New peptidic cysteine protease inhibitors derived from the electrophilic α-amino acid aziridine-2,3-dicarboxylic acid

Article abstract of DOI:10.1021/jm981061z

Three different types of peptides containing aziridine-2,3-dicarboxylic acid (Azi) as an electrophilic α-amino acid at different positions within the peptide chain (type I, N-acylated aziridines with Azi as C-terminal amino acid; type II, N-unsubstituted aziridines with Azi as N-terminal amino acid; type III, N-acylated bispeptidyl derivatives of Azi) have been synthesized and tested as inhibitors of the cysteine proteases papain, cathepsins B, L, and H, and calpains I and II, as well as against several serine proteases, one aspartate, and one metalloprotease. All aziridinyl peptides are specific cysteine protease inhibitors. Papain and cathepsins B and L are inhibited irreversibly, whereas cathepsin H and calpains are inhibited in a non-time- dependent manner. Some compounds turned out to be substrates for serine proteases and for the metalloprotease thermolysin. Remarkable differences can be observed between the three different types of inhibitors concerning stereospecificity, pH dependency of inhibition, selectivity between different cysteine proteases, and the importance of a free carboxylic acid function at the aziridine ring for inhibition. Above all type II inhibitors, aza analogues of the well-known epoxysuccinyl peptides, are potent cysteine protease inhibitors. With the exception of BOC-Leu-Gly-(S,S+R,R)-Azi-(OEt)₂ (28a+b), a highly selective and potent cathepsin L inhibitor, N-acylated aziridines of type I are weaker inhibitors than type II or type III compounds. The observed results can be explained by different binding modes of the three types of inhibitors with respect to their orientation in the S- and S'-binding sites of the enzymes. Furthermore, the presence of a protonated aziridine N modifies the binding mode of type II inhibitors.

Full text of DOI:10.1021/jm981061z

560
J . Med. Chem. 1999, 42, 560-572
New P ep tid ic Cystein e P r otea se In h ibitor s Der ived fr om th e Electr op h ilic
r-Am in o Acid Azir id in e-2,3-d ica r boxylic Acid
Tanja Schirmeister*
Department of Pharmaceutical Chemistry, Pharmaceutical Institute, Albert-Ludwigs-University of Freiburg,
Hermann-Herder-Strasse 9, D-79104 Freiburg, Germany
Received October 23, 1998
Three different types of peptides containing aziridine-2,3-dicarboxylic acid (Azi) as an
electrophilic R-amino acid at different positions within the peptide chain (type I, N-acylated
aziridines with Azi as C-terminal amino acid; type II, N-unsubstituted aziridines with Azi as
N-terminal amino acid; type III, N-acylated bispeptidyl derivatives of Azi) have been synthesized
and tested as inhibitors of the cysteine proteases papain, cathepsins B, L, and H, and calpains
I and II, as well as against several serine proteases, one aspartate, and one metalloprotease.
All aziridinyl peptides are specific cysteine protease inhibitors. Papain and cathepsins B and
L are inhibited irreversibly, whereas cathepsin H and calpains are inhibited in a non-time-
dependent manner. Some compounds turned out to be substrates for serine proteases and for
the metalloprotease thermolysin. Remarkable differences can be observed between the three
different types of inhibitors concerning stereospecificity, pH dependency of inhibition, selectivity
between different cysteine proteases, and the importance of a free carboxylic acid function at
the aziridine ring for inhibition. Above all type II inhibitors, aza analogues of the well-known
epoxysuccinyl peptides, are potent cysteine protease inhibitors. With the exception of BOC-
Leu-Gly-(S,S+R,R)-Azi-(OEt)2 (28a +b), a highly selective and potent cathepsin L inhibitor,
N-acylated aziridines of type I are weaker inhibitors than type II or type III compounds. The
observed results can be explained by different binding modes of the three types of inhibitors
with respect to their orientation in the S- and S′-binding sites of the enzymes. Furthermore,
the presence of a protonated aziridine N modifies the binding mode of type II inhibitors.
Ch a r t 1
In tr od u ction
The papain superfamily of cysteine proteases includes
a variety of enzymes with closely related amino acid
sequences and overall folding structures.1 Among them
are vacuolar plant enzymes (e.g., papain),2 protozoen
enzymes3 (e.g., cruzipain, falcipain), and mammalian
lysosomal cathepsins4 (e.g., cathepsins B, L, H) and
cytoplasmatic calpains.5 The mammalian cysteine pro-
teases are involved in a variety of pathological processes
including dysregulated protein turnover such as mus-
cular dystrophy,6 bone resorption,7 growth and malig-
nancy of tumors,8 and myocardial infarct.9 Therefore
these enzymes are promising targets for the develop-
ment of inhibitors as therapeutic agents.
A number of irreversible and selective cysteine pro-
tease inhibitors have been developed. Most of them
exhibit a peptide segment for recognition by the enzyme
and an electrophilic building block for reaction with the
cysteine residue of the enzyme’s active site as common
structural features. Examples are diazomethyl ke-
tones,10 fluoromethyl ketones,11 acyloxymethyl ke-
tones,12 O-acylhydroxamates,13 vinyl sulfones,14 and
epoxysuccinic acid derivatives.15 With regard to the
requirements for drugs, the latter are one of the most
promising inhibitor classes. Inactivation of cysteine
proteases by epoxysuccinyl peptides proceeds from a
nucleophilic opening of the epoxide ring leading to
alkylated enzymes.16 Development of these inhibitors
was based on the discovery of E-64 (1) (Chart 1) isolated
from an Aspergillus japonicus culture by Hanada et al.
in 1978.17 Cell permeability was improved by replacing
the agmatine by uncharged alkyl groups and by esteri-
fication of the carboxylate function, e.g., Loxistatin (2)
(Chart 1).18 Even if the esters are 100-1000 times less
active than the free acids in vitro, they are used as
prodrugs which are easily resorbed and subsequently
hydrolyzed to their active forms.19 These epoxides have
been shown to interact in an antisubstrate orientation
with the S-subsite of papain and cathepsin B.20 Cysteine
proteases of the papain superfamily prefer substrates
10.1021/jm981061z CCC: $18.00 © 1999 American Chemical Society
Published on Web 02/25/1999

New Peptidic Cysteine Protease Inhibitors
J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 4 561
Sch em e 1
with Phe in the P2-position due to a hydrophobic
binding pocket built up by the amino acids Trp67,
Phe207, Pro68, Ala160, Val133, and Val157 (papain
numbering).21 Epoxysuccinyl peptides such as 1-3
(Chart 1) bind with their Leu residue into this hydro-
phobic binding pocket. Epoxides such as 4 interact with
the S′-subsite of cathepsin B in a substrate-like mode.
In contrast to inhibitors 1-322 they are significantly
more potent in the R,R configuration23 and exhibit a
very good selectivity against cathepsin B.24 This selec-
tivity results from interaction of the proline carboxylate
of 4 with two histidine residues (His110, His111) which
are located at the occluding loop of cathepsin B and are
Ch a r t 2
responsible for its dipeptidyl peptidase activity.25
A
chimeric epoxysuccinyl peptide inhibitor has been pre-
pared by Moroder et al.23 by combining the S′-recogni-
tion element of cathepsin B (Leu-Pro) with the inhib-
iting propeptide sequence of cathepsin B (Leu-Gly-Gly,
sequence portion 46-48 of the propeptide) which spans
the S-binding pockets in an antisubstrate orientation.
This inhibitor is the most potent cathepsin B inhibitor
known so far and exhibits a very good selectivity
between cathepsin B and cathepsin L.
Sch em e 2
The three-membered aziridine ring is closely related
in structure to the epoxide and is also susceptible to ring
opening by nucleophiles.26 J ones et al.27 have synthe-
sized and tested the aziridine analogues 5-7 (Chart 1)
of the well-known cysteine protease inhibitors EtO-Eps-
i
Chart 2) are described. These are classified according
to their structural differences: Type I aziridinyl pep-
tides contain the aziridine moiety as the C-terminal
amino acid. They are aziridines N-acylated with amino
acids or peptides. Type II peptides are aza analogues of
the known epoxysuccinyl peptides and contain the
aziridine-2,3-dicarboxylic acid as the N-terminal amino
acid. In type III peptides the aziridine is located in the
middle of the peptide chain. Although type I and III
peptides are both N-acylated aziridines, experimental
results show differences between these two types which
support their classification as two different inhibitor
types. The particular peptide moieties used for prepara-
tion of the inhibitors have been selected with respect to
the substrate specificities21 of hydrolysis and to known
inhibiting sequences,23 respectively.
Leu-NHiAm (2), HO-Eps-Leu-NHiAm (3), and BuNH-
Eps-Ile(Leu)-ProOH (4). Even though these aziridine
analogues are also irreversible inhibitors of the cysteine
proteases papain, cathepsin B, and cathepsin L, re-
markable differences between the epoxides and the
aziridines have been found with respect to reactivity,
selectivity, stereospecificity, and pH dependency of
inhibition.27 Recently a variety of aziridinyl peptides in
which the Leu residue of compounds 5 and 6 has been
replaced by Phe have been patented by Takeda Chem.
Ind. as powerful cathepsin L inhibitors.28,29 For these
inhibitors naturally occurring epoxide analogues, called
cathestatins, are known.30
In contrast to the epoxide the aziridine ring can
additionally be derivatized at its heteroatom. This offers
a greater variability for structure-activity studies.
Chimeric inhibitors can not only be built by derivati-
zation of the two carboxylates but also by variations at
the aziridine nitrogen. Even though some N-alkylated
aziridinyl peptides are described in the above-mentioned
patents28,29 and some results concerning N-acylated
derivatives could already be obtained in the author’s
laboratory31 this field has yet to be explored systemati-
cally.
P r ep a r a tion of In h ibitor s
The syntheses of the compounds were performed as
depicted in Schemes 1-5. Stereoselective synthesis of
(S,S)- and (R,R)-diethyl aziridine-2,3-dicarboxylate
(10a ,b) was carried out according to previously31,32
described procedures using the reverse configured ep-
oxides (R,R)-8b and (S,S)-8a , respectively, as starting
materials. The latter were prepared from (R,R)- and
(S,S)-diethyl tartrate, respectively.33,34
In the present paper, both the syntheses and the
inhibition profiles of peptides containing aziridine-2,3-
dicarboxylic acid as the electrophilic R-amino acid at
different positions within the peptide chain (types I-III,
Half-esters 11a ,b of the aziridine building blocks were
prepared by alkaline hydrolysis with 1 equiv of LiOH

562 J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 4
Schirmeister
Sch em e 3
dine 13a +b (Scheme 2) were synthesized as diastere-
omeric mixtures (a +b series).
Acylation of the aziridine nitrogen with amino acids
was carried out via symmetric anhydrides, prepared
from 2 equiv of N-protected amino acid and 1 equiv of
DCC, under DMAP catalysis (Scheme 3). This yielded
the type I compounds 17a ,b,31 18a ,31 19a ,31 20a , and
23a +b and the type III inhibitors 21a ,b and 22a ,b.
Nitrogen acylation with di- or tripeptides could be
performed by fragment condensation using mixed an-
hydrides of peptide and isobutyl chloroformiate.35 Since
this method leads to racemization within the peptide
chain, the one-pot EEDQ36 procedure was chosen to
prepare the type I inhibitors 24a , 25a ,b, 26a , 27a +b,
28a +b, and 29a (Scheme 4).
A free carboxylate function at the aziridine ring has
been shown to be essential for inhibition by N-unsub-
stituted derivatives.27 To examine the role of a free
carboxylate group for inhibition by N-acylated deriva-
tives, selective hydrolysis of one ester function of the
N-acylated diethyl ester 17a by alkaline or enzymatic
methods was investigated. For the alkaline procedure
either 1 equiv of LiOH/EtOH or 1 equiv of NaOH/CH3-
CN was used. The enzymatic methods were carried out
using the following enzymes:39 PLE, chymotrypsin,
trypsin, CCL, PPL, LPR, LRA, and LAN. Alkaline
hydrolyses and enzymatic procedures with the serine
hydrolases PLE, trypsin, and chymotrypsin, respec-
tively, led to preferential amide hydrolysis. The reasons
for the sensitivity of the amide bond to alkaline hy-
drolysis are disturbed amide resonance37 and high
acidity of the leaving group. Amides containing an
aziridine as the amino component cannot form meso-
meric structures due to increased ring tension. The pKa
of the aziridinium ion has been determined by J ones et
al.27 with pKa 3.6 for compound 6a . In the author’s
laboratory a pKa of 3.8 ( 0.273 was found for compound
11a . PLE is a serine esterase which can also cleave
amide bonds and which prefers aromatic residues.38
Chymotrypsin and trypsin are serine proteases prefer-
ring hydrophobic and basic amino acids, respectively,
in the P1-position. In both cases, amide and simulta-
neous ester hydrolysis could be observed. Compound
17a was not hydrolyzed by the above-mentioned lipases.
Due to the failure of the selective hydrolysis of one ester
function with maintenance of the aziridide structure,
another synthetic path had to be chosen. The aziridine
building block 13a +b was used as a starting material
to allow ester cleavage by nonhydrolytic procedures. The
N-acylated aziridinyl peptide 30a +b which contains one
free carboxylate function was obtained by N-acylation
of building block 13a +b with BOC-Phe via the sym-
metric anhydride (Scheme 3, compound 23a +b) and
subsequent hydrogenolysis of the benzyl ester (Scheme
5).
Sch em e 4
Sch em e 5
monohydrate in ethanol and subsequent protonation by
ion exchange.27 Peptide coupling at the free carboxylate
function of the aziridine leading to the type II inhibitors
14a ,b and 16a ,b could be achieved by the DPPA
method, whereby N-protection was not necessary.27 Both
stepwise and fragment condensations were carried out
(Scheme 1).
Resu lts
Racemic trans-benzylethyl aziridine-2,3-dicarboxylic
acid (13a +b) was synthesized by Michael type addition
of diphenylsulfimine to the corresponding fumarate 12
(Scheme 2). Most of the aziridinyl peptides derived from
aziridines 10a ,b were prepared as single diastereomers
with S,S (a series) as well as R,R (b series) configured
aziridine ring. Peptides derived from the racemic aziri-
The second-order rate constants for the inactivation
of papain and cathepsins B and L by the aziridinyl
peptides of type I, 17-20 and 23-30, which contain the
aziridine-2,3-dicarboxylate as the C-terminal amino
acid, are shown in Table 1. Since proteases of the papain
family prefer Phe in the P2-position, compounds 17a ,b,
25a ,b, and 26a with a Phe residue at different positions

New Peptidic Cysteine Protease Inhibitors
J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 4 563
Ta ble 1. Second-Order Rate Constants for Inactivation of Papain,a Cathepsin B,b and Cathepsin Lb by Type I Aziridinyl Peptides
ki/Ki (M-1 min-1
)
inhibitor
papain
cathepsin B
cathepsin L
65 ( 12
BOC-Phe-(S,S)-Azi-(OEt)2 (17a )c
BOC-Phe-(R,R)-Azi-(OEt)2 (17b)c
Z-Ala-(S,S)-Azi-(OEt)2 (18a )c
408 ( 16
125 ( 19
18 ( 6
67 ( 17
42 ( 9
16 ( 6
305 ( 26
nd
73 ( 2
BOC-Ala-(S,S)-Azi-(OEt)2 (19a )
43 ( 5c
ndd
nd
nd
Fmoc-Leu-(S,S)-Azi-(OEt)2 (20a )
BOC-Phe-(S,S+R,R)-Azi-(OEt)(OBzl) (23a +b)c
BOC-Phe-(S,S+R,R)-Azi-(OEt)(OH) (30a +b)
65 ( 9c
99 ( 10c
nd
76 ( 9
505
1230
188 ( 6c
kie ) 0.21 ( 0.013
Kif ) 0.41 ( 0.07
297c ( 55
ki ) 0.83 ( 0.019
Ki ) 0.68 ( 0.03
389c ( 11
Z-Gly-Gly-(S,S)-Azi-(OEt)2 (24a )
Z-Ala-Phe-(S,S)-Azi-(OEt)2 (25a )
317
ki ) 0.087 ( 0.011
Ki ) 0.27 ( 0.06
29 ( 10c
177c ( 10
125
ki ) 0.011 ( 0.001
Ki ) 0.090 ( 0.001
22 ( 1
Z-Ala-Phe-(R,R)-Azi-(OEt)2 (25b)c
BOC-Phe-Ala-(S,S)-Azi-(OEt)2 (26a )
21 ( 1
1370
33 ( 4c
469
455 ( 15c
ki ) 0.028 ( 0.0026
Ki ) 0.020 ( 0.009
1232
ki ) 0.020 ( 0.001
Ki ) 0.016 ( 0.001
149
ki ) 0.01 ( 0.001
Ki ) 0.069 ( 0.0012
523 ( 25
ki ) 0.051 ( 0.019
Ki ) 0.11 ( 0.08
281 ( 2c
BOC-Phe-Ala-(S,S+R,R)-Azi-(OEt)(OBzl) (27a +b)
BOC-Leu-Gly-(S,S+R,R)-Azi-(OEt)(OBzl) (28a +b)
Z-Gly-Phe-Ala-(S,S)-Azi-(OEt)2 (29a )c
443 ( 71c
240 ( 61c
421 ( 100
3237
ki ) 0.028 ( 0.0015
Ki ) 0.0088 ( 0.0015
315 ( 48
a
b
pH 6.5. pH 6.0. c Measurements were limited to the linear range, with [I] e Ki, due to solubility problems. Therefore, only the second-
order rate constant could be obtained. Not determined. e ki (min-1). f Ki (mM).
d
within the peptide chain were synthesized to evaluate
the most favorable distance between the electrophilic
building block and P2-recognition element. 20a , 24a ,
and 28a +b are peptides which contain fragments of the
inhibiting procathepsin B sequence Leu-Gly-Gly. These
type I inhibitors are irreversible inhibitors of cathepsins
B and L and papain. This irreversibility is evident from
the time dependency of inhibition in the continuous or
dilution assays which were performed to determine the
second-order rate constants of inhibition. It was con-
firmed by dialysis experiments with papain and inhibi-
tors 17a and 11a as well as with cathepsin L and
inhibitor 28a +b.
Second-order rate constants for these type I inhibitors
are very low and are similar to those found by Albeck
et al.49 for several erythro-peptidyl epoxides. Especially
cathepsin L is inactivated very weakly by type I aziri-
dines. An interesting and surprising exception in this
regard is the inhibition of cathepsin L by mixture
28a +b. Although it is still a weak inhibitor, the second-
order rate constant for this diastereomeric mixture is
about 10 times higher than all other values found for
type I inhibitors. This increase in inhibition results from
a 10 times lower binding constant (Ki) and not from a
higher first-order rate of inactivation (ki). A difference
between the overall foldings of cathepsins B and L and
papain which may be a criterion to explain this out-
standing value is the structure of the S2-pocket of
cathepsin L.47 An additional Met161 residue makes its
pocket more shallow and narrow compared to those of
cathepsin B and papain.
rate constants are reported in Table 2, the S,S diaster-
eomers 17a and 25a are about 7-10 times more active
than their R,R isomers 17b and 25b. On the other hand
this stereospecificity corresponds to the one observed
for the epoxysuccinyl peptides exemplified by com-
pounds 1-3. Another difference between type I aziri-
dines and aziridines without N-derivatization and ep-
oxides 1-3 is the low inhibition improvement by the
free carboxylic acid function at the three-membered
ring. While 6a ,b27 (Table 2) and even the single aziri-
dine building block 11a (Table 5) exhibit an about 300-
600 times higher second-order rate constant for the
inhibition of papain than their corresponding esters,
values for the inhibition by 30a +b are only 3-10 times
higher in comparison to those for 17a ,b or 23a +b.
Interestingly this inhibition improvement is a result of
a higher first-order rate of alkylation (ki) and not of a
lower binding constant (Ki). Taking into consideration
that 30a +b is a free acid mixture of diastereomers with
S,S and R,R configured aziridine ring, the inhibition
constant for the eutomer may be somewhat higher but
nevertheless cannot reach the rates found for N-unsub-
stituted derivatives.
Prolongation of the peptide chain is normally known
to improve inhibition of endoproteinases.40 However,
this is not the case for N-acylated aziridinyl peptides of
type I (inhibitor 29a ) as well as for the N-unsubstituted
ones of type II (inhibitors 16a ,b). The highest inhibition
constants within the type II inhibitors are observed for
R,R configured aziridines coupled to a Leu derivative
(14b and 5b27).
As shown by J ones et al.,27 N-unsubstituted aziridines
are much more potent at low pH values due to proto-
nation of the aziridine nitrogen. The data27 showed that
the neutral form must be at least 100 times less reactive
than the protonated form. The true inhibition constants
The highest inhibition constants for type I derivatives
containing Phe within the peptide moiety are found for
compounds 26a and 27a +b with BOC-Phe-Ala as the
N-terminal peptide chain.
In contrast to N-unsubstituted aziridines of type II,
14a ,b, 5a ,b,27 and 6a ,b,27 for which the second-order

564 J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 4
Schirmeister
Ta ble 2. Second-Order Rate Constants for Inactivation of Papain,a Cathepsin B,b and Cathepsin Lb by Type II Aziridinyl Peptides
ki/Ki (M-1 min-1
cathepsin B
41 ( 8c
)
inhibitor
papain
cathepsin L
3130
ki ) 0.055 ( 0.009
Ki ) 0.018 ( 0.021
16261
EtO-(S,S)-Azi-Leu-OBzl (14a )
214
ki e ) 0.036 ( 0.0015
Ki f ) 0.17 ( 0.02
1533 ( 41c
EtO-(R,R)-Azi-Leu-OBzl (14b)
1607 ( 395c
ki ) 0.44 ( 0.0004
Ki ) 0.027 ( 0.0001
ndd
EtO-(R,R)-Azi-Leu-OH (15b)
EtO-(S,S)-Azi-Leu-Pro-OBzl (16a )
63 ( 18c
227
nd
405
2142
ki ) 0.021 ( 0.0002
Ki ) 0.094 ( 0.0034
466 ( 37c
ki ) 0.039 ( 0.02
ki ) 0.047 ( 0.005
Ki ) 0.022 ( 0.0085
1728
ki ) 0.043 ( 0.0033
Ki ) 0.025 ( 0.0057
Ki ) 0.096 ( 0.01
EtO-(R,R)-Azi-Leu-Pro-OBzl (16b)
188
ki ) 0.025 ( 0.0001
Ki ) 0.13 ( 0.0011
EtO-(S,S)-Azi-Leu-NHiAm (5a )g
EtO-(R,R)-Azi-Leu-NHiAm (5b)g
HO-(S,S)-Azi-Leu-NHiAm (6a )g
HO-(R,R)-Azi-Leu-NHiAm (6b)g
iBuNH-(S,S)-Azi-Leu-Pro-OH (7a )g
iBuNH-(R,R)-Azi-Leu-Pro-OH (7b)g
180
3300
(108 ( 6) × 103
(864 ( 42) × 103
108 ( 6
nih
780 ( 180
402 ( 6
3000 ( 60
nih
a
b
pH 6.5. pH 6.0. c Measurements were limited to the linear range, with [I] e Ki, due to solubility problems. Therefore, only the
second-order rate constant could be obtained. Not determined. e ki (min-1). f Ki (mM). Taken from J ones et al.27 No time-dependent
d
g
h
inhibition.27
Ta ble 3. Second-Order Rate Constants for Inactivation of Papain,a Cathepsin B,b and Cathepsin Lb by Type III Aziridinyl Peptides
ki/Ki (M-1 min-1
)
inhibitor
papain
cathepsin B
cathepsin L
BOC-Phe-(S,S)-(EtO)-Azi-Leu-OBzl (21a )c
BOC-Phe-(R,R)-(EtO)-Azi-Leu-OBzl (21b)
171 ( 40
122 ( 12
219 ( 15
225 ( 70c
257 ( 29c
314
kid ) 0.0099 ( 0.0023
Kie ) 0.031 ( 0.0023
1210 ( 260c
BOC-Phe-(S,S)-(EtO)-Azi-Leu-Pro-OBzl (22a )
BOC-Phe-(R,R)-(EtO)-Azi-Leu-Pro-OBzl (22b)
321
450
ki ) 0.016 ( 0.0019
Ki ) 0.05 ( 0.029
768 ( 125c
ki ) 0.037 ( 0.0022
Ki ) 0.082 ( 0.013
1938 ( 440c
5896 ( 409c
a
b
pH 6.5. pH 6.0. c Measurements were limited to the linear range, with [I] e Ki, due to solubility problems. Therefore, only the second-
order rate constant could be obtained. ki (min-1). e Ki (mM).
d
Ta ble 4. Inhibition Constants for the Non-Time-Dependent Inhibition of Cathepsin H,a Calpain I,b and Calpain IIb
Ki (µM)
inhibitor
cathepsin H
calpain I
calpain II
BOC-Phe-(S,S)-Azi-(OEt)2 (17a )
Z-Ala-(S,S)-Azi-(OEt)2 (18a )
(EtO)-(S,S)-Azi-Leu-OBzl (14a )
(EtO)-(R,R)-Azi-Leu-OBzl (14b)
BOC-Phe-(S,S)-(EtO)-Azi-Leu-OBzl (21a )
BOC-Phe-(R,R)-(EtO)-Azi-Leu-OBzl (21b)
210 ( 19
ni
260 ( 35
ni
93 ( 8
137 ( 11
nic
160 ( 31
nd
ni
nd
19 ( 7
ndd
nd
nd
nd
nd
42 ( 14
a
b
d
pH 6.5. pH 7.5. c No inhibition. Not determined.
Ta ble 5. Second-Order Rate Constants for Inactivation of
N-unsubstituted derivatives can be confirmed as shown
in Table 5. In contrast to N-unsubstituted aziridines the
N-acylated aziridine 17a shows nearly identical inhibi-
tion constants at pH 6.5 and pH 8. Rich et al.41
investigated the pH dependency of inhibition of papain
by a nonionizable amide of 3. This inhibition was found
to depend on two acidic dissociation constants (pKa’s
3.93 and 4.09) of papain. This is in contrast to the
inhibition of papain by E-64 (1) for which an acidic
dependency and an alkaline dependency were found.27
The latter result supports the assumption that His159
may play an essential role for inhibition. Since 17a is
sensitive to alkaline media, determination of second-
order rate constants at pH values above pH 8 is useless.
Table 6 reports the pH dependency of inhibition of
papain by 17a . These values, obtained within a pH
range from 5.2 to 8.0, indicate, however, that inhibition
Papain at Different pH Valuesa
ki/Ki (M-1 min-1
pH 6.5
)
inhibitor
pH 4
pH 8
ndc
10a b
11a b
14a
61 ( 5
24774 ( 7000
786 ( 39b
72 ( 10
11 ( 0.5
4692 ( 1000
780 ( 85
168d
24 ( 5b
17a b
424 ( 32
410 ( 21
a
Substrate, 1.4-1.5 mM L-BAPA; [E] ) 0.5 mg mL-1; [I]: 10a ,
0.58-4.7 mM; 11a , 6.6 µM-0.3 mM; 14a , 0.27-1.36 mM; 17a ,
b
0.092-0.37 mM. Measurements were carried out at the linear
range, where [I] e Ki. Therefore, only the second-order rate
constant was obtained. c Not determined. ki ) 0.033 ( 0.002
d
(min-1), Ki ) 0.20 ( 0.033 (mM).
for the completely protonated inhibitor would be much
higher at pH values below pH 4 which cannot be tested
because of loss of enzyme activity.27 The maximum
activity pH of 4 determined by J ones et al.27 for

New Peptidic Cysteine Protease Inhibitors
J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 4 565
Ta ble 6. pH Dependency of Inhibition of Papain by 17a
Su m m a r y a n d Discu ssion
activity (mU mL-1
)
Even though several similarities, including the ex-
clusive inhibition of cysteine proteases and the irrevers-
ibility of inhibition of papain and cathepsins B and L,
exist between aziridines and epoxides, their respective
behavior as cysteine protease inhibitors shows remark-
able differences. These differences may first of all be
due to the possible protonation of the aziridine nitrogen
in the case of the type II aziridines, thus leading to a
different binding mode with an additional water mol-
ecule not being necessary and with the positioning of
the inhibitor being disturbed by the positively charged
group. Second, aziridines of type II are analogues of the
epoxides concerning their chemical reactivity, but they
are not bioisosters. The epoxide oxygen is a H-bond
acceptor, whereas the aziridine nitrogen in most cases
reacts as a H-bond donor. Thus totally different interac-
tions with the enzyme could be possible. On the other
hand, there exist noteworthy differences between the
different types of aziridinyl peptides. These differences
concern pH dependency, stereospecificity, and selectivity
of inhibition. In contrast to the type II inhibitors
H-bonds cannot be built under participation of the
aziridine N by the N-acylated type I and III inhibitors.
Type I inhibitors which are more active with the S,S
configured aziridine ring and which in this regard
resemble the epoxysuccinyl peptides 1-3 could bind to
the S-binding site in a substrate-like mode. Comparison
of inhibition constants obtained for the Phe-containing
series of inhibitors (compounds 17a ,b and 25a ,b with
26a and 27a +b) leads to the assumption that the Phe
residue of the latter is probably located in the hydro-
phobic S2-pocket as known from the peptidyl chloro-
methyl ketone BPACK.48 The aziridide structure may
possibly lead to a more unfavorable inhibitor conforma-
tion which could impede the attack of the active site
cysteine residue at the aziridine ring carbons and
therefore lead to low alkylation rate constants. A
superimposition of minimized conformations of the
epoxide 2, the type II aziridine 14b, and the type I
inhibitor 26a supports this supposition (Figure 1A).76
However, first results of docking experiments77 per-
formed with papain and 26a indicate that the flexibility
of the molecule nevertheless allows a conformation in
which the Phe residue of this inhibitor can bind into
the hydrophobic S2-pocket and the N-terminal BOC
group can be located within the S3-subsite (Figure 1B).
A comparison of type II compounds 14a ,b and 16a ,b
with the N-acylated type III compounds 21a ,b and
22a ,b shows that the activation of the aziridine ring to
nucleophilic ring opening by N-acylation does in general
not lead to an inhibition improvement. Therefore, the
main reason for the remarkable increase in inhibition
by protonation in the case of type II inhibitors should
be due to improved binding rather than to improved
nucleophilic ring opening. Type III inhibitors are new
chimeric inhibitors, and similar to epoxides with two
peptide chains,23 a tendency for similar second-order
rate constants for the R,R and S,S configured isomers
can be observed. Thus, one can assume that these
inhibitors, like the chimeric epoxides, bind to both the
S- and S′-binding sites. A new aspect which should be
taken into consideration is the possibility that the
aziridine ring can be opened by nucleophiles in two
different ways. Not only is cleavage of the C-N bond
possible but also the cleavage of the C-C bond.50 C-N
cleavage would lead to aspartic acid derivatives, while
C-C cleavage would form glycine derivatives. These two
pH
without 17a a
with 17a a,b
% inhibition
5.2
6.0
6.5
7.0
8.0
6.54 ( 0.2
7.33 ( 0.14
7.38 ( 0.21
7.33 ( 0.15
7.31 ( 0.15
1.86 ( 0.03
1.38 ( 0.1
0.98 ( 0.09
0.75 ( 0.15
0.75 ( 0.2
71.6
81.1
86.7
89.8
89.7
a
b
Mean values from five assays. [I] ) 17 µM; incubation time
30 min; [S] ) 1.49 mM L-BAPA; [E] ) 0.5 mg mL-1
.
by this N-acylated aziridine corresponds to the one
found for the amide of 3 rather than to that found for
E-64 (1).
Activation of aziridines toward nucleophilic ring
opening is possible not only by protonation of the
aziridine nitrogen but also by N-acylation.42 In Table 3
inhibition constants are reported for inhibitors of type
III (21a ,b, 22a ,b). These are bispeptidyl derivatives
which bear the aziridine moiety in the middle of the
peptide chain and which can be regarded as chimeric
inhibitors combining the P1′ and P2′ (Leu-Pro) specific-
ity of cathepsin B with its P2 specificity (Phe). The
N-acylation of tripeptide 16b with BOC-Phe to tet-
rapeptide 22b leads only to a weak inhibition improve-
ment in the case of cathepsin L and papain, but to a
10-fold improvement of inhibition in the case of cathe-
psin B. No or only a slight increase of inhibition is
obtained by N-acylation of the type II inhibitors 14a ,b
and 16a . Again, in accordance with the results found
for type II inhibitors, the R,R diastereomers are the
more potent, but with a very low eudismic ratio.
When the selectivity between cathepsins B and L and
papain was compared, the inhibitors of types II and III
which bear peptide chains at one of the carboxylate
functions exhibit higher activities against cathepsin L
than against cathepsin B or papain. The reverse holds
for the type I inhibitors which are, with the one
exception of mixture 28a +b already mentioned, only
weak cathepsin L inhibitors. Cathepsin H and calpains
are not inhibited time dependently by all types of
inhibitors. If a reduction in enzyme activity can be
observed, Ki values in the upper micromolar range are
also observed (Table 4). This non-time-dependent inhi-
bition may be due to extremely low alkylation rates
which are also known for the inhibition of these enzymes
by epoxysuccinyl peptides.22,43 For a better inhibition
of cathepsin H, inhibitors without an N-terminal pro-
tecting group may be required.
To evaluate the selectivity between proteases with
different mechanisms of hydrolysis (cysteine - serine
- aspartate - metallo) the aziridines 10a , 14b, 17a ,
18a , and 21b were tested against the serine proteases
chymotrypsin, trypsin, and elastase, the aspartate pro-
tease pepsin, and the metalloprotease thermolysin,
respectively. As shown by the above-mentioned hydroly-
sis assays of 17a with several serine hydrolases, this
compound does not react as an inhibitor but as a
substrate for these enzymes. BOC-Phe and compounds
10a and 11a could be isolated and identified by TLC
and IR and NMR spectroscopy. This is in agreement
with earlier reports on ester hydrolysis of aziridinecar-
boxylates by these enzymes.44,45 In contrast to 21b
compound 14b is a substrate for thermolysin which
prefers Leu in the P1′-position.46 In this case 11b and
LeuOBzl could be identified as hydrolysis products.
Neither 10a nor 21b could inhibit pepsin.

566 J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 4
Schirmeister
F igu r e 1. (A) Superimposition of minimized conformations of epoxide 2, type II aziridine 14b, and type I aziridine 26a .76 (B)
Possible binding mode of 26a to the active site of papain.77
possibilities may also play a role in the observed
differences between epoxides and aziridines on the one
hand and between different types of aziridinyl peptides
on the other hand.
The exclusive inhibition of cysteine proteases could
be explained by the enhanced nucleophilicity of the
active site compared to serine proteases as a result of
the thiolate-imidazolium ion pair74 even though the
nucleophilicity of serine and cysteine proteases cannot
be compared that easily. The “hard” nucleophile OH of
the serine proteases attacks the “hard” electrophilic Cd
O center, while the “soft” S- nucleophile attacks the
“soft” ring carbon.75 Another reason for this selectivity
may be the opposite active site geometry of serine
proteases51 which could convert aziridinyl peptides from
inhibitors into substrates for serine proteases. Starting
points to improve the selectivity between different
cysteine proteases are the good activities of type II
inhibitors and the type I mixture 28a +b against cathe-
psin L and the remarkable inhibition improvement by
the N-acylation of 16b with BOC-Phe to 22b in the case
of cathepsin B. A better understanding of the observed
differences between epoxysuccinyl peptides and aziri-
dinyl peptides on the one hand and between the three
types of aziridines on the other hand will require more

New Peptidic Cysteine Protease Inhibitors
J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 4 567
in 40 mL of dichloromethane at 0 °C; 1.65 g (7.7 mmol) of DCC
was added, and the mixture was stirred at 0 °C for 45 min.
Insoluble dicyclohexylurea was filtered off, and 7 mmol of
aziridine was added in 20 mL of dichloromethane at 0 °C
together with a few crystals of DMAP (10 mol %). The reaction
was stirred at 0 °C for 1 h and for 3-5 h at room temperature.
Dicyclohexylurea was filtered off, and the solvent was removed.
The residue was stirred in 10 mL of ethyl acetate for 20 min
and again filtered off. The organic layer was washed with
saturated NaHCO3 solution (20 mL) and brine (20 mL), dried
with MgSO4, and concentrated. The residue was purified for
each individual compound by either flash or hydrostatic
chromatography as described below.
detailed molecular modeling studies as well as X-ray
analysis of enzyme-inhibitor complexes. These studies
are in progress. Nevertheless, the present study shows
that N-unsubstituted aziridinyl peptides and in special
cases N-acylated ones, too, can be highly selective and
potent inactivators of cysteine proteases.
Exp er im en ta l Section
Gen er a l Meth od s. Enzymes were purchased from the
following companies: papain from Carica papaya from Fluka
(BioChemika USP, 76220), cathepsin B from bovine spleen
from Sigma (C 6268), cathepsin H from human liver (219404),
calpain I from porcine erythrocytes (208712) and calpain II
from porcine kidney (208715) from Calbiochem, chymotrypsin
from bovine pancreas (102307), trypsin from bovine pancreas
(124579) and pepsin (7185) from Merck, elastase from porcine
pancreas (20929) from Serva. Cathepsin L from Paramecium
tetraurelia56 was a gift from Prof. J . Schultz, Department of
Pharmaceutical Chemistry, University of Tuebingen, Ger-
many. Pig liver esterase PLE (46063), lipase from Rhizopus
arrhizus LRA (62305), lipase from Aspergillus niger LAN
(62294), lipase from Penicillium roqueforti LPR (62308), and
lipase from hog pancreas PPL (62300) were from Fluka.
Candida cylindraceae lipase type VII CCL (L1754) was from
Sigma. All enzymes were used without further purification.
All substrates, protected amino acids, and di- and tripeptides
were purchased from Bachem. Leu-Pro-OBzl trifluoroacetate
and BOC-Phe-AlaOH were prepared by well-established lit-
erature procedures.24,52 Buffer substances were biochemical
grade and were purchased from Merck. N-Ethylmaleimide was
from Aldrich; E-64 was from Boehringer-Mannheim. Reagent
grade chemicals were purchased from the following companies
and were used without further purification: EEDQ from
Novabiochem, DPSI, NaN3, LiOH‚H2O, and NH4Cl from
Merck, Ph3P, DCC, DMAP, and Et3N fom Fluka, DPPA from
Aldrich, Pd-C 10% type E 10 N/D from Degussa. All solvents
were anhydrous grade and were purchased from Fluka. Brine
refers to a saturated aqueous solution of NaCl. Analytical TLC
was performed on Merck aluminum sheets (silica gel 60 F254).
Compounds that were not visualized by UV light were detected
by spraying with Ehrlich’s reagent (1 g of p-(dimethylamino)-
benzaldehyde, 25 mL of HCl concentrated, 75 mL of MeOH)
followed by heating. Preparative flash column chromatography
was performed using silica gel 60, 40-63 µm, from Merck.
Preparative hydrostatic column chromatography was per-
formed using silica gel 60, 63-200 µm, from Merck. Melting
points are uncorrected and were obtained on a Mel-Temp II
capillary melting point apparatus (Laboratory Devices). Opti-
cal rotations were measured on a Perkin-Elmer 241 polarim-
eter in a thermostated cell. IR spectra were determined in KBr
pellets or with NaCl solution cells on a Perkin-Elmer 841 IR
spectrophotometer. Mass spectra were measured on a Finnigan
MAT 312 (low resolution; EI, 70 eV, 0.8 mA; CI, isobutane or
NH3, 200 eV, 0.5 mA) by Chemisches Laboratorium of the
University of Freiburg or on a Voyager-RP biospectrometry
workstation by PerSeptive Biosystems, Wiesbaden, Germany
(high resolution) (mode, reflector; accelerating voltage, 12 or
15 kV; scans averaged, 15-34; negative ions, off). Elemental
analyses were determined on a Perkin-Elmer elemental ana-
lyzer 240 by Chemisches Laboratorium of the University of
Freiburg. NMR (1H, 13C) spectra were recorded on a Varian
Unity 300 spectrometer (300 and 75.43 MHz, respectively). 1H
NMR chemical shifts are reported in ppm relative to the CHCl3
peak at δ ) 7.26 with CDCl3 as solvent and to the DMSO peak
(δ ) 2.49) with DMSO-d6 as solvent. 13C NMR chemical shifts
are reported in ppm relative to the CHCl3 peak (δ ) 77.00)
with CDCl3 as solvent. All 1H NMR assignments were sup-
ported by homonuclear decoupling experiments or by 2-D
COSY experiments. All 13C NMR assignments were supported
by 2-D HETCOR experiments. Coupling constants (J ) are
reported in hertz (Hz).
N-Acyla tion via Mixed An h yd r id es. The coupling re-
agent EEDQ (426 mg, 1.72 mmol) was added to a solution of
N-protected peptide (1.61 mmol) and aziridine (1.61 mmol) in
15 mL of DMF. The mixture was stirred at room temperature
for 48 h. The solvent was removed in vacuo; the residue was
dissolved in 100 mL of ethyl acetate and extracted with 50
mL of a 5% solution of NaHCO3 and 50 mL of water. The
organic layer was dried with MgSO4 and evaporated. The
residue was purified for each individual compound by either
flash or hydrostatic chromatography as described below.
DP P A-Med ia ted P ep tid e Cou p lin g. A stirred solution of
carboxylic acid (5 mmol) and C-protected amino acid or peptide
(HCl, TFA, or tosylate) (5.2 mmol) in DMF (30 mL) was cooled
in an ice-water bath and treated with DPPA (1.2 mL, 5.5
mmol). A solution of Et3N in 10 mL of DMF (10.5 mmol, 1.43
mL) was added dropwise over a period of 10 min. Stirring and
cooling were continued for 10 h. The mixture was diluted with
ethyl acetate (150 mL) and washed with 25-mL portions of
water (3×), 5% NaHCO3 (1×), and brine (2×). The organic
layer was dried with MgSO4 and evaporated in vacuo. The
residue was purified for each individual compound by either
flash or hydrostatic chromatography as described below.
Ca ta lytic Hyd r ogen olysis of Ben zyl Ester s. An evacu-
ated solution of 0.6 mmol of benzyl ester in 30 mL of MeOH
and 40 mg of Pd-C (10%) was vigorously stirred at room
temperature and atmospheric pressure for 1.5 h (TLC control)
under a slow stream of hydrogen. The catalyst was removed
by filtration over Celite and washed with methanol (50 mL).
The filtrate was evaporated in vacuo, and the residue was
recrystallized.
The following compounds were prepared according to lit-
erature procedures: d ieth yl (2S,3S)-2,3-ep oxysu ccin a te
(8a ),34 d ieth yl (2R,3R)-2,3-ep oxysu ccin a te (8b),34 d ieth yl
(2S,3S)-azir idin e-2,3-dicar boxylate (10a),31 dieth yl (2R,3R)-
a zir id in e-2,3-d ica r boxyla te (10b),31 (2S,3S)-3-(eth oxyca r -
b on yl)a zir id in e-2-ca r b oxylic a cid (11a ),27 (2R,3R)-3-
(et h oxyca r b on yl)a zir id in e-2-ca r b oxylic a cid (11b ),27
ben zyl eth yl fu m a r a te (12),68 d ieth yl (2S,3S)-1-[N-(ter t-
b u t oxyca r b on yl)-(S)-p h en yla la n yl]a zir id in e-2,3-d ica r -
boxylate (17a)(BOC-P h e-(S,S)-Azi-(OEt)2),31 dieth yl (2R,3R)-
1-[N-(ter t-b u t oxyca r b on yl)-(S)-p h en yla la n yl]a zir id in e-
2,3-d ica r b oxyla t e (17b ) (BOC-P h e-(R,R)-Azi-(OE t )2),31
d ieth yl (2S,3S)-1-[N-(ben zyloxyca r bon yl)-(S)-a la n yl]a zir -
id in e-2,3-d ica r b oxyla t e (18a ) (Z-Ala -(S,S)-Azi-(OE t )2),31
a n d d ieth yl (2S,3S)-1-[N-(ter t-bu toxyca r bon yl)-(S)-a la -
n yl]a zir id in e-2,3-d ica r boxyla te (19a ) (BOC-Ala -(S,S)-Azi-
(OEt)2).31
(2S,3S)- a n d (2R,3R)-2-Ben zyl 3-eth yla zir id in e-2,3-d i-
ca r boxyla te (13a +b) (EtO-Azi-OBzl). DPSI69 (5.2 g, 0.024
mol) and 5.9 g of benzyl ethyl fumarate (12) (0.02 mol) were
dissolved in 60 mL of toluene and heated at 80 °C for 24 h.
The mixture was evaporated in vacuo. Column chromatogra-
phy (silica gel 60, cyclohexane/ethyl acetate, 4/1, Rf ) 0.2)
yielded 1.2 g (25%) of 13a +b as a yellowish solid:70 mp 52 °C.
IR (CCl4): 3282 (br), 1733, 1497 cm-1 1H NMR (CDCl3): δ
.
1.30 (t, J ) 7.0 Hz, 3 H), 1.85 (bs, 1 H, NH), 2.90 (d, J ) 2.3
Hz, 1 H, CH-Azi), 2.92 (d, J ) 2.3 Hz, 1 H, CH-Azi), 4.22 (q, J
) 7.0 Hz, 2 H), 5.15 (d, J ) 12.0 Hz, 1 H), 5.25 (d, J ) 12.0
Hz, 1 H), 7.3-7.45 (m, 5 H). 13C NMR (CDCl3): δ 13.81, 35.45,
35.60, 61.71, 67.36, 126.60, 128.22, 128.40, 134.73, 169.22
Gen er a l P r oced u r es. N-Acyla tion via Sym m etr ic An -
h yd r id es. N-Protected amino acid (15.5 mmol) was dissolved

568 J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 4
Schirmeister
(double peak). MS (CI, C4H10): m/z (%) ) 306 (1.1) [M+
C4H9], 250 (100) [M+ + 1]. Anal. (C13H15NO4) C,H,N.
+
liquid: [R]D ) -125.0° (c 0.44, EtOH). IR (ethyl acetate):
21
3275 (br), 1742, 1656, 1539, 1448 cm-1. H NMR (CDCl3): δ
1
0.88 (d, J ) 6.6 Hz, 3 H), 0.92 (d, J ) 6.6 Hz, 3 H), 1.26 (t, J
) 7.1 Hz, 3 H), 1.4-1.52 (m, 2 H, â-H Leu), 1.58-1.7 (m, 1 H,
γ-H Leu), 1.74 (bt, J ) 8.0 Hz, 1 H, NH Azi), 1.9-2.1 (m, 3 H,
â-H Pro, 2 γ-H Pro), 2.2 (m, 1 H, â-H Pro), 2.64 (d, J ) 7.6 Hz,
1 H, Azi), 2.80 (d, J ) 8.5 Hz, 1 H, Azi), 3.56 (m, 1 H, δ-H
Pro), 3.64 (m, 1 H, δ-H Pro), 4.18 (q, J ) 7.1 Hz, 2 H), 4.52 (m,
1 H, R-H Pro), 4.70 (m, 1 H, R-H Leu), 5.04 (d, J ) 12.2 Hz, 1
H), 5.16 (d, J ) 12.2 Hz, 1 H), 6.86 (bd, J ) 8.0 Hz, 1 H, NH
Leu), 7.2-7.5 (m, 5 H). 13C NMR (CDCl3): δ 14.14, 21.78,
23.40, 24.70, 24.99, 20.02, 35.58, 37.49, 41.48, 46.90, 48.40,
58.98, 62.19, 67.03, 128.25, 128.40, 128.63, 135.62, 167.88,
170.50, 170.92, 171.74. MS (EI 70 eV): m/z (%) ) 459 (27) [M+],
227 (100). Anal. (C24H33N3O6) C,H,N.
N-{[(2S,3S)-3-(Eth oxycar bon yl)azir idin -2-yl]car bon yl}-
(S)-leu cin e Ben zyl Ester (14a ) (EtO-(S,S)-Azi-Leu -OBzl).
11a was coupled with L-leucine benzyl ester tosylate using the
general DPPA procedure. The product was purified by column
chromatography on silica gel (ethyl acetate/cyclohexane, 1/1,
Rf ) 0.44) to give 14a (760 mg, 42%) as a yellowish viscous
27.3
liqiud: [R]D
) +47.2° (c 1.4, EtOH). IR (CCl4): 3283 (br),
1739, 1675, 1526 cm-1 1H NMR (CDCl3): δ 0.9 (d, J ) 6.1
.
Hz, 6 H), 1.3 (t, J ) 7.08 Hz, 3 H), 1.46-1.7 (m, 3 H), 1.8 (bt,
J ) 8.5 Hz, 1 H, NH), 2.55 (dd, J ) 2.2, 7.8 Hz, 1 H), 2.87 (dd,
J ) 2.2, 9.0 Hz, 1 H), 4.22 (q, J ) 7.08 Hz, 2 H), 4.55-4.65
(m, 1 H), 5.15 (s, 2 H), 6.78 (bd, J ) 8.8 Hz, 1 H), 7.28-7.4 (m,
5 H). 13C NMR (CDCl3): δ 14.07, 21.73, 22.79, 24.84, 35.44,
37.40, 41.03, 50.36, 62.20, 67.09, 128.20, 128.41, 128.61,
135.29, 167.75, 170.30, 172.22. MS (EI, 70 eV): m/z (%) ) 363
(14) [M+ + 1], 227 (100). Anal. (C19H26N2O5) C,H,N.
Dieth yl (2S,3S)-1-[N-(F lu or en -9-ylm eth oxyca r bon yl)-
(S)-leu cyl]a zir id in e-2,3-d ica r boxyla te (20a ) (F m oc-Leu -
(S,S)-Azi-(OEt)2). N-Acylation of 10a via symmetric anhy-
dride using Fmoc-Leu yielded after column chromatography
(cyclohexane/ethyl acetate, 5/1, Rf ) 0.21) 20a (1.6 g, 44%):
mp 56 °C; [R]D25.3 ) -10.52° (c 1.065, EtOH). IR (CH2Cl2): 3418
N-{[(2R,3R)-3-(Eth oxycar bon yl)azir idin -2-yl]car bon yl}-
(S)-leu cin e Ben zyl Ester (14b) (EtO-(R,R)-Azi-Leu -OBzl).
11b was coupled with L-leucine benzyl ester tosylate using the
general DPPA method. The product was purified by column
chromatography on silica gel (ethyl acetate/cyclohexane, Rf )
1
(br), 1738, 1527 cm-1. H NMR (CDCl3): δ 0.9 (d, J ) 6.6 Hz,
6 H), 1.23 (t, J ) 7.1 Hz, 3 H), 1.25 (t, J ) 7.1 Hz, 3 H), 1.45-
1.8 (m, 3 H), 3.5 (s, 2 H, Azi), 4.15-4.3 (m, 3 H, Fmoc), 4.4 (q,
J ) 7.1 Hz, 4 H), 5.15 (m, 1 H, R-H), 5.3 (bd, J ) 8.0 Hz, 1 H,
0.39) to give 14b (633 mg, 35%) as a yellowish viscous liqiud:
25.2
[R]D
) -110.7° (c 1.4, EtOH). IR (CCl4): 3275 (br), 1742,
NH), 7.2-7.4 (m, 4 H), 7.5-7.6 (m, 2 H), 7.7-7.8 (m, 2 H). 13
C
1670, 1537 cm-1. 1H NMR (CDCl3): δ 0.9 (d, J ) 6.1 Hz, 6 H),
1.25 (t, J ) 7.0 Hz, 3 H), 1.45-1.8 (m, 4 H, NH, â-H, γ-H),
2.62 (d, J ) 2.0 Hz, 1 H), 2.85 (d, J ) 2.0 Hz, 1 H), 4.2 (q, J )
7.0 Hz, 2 H), 4.55-4.65 (m, 1 H), 5.13 (d, J ) 12.0 Hz, 1 H),
5.18 (d, J ) 12.0 Hz, 1 H), 6.65 (bd, J ) 8.5 Hz, 1 H), 7.3-7.4
(m, 5 H).13C NMR (CDCl3): δ 13.83, 21.49, 22.56, 24.58, 35.17,
37.12, 40.72, 50.18, 61.86, 66.80, 127.93, 128.14, 128.35,
135.10, 167.65, 170.02, 171.97. MS (EI, 70 eV): m/z (%) ) 363
(9) [M+ + 1], 227 (100). MS (CI, C4H10): m/z (%) ) 363 (100)
[M+ + 1]. Anal. (C19H26N2O5) C,H,N.
NMR (CDCl3): δ 13.97, 21.74, 22.73, 24.66, 39.89, 41.44, 47.20,
54.37, 62.62, 66.98, 119.96, 125.11, 127.05, 127.68, 141.29,
143.74, 155.79, 166.06, 180.63. MS (EI, 70 eV): m/z (%) ) 522
(10) [M+], 178 (100). MALDI-TOF HRMS: calcd, 545.22637
[M + Na]+; found, 545.2252 ( 0.00174. Anal. (C29H34N2O7)
C,H; N: calcd, 5.36; found, 4.25.
N-{(2S,3S)-1-[[N-(ter t-Bu toxyca r bon yl)-(S)-p h en yla la -
n yl]-3-(et h oxyca r b on yl)a zir id in -2-yl]ca r b on yl}-(S)-leu -
cin e Ben zyl Ester (21a ) (BOC-P h e-(EtO)-(S,S)-Azi-Leu -
OBzl). N-Acylation of 14a (1.6 mmol, 579 mg) via the
symmetric anhydride procedure using BOC-Phe gave, after
N-{[(2R,3R)-3-(Eth oxycar bon yl)azir idin -2-yl]car bon yl}-
(S)-leu cin e (15b) (EtO-(R,R)-Azi-Leu -OH). 14b was hydro-
genolyzed using the general hydrogenolysis procedure. Re-
crystallization from MeOH yielded 95% (155 mg) 15b: mp 66
column chromatography (cyclohexane/ethyl acetate, 1/1, Rf )
27.3
0.7), 21a (347 mg, 95%): mp 50 °C; [R]D
) -10.6° (c 0.25,
1
EtOH). IR (KBr): 3359 (br), 1743, 1710, 1498 cm-1. H NMR
(CDCl3): δ 0.9 (t, J ) 6.3 Hz, 6 H), 1.3 (t, J ) 7.3 Hz, 3 H), 1.4
(s, 9 H), 1.45-1.7 (m, 3 H, Leu), 2.6 (d, J ) 1.95 Hz, 1 H, Azi),
3.0-3.2 (m, 2 H, â-H Phe), 3.28 (d, J ) 1.95 Hz, 1 H, Azi),
4.23 (m, 2 H, OCH2), 4.45 (m, 1 H, R-H Phe), 4.6 (m, 1 H, R-H
Leu), 5.18 (s, 2 H), 5.26 (bd, J ) 8.3 Hz, 1 H, NH Phe), 6.34
(bd, J ) 8.1 Hz, 1 H, NH Leu), 7.12-7.28 (m, 5 H), 7.28-7.4
(m, 5 H). 13C NMR (CDCl3): δ 13.93 (CH3), 21.85, 22.71 (Leu
CH3), 24.91 (Leu), 28.26 (BOC), 39.58 (Azi), 40.57 (Azi), 41.34
(double peak, â-C Leu, Phe), 50.88 (R-C Leu), 57.0 (R-C Phe),
62.63, 67.25 (OCH2), 80.01 (BOC), 126.87, 128.22, 128.48,
128.54, 128.65, 129.54, 135.35, 136.73, 154.87, 164.76, 165.76,
171.93, 178.95. HREIMS (70 eV): calcd, 609.3050; found,
609.3051. Anal. (C33H43N3O8) C,H,N.
23
°C; [R]D ) -63.6° (c 0.55, EtOH). IR (KBr): 3374 (br), 1737,
1670, 1544 cm-1 1H NMR (CDCl3): δ 0.94 (d, J ) 5.9 Hz, 3
.
H), 0.96 (d, J ) 6.11 Hz, 3 H), 1.30 (t, J ) 7.08 Hz, 3 H), 1.5-
1.8 (m, 3 H), 2.71 (d, J ) 2.2 Hz, 1 H), 2.89 (d, J ) 2.2 Hz, 1
H), 4.23 (q, J ) 7.3 Hz, 2 H), 4.5-4.6 (m, 1 H), 5.0 (bs, 2 H,
NH, COOH), 6.76 (bd, J ) 8.5 Hz, 1 H, NH). MS (CI, C4H10):
m/z (%) ) 273 (100) [M+ + 1]. MS (EI, 70 eV): m/z (%) ) 273
(12) [M+ + 1], 227 (100). Anal. (C12H20N2O5‚1/2MeOH) C,N; H:
calcd, 7.69; found, 7.23.
N-{[(2S,3S)-3-(Eth oxycar bon yl)azir idin -2-yl]car bon yl}-
(S)-leu cyl-(S)-p r olin e Ben zyl Ester (16a ) (EtO-(S,S)-Azi-
Leu -P r o-OBzl). DPPA-mediated coupling of 11a with Leu-
Pro-OBzl TFA gave, after purification by flash chromatography
(cyclohexane/ethyl acetate, 1/2, Rf ) 0.19), 51% (1.17 g) of 16a
22
N-{(2R,3R)-1-[[N-(ter t-Bu toxyca r bon yl)-(S)-p h en yla la -
n yl]-3-(et h oxyca r b on yl)a zir id in -2-yl]ca r b on yl}-(S)-leu -
cin e Ben zyl Ester (21b) (BOC-P h e-(EtO)-(R,R)-Azi-Leu -
OBzl). N-Acylation of 14b (1.0 mmol, 362 mg) via the
symmetric anhydride procedure using BOC-Phe gave, after
column chromatography (cyclohexane/ethyl acetate, 1/1, Rf )
as a colorless viscous liquid: [R]D ) -18.3° (c 1.45, EtOH).
IR (ethyl acetate): 3271, 1740, 1631, 1551, 1453 cm-1. 1H NMR
(CDCl3): δ 0.86 (d, J ) 6.35 Hz, 3 H), 0.91 (d, J ) 6.35 Hz, 3
H), 1.27 (t, J ) 7.1 Hz, 3 H), 1.45 (m, 2 H, â-H Leu), 1.5-1.64
(m, 1 H, γ-H Leu), 1.8-2.1 (m, 4 H, â-H Pro, 2 γ-H Pro, NH
Azi), 2.1-2.28 (m, 1 H, â-H Pro), 2.58 (d, J ) 1.95 Hz, 1 H),
2.78 (d, J ) 2.2 Hz, 1 H), 3.5-3.62 (m, 1 H, δ-H Pro), 3.7-3.8
(m, 1 H, δ-H Pro), 4.2 (q, J ) 7.3 Hz, 2 H), 4.54 (m, 1 H, R-H
Pro), 4.74 (m, 1 H, R-H Leu), 5.06 (d, J ) 12.2 Hz, 1 H), 5.16
(d, J ) 12.2 Hz, 1 H), 7.06 (bd, J ) 8.0 Hz, 1 H, NH Leu), 7.35
(m, 5 H). 13C NMR (CDCl3): δ 13.97, 21.63 (Leu), 23.19 (Leu),
24.52 (Leu), 24.76 (Pro), 28.84 (Pro), 35.59 (Azi), 37.35 (Azi),
41.47 (Leu), 46.74 (Pro), 48.55 (Leu), 58.82 (Pro), 61.93 (O-
ethyl), 66.79 (O-benzyl), 128.01, 128.18, 128.43, 135.42, 167.68,
170.00, 170.74, 171.49. MS (CI, C4H9): m/z (%) ) 459 (4) [M+],
368 (100). Anal. (C24H33N3O6) C,H,N.
24.4
0.7), 21b (429 mg, 70%) as a colorless viscous liquid: [R]D
) -34.7° (c 0.36, EtOH). IR (KBr): 3339 (br), 1742, 1498 cm-1
.
1H NMR (CDCl3): δ 0.9 (d, J ) 6.1 Hz, 6 H), 1.28 (t, J ) 7.1
Hz, 3 H), 1.38 (s, 9 H), 1.44-1.66 (m, 3 H, Leu), 3.1-3.26 (m,
2 H, â-H Phe), 3.35 (d, J ) 2.2 Hz, 1 H, Azi), 3.51 (d, J ) 2.2
Hz, 1 H, Azi), 4.22 (q, J ) 7.1 Hz, 2 H), 4.44-4.66 (m, 2 H,
R-H Leu, Phe), 4.92 (bd, J ) 9.0 Hz, 1 H, NH Phe), 5.16 (s, 2
H), 6.4 (bd, J ) 8.0 Hz, 1 H, NH Leu), 7.15-7.4 (m, 10 H). 13
C
NMR (CDCl3): δ 13.88, 21.75, 22.66, 24.76, 28.15, 38.22, 40.94,
41.19, 41.37, 50.84, 56.16, 62.50, 67.16, 80.09, 126.81, 128.15,
128.44, 128.57, 129.48, 135.14, 136.11, 154.87, 164.93, 165.81,
172.02, 179.26. MS (EI, 70 eV): m/z (%) ) 609 (31) [M+], 418
(100). Anal. (C33H43N3O8) C,H,N.
N-{[(2R,3R)-3-(Eth oxycar bon yl)azir idin -2-yl]car bon yl}-
(S)-leu cyl-(S)-p r olin e Ben zyl Ester (16b) (EtO-(R,R)-Azi-
Leu -P r o-OBzl). DPPA-mediated coupling of 15b with Pro-
OBzl HCl gave after purification by column chromatography
(ethyl acetate, Rf ) 0.56) 16b (1.3 g, 57%) as a colorless viscous
N-{(2S,3S)-1-[[N-(ter t-Bu toxyca r bon yl)-(S)-p h en yla la -
n yl]-3-(eth oxycar bon yl)azir idin -2-yl]car bon yl}-(S)-leu cyl-

New Peptidic Cysteine Protease Inhibitors
J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 4 569
(S)-p r olin e Ben zyl Ester (22a ) (BOC-P h e-(EtO)-(S,S)-Azi-
Leu -P r o-OBzl). N-Acylation of 16a (1.5 mmol, 688 mg) with
BOC-Phe via the symmetric anhydride method gave, after
Gly by the general EEDQ procedure gave after column
chromatography (ethyl acetate, Rf ) 0.64) 24a (320 mg, 49%)
24.9
as a colorless viscous liquid: [R]D
) +15.8° (c 0.92, EtOH).
1
purification by flash chromatography (cyclohexane/ethyl ac-
IR (CH2Cl2): 3395 (br), 1737, 1532 cm-1. H NMR (CDCl3): δ
1.2 (t, J ) 7.1 Hz, 6 H), 3.38 (s, 2 H, Azi), 3.8-4.08 (m, 4 H,
Gly), 4.15 (q, J ) 7.1 Hz, 4 H), 5.02 (s, 2 H), 5.72 (bt, J ) 5.8
Hz, 1 H, NH), 6.95 (bs, 1 H, NH), 7.25 (m, 5 H). 13C NMR
(CDCl3): δ 13.93, 39.72, 43.71, 44.32, 62.77, 67.20, 128.06,
128.17, 128.50, 136.09, 156.52, 166.00, 169.06, 176.54. MS (EI,
70 eV): m/z (%) ) 435 (28) [M+], 188 (100). MALDI-TOF
HRMS: calcd, 458.1539 [M + Na]+; found, 458.16082 (
0.00637. Anal. (C20H25N3O8) H,N; C: calcd, 55.17; found, 54.23.
25.1
etate, 1/1, Rf ) 0.6), 560 mg (53%) of 22a : mp 80 °C; [R]D
)
-46.7° (c 0.51, EtOH). IR (KBr): 3275 (br), 1744, 1713, 1632,
1498, 1455 cm-1. H NMR (CDCl3): δ 0.88 (d, J ) 6.35 Hz, 3
1
H, δ-H Leu), 0.92 (d, J ) 6.34 Hz, 3 H, δ-H Leu), 1.24 (t, J )
7.1 Hz, 3 H), 1.35 (s, 9 H, BOC), 1.42-1.8 (m, 3 H, â-H, γ-H
Leu), 1.85-2.1 (m, 3 H, 1 â-H, 2 γ-H Pro), 2.1-2.3 (m, 1 H,
â-H, Pro), 2.94-3.28 (m, 2 H, â-H Phe), 3.35 (d, J ) 2.0 Hz, 1
H, Azi), 3.45 (d, J ) 2.0 Hz, 1 H, Azi), 3.52-364 (m, 1 H, δ-H,
Pro), 3.66-3.80 (m, 1 H, δ-H, Pro), 4.12-4.24 (m, 2 H, OCH2),
4.4 (m, 1 H, R-H Phe), 4.58 (m, 1 H, R-H Pro), 4.76 (m, 1 H,
R-H Leu), 5.08 (d, J ) 12.2 Hz, 1 H), 5.20 (d, J ) 12.2 Hz, 1
H), 5.36 (bd, J ) 7.81, 1 H, NH Phe), 6.81 (bd, J ) 8.7 Hz, 1
H, NH Leu), 7.2-7.4 (m, 10 H). 13C NMR (CDCl3): δ 13.9
(OCH2CH3), 21.61, 23.28 (δ-C Leu), 24.58 (γ-C Pro), 24.79 (γ-C
Leu), 28.19 (BOC), 28.90 (â-C Pro), 39.67 (â-C Phe), 40.06 (Azi),
40.72 (Azi), 41.57 (â-C Leu), 46.83 (δ-C Pro), 48.96 (R-C Leu),
57.11 (R-C Phe), 58.92 (R-C Pro), 62.51 (OCH2), 66.91 (OCH2-
Ph), 79.48 (BOC), 126.67, 128.10, 128.30, 128.42, 128.52,
129.65, 135.47 (double peak), 154.71, 164.62, 165.97, 170.57,
171.51, 179.35. MS (CI, C4H9): m/z (%) ) 706 (100) [M+]. Anal.
(C38H50N4O9) C,H; N: calcd, 7.93; found, 7.46.
Dieth yl (2S,3S)-1-[N-(Ben zyloxyca r bon yl)-(S)-a la n yl-
(S)-p h en yla la n yl]a zir id in e-2,3-d ica r boxyla te (25a ) (Z-
Ala -P h e-(S,S)-Azi-(OEt)2. N-Acylation of 10a (5 mmol, 935
mg) with Z-Ala-Phe by the general EEDQ procedure gave, after
column chromatography (cyclohexane/ethyl acetate, 2/1, Rf )
25.1
0.14), 25a (1.17 g, 42.8%) as a yellowish viscous liquid: [R]D
) +9.2 (c 0.82, EtOH). IR (CH2Cl2): 3423 (br), 1738, 1523 cm-1
.
1H NMR (CDCl3): δ 1.22 (d, J ) 6.6 Hz, 3 H, Ala), 1.28 (t, J
) 7.1 Hz, 6 H), 3.06-3.34 (m, 2 H, Phe), 3.5 (s, 2 H, Azi), 4.18-
4.3 (m, 5 H, R-H Ala, 2 OCH2), 4.9-5.02 (m, 1 H, R-H Phe),
5.08 (d, J ) 12.2 Hz, 1 H), 5.15 (d, J ) 12.2 Hz, 1 H), 5.25 (bs,
1 H, NH Ala), 6.44 (bd, J ) 8.7 Hz, 1 H, NH Phe), 7.16-7.4
(m, 10 H). 13C NMR (CDCl3): δ 13.91, 18.61, 37.77, 40.25,
50.48, 54.39, 62.56, 67.02, 126.98, 128.01, 128.43, 128.48,
129.52, 135.73, 136.12, 155.74, 165.87, 171.70, 178.46. MS (EI,
70 eV): m/z (%) ) 539 (63) [M+], 91 (100). Anal. (C28H33N3O8)
C,H,N.
N-{(2R,3R)-1-[[N-(ter t-Bu toxyca r bon yl)-(S)-p h en yla la -
n yl]-3-(eth oxycar bon yl)azir idin -2-yl]car bon yl}-(S)-leu cyl-
(S)-p r olin e Ben zyl Ester (22b) (BOC-P h e-(EtO)-(R,R)-Azi-
Leu -P r o-OBzl). N-Acylation of 16b (0.15 mmol, 69 mg) with
BOC-Phe via the symmetric anhydride method gave, after
Dieth yl (2R,3R)-1-[N-(Ben zyloxyca r bon yl)-(S)-a la n yl-
(S)-p h en yla la n yl]a zir id in e-2,3-d ica r boxyla te (25b) (Z-
Ala -P h e-(R,R)-Azi-(OEt)2). N-Acylation of 10b (3.3 mmol,
617 mg) with Z-Ala-Phe by the general EEDQ procedure gave,
after purification by column chromatography (cyclohexane/
ethyl acetate, 3/1, Rf ) 0.15), 25b (410 mg, 23.1%): mp 57 °C;
[R]D25 ) -35.78° (c 1.034, EtOH). IR (ethyl acetate): 3325 (Br),
1738, 1675, 1520, 1455 cm-1. 1H NMR (CDCl3): δ 1.26 (d, J )
6.6 Hz, 3 H, Ala), 1.27 (t, J ) 7.1 Hz, 6 H), 3.0-3.35 (m, 2 H,
Phe), 3.5 (s, 2 H, Azi), 4.16-4.25 (m, 5 H, R-H Ala, 2 OCH2),
4.90-5.0 (m, 1 H, R-H Phe), 5.05-5.2 (m, 3 H, OCH2, NH Ala),
6.6 (bd, J ) 8.1 Hz, 1 H, NH Phe), 7.1-7.4 (m, 10 H). 13C NMR
(CDCl3): δ 13.98, 18.16, 37.81, 40.27, 50.56, 54.44, 62.58, 67.19,
127.00, 128.17, 128.28, 128.46, 128.57, 129.54, 135.73, 136.07,
155.80, 165.83, 171.56, 178.32. MS (CI, C4H10): m/z (%) ) 540
(5) [M+ + 1], 353 (100). Anal. (C28H33N3O8): C,H,N.
purification by flash chromatography (cyclohexane/ethyl ac-
24.7
etate, 1/1, Rf ) 0.7), 60 mg (57%) of 22b: mp 77 °C; [R]D
)
.
-61.4° (c 0.3, EtOH). IR (CH2Cl2): 3413 (br), 1742, 1641 cm-1
1H NMR (CDCl3): δ 0.9 (d, J ) 6.6 Hz, 3 H), 0.96 (d, J ) 6.6
Hz, 3 H), 1.26 (t, J ) 7.1 Hz, 3 H), 1.36 (s, 9 H), 1.44-1.76 (m,
3 H, Leu), 1.88-2.1 (m, 3 H, Pro), 2.1-2.3 (m, 1 H, Pro), 3.0-
3.24 (m, 2 H, Phe), 3.4 (d, J ) 1.96, 1 H, Azi), 3.47 (d, J ) 1.96
Hz, Azi), 3.58 (m, 1 H, Pro), 3.76 (m, 1 H, Pro), 4.2 (q, J ) 7.1
Hz, 2 H), 4.50-4.66 (m, 2 H, R-H Phe, Pro), 4.74 (m, 1 H, R-H
Leu), 5.0 (bd, J ) 9.0 Hz, 1 H, NH Phe), 5.09 (d, J ) 12.2 Hz,
1 H), 5.2 (d, J ) 12.2 Hz, 1 H), 6.7 (bd, J ) 8.0 Hz, 1 H, NH),
7.14-7.4 (m, 10 H). 13C NMR (CDCl3): δ 13.87, 21.68, 23.26,
24.55, 24.83, 28.16, 28.89, 38.42, 41.18, 41.31, 41.42, 46.81,
49.12, 56.25, 58.89, 62.50, 66.90, 79.89, 126.82, 128.10, 128.30,
128.43, 128.52, 129.56, 135.46, 136.13, 154.78, 164.88, 165.92,
170.57, 171.51, 179.52. MS (EI, 70 eV): m/z (%) ) 706 (46)
[M+], 206 (100). Anal. (C38H50N4O9) C,H; N: calcd, 7.93; found,
7.40.
Dieth yl (2S,3S)-1-[N-(ter t-Bu toxyca r bon yl)-(S)-p h en y-
lalan yl-(S)-alan yl]azir idin e-2,3-dicar boxylate (26a) (BOC-
P h e-Ala -(S,S)-Azi-(OEt)2). N-Acylation of 10a (1.5 mmol, 280
mg) with BOC-Phe-Ala by the general EEDQ method yielded,
after purification by column chromatography (cyclohexane/
(2S,3S)- a n d (2R,3R)-2-Ben zyl 3-Eth yl 1-[N-(ter t-Bu -
toxyca r bon yl)-(S)-p h en yla la n yl]a zir id in e-2,3-d ica r boxy-
la te (23a +b) (BOC-P h e-(S,S+R,R)-Azi-(OEt)(OBzl). N-
Acylation of 13a +b (2 mmol, 498 mg) with BOC-Phe via the
symmetric anhydride procedure gave, after column chroma-
tography (cyclohexane/ethyl acetate, 2/1, Rf ) 0.65) 23a +b
ethyl acetate, 3/1, Rf ) 0.15), 26a (580 mg, 77%): mp 99 °C;
23.8
[R]D
) -4.7° (c 0.48, EtOH). IR (ethyl acetate): 3282 (br),
1734, 1446 cm-1 1H NMR (CDCl3): δ 1.25 (t, J ) 7.1 Hz, 6
.
H), 1.35 (s, 9 H, BOC), 1.4 (d, J ) 6.6 Hz, 3 H, Ala), 2.9-3.1
(m, 2 H, Phe), 3.4 (s, 2 H, Azi), 4.1-4.38 (m, 5 H, R-H Phe, 2
OCH2), 4.5 (m, 1 H, R-H Ala), 4.96 (bs, 1 H, NH Phe), 6.7 (bd,
J ) 7.8 Hz, 1 H, NH Ala), 7.1-7.3 (m, 5 H). 13C NMR (CDCl3):
δ 13.98 (OCH2CH3), 17.92 (â-C Ala), 28.15 (BOC), 39.75 (â-C
Phe), 40.03 (Azi), 49.23 (R-C), 49.94 (R-C), 62.56 (OCH2), 80.0
(BOC), 126.75, 128.47, 129.35, 136.54, 156.5 (BOC), 165.93,
170.34, 179.96. MS (CI, C4H9): m/z (%) ) 505 (100) [M+]. Anal.
(C25H35N3O8) C,H; N: calcd, 8.31; found, 7.43.
(2S,3S)- a n d (2R,3R)-2-Ben zyl 3-Eth yl 1-[N-(ter t-Bu -
toxyca r bon yl)-(S)-p h en yla la n yl-(S)-a la n yl]a zir id in e-2,3-
dicar boxylate (27a+b) (BOC-P h e-Ala-(S,S+R,R)-Azi-(OEt)-
(OBzl)). N-Acylation of 13a +b (0.4 mmol, 100 mg) with BOC-
Phe-Ala by the general EEDQ procedure gave, after column
(23a (S,S)/23b (R,R) ) 5/4) (850 mg, 86%) as a yellowish
22.7
viscous liquid: [R]D
) -10.2° (c 0.71, EtOH). IR (ethyl
acetate): 3377 (br), 1743, 1712, 1497, 1455 cm-1 1H NMR
.
(CDCl3): δ 1.28 (t, J ) 7.1 Hz, 3 H), 1.36 and 1.38 (s, together
9 H), 3.0-3.3 (m, 2 H, â-H Phe), 3.35 (S,S) and 3.49 (R,R)(d,
J ) 2.2 Hz, together 1 H, CH Azi), 3.40 (S,S) and 3.53 (R,R)
(d, J ) 2.2 Hz, together 1 H, CH Azi), 4.22 (q, J ) 7.1 Hz, 2
H), 4.5 (S,S) and 4.63 (R,R) (m, together 1 H, R-H Phe), 4.93
(R,R) and 5.1 (S,S) (bd, J ) 9.0 Hz, together 1 H, NH), 5.2 (m,
2 H), 7.2-7.45 (m, 10 H). 13C NMR (CDCl3): δ 13.75, 28.02,
28.06 (BOC), 38.09, 39.03, 39.61, 39.87, 39.93, 40.21 (â-C Phe,
Azi), 55.97, 56.37 (R-C Phe), 62.31, 62.36, 67.91, 67.93 (OCH2),
79.48, 79.64 (BOC), 126.57, 126.60, 128.14, 128.22, 128.30,
128.40, 128.43, 128.48, 128.52, 128.58, 129.43, 134.38, 134.41,
136.01, 136.11, 154.60, 154.75, 165.67, 165.74, 178.78, 178.99.
MS (CI, C4H9): m/z (%) ) 496 (75) [M+], 405 (100). Anal.
(C27H32N2O7) C,H,N.
chromatography (cyclohexane/ethyl acetate, 3/1, Rf ) 0.13),
23.3
27a +b (150 mg, 66%) (1.3/1): mp 115 °C; [R]D
) -11.4° (c
0.37, EtOH). IR (ethyl acetate): 3305 (br), 1741, 1720, 1664,
1524, 1500, 1455 cm-1. H NMR (CDCl3): δ 1.2-1.45 (m, 15
1
Dieth yl(2S,3S)-1-[N-(Ben zyloxyca r bon yl)glycylglycyl]-
a zir id in e-2,3-d ica r boxyla te (24a ) (Z-Gly-Gly-(S,S)-Azi-
(OEt)2. N-Acylation of 10a (1.5 mmol, 280 mg) with Z-Gly-
H), 3.0-3.15 (m, 2 H), 3.4-3.55 (m, 2 H), 4.15-4.3 (m, 3 H,
R-H Phe, OCH2), 4.55 und 4.7 (m, together 1 H, R-H Ala), 4.9
(bs, 1 H, NH Phe), 5.2 (s, 2 H), 6.55 and 6.65 (bs, together 1

570 J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 4
Schirmeister
H, NH Ala), 7.1-7.3 (m, 10 H). 13C NMR (CDCl3): δ 13.9,
18.18, 26.90, 28.21 (BOC), 39.70, 39.99 (â-C Phe), 40.05, 40.23
(Azi), 49.51, 50.01 (R-C Phe), 62.54, 62.72 (OCH2), 68.16, 68.29
(OCH2Ph), 80.0 (BOC), 126.97, 128.63, 128.68, 128.72, 128.74,
128.81, 129.14, 129.37, 129.43, 134.46, 134.55, 136.35, 136.44,
156.0, 165.78, 165.82, 165.88, 165.95, 170.36, 170.61, 179.26,
179.96. MS (CI, C4H9): m/z (%) ) 567 (62) [M+], 476 (100).
MALDI-TOF HRMS: calcd, 590.2478 [M + Na]+; found,
590.2464 ( 0.0010. Anal. (C30H37N3O8) H,N; C: calcd, 63.48;
found, 64.15.
(2S,3S)- a n d (2R,3R)-2-Ben zyl 3-Eth yl 1-[N-(ter t-Bu -
toxyca r bon yl)-(S)-leu cylglycyl]a zir id in e-2,3-d ica r boxy-
la te (28a +b) (BOC-Leu -Gly-(S,S+R,R)-Azi-(OEt)(OBzl)).
N-Acylation of 13a +b (1.6 mmol, 398 mg) with BOC-Leu-Gly
via the EEDQ method gave, after flash chromatography
mM DTT, 0.005% Brij 35; cathepsin B55 or cathepsin L/Z-Phe-
Arg-AMC,55-57 50 mM citrate buffer, pH 6.0, 2.5 mM DTT, 5
mM EDTA, 200 mM NaCl, 0.005% Brij 35; calpain I or II/Suc-
Leu-Tyr-AMC,58 50 mM TRIS buffer, pH 7.5, 5 mM CaCl2, 2.5
mM EDTA, 2.5 mM DTT; trypsin/L-BAPA59 or chymotrypsin/
Ac-Phe-pNA,60 0.1 M phosphate buffer, pH 7.6; elastase/Ac-
Ala-Ala-Ala-pNA,61 0.1 M TRIS buffer, pH 8.0; thermolysin/
FAGLA,62 0.1 M TRIS buffer, pH 7.22, 10 mM CaCl2; pepsin/
Ac-Phe-3,5-diiodo-TyrOH,63 0.22 mM acetic acid. Assay buffers
for the inactivation of papain at different pH values were 50
mM citrate for pH 4-6, phosphate for pH 6-8, and borate for
pH 8-9. All kinetic experiments were done at 25 °C. Enzyme
solutions were prepared fresh by incubating the enzymes in
the individual reaction buffers at 25 °C for 30 min. Substrates
and inhibitors were dissolved in DMSO, the final concentra-
tions at measurements not being higher than 12% DMSO.
Substrate solutions were prepared by diluting a stock solution
with buffer, and inhibitor solutions were diluted with water.
Substrate hydrolysis in assays using fluorogenic substrates
was monitored by the increase of fluorescence at 460 nm
(irradiation at 380 nm) and in assays using nitroanilide
substrates by absorption increase at 405 nm. Thermolysin was
assayed by absorption decrease at 345 nm. Pepsin was assayed
(cyclohexane/ethyl acetate, 3/1, Rf ) 0.14), 28a +b (1.5/1) (430
24.2
mg, 51.7%): mp 88 °C; [R]D
) -18.3° (c 0.59, EtOH). IR
(ethyl acetate): 3338 (br), 1741, 1515, 1455 cm-1 1H NMR
.
(CDCl3): δ 0.9 (d, J ) 6.6 Hz, 3 H), 0.92 (d, J ) 6.6 Hz, 3 H),
1.25 und 1.28 (t, J ) 7.1 Hz, together 3 H), 1.45 (m, 11 H,
BOC, â-H Leu), 1.6-1.8 (m, 1 H, γ-H Leu), 3.45-3.55 (m, 2
H, Azi), 3.95-4.10 (m, 1 H, 1 R-H Gly), 4.15-4.35 (m, 4 H,
R-H Leu, 1 R-H Gly, OCH2), 4.9 (bs, 1 H, NH Leu), 5.2 (s, 2
H), 6.7 (bs, 1 H, NH Gly), 7.2-7.4 (m, 5 H). 13C NMR (CDCl3):
δ 13.94, 14.08 (OCH2CH3), 21.74, 22.97 (CH3 Leu), 24.70 (γ-C
Leu), 28.26 (BOC), 39.63, 39.84 (Azi), 41.28 (â-C Leu), 43.77
(R-C Gly), 53.02 (R-C Leu), 61.43, 62.73 (OCH2), 68.31 (OCH2-
Ph), 80.0 (BOC), 126.98, 128.64, 128.72, 128.83, 129.12, 130.99,
134.36, 156.0 (BOC), 165.87 (double peak), 172.57, 176.45. MS
(CI, C4H9): m/z (%) ) 519 [M+], 428 (100). Anal. (C26H37N3O8)
C,H; N: calcd, 8.09; found, 7.64.
by absorption increase at 570 nm. The inactivation rates (kobs
)
for different inhibitor concentrations in the presence of the
substrate were determined according to the continuous method
of Tian and Tsou64 by monitoring the product released from
hydrolysis of the substrate in the presence of the inhibitor as
a
function of time (fluorescence or absorption ) A(1 -
exp(-kobst)) + B) until complete inactivation of the enzyme
(typically 5-60 min), with steady-state conditions established
during inactivation time. This was done at constant enzyme
and various (3-7) inhibitor concentrations, respectively. For
weak inhibitors where complete inactivation took longer than
60 min, kobs values were determined by the dilution assay of
Kitz and Wilson.65 Thereby enzyme and inhibitor were incu-
bated. After an incubation time of 5-60 min (5-7 values), the
incubation mixture was diluted by adding substrate and buffer
and the residual enzyme activity [E] was measured. kobs values
were then calculated using the equation [E] ) [E0] exp(-kobst).
These experiments were repeated for 3-7 inhibitor concentra-
tions. Fitting of the kobs values, obtained by either continuous
or dilution assays, against the inhibitor concentrations to the
hyperbolic equation kobs ) ki[I]/Kiapp + [I] gave the individual
values of Kiapp and ki. The Kiapp values were corrected to zero
substrate concentration by the term 1 + [S]/Km in the equation
Diet h yl (2S,3S)-1-[N-(Ben zyloxyca r b on yl)glycyl-(S)-
ph en ylalan yl-(S)-alan yl]azir idin e-2,3-dicar boxylate (29a)
(Z-Gly-P h e-Ala -(S,S)-Azi-(OEt)2). N-Acylation of 10a (0.4
mmol, 74 mg) with Z-Gly-Phe-Ala using the general EEDQ
procedure gave, after purification by column chromatography
(cyclohexan/ethyl acetate, 3/1, Rf ) 0.1), 29a (110 mg, 46%):
24.6
mp 89 °C; [R]D
) +6.3° (c 0.5, EtOH). IR (ethyl acetate):
3305 (br), 1738, 1656, 1455 cm-1
.
H NMR (CDCl3): δ 1.0-
1
1.4 (m, 9 H, â-H Ala, 2 CH3), 3.08 (m, 2 H, â-H Phe), 3.45 (s,
2 H, Azi), 3.75-4.0 (m, 2 H, Gly), 4.2 (q, J ) 7.1 Hz, 4 H,
OCH2), 4.6 (m, 1 H, R-H Ala), 4.8 (m, 1 H, R-H Phe), 5.1 (s, 2
H, OCH2Ph), 5.7 (bs, 1 H, NH Gly), 6.75 (bd, J ) 8.8 Hz, 1 H,
NH Phe), 6.88 (bs, 1 H, NH Ala), 7.1-7.5 (m, 10 H). 13C NMR
(CDCl3): δ 13.92 (CH2CH3), 17.84 (CH3 Ala), 38.24 (â-C Phe),
40.10 (Azi), 44.66 (Gly), 49.41 (R-C Ala), 54.25 (R-C Phe), 62.56
(OCH2), 67.22. (OCH2Phe), 127.02, 128.08, 128.20, 128.50,
128.63, 129.29, 136.11, 136.23, 155.0, 166.08, 169.27, 170.02,
179.36. MS (CI, C4H10): m/z (%) ) 596 (93) [M+], 488 (100).
Anal. (C30H36N4O9) C,H,N.
Ki ) Kiapp/(1 + [S]/Km). The second-order rate constants k2nd
)
ki/Ki were directly calculated from the individual constants.
In cases where no saturation kinetics were achieved ([I] < Ki)
due to limited solubility of the inhibitors or high Ki values,
the second-order rate constants were calculated from linear
regression to the equation kobs ) kapp[I] and corrected to zero
substrate concentration from the equation k2nd ) kapp(1 +
[S]/Km). Ki values for the non-time-dependent inhibition of
cathepsin H and calpains I and II were obtained by Dixon
plots66 using the equation v0/vi ) 1 + [I]/Kiapp and corrected to
zero substrate concentration from Ki ) Kiapp/(1 + [S]/Km). The
following Km values were used: papain/Z-Phe-Arg-AMC 0.09
mM, papain/L-BAPA 2.5 mM, cathepsin B/Z-Phe-Arg-AMC
0.15 mM, cathepsin L/Z-Phe-Arg-AMC 6.5 µM, cathepsin H/H-
Arg-AMC 0.15 mM, calpain I/Suc-Leu-Tyr-AMC 0.4 mM,
calpain II/Suc-Leu-Tyr-AMC 0.22 mM. Enzyme concentrations
(2S,3S)- a n d (2R,3R)-1-[N-(ter t-Bu toxyca r bon yl)-(S)-
p h en yla la n yl]-3-(et h oxyca r b on yl)a zir id in e-2-ca r boxyl-
ic Acid (30a+b) (BOC-P h e-(S,S+R,R)-Azi-(OEt)(OH)). Cata-
lytic hydrogenolysis of 23a +b by the general method described
above gave, after recrystallization from acetonitrile, 30a +b
(1/1) (234 mg, 96%): mp 83 °C; [R]D25.1 ) -1.0° (c 0.49, EtOH).
IR (ethyl acetate): 3376 (br), 1742, 1720, 1497, 1455 cm-1. 1H
NMR (CDCl3): δ 1.1-1.5 (m, 12 H), 3.1 (m) and 3.3 (m)
(together 2 H, â-H Phe), 3.4-3.6 (m, 2 H, Azi), 4.23 (q, J ) 7.1
Hz, 2 H), 4.4 (m) and 4.6 (m) (together 1 H, R-H Phe), 5.15
(bd, J ) 8.8 Hz) and 5.3 (bd, J ) 8.9 Hz) (together 1 H, NH),
7.0-7.4 (m, 5 H), 9.5 (bs, 1 H, COOH).13C NMR (CDCl3): δ
13.81 (CH3), 27.98, 28.05 (BOC), 38.31 (double peak), 38.83,
39.19, 39.61, 40.66 (â-C Phe, Azi), 56.18 (R-C Phe), 62.17, 62.25
(OCH2), 80.31 (BOC), 126.64, 126.83, 128.26, 129.09, 129.19,
129.41, 135.79, 135.99, 155.19 (double peak), 166.14 (double
peak), 169.00 (double peak), 178.96, 179.60. MS (CI, C4H10):
m/z (%) ) 463 (4) [M+ + C4H9], 407 (46) [M+ + 1], 351 (100).
Anal. (C20H26N2O7) H,N; C: calcd, 59.10; found, 60.30.
Kin etic Mea su r em en ts. Kinetic measurements were car-
ried out on a Pharmacia LKB Ultrospec III photometer and
on a Perkin-Elmer LS-3B fluorescence spectrometer, respec-
tively. The following assay buffers were used: papain/L-
BAPA,53 50 mM phosphate buffer, pH 6.5, 5 mM EDTA, 5 mM
cysteine; papain/Z-Phe-Arg-AMC54 and cathepsin H/H-Arg-
AMC,55 50 mM phosphate buffer, pH 6.5, 2.5 mM EDTA, 2.5
were: papain 0.48 µg mL-1, cathepsin B 0.015 µg mL-1
,
cathepsin L 0.05 µg mL-1, cathepsin H 0.082 µg mL-1, calpain
I 3.8 µg mL-1, calpain II 0.04 mg mL-1, elastase 7.5 µg mL-1
chymotrypsin 0.15 mg mL-1, trypsin 3 µg mL-1, thermolysin
8.9 µg mL-1, pepsin 0.6 mg mL-1. The kinetic constants were
obtained by nonlinear or linear regression analysis using the
program GraFit.67
,
Dia lysis Exp er im en ts. Dialysis experiments were carried
out as follows. Papain (0.52 mg mL-1) was inactivated by 17a
(0.37 mM) and 11a (0.13 mM), respectively (60 min incubation
time, each) and then subjected to dialysis (visking dialysis
tubing type 27/32, Serva) against reaction buffer (60 min,
each). No recovery of enzyme activity (substrate: L-BAPA, 1.18
mM) was detected, whereas control enzymes maintained 70%

New Peptidic Cysteine Protease Inhibitors
J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 4 571
and 92% activity, respectively. Cathepsin L (0.07 µg mL-1) was
inactivated by 28a +b (50 µM) (60 min incubation time) and
then subjected to dialysis against reaction buffer (60 min). No
recovery of enzymatic activity was detected, whereas the
control enzyme maintained 52% activity.
(10) Green, G.; Shaw, E. Peptidyl diazomethyl ketones are specific
inactivators of thiol proteinases. J . Biol. Chem. 1981, 256, 1923-
1928.
(11) Angliker, H.; Anagli, J .; Shaw, E. Inactivation of calpain by
peptidyl fluoromethyl ketones. J . Med. Chem. 1992, 35, 216-
220.
Stoich iom etr y of In a ctiva tion of P a p a in by 11a . A
papain solution (2.76 µM, as determined by active site titration
with E-64 as described in ref 55) was titrated with 11a (three
values, substrate: L-BAPA, 1.18 mM, pH 6.5) by adding
increasing amounts of the inhibitor solution ([I] ) 0.528 µM-
1.056 mM) to the papain solution. After each aliquot of
inhibitor was added, residual enzyme activity [E] was moni-
tored. The intercept with the x-axis of a plot of [E] vs [I] gave
the active site concentration; 1.11 ( 0.4 equiv of 11a was
required to totally inactivate 1 equiv of papain.
Standard inhibitors were N-ethylmaleimide71 for the dilu-
tion assay and E-6472 for the continuous assay, respectively.
Abbr evia tion s. Amino acids are written in the three-letter
codes and are L configured. Others: Azi (aziridine-2,3-dicar-
boxylic acid), Eps (epoxysuccinic acid), E-64 (1-[N-[[(L-3-trans-
carboxyoxiran-2-yl)carbonyl]-L-leucyl]amino]-4-guanidinobu-
tane), DPSI (diphenyl sulfimide), EEDQ (1-(ethoxycarbonyl)-
2-ethoxy-1,2-dihydroquinoline), DPPA (diphenyl phosphorazi-
date), DCC (dicyclohexylcarbodiimide), DMAP (4-(dimethyl-
amino)pyridine), PLE (pig liver esterase), CCL (Candida
Cylindraceae lipase), PPL (porcine pancreatic lipase), LPR
(lipase Penicillium roquefortii), LRA (lipase Rhizopus arrhi-
zus), LAN (lipase Aspergillus niger), BPACK ((benzyloxycar-
bonyl)phenylalanyl-alanine chloromethyl ketone), BAPA (ben-
zoylarginyl p-nitroanilide), AMC ((aminomethyl)coumarin),
pNA (p-nitroanilide), Tris (tris(hydroxymethyl)aminomethane),
FAGLA (3-(2-furyl)acryloyl-glycyl-leucine amide), DTT (dithio-
threitol), EDTA (ethylenedinitrilotetraacetic acid), Suc (suc-
cinyl).
(12) Krantz, A.; Copp, L.; Coles, P.; Smith, R.; Heard, S. Peptidyl
(acyloxy)methyl ketones and the quiescent affinity label con-
cept: The departing group as a variable structural element in
the design of inactivators of cysteine proteinases. Biochemistry
1991, 30, 4678-4687.
(13) Bro¨mme, D.; Schierhorn, A.; Kirschke, H.; Wiederanders, B.;
Barth, A.; Fittkau, A.; Demuth, H.-U. Potent and selective
inactivation of cysteine proteinases with N-peptidyl-O-acyl hy-
droxylamines. Biochem. J . 1989, 263, 861-866.
(14) Palmer, J .; Rasnick, D.; Klaus, J .; Bro¨mme, D. Vinyl sulfones
as mechanism-based cysteine protease inhibitors. J . Med. Chem.
1995, 38, 3193-3196.
(15) Drugs Future 1994, 19, 1039-1040.
(16) Yabe, Y.; Guillaume, D.; Rich, D. Irreversible inhibition of papain
by epoxysuccinyl peptides. 13C NMR characterization of the site
of alkylation. J . Am. Chem. Soc. 1988, 110, 4043-4044.
(17) Hanada, K.; Tamai, M.; Yamagishe, M.; Ohmura, S.; Sawada,
J .; Tanaka, I. Isolation and characterization of E-64, a new thiol
protease inhibitor. Agric. Biol. Chem. 1978, 42, 523-528.
(18) (a) Noda, T.; Isogal, K.; Katunuma, N.; Tarumoto, Y.; Ohzeki,
M. Effects on cathepsins B, H and D in pectoral muscle of
dystrophic chickens (line 413) of in vivo administration of E-64-c
(N-[N-(L-3-transcarboxyoxirane-2-carbonyl)-L-leucyl]-3-methyl-
butylamine). J . Biochem. 1981, 90, 893-896. (b) Drugs Future
1986, 11, 927-930.
(19) Tamai, M.; Matsumoto, K.; Ohmura, S.; Koyama, I.; Ozawa, Y.;
Hanada, K. In vitro and in vivo inhibition of cysteine proteinases
by EST, a new analogue of E-64. J . Pharmacobio-Dyn. 1986, 9,
672-677.
(20) Varughese, K.; Ahmed, F.; Carey, P.; Hasnain, S.; Huber, C.;
Storer, A. Crystal structure of a papain-E-64 complex. Biochem-
istry 1989, 28, 1330-1332.
(21) Brocklehurst, K.; Willenbrock, F.; Salih, E. Cysteine proteinases
In Hydrolytic Enzymes, New Comprehensive Biochemistry; Neu-
berger, A., Brocklehurst, K., Eds.; Elsevier: Amsterdam, New
York, 1987; Vol. 16, pp 39-158.
(22) Barrett, A.; Kembhavi, A.; Brown, M.; Kirschke, H.; Knight, C.;
Tamai, M.; Hanada, K. L-trans-Epoxysuccinyl-leucylamido(4-
guanidino)butane (E-64) and its analogues as inhibitors of
cysteine proteinases including cathepsins B, H and L. Biochem.
J . 1982, 201, 189-198.
(23) Schaschke, M.; Assfalg-Machleidt, I.; Machleidt, W.; Moroder,
L. Bis-peptidyl-derivatives of trans-epoxysuccininc acid as in-
hibitors of cathepsins. Poster presentation at the International
Conference on Protease Inhibitors (ICPI), Kyoto, J apan, 1997.
(24) Gour-Salin, B.; Lachance, P.; Plouffe, C.; Storer, A.; Me´nard, R.
Epoxysuccinyl dipeptides as selective inhibitors of cathepsin B.
J . Med. Chem. 1993, 36, 720-725.
(25) Turk, D.; Podobnik, M.; Popovic, T.; Katunuma, N.; Bode, W.;
Huber, R.; Turk, V. Crystal structure of cathepsin B inhibited
with CA030 at 2.0-Angstrom resolution: A basis for the design
of specific epoxysuccinyl inhibitors. Biochemistry 1995, 34,
4791-4794.
(26) Hata, Y.; Watanabe, M. Reaction of aziridine-carboxylic acids
with thiols in aqueous solution. The formation of â-amino acids.
Tetrahedron 1987, 43, 3881-3888.
(27) Martichonok, V.; Plouffe, C.; Storer, A.; Me´nard, R.; J ones, J .
B. Aziridine analogues of [[trans-(epoxysuccinyl)-L-leucyl]amino]-
4-guanidinobutane (E-64) as inhibitors of cysteine proteases. J .
Med. Chem. 1995, 38, 3078-3085.
(28) Tsubotani, S.; Takizawa, M.; Shirasaki, M.; Mizoguchi, J .;
Shimizu, Y. (Takeda Chem. Industries Ltd.) Aziridine deriva-
tives, their production and use. WO 95/28416, 1995.
(29) Tsuboya, S.; Takizawa, M.; Hattori, M.; Shirasaki, M. (Takeda
Chem. Industries Ltd.) Preparation of aziridine dicarboxylic acid
derivatives as cathepsin L inhibitors. J P09221470, 1996.
(30) Woo, J .-T.; Ono, H.; Tsuji. Biosci. Biotechnol. Biochem. 1995,
59, 350-352.
(31) Schirmeister, T. Aziridine-2,3-dicarboxylic acid derivatives as
inhibitors of papain. Arch. Pharm. Pharm. Med. Chem. 1997,
329, 239-244.
(32) Legters, J .; Thijs, L.; Zwanenburg, B. Synthesis of naturally
occurring (2S,3S)-(+)-aziridine-2,3-dicarboxylic acid. Tetrahedron
1991, 47, 5287-5294.
(33) Mori, K.; Iwasawa, H. Stereoselective synthesis of optically active
forms of δ-multistriatin, the attractant for european populations
of the smaller european elm bark beetle. Tetrahedron 1980, 36,
87-90.
(34) Corey, E.; Lansbury, P., J r. Stereochemical course of 5-lipoxy-
genation of arachidonate by rat basophil leukemic cell (RBL-1)
and potato enzymes. J . Am. Chem. Soc. 1983, 105, 4093-4094.
Ack n ow led gm en t. Financial support of this work
by the Deutschen Forschungsgemeinschaft DFG and the
Faculty of Chemistry and Pharmacy, University of
Freiburg, Germany, is gratefully acknowledged. I would
like to thank Prof. J . Schultz, Department of Pharma-
ceutical Chemistry, University of Tuebingen, Germany,
for a gift of cathepsin L; Dr. Beck and Dr. J endralla,
Fa. Hoechst AG, Frankfurt, Germany, for chromato-
graphic purification of compounds 10a ,b, respectively,
in large scale; and Dr. Waidelich, Perseptive Biosys-
tems, Wiesbaden, Germany, for HR mass spectroscopy
of several compounds.
Refer en ces
(1) Schirmeister, T.; Otto, H.-H. Cysteine proteases and their
inhibitors. Chem. Rev. 1997, 97, 133-171.
(2) Drenth, J .; J ansonius, J . N.; Koekoeck, R.; Wolters, B. G. The
structure of papain. In The Enzymes; Boyer, P. D., Ed.; Academic
Press: New York, 1971; Vol. 3, pp 485-499.
(3) Rosenthal, P.; McKerrow, J .; Aikawa, M.; Nagasawa, H.; Leech,
J . A malarial cysteine proteinase is necessary for hemoglobin
degradation by Plasmodium falciparum. J . Clin. Invest. 1988,
82, 1560-1566.
(4) Rawlings, N. D.; Barrett, A. J . Families of cysteine peptidases.
Methods Enzymol. 1994, 244, 461-486.
(5) Murachi, T. Calpain and calpastatin. Trends Biochem. Sci. 1983,
167-169.
(6) Kawashima, S.; Nakamura, M.; Hayashi, M. Activities of
calcium-activated neutral proteases and its endogenous inhibitor
in skeletal muscle of dystrophic hamster. Biol. Chem. Hoppe-
Seyler 1990, 371 (Suppl.), 205-210.
(7) Drake, F. H.; Dodds, R. A.; J ames, I. E.; Connor, J . R.; Debouck,
C.; Richardson, S.; Lee-Rykaczewski, E.; Coleman, L.; Rieman,
D.; Barthlow, R.; Hastings, G.; Gowen, M. Cathepsin K, but not
cathepsins B, L, or S, is abundantly expressed in human
osteoclasts. J . Biol. Chem. 1996, 271, 12511-12516.
(8) Calkins, C.; Sloane, B. Mammalian cysteine protease inhibi-
tors: Biochemical properties and possible roles in tumor pro-
gression. Biol. Chem. Hoppe-Seyler 1995, 376, 71-80.
(9) Bolli, R.; Cannon, R.; Speir, E.; Goldstein, R.; Epstein, S. Role
of cellular proteinases in acute myocardial infarction. I. Pro-
teolysis in nonischemic and ischemic rat myocardium and the
effects of antipain, leupeptin, pepstatin, and chymostatin ad-
ministered in vivo. J . Am. Coll. Cardiol. 1983, 2, 681-688.

572 J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 4
Schirmeister
(35) Bodanszky, M.; Bodanszky, A. The Practice of Peptides Synthesis;
Springer: Berlin, 1994; pp 93-94.
(36) Belleau, B.; Malek, G. A new convenient reagent for peptide
synthesis. J . Am. Chem. Soc. 1968, 90, 1651-1652.
(37) Krohn, K.; Wolf, U. Chemie der Heterocyclen; Teubner: Stuttgart,
1994; pp 25-26.
(59) Christensen, U.; Muellertz, S. Mechanism of reaction of human
plasmin with R-N-Benzoyl-L-arginine-p-nitroanilide (Titration of
the enzyme). Biochim. Biophys. Acta 1974, 334, 187-190.
(60) Petkov, D. Biochim. Biophys. Acta 1978, 527, 131-135.
(61) Feinstein, G.; Kupfer, A.; Sokolovsky, M. N-Acetyl-(L-Ala)3-p-
nitroanilide as a new chromogenic substrate for elastase. Bio-
chem. Biophys. Res. Commun. 1973, 50, 1020-1026.
(62) Feder, J . A spectrophotometric assay for neutral protease.
Biochem. Biophys. Res. Commun. 1968, 32, 326-332.
(63) J ackson, W.; Schlamowitz, M.; Shaw, A. Kinetics of the pepsin-
catalyzed hydrolysis of N-acetyl-L-phenylalanyl-L-diiodotyrosine.
Biochemistry 1965, 4, 1537-1543.
(64) Tian, W.-X.; Tsou, C.-L. Determination of the rate constant of
enzyme modification by measuring the substrate reaction in the
presence of the modifier. Biochemistry 1982, 21, 1028-1032.
(65) Kitz, R.; Wilson, I. Esters of methanesulfonic acid as irreversible
inhibitors of acetylcholinesterase. J . Biol. Chem. 1962, 237,
3245-3249.
(66) Sasaki, T.; Kishi, M.; Saito, M.; Tanaka, T.; Higuchi, N.;
Kominami, E.; Katunuma, N.; Murachi, T. Inhibitory effect of
di- and tripeptidyl aldehydes on calpains and cathepsins. J .
Enzyme Inhib. 1990, 3, 195-201.
(38) Ohno, M.; Otsuka, M. Chiral synthons by ester hydrolysis
catalyzed by Pig Liver Esterase. Org. React. 1989, 37, 1-55.
(39) All hydrolysis assays were carried out in a shaking water bath
at 32 °C with 150 mg of 17a (0.34 mmol) in 15 mL of phosphate
buffer, pH 8, 0.05 M, containing 10% acetonitrile. The following
amounts of enzyme were used: PLE 200 µL, CCL 500 mg, PPL
500 mg, trypsin 100 mg, chymotrypsin 100 mg, LPR 100 mg,
LRA 50 mg, LAN 100 mg. Assays with PLE were additionally
carried out at pH 7. Assays without enzyme were carried out at
pH 7 and pH 8. All assays were run over a period of 24 h and
were controlled by TLC (cyclohexane/ethyl acetate, 1/1) and pH
measurement. If hydrolysis took place, the pH was adjusted to
the original value (pH 7 and pH 8, respectively) by adding 0.1
N NaOH (pH-stat conditions, SM-Titrino 702 Metrohm). Work-
up was as follows: extraction with ethyl acetate (3 × 10 mL) at
pH 8 (pH 7), adjustment of the water layer to pH 2, extraction
with ethyl acetate (3 × 10 mL) at pH 2. The organic layers were
dried with MgSO4 and evaporated in vacuo. Without enzymes
no hydrolysis of 17a took place.
(40) Dolle, R.; Hoyer, D.; Prasad, C.; Schmidt, S.; Helaszek, C.; Miller,
R.; Ator, M. P1 Aspartate-based peptide R-((2,6-dichlorobenzyl)-
oxy)methyl ketones as potent time-dependent inhibitors of
Interleukin-1â-Converting Enzyme. J . Med. Chem. 1994, 37,
563-564.
(41) Meara, J . P.; Rich, D. H. Mechanistic studies on the inactivation
of papain by epoxysuccinyl inhibitors. J . Med. Chem. 1996, 39,
3357-3366.
(42) Gilchrist, T. Heterocyclenchemie; VCH: Weinheim, 1995; pp
355-357.
(43) Parkes, C.; Kembhavi, A.; Barrett, A. Calpain inhibition by
peptide epoxides. Biochem. J . 1985, 230, 509-516.
(44) Renold, P.; Tamm, C. Stereoselective hydrolysis of dimethyl
aziridin-2,3-dicarboxylates with pig liver esterase (PLE). Tet-
rahedron Asymmetry 1993, 4, 2295-2298.
(45) Bucciarelli, M.; Forni, A.; Moretti, I.; Prati, F.; Torre, G. Candida
cylindracea lipase-catalyzed hydrolysis of methyl aziridine-2-
carboxylates and -2,3-dicarboxylates. Tetrahedron Asymmetry
1993, 4, 903-906.
(46) Feder, J .; Schuck, J . M. Studies on the Bacillus subtilis Neutral-
Protease- and Bacillus thermoproteolyticus Thermolysin-cata-
lyzed hydrolysis of dipeptide substrates. Biochemistry 1970, 9,
2784-2791.
(67) GraFit, version 3.0; Erithacus Software Ltd.: London, 1992.
(68) Zaderenko, P.; Lo´pez, M. C.; Ballesteros, P. Addition of azoles
and amines to unsymmetrical fumaric esters. J . Org. Chem.
1996, 61, 6825-6828.
(69) Furukawa, N.; Oae, S. The Michael type addition of free
sulfilimine. Synthesis 1976, 1, 30-32.
(70) Main product (2.5 g, 50%) of this reaction is a 2/1 mixture of
enamines 31 and 32, isolated by column chromatography (cy-
clohexane/ethyl acetate, 5/1, Rf ) 0.5). IR (ethyl acetate): 3485
(br), 3350 (br), 3034 (br), 1729, 1678, 1621, 1556, 1498, 1455
cm-1 1H NMR (CDCl3, 50 °C): δ 1.3 (t, J ) 7.1 Hz, 32) and t
.
(1.35, J ) 7.1 Hz, 31) (together 3 H), 4.2 (q, J ) 7.1 Hz, 32) and
4.35 (q, J ) 7.1 Hz, 31) (together 2 H), 5.2 (s, 31) and 5.3 (s, 32)
(together 2 H), 5.58 (s, 32) and 5.61 (s, 31) (together 1 H), 6.5
(bs, 2 H), 7.2-7.45 (m, 5 H). 13C NMR (CDCl3): δ 14.23 (31),
14.63 (32) (CH3), 59.78 (32), 62.65 (31) (OCH2), 65.62 (31), 68.20
(32) (OCH2Ph), 88.61 (31), 89.45 (32) (HC)C), 128.25, 128.41,
128.60, 128.76, 128.89, 128.93, 135.10, 136.82, 146.38 (32),
147.03 (31) (HC)C), 163.80 (31), 163.84 (32), 169.86 (31), 170.05
(32). Anal. (C13H15NO4) C,H,N. Assignment of NMR signals was
carried out by INEPT long-range NMR spectroscopy.
(47) Fujishima, A.; Imai, Y.; Nomura, T.; Fujisawa, Y.; Yamamoto,
Y.; Sugawara, T. The crystal structure of human cathepsin L
complexed with E-64. FEBS Lett. 1997, 407, 47-50.
(48) Drenth, J .; Kalk, H.; Swen, H. Binding of chloromethyl ketone
substrate analogues to crystalline papain. Biochemistry 1976,
15, 3731-3738.
(71) A k2nd ) 149 ( 19 M-1 min-1 was found for inhibition of papain
by N-ethylmaleimide (Anderson, B.; Vasini, E. Biochemistry
1970, 9, 2248-3352: k2nd ) 162 M-1 min-1).
(49) Albeck, A.; Fluss, S.; Persky, R. Peptidyl epoxides: Novel
selevtive inactivators of cysteine proteases. J . Am. Chem. Soc.
1996, 118, 3591-3596.
(72) A k2nd ) 3.4 ( 0.5 × 107 M-1 min-1 was found for inhibition of
(50) Claus, P. Azomethinylide. In Methoden Org. Chem. (Houben-
Weyl); Klamann, D., Hagemann, H., Eds.; Thieme: Stuttgart,
1990; Vol. E14b, Part 1, pp 88-91.
papain by E-64 (ref 1: 2.2-3.8 × 107 M-1 min-1).
- pKa1
(73) The following equation was used:67 k2nd ) (limit × 10(pH
))/
a1-pka2
(10(2pH-pk
) + 10(pH-pk ) + 1). The following values were found
a1
(51) Garavito, R. M.; Rossmann, M. G.; Argos, P.; Eventoff, W.
Convergence of active center geometries. Biochemistry 1977, 16,
5065-5071.
for 11a : pKa1 ) 4.8 ( 0.2, pKa2 ) 3.8 ( 0.2, limit ) 6043 M-1
s-1
.
(74) Demuth, H. Recent developments in inhibiting cysteine and
serine proteases. J . Enzyme Inhib. 1990, 3, 249-278.
(75) Ho, T.-L. The hard soft acids bases (HSAB) principle and organic
chemistry. Chem. Rev. 1975, 75, 1-20.
(52) Bodanszky, M.; Bodanszky, A. The Practice of Peptide Synthesis,
2nd ed.; Springer: Heidelberg, 1994; pp 104-105.
(53) Mole, J . E.; Horton, H. R. Kinetics of papain-catalyzed hydrolysis
of R-N-benzoyl-L-arginine-p-nitroanilide. Biochemistry 1973, 12,
816-822.
(76) Structures have been minimized using the Systematic Search
program from Sybyl 6.4 (Tripos Associates Inc., St. Louis, MO,
1997). Energy minimization: Tripos force field, convergence
criterion 0.001 kcal mol-1. The superposition has been performed
using the program Multifit from Sybyl 6.4.
(77) The docking of 26a into the active site of papain has been
performed using the program FlexiDock from the Sybyl 6.4
Biopolymer Module (Tripos Associates Inc., St. Louis, MO).
Energy minimization: Tripos force field, Gasteiger-Hueckel
charges on ligand, Kollmann charges on protein. As FlexiDock
pocket a region of 0.4 nm around the amino acids Cys25, Tyr67,
Pro68, Trp69, Val133, Val157, His159, Ala160, and Phe207 of
papain has been defined. The structure of papain has been taken
from the Brookhaven Protein Databank (entry 1pe6).
(54) Barrett, A. Biochem. J . 1980, 187, 909-913.
(55) Barrett, A.; Kirschke, H. Cathepsin B, cathepsin H, and cathe-
psin L. Methods Enzymol. 1981, 80, 535-561.
(56) Vo¨lkel, H.; Kurz, U.; Linder, J .; Klumpp, S.; Gnau, V.; J ung, G.;
Schultz, J . Cathepsin L is an intracellular and extracellular
protease in Paramecium tetraurelia. Purification, cloning, se-
quencing and specific inhibition by its expressed propeptide. Eur.
J . Biochem. 1996, 238, 198-206.
(57) Gour-Salin, B.; Lachance, P.; Bonneau, P.; Storer, A.; Kirschke,
H.; Broemme, D. E-64 Analogues as inhibitors of cathepsin L
and cathepsin S: Importance of the S2-P2 interactions for
potency and selectivity. Bioorg. Chem. 1994, 22, 227-241.
(58) Sasaki, T.; Kikuchi, T.; Yumoto, N.; Yoshimura, N.; Murachi,
T. Comparative specificity and kinetic stucies on porcine calpain
I and calpain II with naturally occurring peptides and synthetic
fluorogenic substrates. J . Biol. Chem. 1984, 259, 12489-12494.
J M981061Z