methyl ester (10). Compound
10 was synthesized by coupling of (2S,3R)-3-benzyl-
4.2.14.
(2S,3R)-3-Amino-2-hydroxy-4-phenylbutanoyl-
(S)-proline methyl ester hydrobromide (13). Compound 1
(100 mg, 0.23 mmol) was suspended in HBr/AcOH
(33%, 0.5 ml) at room temperature. After completion of
deprotection, controlled by TLC, the product was precip-
itated by addition of dry diethyl ether, filtered off, washed
several times with diethyl ether, and recrystallized from
MeOH/diethyl ether. Yield: 63 mg (52%, cis/trans-iso-
mers); decomposition >104 ꢁC; elemental analysis Calcd
oxycarbonylamino-2-hydroxy-5-methylhexanoic
(220 mg, 0.74 mmol) and (S)-proline methyl ester hydro-
acid
chloride (135 mg, 0.82 mmol). Yield: 240 mg (79%, cis/
trans-isomers); mp 112–117 ꢁC; TLC: Rf = 0.60
20
D
(CHCl3/MeOH 9:1); ½aꢂ ꢁ39.4 (c 1.1, CHCl3); elemen-
tal analysis Calcd (%) for C21H30N2O6 (406.48): C,
62.05; H, 7.44; N, 6.52. Found: C, 61.88; H, 7.48; N,
6.52; MS ESI: m/z = 407.24 [M+H]+; 13C NMR
(500 MHz, CDCl3, 27 ꢁC, TMS) major isomer:
d = 22.3 (CeH3-AHMH), 23.2 ðCeH30 -AHMHÞ, 24.7
(CdH-AHMH), 25.4 (CcH2-Pro), 28.6 (CbH2-Pro), 40.9
(CcH2-AHMH), 46.9 (CdH-Pro), 50.2 (CbH-AHMH),
52.2 (OCH3), 59.5 (CaH-Pro), 66.9 (CH2-Z), 70.8
(CaH-AHMH), 128.0 (CH-Ph), 128.1 (2 CH-Ph), 128.4
(2 CH-Ph), 136.7 (CH-Ph), 156.0 (CO-Z), 170.8 (CO-
1
2
(%) for C16H22N2O4 ꢃ HBr ꢃ H2O (396.28): C, 48.49; H,
6.10; N, 7.07. Found: C, 48.35; H, 6.00; N, 6.83; MS
ESI: m/z = 307.21 [M+H]+; 13C NMR (500 MHz,
[D6]DMSO, 27 ꢁC, DMSO) major isomer: d = 24.5
(CcH2-Pro), 28.5 (CbH2-Pro), 34.6 (CcH2-AHPB), 46.6
(CdH2-Pro), 52.0 (OCH3), 54.2 (CbH-AHPB), 58.8
(CaH-Pro), 66.8 (CaH-AHPB), 127.1 (CH-Ph), 128.6 (2
CH-Ph), 129.4 (2 CH-Ph), 136.0 (CH-Ph), 169.1
(CO-AHPB), 172.0 (CO-Pro); minor isomer: d = 24.4
(CcH2-Pro), 28.4 (CbH2-Pro), 35.0 (CcH2-AHPB), 46.3
(CdH2-Pro), 51.0 (OCH3), 53.6 (CbH-AHPB), 58.6
(CaH-Pro), 66.1 (CaH-AHPB), 127.0 (CH-Ph), 128.7
(2 CH-Ph), 129.5 (2 CH-Ph), 135.9 (CH-Ph), 169.3
(CO-AHPB), 172.6 (CO-Pro).
AHMH), 171.7 (CO-Pro); minor isomer: d = 22.0
0
(CeH3-AHMH), 22.9 (CeH3 -AHMH), 24.7 (CdH-
AHMH), 25.4 (CcH2-Pro), 31.4 (CbH2-Pro), 41.3
(CcH2-AHMH), 47.4 (CdH-Pro), 50.1 (CbH-AHMH),
52.8 (OCH3), 58.4 (CaH-Pro), 66.9 (CH2-Z), 71.1
(CaH-AHMH), 127.8 (CH-Ph), 128.1 (2 CH-Ph), 128.4
(2 CH-Ph), 136.6 (CH-Ph), 156.1 (CO-Z), 170.7 (CO-
AHMH), 172.1 (CO-Pro).
4.2.15.
(2S,3R)-3-Amino-2-hydroxy-4-phenylbutanoyl-
(S)-proline (14) and (3S,4R,7S)-4-benzyl-3-hydroxy-1,5-
diazabicyclo-[5.3.0]-decan-2,6-dion (15). Compound 1
(374 mg, 0.85 mmol) was dissolved in MeOH. The cata-
lyst Pd/C was added. Subsequently, the mixture was
hydrogenated at room temperature. After completion
of the reaction controlled by TLC, the catalyst was deac-
tivated and filtered off. The solvent was evaporated, and
compounds 14 and 15 were isolated by fractionated
crystallization from methanol/diethyl ether and dried
over P4O10.
4.2.12. (2S,3R)-3-Benzyloxycarbonylamino-2-hydroxy-5-
methylhexanoic acid thiazolidide (11). Compound 11 was
synthesized by coupling of (2S,3R)-3-benzyloxycarbon-
ylamino-2-hydroxy-5-methylhexanoic acid (380 mg,
1.3 mmol) and thiazolidine (130 mg, 1.4 mmol). Yield:
280 mg (60%, E/Z-isomers); mp 141–144 ꢁC; TLC:
20
D
Rf = 0.65 (CHCl3/MeOH 9:1); ½aꢂ
ꢁ42.0 (c 1.1,
CH3OH); elemental analysis Calcd (%) for
C18H26N2O4S (366.48): C, 58.99; H, 7.15; N, 7.64; S,
8.75. Found: C, 58.85; H, 7.21; N, 7.10; S, 8.63; MS
ESI: m/z = 367.17 [M+H]+; 13C NMR (500 MHz,
CDCl3, 27 ꢁC, TMS) major isomer: d = 22.3 (CeH3-
AHMH), 23.1 ðCeH03-AHMHÞ, 24.8 (CdH-AHMH),
31.0 (C5H2-Thia), 41.5 (CcH2-AHMH), 48.0 (C2H2-
Thia), 48.7 (C4H2-Thia), 50.1 (CbH-AHMH), 66.7
(CH2-Z), 71.7 (CaH-AHMH), 127.8 (CH-Ph), 128.0 (2
CH-Ph), 128.5 (2 CH-Ph), 136.6 (CH-Ph), 156.1 (CO-
Z), 170.0 (CO-AHMH); minor isomer: d = 22.3 (CeH3-
AHMH), 23.1 ðCeH03-AHMHÞ, 24.8 (CdH-AHMH),
29.0 (C5H2-Thia), 41.5 (CcH2-AHMH), 47.6 (C4H2-
Thia), 49.2 (C2H2-Thia), 50.1 (CbH-AHMH), 66.7
(CH2-Z), 71.5 (CaH-AHMH), 127.8 (CH-Ph), 128.0
(2 CH-Ph), 128.5 (2 CH-Ph), 136.6 (CH-Ph), 156.1
(CO-Z), 169.7 (CO-AHMH).
4.2.15.1. (2S,3R)-3-Amino-2-hydroxy-4-phenylbuta-
noyl-(S)-proline (14). Yield: 87 mg (35%); mp 126–
137 ꢁC;
elemental
analysis
Calcd
(%)
for
C15H20N2O4 ꢃ 1 12 H2O (319.355): C, 56.42; H, 7.26; N,
8.77. Found: C, 56.68; H, 7.16; N, 8.53; MS ESI:
m/z = 293.15 [M+H]+; 13C NMR (500 MHz, [D6]DMSO,
27 ꢁC; DMSO) d = 24.5 (CcH2-Pro), 28.6 (CbH2-Pro),
34.4 (CdH2-AHPB), 46.6 (CdH2-Pro), 54.4 (CbH-
AHPB), 58.9 (CaH-Pro), 66.7 (CaH-AHPB), 127.0
(CH-Ph), 128.7 (2 CH-Ph), 129.5 (2 CH-Ph), 136.2
(CH-Ph), 169.0 (CO-AHPB), 173.0 (COOH).
4.2.15.2. (3S,4R,7S)-4-Benzyl-3-hydroxy-1,5-diazabi-
cyclo-[5.3.0]-decan-2,6-dion (15). Yield: 75 mg (27%);
mp 135–143 ꢁC; elemental analysis Calcd (%) for
1
4.2.13. (2R,3R)-3-Benzyloxycarbonylamino-2-hydroxy-5-
methylhexanoyl-(S)-proline methyl ester (12). Compound
12 was synthesized by coupling of (2R,3R)-3-benzyl-
oxycarbonylamino-2-hydroxy-5-methylhexanoic acid
(140 mg, 0.47 mmol) and (S)-proline methyl ester hydro-
C15H18N2O3 ꢃ H2O (283.32): C, 63.59; H, 6.76; N,
9.89. Found: 2C, 63.42; H, 6.74; N, 9.58; MS ESI:
m/z = 275.13 [M+H]+; 1H NMR (500 MHz, [D6]DMSO,
30 ꢁC, TMS) d = 1.64 (m, 2H, C9H2), 2.08 (A, 1H,
C8H2), 2.33 (B, 1H, C8H2), 2.85 (A, 1H, CH2-benzyl),

4816
A. Sto¨ckel-Maschek et al. / Bioorg. Med. Chem. 13 (2005) 4806–4820
2.93 (B, 1H, CH2-benzyl), 3.55 (m, 2H, C10H2), 3.58 (m,
4.2.18.
(2S,3R)-3-Amino-2-hydroxy-4-phenylbutanoic
3
1H, C4H), 3.81 (d, 1H, J(H,H) = 4.3, C3H), 4.99 (m,
acid thiazolidide hydrobromide (18). Compound 18 was
obtained by acidolytic deprotection of compound 4
(140 mg, 0.35 mmol), according to compound 13. Yield:
100 mg (82%, E/Z-isomers); mp 98–106 ꢁC; ½aꢂ +7.9
3
1H, C7H), 6.33 (d, 1H, J(H,H) = 4.3 Hz, OH), 7.23–
7.33 (m, 6H, phenyl, NH); 13C NMR (500 MHz,
[D6]DMSO, 30 ꢁC, TMS) d = 21.8 (C9H2), 29.2
(C8H2), 38.0 (CH2-benzyl), 46.8 (C10H2), 56.9 (C4H),
57.3 (C7H), 70.8 (C3H), 126.1 (CH-Ph), 128.5 (2 CH-
Ph), 129.2 (2 CH-Ph), 137.2 (C-Ph), 169.3 (C2), 169.8
(C6).
20
D
(c 1, CH3OH); elemental analysis Calcd (%) for
C13H18N2O2S ꢃ HBr ꢃ 1 14 AcOH (422.34): C, 44.08; H,
5.73; N, 6.63; S, 7.59. Found: C, 43.91; H, 5.53; N,
7.12; S, 7.27; MS ESI: m/z = 266.97 [M+H]+; 13C
NMR (500 MHz, [D6]DMSO, 27 ꢁC, TMS) major iso-
mer: d = 30.6 (C5H2-Thia), 35.0 (CcH2-AHPB), 48.1
(C4H2-Thia), 48.3 (C2H2-Thia), 53.7 (CbH-AHPB),
67.0 (CaH-AHPB), 126.9 (CH-Ph), 128.5 (2 CH-Ph),
129.3 (2 CH-Ph), 136.0 (CH-Ph), 168.6 (CO); minor iso-
mer: d = 28.4 (C5H2-Thia), 35.0 (CcH2-AHPB), 47.5
(C2H2-Thia), 48.7 (C4H2-Thia), 53.7 (CbH-AHPB),
64.8 (CaH-AHPB), 126.9 (CH-Ph), 128.5 (2 CH-Ph),
29.3 (2 CH-Ph), 136.0 (CH-Ph), 168.3 (CO).
4.2.16.
(2S,3R)-3-Amino-2-hydroxy-4-phenylbutanoyl-
(S)-prolyl-(S)-phenylalanine methyl ester hydrochloride
(16). Compound 2 (200 mg, 0.34 mmol) was dissolved
in MeOH. The catalyst Pd/C was added. Subsequently,
the mixture was hydrogenated at room temperature.
After completion of the reaction controlled by TLC,
the catalyst was deactivated and filtered off. The crude
product, obtained by evaporation, was treated with
HCl/MeOH (4.6 m). Compound 16 was precipitated
by dry diethyl ether, filtered off, and dried in vacuo
4.2.19. (2R,3R)-3-Amino-2-hydroxy-4-phenylbutanoyl-
(S)-proline trifluoroacetate (19). Compound 5 (205 mg,
0.46 mmol) was dissolved in MeOH. The catalyst Pd/C
was added. Subsequently, the mixture was hydrogenated
at room temperature. After completion of the reaction
controlled by TLC, the catalyst was deactivated and
filtered off. After hydrogenolytic deprotection, the
crude product, obtained by evaporation, was dissolved
in dioxane (8 ml) and aqueous NaOH (0.1 m, 8 ml).
The reaction mixture was stirred at room temperature.
After completion of saponification, the reaction mixture
was neutralized with 1 M HCl and evaporated to dry-
ness. Purification was succeeded by RP-HPLC (column
RP-8, HiBarꢂ 25 lm, 250 mm Merck) with acetonitrile–
water (45/55, v/v) containing trifluoroacetic acid (0.04%,
v/v) as eluent. Compound 19 became amorphous. Yield:
147 mg (76%); decomposition >139 ꢁC; elemental analy-
over P4O10. Yield: 95 mg (57%, cis/trans-isomers); mp
20
D
98–106 ꢁC; ½aꢂ ꢁ30.8 (c 1, CH3OH); elemental analy-
1
4
sis Calcd (%) for C25H31N3O5 ꢃ HCl ꢃ H2O (494.50): C,
60.72; H, 6.62; N, 8.50. Found: C, 60.67; H, 6.59; N,
8.17; MS ESI: m/z = 454.48 [M+H]+; 13C NMR
(500 MHz, [D6]DMSO, 27 ꢁC, DMSO) major isomer:
d = 24.2 (CcH2-Pro), 29.0 (CbH2-Pro), 34.5 (CcH2-
AHPB), 36.5 (CbH2-Phe), 46.8 (CdH2-Pro), 51.8
(OCH3), 53.8 (CbH-AHPB), 54.4 (CaH-Phe), 59.5
(CaH-Pro), 66.4 (CaH-AHPB), 126.6 (CH-Ph), 127.0
(CH-Ph), 128.2 (2 CH-Ph), 128.6 (2 CH-Ph), 129.3 (2
CH-Ph), 129.5 (2 CH-Ph), 136.2 (CH-Ph), 136.3
(CH-Ph), 169.1 (CO-AHPB), 171.5 (CO-Pro), 171.8
(CO-Phe); minor isomer: d = 23.9 (CcH2-Pro), 28.3
(CbH2-Pro), 34.5 (CcH2-AHPB), 36.5 (CbH2-Phe),
46.5 (CdH2-Pro), 51.9 (OCH3), 53.8 (CbH-AHPB),
54.4 (CaH-Phe), 59.4 (CaH-Pro), 66.2 (CaH-AHPB),
126.6 (CH-Ph), 127.0 (CH-Ph), 128.2 (2 CH-Ph),
128.6 (2 CH-Ph), 129.3 (2 CH-Ph), 129.5 (2 CH-Ph),
136.2 (CH-Ph), 136.3 (CH-Ph), 169.2 (CO-AHPB),
171.5 (CO-Pro), 171.9 (CO-Phe).
sis Calcd (%) for C15H20N2O4 ꢃ 1 12 TFA ꢃ H2O (472.38):
1
C, 45.77; H, 4.80; N, 5.93. Found: C, 45.279; H, 5.06; N,
5.72; MS ESI: m/z = 293.16 [M+H]+; 13C NMR
(500 MHz, [D6]DMSO, 27 ꢁC, DMSO) d = 24.3 (CcH2-
Pro), 28.3 (CbH2-Pro), 33.5 (CcH2-AHPB), 46.3
(CdH2-Pro), 54.0 (CbH-AHPB), 58.8 (CaH-Pro), 67.6
(CaH-AHPB), 126.8 (CH-Ph), 128.5 (2 CH-Ph), 129.3
(2 CH-Ph), 136.1 (CH-Ph), 168.3 (CO-AHPB), 172.9
(COOH).
4.2.17.
(2S,3R)-3-Amino-2-hydroxy-4-phenylbutanoic
acid pyrrolidide hydrochloride (17). Compound 17 was
obtained from compound 3 (170 mg, 0.44 mmol) by
the same procedure described for 16. Yield: 78 mg
20
(62%); mp 167–172 ꢁC; ½aꢂ ꢁ11.7 (c 1, CH3OH);
4.2.20.
(2R,3R)-3-Amino-2-hydroxy-4-phenylbutanoic
D
elemental analysis Calcd (%) for C14H20N2O2ꢃ
acid thiazolidide hydrobromide (20). Compound 20 was
obtained by acidolytic deprotection of 6 (130 mg,
0.32 mmol) by the same procedure as described for 13.
Yield: 82 mg (75%, E/Z-isomers); mp 106–114 ꢁC; ½aꢂ
1
4
HCl ꢃ H2O (289.29): C, 58.13; H, 7.49; N, 9.68. Found:
C, 58.20; H, 7.33; N, 9.41; MS ESI: m/z = 249.23
[M+H]+; 1H NMR (500 MHz, [D6]DMSO, 30 ꢁC,
DMSO) d = 1.73 (m, 2H, C3H2-Pyrr), 1.77 (m, 2H,
C4H2-Pyrr), 2.89 (A, 1H, CcH2-AHPB), 2.97 (B, 1H,
CcH2-AHPB), 3.24 (A, 1H, C5H2-Pyrr; m, 2H, C2H2-
Pyrr), 3.49 (B, 1H, C5H2-Pyrr), 3.56 (m, 1H, CbH-
AHPB), 4.16 (m, 1H, CaH-AHPB), 6.47 (d, 1H,
3J(H,H) = 6.0 Hz, OH), 7.25–7.35 (m, 5H, phenyl),
8.01 (br s, 2H, NH2); 13C NMR (500 MHz, [D6]DMSO,
27 ꢁC, DMSO) d = 23.5 (C4H2-Pyrr), 25.5 (C3H2-Pyrr),
35.0 (CcH2-Pyrr), 45.7 (C2H2-Pyrr), 45.8 (C5H2-Pyrr),
53.9 (CbH-AHPB), 66.1 (CaH-AHPB), 126.9 (CH-Ph),
128.5 (2 CH-Ph), 129.4 (2 CH-Ph), 136.3 (CH-Ph),
168.8 (CO).
20
D
ꢁ2.0 (c 1, CH3OH); elemental analysis Calcd (%) for
C13H19N2O2S ꢃ HBr 12 AcOH (378.31): C, 44.45; H, 5.86;
N, 7.40; S, 8.48. Found: C, 44.81; H, 5.62; N, 7.35; S,
8.23; MS ESI: 267.15 [M+H]+; 13C NMR (500 MHz,
CD3OD, 27 ꢁC, CH3OH) major isomer: d = 31.8
(C5H2-Thia), 35.6 (CcH2-AHPB), 50.2 (C2H2-Thia),
56.7 (CbH-AHPB), 68.6 (CaH-AHPB), 128.6 (CH-Ph),
128.9 (CH-Ph), 130.3 (CH-Ph), (CH-Ph), 135.8 (CH-
Ph), 136.6 (CH-Ph), 169.9 (CO); minor isomer:
d = 29.8 (C5H2-Thia), 35.5 (CcH2-AHPB), 49.9 (C2H2-
Thia), 56.7 (CbH-AHPB), 68.5 (CaH-AHPB), 128.6
(CH-Ph), 128.9 (CH-Ph), 130.3 (CH-Ph), 135.8

A. Sto¨ckel-Maschek et al. / Bioorg. Med. Chem. 13 (2005) 4806–4820
4817
(CH-Ph), 136.6 (CH-Ph), 169.7 (CO). For both isomers
the signal of C4H2 is buried under the CH3OH signal at
49.0 ppm.
mental analysis Calcd (%) for C13H18N2O2Sꢃ
1 14 HBr ꢃ AcOH (427.55): C, 42.14; H, 5.48; N, 6.55; S,
7.52. Found: C, 42.42; H, 5.30; N, 6.99; S, 7.49; MS
ESI: m/z = 266.97 [M+H]+; 13C NMR (500 MHz,
[D6]DMSO, 27 ꢁC, DMSO) major isomer: d = 30.7
(C5H2-Thia), 35.1 (CcH2-AHPB), 48.3 (C4H2-Thia),
48.4 (C2H2-Thia), 53.7 (CbH-AHPB), 67.0 (CaH-
AHPB), 127.0 (CH-Ph), 128.6 (2 CH-Ph), 129.3 (2
CH-Ph), 136.0 (CH-Ph), 168.7 (CO); minor isomer:
d = 28.5 (C5H2-Thia), 35.1 (CcH2-AHPB), 47.6 (C2H2-
Thia), 48.9 (C4H2-Thia), 53.7 (CbH-AHPB), 67.0
(CaH-AHPB), 127.0 (CH-Ph), 128.6 (2 CH-Ph), 129.3
(2 CH-Ph), 136.0 (CH-Ph), 168.3 (CO).
4.2.21.
(2R,3S)-3-Amino-2-hydroxy-4-phenylbutanoyl-
(S)-proline trifluoroacetate (21) and (3R,4S,7S)-4-benzyl-
3-hydroxy-1,5-diazabicyclo-[5.3.0]-decan-2,6-dion (22).
From compound 7 (154 mg, 0.35 mmol), the N-terminal
protecting group was removed by catalytic hydrogenol-
ysis, following the same procedure as described for 14
and 15. Compound 22 was isolated by crystallization
from MeOH/diethyl ether. Compound 22 was filtered
off and the solvent was evaporated. Dioxane (6 ml)
and 0.1 M aqueous NaOH (6 ml) were added to the
remaining residue. After completion of saponification,
compound 21 was isolated and purified by RP-HPLC
as described for 19.
4.2.23.
(2S,3S)-3-Amino-2-hydroxy-4-phenylbutanoyl-
(S)-proline methyl ester (24). Dioxane (3 ml) and aque-
ous NaOH (0.1 M, 3 ml) were added to 9 (140 mg,
0.32 mmol). The reaction mixture was stirred at room
temperature. After completion of saponification, the
dioxane was removed under reduced pressure and the
residue was diluted with water (10 ml). The aqueous
solution was extracted with ethyl acetate to remove
remaining 9 and, subsequently, acidified to pH 2–3 with
1 M HCl. The C-terminal free dipeptide was extracted
with ethyl acetate. The combined extracts were washed
with brine and dried over MgSO4. The solvent was re-
moved under reduced pressure. The N-terminal depro-
tection was carried out by Pd-catalyzed hydrogenolysis
as described for 14 and 15. (2S,3S)-AHPB-(S)-Pro-
OMe was obtained, instead of the expected acid. Ester-
ification by the solvent methanol took place. Yield:
4.2.21.1. (2R,3S)-3-Amino-2-hydroxy-4-phenylbuta-
noyl-(S)-proline trifluoroacetate (21). Yield: 72 mg
20
D
0.8, CH3OH); elemental analysis Calcd (%) for
(49%, cis/trans-isomers); mp 91–99 ꢁC; ½aꢂ ꢁ27.8 (c
1
C15H20N2O4 ꢃ 1 12 TFA ꢃ H2O (472.38): C, 45.77; H,
2
4.80; N, 5.93. Found: C, 45.44; H, 4.97; N, 5.92; MS
ESI: m/z = 293.16 [M+H]+; 13C NMR (500 MHz,
[D6]DMSO, 27 ꢁC, TMS) major isomer: d = 28.4
(CbH2-Pro), 34.9 (CcH2-AHPB, CcH2-Pro), 46.2
(CdH2-Pro), 53.7 (CbH-AHPB), 58.7 (CaH-Pro), 66.1
(CaH-AHPB), 126.9/127.0 (CH-Ph), 128.6/128.7 (2
CH-Ph), 129.4/129.5 (2 CH-Ph), 136.0/136.1 (CH-Ph),
169.1 (CO-AHPB), 173.0 (COOH); minor isomer:
d = 28.5 (CbH2-Pro), 34.9 (CcH2-AHPB, CcH2-Pro),
46.4 (CdH2-Pro), 53.6 (CbH-AHPB), 58.9 (CaH-Pro),
66.8 (CaH-AHPB), 126.9/127.0 (CH-Ph), 128.6/128.7
(2 CH-Ph), 129.4/129.5 (2 CH-Ph), 136.0/136.1 (CH-
Ph), 169.0 (CO-AHPB), 173.3 (COOH).
60 mg
(61%,
cis/trans-isomers);
decomposition
>121 ꢁC; elemental analysis Calcd (%) for C16H22N2O4Æ-
H2O (324.37): C, 59.24; H, 7.46; N, 8.64. Found: C,
59.28; H, 7.49; N, 8.43; MS ESI: m/z = 307.16
[M+H]+; 1H NMR (500 MHz, [D6]DMSO, 30 ꢁC,
TMS) major isomer: d = 1.90 (m, 4H, CbH2-Pro,
CcH2-Pro), 2.91 (m, 2H, CcH2-AHPB), 3.51 (m, 1H,
CbH-AHPB), 3.64 (B, CdH2-Pro), 3.67 (s, 3H, OCH3),
4.33 (m, 2H, CaH-AHPB, CaH-Pro), 6.17 (br s, 1H,
OH), 7.22–7.37 (m, 5H, phenyl), 7.95 (br s, 2H, NH2);
minor isomer: d = 1.90 (m, 2H, CcH2-Pro), 2.23 (m,
2H, CbH2-Pro), 2.79 (m, 2H, CcH2-AHPB), 3.51 (m,
1H, CbH-AHPB), 3.64 (B, 1H, CdH2-Pro), 3.67 (s, 3H,
OCH3), 4.33 (m, 2H, CaH-AHPB, CaH-Pro), 6.17 (br
s, 1H, OH), 7.22–7.37 (m, 5H, phenyl), 7.95 (br s, 2H,
NH2). For both isomers, the signal of A (1H) of
CdH2-proline is buried under the water signal at
3.29 ppm.
4.2.21.2. (3R,4S,7S)-4-Benzyl-3-hydroxy-1,5-diazabi-
cyclo-[5.3.0]-decan-2,6-dion (22). Yield: 15 mg (16%);
decomposition >234 ꢁC; elemental analysis Calcd (%)
1
for C15H18N2O3 ꢃ H2O (283.32): C, 63.59; H, 6.76; N,
9.89. Found: C, 263.65; H, 6.72; N, 9.65; MS ESI:
m/z = 275.10 [M+H]+; 1H NMR (500 MHz, [D6]DMSO,
30 ꢁC, TMS) d = 1.69 (m, 2H, C9H2), 1.95 (A, 1H,
C8H2), 2.30 (B, 1H, C8H2), 2.66 (A, 1H,
2J(H,H) = 13.7 Hz, CH2-benzyl), 2.88 (B, 1H,
2J(H,H) = 13.6, CH2-benzyl), 3.48 (B, 1H, C10H2),
3.91 (m, 1H, C4H), 4.16 (m, 1H, C3H), 4.57 (t, 1H,
3J(H,H) = 7.3 Hz, C7H), 5.44 (d, 1H, 3J(H,H) = 5.1 Hz,
OH), 7.18–7.30 (m, 5H, phenyl), 7.74 (d, 1H,
3J(H,H) = 6.3 Hz, NH). The signal of the second proton
of C10H2 is buried under the water signal from solvent at
3.29 ppm. 13C NMR (500 MHz, [D6]DMSO, 30 ꢁC,
TMS) d = 21.9 (C9H2), 28.3 (C8H2), 36.3 (CH2-benzyl),
47.6 (C10H2), 54.1 (C4H), 57.5 (C7H), 70.5 (C3H),
126.2 (CH-Ph), 128.1 (2 CH-Ph), 129.5 (2 CH-Ph),
138.3 (C-Ph), 169.5 (C2), 169.8 (C6).
4.2.24. (2S,3R)-3-Amino-2-hydroxy-5-methylhexanoyl-
(S)-proline trifluoroacetate (25). The deprotection of 10
(100 mg, 0.25 mmol) to yield compound 25 succeeded
by the procedure as described for 19. Yield: 57 mg
(48%, cis/trans-isomers); mp 117–122 ꢁC; elemental
1
analysis Calcd (%) for C12H22N2O4 ꢃ TFA ꢃ 1 12 H2O
4
(427.87): C, 40.70; H, 6.18; N, 6.55. Found: C, 40.63;
H, 6.25; N, 6.39; MS ESI: m/z = 259.20 [M+H]+; 13C
NMR (500 MHz, [D6]DMSO, 27 ꢁC, DMSO)0 major iso-
mer: d = 22.0 ðCeH3-AHMHÞ, 22.7 (CeH3 -AHMH),
23.3 (CdH-AHMH), 24.7 (CcH2-Pro), 28.6 (CbH2-Pro),
37.6 (CcH2-AHMH), 46.5 (CdH2-Pro), 51.0 (CbH-
AHMH), 59.6 (CaH-Pro), 67.9 (CaH-AHMH), 169.1
4.2.22.
(2R,3S)-3-Amino-2-hydroxy-4-phenylbutanoic
acid thiazolidide hydrobromide (23). Compound 23 was
obtained by deprotection of compound 8 (59 mg,
0.15 mmol) according to 13. Yield: 40 mg (77%, E/Z-iso-
20
mers); mp 108–116 ꢁC; ½aꢂ +4.6 (c 1.1, CH3OH); ele-
D

4818
A. Sto¨ckel-Maschek et al. / Bioorg. Med. Chem. 13 (2005) 4806–4820
(CO-AHMH), 173.2 (COOH); minor isomer: d = 22.0
(CeH3-AHMH), 22.8 ðCeH03-AHMHÞ, 23.3 (CdH-
AHMH), 24.7 (CcH2-Pro), 30.6 (CbH2-Pro), 38.6
(CcH2-AHMH), 46.2 (CdH-Pro), 50.7 (CbH-AHMH),
57.4 (CaH-Pro), 68.1 (CaH-AHMH), 170.0 (CO-
AHMH), 173.5 (COOH).
4.3.2.2. Rat APP. Rat intestine APP (25 ll) was
incubated with Tricine–citrate buffer (275 ll, 40 mM
Tricine, 0.6 mM citrate, pH 7.4) containing MnCl2
(1.25 mM) for 1 h at 30 ꢁC.28 The activated enzyme
was diluted with buffer (1.2 ml). The reaction mixture
consisted of Tricine buffer (40 mM, pH 7.4), MnCl2
(2.4 mM), various concentrations of Lys(Abz)-Pro-
Pro-4NA (4–10 lM), and different concentrations of
inhibitor around the expected Ki value. The reaction
was initiated by adding activated enzyme (10 ll). The
final concentration of APP was 0.24 and 0.12 nM for
mixed-type inhibition and slow-binding inhibition,
respectively (MW: 96 kDa).31
4.2.25.
(2S,3R)-3-Amino-2-hydroxy-5-methylhexanoic
acid thiazolidide hydrobromide (26). Compound 26 was
obtained by deprotection of 11 (40 mg, 0.11 mmol)
according to compound 13. Yield: 33 mg (96%, E/Z-iso-
20
mers); mp 161–165 ꢁC; ½aꢂ ꢁ6.1 (c 0.8, CH3OH); elemen-
D
tal analysis Calcd (%) for C10H20N2O2SÆHBrÆAcOH
(373.31): C, 38.61; H, 6.75; N, 7.50; S, 8.59. Found: C,
38.67; H, 6.83; N, 7.23; S, 8.53; ESI MS: m/z = 233.04
[M+H]+; 1H NMR (500 MHz, [D6]DMSO, 27 ꢁC,
TMS) major isomer: d = 0.88 (d, 6H, 3J(H,H) = 6.49 Hz,
CeH3, CeH30 -AHMH), 1.41 (m, 2H, CcH2-AHMH), 1.73
(m, 1H, CdH-AHMH), 3.11 (t, 2H, 3J(H,H) = 6.24 Hz,
C4H2-Thia), 3.82 (A, 1H, C5H2-Thia), 3.91 (B, 1H, C5H2-
Thia), 4.31 (br s, 1H, CaH-AHMH), 4.48 (A, 1H,
2J(H,H) = 10.22 Hz, C2H2-Thia), 4.51 (B, 1H,
2J(H,H) = 10.24 Hz, C2H2-Thia), 6.38 (br s, 1H OH)
7.69 (br s, 2H, NH2); minor isomer: d = 0.88 (d, 6H,
3J(H,H) = 6.49 Hz, CeH3, CeH03-AHMH), 1.41 (m, 2H,
CcH2-AHMH), 1.73 (m, 1H, CdH-AHMH), 3.02 (t, 2H,
3J(H,H) = 6.42 Hz, C4H2-Thia), 3.69 (dm, 2H, C5H2-
Thia), 4.35 (br s, 1H, CaH-AHMH), 4.66 (A, 1H,
2J(H,H) = 9.14 Hz, C2H2-AHMH), 4.72 (B, 1H,
2J(H,H) = 9.12 Hz, C2H2-Thia), 6.38 (br s, 1H OH)
7.69 (br s, 2H, NH2). For both isomers, the signal of
CbH-AHMH is buried under the water signal at
3.22 ppm.
4.3.3. Determination of inhibition constants. All investiga-
tions were started by adding the enzyme and were run in
duplicates. In general, the reaction velocities for both
inhibition mechanisms were calculated over a time inter-
val in which less than 10% cleavage of substrate took
place.
4.3.3.1. Mixed-type inhibition. In the case of mixed-
type inhibition, steady-state kinetics were analyzed using
the equation
V max½Sꢂ
v ¼
;
ð1Þ
½Iꢂ
Kmð1 þ K½IꢂÞ þ ½Sꢂð1 þ
Þ
aKi
i
where Ki is the competitive inhibition constant and Ki
multiplied with factor a represents the uncompetitive
inhibition constant.18 Thus, the equation is applicable
to the three classical reversible inhibition mechanisms,
namely the competitive, uncompetitive, and noncompet-
itive inhibition. However, the Dixon plot (Eq. 2) of the
velocity equation (Eq. 1) does not allow to distinguish
between pure competitive and linear mixed-type inhibi-
tion, a special case of noncompetitive inhibition. In both
cases, the straight lines have a common point of inter-
section within the second quadrant. For this reason,
slopes and intercepts from Dixon plot (Eq. 2) were rep-
lotted versus 1/[S] to determine the inhibition mecha-
nism and the constants Ki, as well as, a (Eqs. 3 and 4).
4.3. Kinetic studies
4.3.1. Enzymes. Recombinant APP from E. coli was a
gift from T. Yoshimoto, Nagasaki University, Japan.
Membrane-bound rat APP was isolated from intestine,
as described previously.28 Lys(Abz)-Pro-Pro-4NA
(Abz, 2-aminobenzoyl; 4NA, 4-nitroanilide) was synthe-
sized, according to standard procedures.29
ꢀ
ꢁ
aKm
ð1 þ ½Sꢂ Þ
1
1
Km
½Sꢂ
4.3.2. Enzyme assays. The activity of APP was deter-
mined by use of the substrate Lys(Abz)-Pro-Pro-4NA
(APPE. coli Km = 40.7 lM, APPrat Km = 3.6 lM) moni-
toring the fluorescence of released Lys(Abz) at an
excitation wavelength of 310 nm and an emission wave-
length of 410 nm at 30 ꢁC on a Perkin-Elmer LS-50B
luminescence spectrometer.29
¼
½Iꢂ þ
1 þ
;
ð2Þ
ð3Þ
ð4Þ
v
aKiV max
V max
Km
1
1
slope ¼
þ
;
KiV max ½Sꢂ aKiV max
4.3.2.1. Bacterial APP. At first, lyophilized E. coli
APP (0.1 mg) was dissolved in Tricine–citrate buffer
(3 ml, 40 mM Tricine, 0.6 mM citrate, pH 7.4) contain-
ing MnCl2 (3 mM). APP was preincubated for 1 h at
30 ꢁC. A typical reaction mixture (1 ml) consisted of Tris
buffer (40 mM, pH 7.4), MnCl2 (0.75 mM), and
Lys(Abz)-Pro-Pro-4NA in various concentrations (4–
15 lM) and different concentrations of inhibitor. The
reaction was initiated by adding activated APP (10 ll).
The final enzyme concentration was 24 nM in the case
of mixed-type inhibition and 5 nM for slow-binding
inhibition (MW: 49,650 Da).30
Km
1
1
intercept ¼
þ
.
V max ½Sꢂ V max
The kinetic data were calculated using the FLWinLab
software 1.10 (Perkin-Elmer, Germany) and GraFit 3.0
(Erithacus Software Ltd, England).
4.3.3.2. Slow-binding inhibition. The single progress
curves of a slow-binding inhibition are described by
the following Eq. 5.20,32
P ¼ vs ꢃ t þ ðvi ꢁ vsÞ ꢃ ð1 ꢁ eꢁk Þ=kobs þ d.
ð5Þ
obsꢃt

A. Sto¨ckel-Maschek et al. / Bioorg. Med. Chem. 13 (2005) 4806–4820
4819
Roderick, S. L.; Huber, R.; Matthews, B. W. Proc. Nat.
Acad. Sci. U.S.A. 1994, 91, 2473–2477.
6. Yaron, A.; Naider, F. Crit. Rev. Biochem. Mol. Biol. 1993,
28, 31–81.
7. Kitamura, S.-I.; Carbini, L. A.; Carretero, O. A.; Simmons,
W. H.; Scicli, A. G. Br. J. Pharmacol. 1995, 114, 6–7.
8. Prechel, M. M.; Orawski, A. T.; Maggiora, V. W.;
Simmons, H. J. Pharmacol. Exp. Ther. 1995, 275, 1136–
1142.
9. (a) Nicolaou, K. C.; Dai, W.-M.; Guy, R. K. Angew.
Chem. 1994, 106, 38–69; (b) Nicolaou, K. C.; Guy, R. K.
Angew. Chem. 1995, 107, 2247–2259; (c) Umezawa, H.;
Aoyagi, T.; Suda, H.; Hamada, M.; Takeuchi, T. J.
Antibiot. 1976, 29, 97–99; (d) Mimoto, T.; Imai, J.;
Hattori, N.; Takahashi, O.; Kisanuki, S.; Nagano, Y.;
Shintani, M.; Hayashi, H.; Sakikawa, H.; Akaji, K.; Kiso,
Y. Chem. Pharm. Bull. 1991, 39, 2465–2467; (e) Mimoto,
T.; Imai, J.; Tanaka, S.; Hattori, N.; Kisanuki, S.; Akaji,
K.; Kiso, Y. Chem. Pharm. Bull. 1991, 39, 3088–3090.
10. (a) Wilkes, S. H.; Prescott, J. M. J. Biol. Chem. 1985, 260,
13154–13162; (b) Yoshida, S.; Nakamura, Y.; Naganawa,
H.; Aoyagi, T.; Takeuchi, T. J. Antibiot. 1990, XLIII, 149–
153; (c) Hecker, S. J.; Werner, K. M. J. Org. Chem. 1993,
58, 1762–1765.
Eq. 5 allows the determination of the initial velocity vi,
the steady-state velocity vs, and the apparent first-order
rate constant kobs for the approach of the steady state.
The constant d is the displacement from zero at t = 0.
P is the concentration of product at time t.
Plots of vi versus [I] and vs versus [I] according to Eq. 6
allowed the determination of the dissociation constant
Ki and the overall inhibition constant Kꢀi , respectively.
V max ꢃ ½Sꢂ
v ¼
.
ð6Þ
½Sꢂ þ Kmð1 þ ½Iꢂ=KiÞ
Eqs. 7 and 8 were used for determination of the detailed
type of slow-binding inhibition, as well as the character-
istic constants koff and kon, respectively.20,32 The con-
stants kon and koff represent the forward and reverse
rate constants, respectively, for a slow conversion of
the initial EI-complex into the tight-binding complex
EI*.
½IꢂKi
kobs ¼ koff þ k
;
ð7Þ
on 1 þ ½Sꢂ=Km þ ½Iꢂ=Kꢀ
i
11. Kim, H.; Lipscomb, W. N. Adv. Enzymol. 1994, 68, 153–
213.
ꢀ
ꢁ
12. Rich, D. H.; Moon, B. J.; Harbeson, S. J. Med. Chem.
1984, 27, 417–422.
13. Yoshimoto, T.; Orawski, A. T.; Simmons, W. H. Arch.
Biochem. Biophys. 1994, 311, 28–34.
kon
koff
Ki
Kiꢀ
¼
ꢁ 1:
ð8Þ
Moreover, the Dixon plots (1/vs versus [I]) and 1/vi ver-
sus [I], (Eq. 2) and the replots of the slopes from Dixon
plots versus 1/[S] (Eq. 3) were used to prove the compet-
itive character of the slow-binding inhibitors.
14. Fehrentz, J.-A.; Castro, B. Synthesis 1983, 676–678.
15. Herranz, R.; Castro-Pichel, J.; Vinuesa, S.; Garcia-Lopez,
M. T. J. Org. Chem. 1990, 55, 2232–2234.
16. (a) Sykes, B. D.; Robertson, E. B.; Dunford, H. B.;
Konasewich, D. Biochemistry 1966, 5, 697–701; (b) Johne,
S.; Gro¨ger, D. Pharmazie 1977, 32, 1–15.
Acknowledgments
17. Sakurai, M.; Higashida, S.; Sugano, M.; Komai, T.; Yagi,
R.; Ozawa, Y.; Handa, H.; Nishigaki, T.; Yabe, Y. Bioorg.
Med. Chem. 1994, 2, 807–825.
18. Segel, I. H. In Enzyme kinetics. Behavior and analysis of
rapid equilibrium and steady-state enzyme systems; John
Wiley, Wiley Classics Library, 1993, pp 161–176.
19. Maggiora, L. L.; Orawski, A. T.; Simmons, W. H. J. Med.
Chem. 1999, 42, 2394–2402.
The authors gratefully acknowledge the assistance of
Dr. A. Schierhorn for mass spectrometry and Dr. C.
Mrestani-Klaus for NMR spectroscopy. This work
was supported by the Deutsche Forschungsgemeinschaft
(SFB 387), the Fond der Chemischen Industrie, as well
as by a grant to A.S.-M. from the Dechema e.V.
20. Morrison, J. F.; Walsh, C. T. Adv. Enzymol. 1988, 61,
201–301.
21. Aoyagi, T.; Wada, T.; Yamamoto, K.; Kojima, F.; Nagai,
M.; Harada, S.; Umezawa, H. J. Appl. Biochem. 1984, 6,
212–221.
References and notes
1. Yaron, A.; Mlynar, D. Biochem. Biophys. Res. Commun.
1968, 32, 658–663.
22. Sto¨ckel-Maschek, A.; Mrestani-Klaus, C.; Stiebitz, B.;
Demuth, H.-U.; Neubert, K. Biochim. Biophys. Acta 2000,
1479, 15–31.
23. (a) Lowther, W. T.; Matthews, B. W. Biochim. Biophys.
Acta 2000, 1477, 157–167; (b) Lowther, W. T.; Matthews,
B. W. Chem. Rev. 2002, 102, 4581–4607.
24. Maher, M. J.; Ghosh, M.; Grunden, A. M.; Menon, A. L.;
Adams, M. W. W.; Freeman, H. C.; Guss, J. Biochemistry
2004, 43, 2771–2783.
25. (a) Stra¨ter, N.; Lipscomb, W. N. Biochemistry 1995, 34,
14792–14800; (b) Kim, H.; Lipscomb, W. N. Biochemistry
1993, 32, 8465–8478.
2. (a) Lasch, J.; Koelsch, R.; Ladhoff, A.-M.; Hartrodt, B.
Biomed. Biochim. Acta 1986, 45, 833–843; (b) Hooper, N.
M.; Hryszko, J.; Turner, A. J. Biochem. J. 1990, 267, 509–
515; (c) Harbeck, H.-T.; Mentlein, R. Eur. J. Biochem.
1991, 198, 451–458; (d) Mars, I.; Monnet, V. Biochim.
Biophys. Acta 1995, 1243, 209–215.
3. (a) Hooper, N. M.; Turner, A. J. FEBS Lett. 1988, 229,
340–344; (b) Simmons, W. H.; Orawski, A. T. J. Biol.
Chem. 1992, 267, 4897–4903; (c) Rusu, I.; Yaron, A. Eur.
J. Biochem. 1992, 210, 93–100.
4. (a) Wilce, M. C. J.; Bond, C. S.; Dixon, N. E.; Freeman,
H. C.; Guss, J. M.; Lilley, P. E.; Wilce, J. A. Proc. Natl.
Acad. Sci. U.S.A. 1998, 95, 3472–3477; (b) Zhang, L.;
Crossley, M. J.; Dixon, N. E.; Ellis, P. J.; Fisher, M. L.;
King, G. F.; Lilley, P. E.; MacLachlan, D.; Pace, R. J.;
Freeman, H. C. J. Biol. Inorg. Chem. 1998, 3, 470–483.
5. (a) Rawlings, N. D.; Barrett, A. J. Methods Enzymol.
1995, 248, 183–228; (b) Bazan, J. F.; Weaver, L. H.;
26. Graham, S. C.; Maher, M. J.; Simmons, W. H.; Freeman,
H. C.; Guss, J. M. Acta Cryst. Sect. D 2004, 60, 1770–
1779.
27. Li, J.-Y.; Chen, L.-L.; Cui, Y.-M.; Luo, Q.-L.; Li, J.; Nan,
F.-J.; Ye, Q.-Z. Biochem. Biophys. Res. Commun. 2003,
307, 172–179.
28. Lasch, J.; Koelsch, R.; Steinmetzer, T.; Neumann, U.;
Demuth, H.-U. FEBS Lett. 1988, 227, 171–174.

4820
A. Sto¨ckel-Maschek et al. / Bioorg. Med. Chem. 13 (2005) 4806–4820
29. Sto¨ckel-Maschek, A.; Stiebitz, B.; Koelsch, R.; Neubert,
K. Anal. Biochem. 2003, 322, 60–67.
31. Gottwald, S.; Martin-Luther-Universita¨t Halle-Witten-
berg, Thesis, 1996.
30. Yoshimoto, T.; Tone, H.; Honda, T.; Osatomi, K.;
Kobayashi, R.; Tsuru, D. J. Biochem. 1989, 105, 412–416.
32. Szedlacsek, S. E.; Duggleby, R. G. Methods Enzymol.
1995, 249, 144–180.