Aza-peptide epoxides: A new class of inhibitors selective for clan CD cysteine proteases

Article abstract of DOI:10.1021/jm025581c

Aza-peptide epoxides, a new class of irreversible protease inhibitors, are specific for the clan CD cysteine proteases. The inhibitors have second-order rate constants up to 10⁵ M^-1 s^-1, with the most potent epoxides having the S,S stereochemistry. The aza-Asn derivatives are effective legumain inhibitors, while the aza-Asp epoxides were specific for caspases. The inhibitors have little or no inhibition with other proteases such as chymotrypsin, papain, or cathepsin B.

Full text of DOI:10.1021/jm025581c

4958
J . Med. Chem. 2002, 45, 4958-4960
Aza -P ep tid e Ep oxid es: A New Cla ss of
In h ibitor s Selective for Cla n CD Cystein e
P r otea ses
J uliana L. Asgian,† Karen Ellis J ames,†
Zhao Zhao Li,† Wendy Carter,‡ Alan J . Barrett,‡
J owita Mikolajczyk,§ Guy S. Salvesen,§ and
J ames C. Powers*,†
F igu r e 1. Structure of the natural product E-64.
School of Chemistry and Biochemistry, Georgia Institute of
Technology, Atlanta, Georgia 30332-0400, MRC Molecular
Enzymology Laboratory, The Babraham Institute,
Babraham, Cambridgeshire CB2 4AT, U.K., and
Program in Apoptosis and Cell Death Research,
The Burnham Institute, 10901 North Torrey Pines Road,
La J olla, California 92037
F igu r e 2. Aza-peptide epoxide design.
Received September 10, 2002
Abstr a ct: Aza-peptide epoxides, a new class of irreversible
protease inhibitors, are specific for the clan CD cysteine
proteases. The inhibitors have second-order rate constants up
to 105 M-1 s-1, with the most potent epoxides having the S,S
stereochemistry. The aza-Asn derivatives are effective legu-
main inhibitors, while the aza-Asp epoxides were specific for
caspases. The inhibitors have little or no inhibition with other
proteases such as chymotrypsin, papain, or cathepsin B.
F igu r e 3. Synthesis of aza-peptide epoxides. Reagents are
the following: (i) BrCH2COOEt, NMM, DMF, NH3/MeOH, 0.1
equiv of NaCN, DMF; (ii) BrCH2COO-tBu, NMM DMF; (iii) 3,
EDC, HOBt, DMF or NMM, IBCF, DMF; (iv) TFA (X ) O-tBu).
Cysteine proteases have been grouped into families
and clans by Barrett and Rawlings.1-3 Most cysteine
proteases are members of the papain clan (clan CA),
which consists of enzymes such as calpains, cathepsins,
and papain. Clan CD is smaller but has several impor-
tant enzymes, including caspases, legumain, gingipain,
clostripain, and separase. The caspases, cysteine aspar-
tate specific proteases, are a family of >15 members that
are involved in apoptosis (programmed cell death).
Caspases are associated with inflammatory disorders
(IL-1â processing) and neurodegenerative diseases such
as stroke, ALS, Alzheimer’s, and Parkinson’s. Gingipain,
from Porphyromonas gingivalis, causes tissue damage
in periodontal disease, legumain is associated with
antigen processing and immune disorders, while clos-
tripain is involved in bacterial infections. Thus, the
design of new specific inhibitors for clan CD cysteine
proteases could lead to the development of potential
drugs.
A variety of other epoxide inhibitors have been
reported in the literature as irreversible inhibitors of
cysteine proteases. Epoxysuccinate derivatives related
to the natural product E-64 (Figure 1), first isolated
from Aspergillus japonicus,4 have been developed as
highly reactive inhibitors for many clan CA cysteine
proteases including papain, cruzain, and cathepsin B.5-8
Subsequently, aminoacyl epoxides and R,â-epoxyketones
have been developed as inhibitors for serine, threonine,
and cysteine proteases. Unfortunately, E-64 derivatives
are ineffective inhibitors of clan CD proteases.3
Aza-peptide epoxides are a new class of protease
inhibitors, which are highly specific for cysteine pro-
teases of clan CD. Aza-peptide epoxides were designed
to closely resemble an extended peptide substrate (1,
Figure 2) with the placement the carbonyl group of the
epoxide moiety in a location identical to that of the
carbonyl of the scissile peptide bond in a substrate (2,
Figure 2). This design allowed the peptide chain of the
inhibitor to exactly match that of a good substrate up
to the scissile peptide carbonyl group. Conversion of the
R-carbon of an amino acid residue into a nitrogen results
in the formation of an aza-peptide,9 which allows the
ready synthesis of a variety of derivatives. We also
hypothesized that the aza-peptide epoxide binding mode
would be opposite that of E-64 and E-64 derivatives,
which bind in the reverse direction, relative to the
substrate binding mode, in papain and cathepsin B. In
addition, we hypothesized that the epoxysuccinate
derivatives of aza-peptides would be more reactive than
simple R-aminoacyl epoxides.
Aza-peptide epoxides were synthesized from peptidyl
hydrazides (3, Figure 3). Introduction of the aza-amino
acid side chain can be accomplished either by reductive
amination with the appropriate aldehyde or by alkyla-
tion. The aza-Asn derivatives (6b) were obtained by
alkylation of 3 with ethyl bromoacetate to form 4a ,
followed by ammonolysis to the amide (4b). The Asp side
chain was introduced by alkylation of 3 with tert-butyl
bromoacetate to form the substituted hydrazide (4c).
The substituted hydrazides (4b and 4c) were coupled
to the respective epoxysuccinate moiety (5) using EDC
* To whom correspondence should be addressed. E-mail:
james.powers@chemistry.gatech.edu. Phone: 404-894-4038. Fax: 404-
894-2295.
† Georgia Institute of Technology.
‡ The Babraham Institute.
§ The Burnham Institute.
10.1021/jm025581c CCC: $22.00 © 2002 American Chemical Society
Published on Web 10/15/2002

Letters
J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 23 4959
Ta ble 1. Second-Order Rate Constants for the Inhibition of Cysteine Proteases by Aza-peptide Epoxides
kobs/[I] c (M-1 s-1
caspase-1 caspase-3 caspase-6 caspase-8 legumain papain cathepsin B
)
inhibitora
EPb
9
PhPr-Val-Ala-AAsp-EP-COOCH2Ph trans
36200
6130
58
<10
52
<10
7100
<10
7720
3050
<10
NI
63
<10
6600
<10
8620
1200
<10
NI
21
NI
46
NI
36
<10
NI
NI
NI
NI
<10
NI
NI
NI
NI
NI
<10
NI
10 PhPr-Val-Ala-AAsp-EP-CH2CH2Ph
11 Cbz-Asp-Glu-Val-AAsp-EP-COOEt
12
13 Cbz-Leu-Glu-Thr-AAsp-EP-COOEt
14
15 Cbz-Ala-Ala-AAsn-EP-COOEt
16
trans
S,S
R,R
S,S
R,R
S,S
R,R
cis
205300
30560
<10
214
<10
15
NI
43000
25200
142
NI
<10
<10
NI
13
21
17
NI
<1
NI
18 Cbz-Leu-Leu-ALeu-EP-COOEt
19
20
S,S
R,R
trans
NI
NI
NI
NI
NI
NI
NI
NI
35
NI
a
b
PhPr ) Ph-CH2-CH2-CO-; EP ) epoxide; AAsp ) aza-Asp; AAsn ) aza-Asn; Aleu ) aza-Leu; Cbz ) Ph-CH2-CO-. The trans
epoxide is a mixture of S,S and R,R, and the cis epoxide is a mixture of R,S and S,R. c Reference 18. NI ) no inhibition.
and HOBt, the mixed anhydride method, or the penta-
fluorophenol method to give the desired aza-peptide
epoxides (6b, 6c, 7c, and 8c). The blocked aza-Asp
derivatives were deprotected using TFA to yield 6d , 7d ,
and 8d .
The Asp aza-peptide inhibitor sequences are based on
sequences derived from caspase substrate cleavage sites
F igu r e 4. Mechanism of inhibition of caspases by aza-peptide
or obtained by peptide mapping of caspases with librar-
ies of AMC substrates (Table 1). The VAD sequence for
caspase-1 is the cleavage sequence in IL-1â.10 The
DEVD and LETD sequences are optimal sequences for
caspase-3 and caspase-8, respectively.11 The VAD aza-
peptide epoxides, 9 and 10, are effective inhibitors of
caspase-1, with the epoxysuccinate derivative 9 having
a kobs/[I] value of 36 200 M-1 s-1. The DEVD aza-peptide
inhibitors 11 and 12 were potent inhibitors of caspase-3
(kobs/[I] for 11 is 205 300 M-1 s-1), while the LETD
epoxides 13 and 14 were potent caspase-6 and -8
inhibitors. The Asn derivatives 15 and 16 were effective
and specific legumain inhibitors.12
The inhibitors show little cross reactivity with other
enzymes in clan CD. The epoxides designed for caspases
do not inhibit or poorly inhibit other caspases and
legumain. The selectivity of 9 for caspase-1 was >574-
fold compared to the other enzymes tested. Similarly,
compound 13 was >210-fold more reactive with caspase-6
and -8 relative to the other caspases. The legumain
inhibitors, 15 and 16, had essentially no reactivity
toward caspases.
Aza-peptide epoxides were specifically designed with
the appropriate sequences for clan CA proteases such
as papain, cathepsin B, and calpains. Only one se-
quence, 18, is shown in Table 1. This inhibitor and
similar structures (not shown) were tested for their
ability to inhibit clan CA cysteine proteases. They
showed little to no inhibition of papain, cathepsin B,
and chymotrypsin. Compound 18, which was designed
as a proteasome inhibitor,13 does not inhibit the pro-
teasome or caspases and poorly inhibits papain and
cathepsin B. Clearly, this design is specific for clan CD,
since aza-peptide epoxides do not work well with clan
CA proteases.
The stereochemistry at the epoxide moiety plays an
important role in the potency of the inhibitor. The order
of reactivity for the epoxide stereochemistry is S,S >
R,R > trans > cis. Stereochemistry is very important
with the caspase inhibitors 11 and 12 where the S,S
isomer is 6.7-fold more reactive with caspase-3 and over
epoxides.
600-fold more reactive with caspase-6 and caspase-8.
Stereochemistry is less significant (2- to 8-fold differ-
ence) with the caspase-6, caspase-8 (13 and 14), and
legumain inhibitors (15 and 16). The more potent form
of E-64 also possesses S,S stereochemistry at the
epoxide moiety, while the R,R isomer is less reactive
toward cysteine proteases.5
X-ray crystal structures of caspase-1 inhibited by
PhPr-Val-Ala-AAsp-EP-COOCH2Ph and PhPr-Val-Ala-
AAsp-EP-CH2CH2Ph have been determined (Ron Rubin,
unpublished results). Nucleophilic attack by the active
site Cys 285 of caspase-1 occurs at the C-2 position of
the epoxide ring (Figure 4, R′ ) COOBzl or CH2CH2-
Ph). This site of attack was very surprising because we
expected the point of attack to be the C-3 position. In
the case of transition-state inhibitors such as peptide
aldehydes, the active site cysteine adds to the aldehyde
carbonyl group that is equivalent to the scissile peptide
carbonyl group in a substrate. Clearly the active site
cysteine can attack functional groups quite distant from
the side chain of the P1 aza-amino acid residue.
The selectivity of aza-peptide epoxides for clan CD
cysteine proteases is a novel feature of these inhibitors.
Other epoxide inhibitors, such as E-64 derivatives, are
highly reactive for clan CA cysteine proteases and poorly
reactive or nonreactive with clan CD cysteine proteases.
Enzymes in clan CD and clan CA have different
polypeptide folds and varying active site topologies.14-17
The caspase substrate binding region is quite open and
shallow with several binding pockets, while papain has
a canyon-like binding site. The more open active site in
the caspases along with a strict specificity at P1 residue
allows the aza-Asp residue to bind in the S1 pocket and
positions the epoxide close to His 237 and Cys 285 where
covalent bond formation can occur. In papain the active
site is more constricted (crowded) and specificity is
determined by interaction with the S2 subsite. A loop
structure on the wall of the canyon, which contains the
active site histidine, restricts possible binding modes of

4960 J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 23
Letters
(6) Parkes, C.; Kembhavi, A. A.; Barrett, A. J . Calpain inhibition
by peptide epoxides. Biochem. J . 1985, 230, 509-516.
(7) Roush, W. R.; Hernandez, A. A.; McKerrow, J . H.; Selzer, P. M.;
Hansell, E.; Engel, J . C. Design, synthesis and evaluation of
D-homophenylalanyl epoxysuccinate inhibitors of the trypano-
somal cysteine protease cruzain. Tetrahedron 2000, 56, 9747-
9762.
(8) Schaschke, N.; Assfalg-Machleidt, I.; Machleidt, W.; Turk, D.;
Moroder, L. E-64 analogues as inhibitors of cathepsin B. On the
role of the absolute configuration of the epoxysuccinyl group.
Bioorg. Med. Chem. 1997, 5, 1789-1797.
(9) Nomenclature of aza-peptides: AAsp ) aza-Asp; AAsn ) aza-
Asn; ALeu ) aza-Leu; EP ) epoxide/oxirane -C2H2O-.
(10) Thornberry, N. A.; Peterson, E. P.; Zhao, J . J .; Howard, A. D.;
Griffin, P. R.; Chapman, K. T. Inactivation of interleukin-1 beta
converting enzyme by peptide (acyloxy)methyl ketones. Bio-
chemistry 1994, 33, 3934-3940.
(11) Thornberry, N. A. Caspases: key mediators of apoptosis. Chem.
Biol. 1998, 5, R97-R103.
(12) Chen, J .-M.; Dando, P. M.; Rawlings, N. D.; Brown, M. A.; Young,
N. E.; Stevens, R. A.; Hewitt, E.; Watts, C.; Barrett, A. J .
Cloning, isolation, and characterization of mammalian legumain,
an asparaginyl endopeptidase. J . Biol. Chem. 1997, 272, 8090-
8098.
(13) Bogyo, M.; McMaster, J . S.; Gaczynska, M.; Tortorella, D.;
Goldberg, A. L.; Ploegh, H. Covalent modification of the active
site threonine of proteasomal beta subunits and the Escherichia
coli homologue HslV by a new class of inhibitors. Proc. Natl.
Acad. Sci. U.S.A. 1997, 94, 6629-6634.
(14) Watt, W.; Koeplinger, K. A.; Mildner, A. M.; Heinrikson, R. L.;
Tomasselli, A. G.; Watenpaugh, K. D. The atomic-resolution
structure of human caspase-8, a key activator of apoptosis.
Struct. Fold. Des. 1999, 7, 1135-1143.
(15) Wilson, K. P.; Black, J . A. F.; Thomson, J . A.; Kim, E. E.; Griffith,
J . P.; Navia, M. A.; Murcko, M. A.; Chambers, S. P.; Aldape, R.
A.; Raybuck, S. A.; Livingston, D. J . Structure and mechanism
of interleukin-1-beta converting-enzyme. Nature 1994, 370, 270-
275.
(16) Rotonda, J .; Nicholson, D. W.; Fazil, K. M.; Gallant, M.; Gareau,
Y.; Labelle, M.; Peterson, E. P.; Rasper, D. M.; Ruel, R.;
Vaillancourt, J . P.; Thornberry, N. A.; Becker, J . W. The three-
dimensional structure of apopain/CPP32, a key mediator of
apoptosis. Nat. Struct. Biol. 1996, 3, 619-625.
(17) Schroder, E.; Phillips, C.; Garman, E.; Harlos, K.; Crawford, C.
X-ray crystallographic structure of a papain leupeptin complex.
FEBS Lett. 1993, 315, 38-42.
(18) Inhibitors were asayed with recombinant caspase-1 in a 100 mM
Hepes, 20% glycerol v/v, 0.5 mM EDTA, 5 mM DTT buffer with
Ac-YVAD-AMC as the substrate. Recombinant caspase-3, -6, and
-8 were assayed in buffer (50 mM Hepes, 100 mM NaCl, 0.1%
(w/v) CHAPS, sucrose 10% (w/v), and 10 mM DTT, at pH 7.4)
with Z-DEVD-AFC as the substrate. Legumain, purified from
pig kidney tissue, was assayed in buffer (39.5 mM citric acid,
121 mM Na2HPO4, pH 5.8, containing 1 mM EDTA, 1 mM
TCEP, 0.01% CHAPS) with Z-AAN-AMC as the substrate.
Papain was assayed in buffer (2.5 mM Hepes, 2.5 mM EDTA,
pH 7.5) with Z-FR-pNA as the substrate. Cathepsin B, from
human liver, was assayed in buffer (0.1 M K2PO3, 1.25 mM
EDTA, 0.01% Brij 35, pH 6.0) with Z-RR-AMC as the substrate.
For more detailed information, see Supporting Information.
inhibitors. We propose that the aza-peptide expoxides
cannot bind in a suitable orientation for effective
irreversible inhibition because of the inability of the
fairly rigid aza-peptide epoxysuccinate moiety to bind
properly near the catalytic residues of papain and other
clan CA cysteine proteases.
Aza-peptide epoxides have the advantage of being
easily extended in the P′ direction, allowing interactions
with the S′ subsites of the enzyme. Currently, we are
trying to refine the P′ portion of the inhibitors to obtain
greater specificity. We are also extending this class of
inhibitors to gingipain and clostripain.
Ack n ow led gm en t. This work was supported by
grants from the National Institute of General Medical
Sciences (Grants GM54401 and GM61964) and in part
by a grant from BASF Biotechnology (now part of
Abbott) to J .C.P., and a NSF grant (Grant NS37878) to
G.S.S. J .L.A. acknowledges a fellowship from the Mo-
lecular Design Institute under prime contract from the
Office of Naval Research. K.E.J . acknowledges a fellow-
ship from the Center for the Study of Women, Science,
and Technology (WST) at Georgia Tech. We thank Dr.
Ken Brady (BASF) for the caspase-1 assay, and Dr. M.
Orlowski at Mount Sinai School of Medicine, NY, for
the proteasome assays.
Su p p or tin g In for m a tion Ava ila ble: Synthetic proce-
dures, final product characterizations, and enzyme assays.
This material is available free of charge via the Internet at
http://pubs.acs.org.
Refer en ces
(1) Barrett, A. J .; Rawlings, N. D.; Woessner, J . F. Handbook of
Proteolytic Enzymes; Academic Press: New York, 1998.
(2) Barrett, A. J .; Rawlings, N. D. Evolutionary lines of cysteine
peptidases. Biol. Chem. 2001, 382, 727-733.
(3) Chen, J . M.; Rawlings, N. D.; Stevens, R. A.; Barrett, A. J .
Identification of the active site of legumain links it to caspases,
clostripain and gingipains in a new clan of cysteine endopepti-
dases. FEBS Lett. 1998, 441, 361-365.
(4) Hanada, K.; Tamai, M.; Yamagishi, M.; Ohmura, S.; Sawada,
J .; Tanaka, I. Studies on thiol protease inhibitors. Part I.
Isolation and characterization of E-64, a new thiol protease
inhibitor. Agric. Biol. Chem. 1978, 42, 523-528.
(5) Barrett, A. J .; Kembhavi, A. A.; Brown, M. A.; Kirschke, H.;
Knight, C. G.; 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.
J M025581C