Calpain inhibitors based on the quiescent affinity label concept: High rates of calpain inactivation with leaving groups derived from N-hydroxy peptide coupling reagents

Article abstract of DOI:10.1016/S0960-894X(00)00451-0

A series of irreversible inhibitors of recombinant calpain has been synthesized and their rates of inactivation have been evaluated against calpain and cathepsin B, respectively. The design of the inhibitors was based on the quiescent affinity label concept. By choosing the appropriate affinity group and by employing leaving groups derived from N-hydroxy coupling reagents, good inhibitors of calpain with high rates of inactivation have been identified. However, poor aqueous stability limits their therapeutic utility. (C) 2000 Elsevier Science Ltd.

Full text of DOI:10.1016/S0960-894X(00)00451-0

Bioorganic & Medicinal Chemistry Letters 10 (2000) 2315±2319
Calpain Inhibitors Based on the Quiescent Anity Label Concept:
High Rates of Calpain Inactivation with Leaving Groups Derived
from N-Hydroxy Peptide Coupling Reagents
Rabindranath Tripathy,* Mark A. Ator and John P. Mallamo
Cephalon, Inc., 145 Brandywine Parkway, West Chester, PA 19380, USA
Received 26 May 2000; accepted 4 August 2000
AbstractÐA series of irreversible inhibitors of recombinant calpain has been synthesized and their rates of inactivation have been
evaluated against calpain and cathepsin B, respectively. The design of the inhibitors was based on the quiescent anity label con-
cept. By choosing the appropriate anity group and by employing leaving groups derived from N-hydroxy coupling reagents, good
inhibitors of calpain with high rates of inactivation have been identi®ed. However, poor aqueous stability limits their therapeutic
utility. # 2000 Elsevier Science Ltd. All rights reserved.
Introduction
label concept,5 which show high rates of inactivation of
recombinant calpain with unconventional leaving
groups derived from N-hydroxy peptide coupling
reagents.
The calpains are a family of cytosolic calcium-activated
cysteine proteases, which exist in mammalian cells as both
ubiquitous and tissue-speci®c forms. The best character-
ized members of the family are the ubiquitous enzymes
calpain I and calpain II, heterodimeric proteases that
contain di€erent catalytic subunits but a common reg-
ulatory subunit, and which can be distinguished bio-
chemically by the concentration of calcium required for
activation.1 The calpains have been postulated to be
involved in pathology of several diseases, suggesting
that calpain inhibitors might have clinical utility in a
variety of disorders.2 The design of inhibitors of calpain
is being pursued from many directions, which are cate-
gorized into two distinct classes. Chemical entities com-
prising peptidyl aldehydes and a-ketoamides are
reversible3 inhibitors, while ¯uoromethylketones, diazo-
methylketones, acyloxymethyl ketones, sulphoniummethyl
ketones, peptidyl epoxides (E-64), vinyl sulfones and
haloacylhydrazines are irreversible4 inhibitors of
cysteine proteases. The design of e€ective irreversible
inhibitors of calpain is challenging, as the enzyme is
inert towards many of the di€erent classes of known
cysteine protease inhibitors. Generally a relatively long
peptide sequence (typically a tripeptide) or a dipeptide
with a modi®ed capping group4g is required to attain
satisfactory rates of calpain inactivation Herein, we
report on calpain inhibitors utilizing the quiescent anity
We have used a dipeptidyl methyl ketone (1) that bears a
leaving group (X) at the a-position as our basic molecular
skeleton.6 The peptidyl segment serves as the anity
group, while the leaving group can be systematically
varied to explore and optimize the rate of enzyme inac-
tivation. Calpain becomes inactivated as its catalytic
thiol group forms a covalent bond to the inhibitor
through displacement of the leaving group via direct or
indirect attack at the carbon bearing the leaving group.5
Although Krantz and co-workers have shown that pep-
tidyl acyloxymethyl ketones work extremely well as
inhibitors of cathepsin B, they provide only a modest
degree of time-dependent inactivation of chicken
smooth muscle calpain II, even with an anity group
optimized for calpain.5f
In their paper,5f Krantz et al. discount the possibility that
the poor reactivity of calpain towards acyloxymethyl
*Corresponding author. Fax: +1-610-344-0065; e-mail: rtripath@
cephalon.com
0960-894X/00/$ - see front matter # 2000 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(00)00451-0

2316
R. Tripathy et al. / Bioorg. Med. Chem. Lett. 10 (2000) 2315±2319
ketones was due to the weak nucleophilicity of the cal-
pain active site thiolate group, since it is quite reactive
towards peptidyl ¯uoromethyl ketones. Intolerance of
the steric bulk of substituted benzoates or the absence of
a residue appropriately positioned in the calpain active
site to protonate the leaving group was postulated to
result in the lack of reactivity of calpain toward benzo-
ate-based inhibitors. We envisioned that steric incom-
patibility of benzoate leaving groups in the calpain
active site could be the principal problem. We chose to
explore this by choosing departing groups with inherent
conformational and binding7 properties di€erent from
the benzoates. Such properties might be helpful in
pointing the leaving group toward a more accom-
modating region of the enzyme or could place it in the
recognition domain of a proton-donating group in the
active site, which would facilitate its expulsion as a
neutral species. After considerable experimentation, we
identi®ed a novel class of leaving groups derived from
N-hydroxy peptide coupling reagents (Table 1). The
anions of N-hydroxy succinimides (HOSu) and benzo-
triazoles (HOBt) are not conventional leaving groups in
organic synthesis (i.e., to facilitate nucleophilic attack at
an isolated sp3 carbon). However, their role as leaving
groups in active esters during peptide synthesis are well
known. Moreover, like substituted benzoates they have
low pKa values8 (pKa value of HOBt is 4.3), which has
been postulated to correlate the rate of inactivation of
cathepsin B by peptidyl acyloxymethyl ketones.5
Assays for human liver cathepsin B (Calbiochem) were
performed by a similar procedure employing published
assay conditions.12
Results and Discussion
Table 1 records the second-order rate constants for inac-
tivation of calpain and cathepsin B. The inhibitors 1a±c
containing a succinimide or a substituted succinimide
analogue as a leaving group were weak inactivators, with
1a having the highest rate of inactivation. However, the
benzotriazoloxy 1d was an extremely rapid time-depen-
dent inactivator of calpain. Inhibitor 1d has one of the
highest rates of inactivation utilizing a dipeptide anity
group reported so far for inhibitors of calpain. For
comparison, 1e, a ¯uoromethyl ketone bearing the same
anity group shows a rate of calpain inactivation of
136,300 M 1 s
.
1 4g By modifying the capping group of 1e
or by using a tripeptide anity group, inhibitors with
1
high rates of calpain inactivation (up to 290,000 M 1 s
)
have been made.4g Other examples with high rates of cal-
pain inactivation are from a dipeptidyl methyl phosphate13
1
(365,000 M
(230,000 M
s
s
1), a tripeptidyl diazomethyl ketone4a
1) and a tripeptidyl sulfonium methyl
1
ketone5f (>200,000 M
s
1). The inactivation by 1d
1
appears to be irreversible, since no catalytic activity was
recovered following removal of inhibitor by repeated
ultra®ltration through a Centicon 30. Based on initial
velocities, a Ki of approximately 35 nM was determined
for 1d, indicating that it binds well to calpain.
Chemistry
Modi®cation of the P1 position of the inhibitor's pep-
tidyl anity group resulted in compounds 1f and 1g,
which display a decrease in inactivation rate analogous
to that observed with the ¯uoromethyl ketones4g corre-
sponding to 1d and 1g. Increasing the length of the a-
nity group to a tripeptide (1h) yielded an approximately
twofold increase in the rate of inactivation. Compounds
1i and 1j were prepared by incorporating the related
peptide coupling reagents 7- and 4-HOAt,14 respec-
tively. Both inhibitors maintain a respectable level of
activity toward calpain.
The synthesis of inhibitors was accomplished from pep-
tidyl bromomethyl ketones. A representative example is
shown in Scheme 1. The mixed anhydride of dipeptide
acid, Z-Leu-Phe-OH (2) was treated with diazomethane
to generate the diazoketone, which was decomposed in
the presence of HBr/HOAc to generate bromoketone 3.
Reaction of 3 and HOBt in DMF with silver(I) oxide or
KF as a base provides inhibitor 1d, which was puri®ed
by crystallization.9
It seems that the role of the N±O bond, presumably its
conformation,15 is extremely crucial for optimal calpain
inactivation rate. Replacement of the benzotriazoloxy
group with the corresponding benzotriazolyl (1k) pro-
vides evidence in that respect. Though benzotriazolyl
anion is considered as an extremely good leaving group
in organic synthesis,16 compound 1k was found to be a
poor inhibitor of calpain. From comparison of the
Biology
The inhibitory activity of the compounds 1a±j was
determined using recombinant human calpain I, pre-
pared as described by Meyer and co-workers.10 Second-
order rate constants for inactivation were determined by
analysis of the progress curves obtained in the presence
of substrate (Suc-Leu-Tyr-MNA)11 and inhibitor.
Scheme 1.

R. Tripathy et al. / Bioorg. Med. Chem. Lett. 10 (2000) 2315±2319
2317
results from 1d with 1k, we speculate that, apart from a
good degree of departing and binding7 potential for the
leaving groups and contributions from the peptidyl a-
nity group, the origin of high reaction rates for com-
pounds such as 1d could be ascribed to the role of the
N±O bond which is believed to play a pivotal role in
providing a stereoelectronically favorable orientation of
the leaving group in the calpain active site.
The selectivity of the benzotriazoloxy class of inhibitors
with respect to inhibition of cathepsin B is also shown in
Table 1. Although these compounds are good cathepsin B
inhibitors, the degree of selectivity can be in¯uenced by
the nature of the peptide anity group, as evidenced by 1d
versus 1f. The nature of the benzotriazoloxy group
appears to have a less signi®cant e€ect on selectivity, at
least for this limited set of analogues.
Table 1. Inhibition of calpain and cathepsin B by inhibitors 1a±k
Inhibitor
X
Y
R
Calpain I
kobs/[I](M 1 s
Cathepsin B
1
Calpain/cathepsin B
1
1
)
kobs/[I](M
s )
a
1a
Cbz
Benzyl
5800
1000
Ð
1b
1c
Cbz
CBz
Benzyl
Benzyl
6900
0.14
b
a
Ð
1d
1e
1f
Cbz
Cbz
CBz
Benzyl
Benzyl
i-Butyl
320,000
136,300c
175,000
118,000
300c
2.7
454.3
9.0
F
19,400
1g
1h
1i
Cbz
Cbz-Leu
Cbz
Ethyl
Benzyl
Benzyl
Benzyl
Benzyl
63,000
524,000
184,000
443,000
0.14
Ð
b
107,000
1.7
0.67
Ð
1j
Cbz
57,000
85,500
d
a
1k
Cbz
aNot determined.
b32% inhibition at 10 mm following 5 min incubation.
cSee ref 4g.
d83% inhibition at 10 mm following 5 min incubation.

2318
R. Tripathy et al. / Bioorg. Med. Chem. Lett. 10 (2000) 2315±2319
A limitation of the peptidyl benzotriazoloxy methyl
ketones is their modest stability in aqueous solution at
neutral pH.17 Estimates of stability were obtained by
preincubating inhibitor in assay bu€er for various times.
Progress curve experiments were then initiated by the
addition of calpain, substrate and calcium, and the
resulting inactivation rates were used as a measure of
remaining inhibitor concentration. The half-life for
compound 1d was approximately 5 min. As a control, the
¯uoromethyl ketone 1e was tested by the above method
and shown to demonstrate no measurable decomposi-
tion over 60 min. There was no correlation between the
inactivation rates for calpain and cathepsin B and the
half-life for loss of inhibitor. Because of the limited sta-
bility of the series, all inactivation rates were determined
as rapidly as possible after addition of inhibitor to the
assay solution. Due to poor aqueous stability, the mea-
sured inactivation rates for 1d and other analogues (1f±j)
represent lower limits.
Akiyama, A.; Barrett, R., III; Carroll, R. M.; Straub, J. A.;
Tkacz, J. N.; Wu, C.; Musso, G. J. Med. Chem. 1994, 37,
2918. (g) Li, Z.; Patil, G.; Golubski, Z. E.; Hori, H.; Tehrani,
K.; Foreman, J. E.; Eveleth, D. D.; Bartus, R. T.; Powers, J.
C. J. Med. Chem. 1993, 36, 3472. (h) Li, Z.; Ortega-Vilain, A.;
Patil, G. S.; Chu, D.; Foreman, J. E.; Eveleth, D. D.; Powers,
J. C. J. Med. Chem. 1996, 39, 4089.
4. (a) Crawford, C.; Mason, R. W.; Wikstrom, P.; Shaw, E.
Biochem. J. 1988, 253, 751. (b) Angliker, H.; Anagli, J.; Shaw,
E. J. Med. Chem. 1992, 35, 216. (b) Rasnick, D. Anal. Bio-
chem. 1985, 149, 461. (c) Parkes, C.; Kembhavi, A. A.; Barrett,
A. J. Biochem. J. 1985, 230, 509. (d) McGowan, E. B.; Becker,
E.; Detwiler, T. C. Biochem. Biophys. Res. Commun. 1989, 158,
432. (e) Huang, Z.; McGowan, E. B.; Detwiler, T. C. J. Med.
Chem. 1992, 35, 2048. (f) Harris, A. L.; Gregory, J. S.; May-
cock, A. L.; Graybill, T. L.; Osifo, I. K.; Schmidt, S. L.; Dolle,
R. E. Bioorg. Med. Chem. Lett. 1995, 5, 393. (g) Chatterjee, S.;
Ator, M.; Bozyczko-Coyne, D.; Josef, K.; Wells, G.; Tripathy,
R.; Iqbal, M.; Bihovsky, R.; Senadhi, S. E.; Mallya, S.;
O'Kane, T. M.; McKenna, B. A.; Siman, R.; Mallamo, J. P. J.
Med. Chem. 1997, 40, 3820. (h) Palmer, J. T.; Rasnick, D.;
Klaus, J. L.; Bromme, D. J. Med. Chem. 1995, 38, 3193. (i)
Giordano, C.; Calabretta, R.; Gallina, C.; Consalvi, V.; Scan-
durra, R.; Noya, F. C.; Franchini, C. Eur. J. Med. Chem.
1993, 28, 297.
5. Krantz, A. In Advances in Medicinal Chemistry; Maryano€,
B. E., Ed.; JAI: Greenwich, CT, 1992; Vol. 1, pp 235±261. (b)
Krantz, A.; Copp, L. J.; Coles, P. J.; Smith, R. A.; Heard, S.
B. Biochemistry 1991, 30, 4678. (c) Robinson, V. J.; Pauls, H.
W.; Coles, P. J.; Smith, R. A.; Krantz, A. Bioorg. Chem. 1992,
20, 42. (d) Smith, R. A.; Copp, L. J.; Cole, P. J.; Pauls, H. W.;
Robinson, V. J.; Spencer, R. W.; Heard, S. B.; Krantz, A. J.
Am. Chem. Soc. 1988, 110, 4429. (e) Wagner, B. M.; Smith, R.
A.; Coles, P. J.; Copp, L. J.; Ernest, M. J.; Krantz, A. J. Med.
Chem. 1994, 37, 1833. (f) Pliura, D. H.; Bonaventura, B. J.;
Smith, R. A.; Coles, P. J.; Krantz, A. Biochem. J. 1992, 288,
759.
In conclusion, we have synthesized a novel class of irre-
versible inhibitors of calpain containing a relatively
short dipeptide anity groups. High inactivation rates
were obtained against two cysteine proteases through
manipulation of the peptide address and recognition
elements within the leaving group. However, poor aqu-
eous stability of those compounds severely limits their
therapeutic utility.
Acknowledgements
The authors would like to thank Dr. Je€ry Vaught for
his support and encouragement. We would also like to
express our sincere thanks to Mrs. Shobha E. Senadhi
and Dr. Satish Mallya for carrying out enzyme assays
and to Drs. Chakrapani Subramanyam, Ming Tao and
Ron Bihovsky for their contributions towards stability
studies.
6. Leucine at p2 of the peptide anity group was selected as
calpain prefers a Val or Leu at p2. See ref 1.
7. The leaving group of the inhibitor might be providing
additional prime site binding interactions with the enzyme.
Inhibitors of cathepsin K (a cysteine protease) with prime site
binding have been identi®ed. See Yamashita, D. S.; Smith, W.
W., Zhao, B.; Janson, C. A.; Tomaszek, T. A.; Bossard, M. J.;
Levy, M. A.; Oh, H.-J.; Carr, T. J.; Thompson, S. K.; Ijames,
C. F.; Carr, S. A.; McQueney, M.; D'Alessio, K. J.; Ame-
gadzie, B. Y.; Hanning, C. R.; Abdel-Meguid, S.; DesJarlais,
R. L.; Gleason, J. G.; Veber, D. F. J. Am. Chem. Soc. 1997,
119, 11351.
8. Barany, G.; Merri®eld, R. B. In The Peptides, Analysis,
Synthesis, Biology; Gross, E., Meienhofer, J., Eds.; Academic:
Orlando, 1980; pp 3±254.
9. Inhibitors (1a±d and 1f±j) were either puri®ed by crystal-
lization or by reverse-phase HPLC as most of them are sensi-
tive to silica gel chromatography.
References and Notes
1. Croall, D. E.; DeMartino, G. E. Physiolog. Rev. 1991, 71,
813 and references therein. (b) Sorimachi, H.; Saido, T. C.;
Suzuki, K. FEBS Lett. 1994, 343, 1.
2. Hanzlik, R. P.; Jacober, S. P.; Zygmunt, J. Biochim. Bio-
phys. Acta 1991, 1073, 33.
3. (a) Mehdi, S.; Angelastro, M. R.; Wiseman, J. S.; Bey, P.
Biochem. Biophys. Res. Commun. 1988, 157, 1117. (b) Tsuji-
naka, T.; Kajiwara, Y.; Kambayashi, J.; Sakon, M.; Higuchi,
N.; Tanaka, T.; Mori, T. Biochem. Biophys. Res. Commun.
1988, 153, 1201. (c) Iqbal, M.; Messina, P. A.; Freed, B.; Das,
M.; Chatterjee, S.; Tripathy, R.; Tao, M.; Josef, K. A.; Dem-
bofsky, B.; Dunn, D.; Grith, E.; Siman, R.; Senadhi, S. E.;
Biazzo, W.; Bozyczko-Coyne, D.; Meyer, S. L.; Ator, M. A.;
Bihovsky, R. Bioorg. Med. Chem. Lett. 1997, 7, 539. (d) Har-
ris, A. L.; Gregory, J. S.; Maycock, A. L.; Graybill, T. L.;
Osifo, I. K.; Schmidt, S. S.; Dolle, R. E. Bioorg. Med. Chem.
Lett. 1995, 5, 393. (e) Chatterjee, S.; Gu, Z.-Q., Dunn, D.;
Tao, M.; Josef, K.; Tripathy, R.; Bihovsky, R.; Senadhi, S. E.;
O'Kane, T. M.; McKenna, B. A.; Mallya, S.; Ator, M. A.;
Bozyczko-Coyne, D.; Siman, R.; Mallamo, J. P. J. Med.
Chem. 1998, 41, 2663. (f) Harbeson, S. L.; Abelleira, S. M.;
10. Meyer, S. L.; Bozyczko-Coyne, D.; Mallya, S. K.; Spais,
C. M.; Bihovsky, R.; Kawooya, J. K.; Lang, D. M.; Scott, R.
W.; Siman, R. Biochem. J. 1996, 314, 511.
11. MNA=b-methoxy naphthylamine.
12. (a) Tian, W.; Tsou, C. Biochemistry 1982, 21, 1028. (b)
Assays for inactivation of calpain contained 50 mM Tris±Cl
(pH 7.5), 50 mM NaCl, 1 mM EDTA, 1 mM EGTA, 5 mM
b-mercaptoethanol, 0.2 mM Suc-Leu-Tyr-MNA, 10 nM recom-
binant human calpain I, 3% DMSO and varying concentra-
tions of inhibitor and were initiated by the addition of 5 mM
CaCl2. Reactions were performed at ambient temperature in
single cuvettes with the increase in ¯uorescence (lex=340 nm,
l
em=440 nm) recorded continuously on
LS50B spectro¯uorimeter (Norwalk, CT, USA) and were
a Perkin±Elmer

R. Tripathy et al. / Bioorg. Med. Chem. Lett. 10 (2000) 2315±2319
2319
monitored until there was no further product generated in
inhibitor-containing assays. Inhibitor concentrations were at
least 10-fold greater than the enzyme concentration in all
cases. Values of kobs, the pseudo ®rst-order rate constant for
inactivation, were calculated from plots of ¯uorescence versus
time by nonlinear regression (Sigma Plot) to the exponential
eq (1)5b
methods that the unusual N±O bond in calicheamicin with its
distinctive torsional angle preference holds the two halves of
the molecule in a shape suitable for DNA binding. Kahne, D.;
Gupta, V.; Gange, D.; Walker, S. J. Am. Chem. Soc. 1994,
116, 3197.
16. (a) Staab, H. A.; Seel, G. Liebigs Ann. Chem. 1958, 612,
187. (b) Katritzky, A. R.; Rachwal, S.; Hitchings, G. J. Tet-
rahedron 1991, 47, 2683. (c) Shalaby, M. A.; Grote, C. W.;
Rapoport, H. J. Org. Chem. 1996, 61, 9045.
17. We postulate an elimination pathway for the decomposi-
tion of these compounds (1d and others) in aqueous media
which is evidenced by the detection of benzotriazole (5) as one
of the products. Though the major elimination product, the
keto-aldehyde 4, could not be isolated, it has been indepen-
dently prepared and was found not to be a time-dependent
inactivator of calpain. Therefore, it is reasonable to assume
that intact 1d as well as 1f±j are inactivators of calpain and
cathepsin B.
à †
y ˆ Aeꢀ-k
‡ B
ꢀ1†
obs
t
where y is the ¯uorescence at time t (Ft), A is the amplitude of
the reaction (F0 F) and B is the maximal amount of the pro-
duct formed when the enzyme is completely inactivated (F).
The apparent second-order rate constant for inactivation was
calculated from the slope of a plot of kobs versus inhibitor
concentration as (kobs/I)Ã(1 + S/Km), correcting for the e€ect
of substrate on the inactivation rate. (b) Rates of inactivation
of cathepsins B were determined under the assay conditions
described by Krantz et al.5b
13. Wells, G. J.; Bihovsky, R. Exp. Opin. Ther. Patents 1998,
8, 1707 and references therein.
14. Carpino, L. A. J. Am. Chem. Soc. 1993, 115, 4397.
15. For N±O bond conformational barrier and its role in
biological systems: Kahne et al. have shown via computational