Peptidyl aldehyde inhibitors of calpain incorporating P2-proline mimetics

Article abstract of DOI:10.1016/S0960-894X(03)00021-0

Four new peptidyl aldehydes bearing proline mimetics at the P₂-position were synthesized and studied as inhibitors of calpain I, cathepsin B, and selected serine proteases. The ring size of the P₂-constraining residue influenced the

Full text of DOI:10.1016/S0960-894X(03)00021-0

Bioorganic & Medicinal Chemistry Letters 13 (2003) 783–784
Peptidyl Aldehyde Inhibitors of Calpain Incorporating
P2-Proline Mimetics
Isaac O. Donkor,a,* Rajani Korukonda,a Tien L. Huangb and Louis LeCour, Jr.b
aDepartment of Pharmaceutical Sciences, University of Tennessee Health Science Center, Memphis, TN 38163, USA
bCollege of Pharmacy, Xavier University of Louisiana, New Orleans, LA70125, USA
Received 11 November 2002; revised 7 January 2003; accepted 10 January 2003
Abstract—Four new peptidyl aldehydes bearing proline mimetics at the P2-position were synthesized and studied as inhibitors of
calpain I, cathepsin B, and selected serine proteases. The ring size of the P2-constraining residue influenced the inhibitory potency
and selectivity of the compounds for calpain I compared to the other proteases.
# 2003 Elsevier Science Ltd. All rights reserved.
The calcium-activated intracellular neutral cysteine
protease, calpain I, has been implicated in a number of
pathological conditions including cardiac ischemia and
cerebral ischemia.1À5 Inhibitors of calpain I are there-
fore of interest as therapeutic agents and as biomedical
tools. A variety of peptide-based reversible inhibitors
(e.g., aldehyde and a-ketocarbonyl) and irreversible
inhibitors (e.g., expoxysuccinate, diazomethyl ketone,
acyloxymethyl ketone, and halomethyl ketone) of cal-
pain have been reported.5,6 Structure–activity relation-
ship studies of the active site of calpain suggest that the
S1-subsite of calpain can accommodate a wide variety of
amino acids but the S2-subsite of the enzyme appears to
have a strict requirement for either l-Val or l-Leu.5
However, recent studies7,8 have demonstrated that cal-
pain can accommodate d-amino acids at the P2-position
of inhibitors. In one such study, Tripathy et al.8 incor-
porated d-proline at the P2 position (4b) to obtain
potent inhibitors of the enzyme. Peet et al.9 have also
reported peptidyl aldehydes with constrained P1–P2
units. As a part of our interest in probing the subsite
specificities of calpain I,10,11 we synthesized 4a–d
(Scheme 1) incorporating proline mimetics at the P2
position of the inhibitors with the objective of investi-
gating the effect of the ring size of the P2 constraining
residue on calpain I inhibition.
d (27–78%), which were coupled with l-phenylalaninol
in the presence of HOBt and BOP to give dipeptides 3a–
d (53–82%). Dess–Martin oxidation of the dipeptides
afforded target compounds 4a–d (62–95%). Compound
5 was synthesized in 63% yield using l-leucine methyl-
ester in place of the proline mimetics. The products were
purified by column chromatography on silica gel and/or
recystallization (4a, 4d, and 5) from hexanes/CH2Cl2.
The diastereomeric ratios (see Table 1) of the com-
pounds were determined by HPLC using a C18 reversed
phase column with 6.2% aqueous TFA and acetonitrile
as the mobile phase.
Tripathy et al.8 have shown that 4b and its isomer with
l-proline at the P2 position are equipotent inhibitors of
calpain I. Compounds 4b and 4c, which could not be
resolved into pure isomers were tested as diastereomeric
mixtures. Peptidyl aldehydes are known to inhibit serine
and cysteine proteases.5 The compounds were therefore
evaluated13 as inhibitors of porcine erythrocyte calpain
I and selected serine proteases (trypsin, thrombin and
factor Xa) to determine selectivity for calpain I inhibi-
tion compared to serine protease inhibition. The com-
pounds were poor inhibitors of serine proteases. They
displayed IC50 values14 greater than 100 mM (data not
shown) versus the serine proteases. On the contrary, the
compounds were potent inhibitors of calpain I with Ki
values in the submicromolar range (Table 1). Thus, 4a–d
are selective inhibitors of cysteine proteases compared
to serine proteases. The basis for the selectivity may be
attributed to differences in the address regions of the
enzymes as well as the high nucleophilicity of the cata-
lytic site cysteine residue of cysteine proteases.
The procedure used to synthesize 4a–d is summarized in
Scheme 1.12 Tosylation of the proline mimetics gave 2a–
*Corresponding author. Tel.: +1-901-448-7736; fax: +1-901-448-
6828; e-mail: idonkor@utmem.edu
0960-894X/03/$ - see front matter # 2003 Elsevier Science Ltd. All rights reserved.
doi:10.1016/S0960-894X(03)00021-0

784
I. O. Donkor et al. / Bioorg. Med. Chem. Lett. 13 (2003) 783–784
potent than 5. This suggests that constraining the P2
residue is detrimental to calpain I inhibition despite
enhanced selectivity for calpain I compared to cathepsin
B. A similar effect was observed when the l-leucine
residue at the P2-position of calpain I inhibitors was
replaced with 2,3-methanoleucine stereoiomers.10
Acknowledgements
The study was supported by grant 5 K14HL03536 from
the U.S. Public Health Service.
References and Notes
Scheme 1. Reagents and conditions: (a) NaOH/H2O/THF; (b) phe-
nylalaninol hydrochloride, NMM, HOBt, BOP, DMF; (c) Dess–Mar-
tin reagent.
1. Sorimachi, H.; Saido, T. C.; Suzuki, K. FEBS Lett. 1994,
343, 1.
2. Bartus, R. T. Neurocientist 1997, 3, 314.
3. Wang, K. K. W.; Yuen, P.-W. Trends Pharm. Sci. 1994, 15,
412419.
4. Wang, K. K. W.; Yuen, P.-W. Adv. Pharm. 1997, 37, 117.
5. Donkor, I. O. Curr. Med. Chem. 2000, 7, 1171.
6. Wells, G. J.; Bihovsky, R. Exp. Opin. Ther. Pat. 1998, 8,
1707.
7. Chatterjee, S.; Gu, Z.-Q.; Dunn, D.; Tao, M.; Josep, 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.
8. Tripathy, R.; Gu, Z.-Q.; Dunn, D.; Senadhi, S. E.; Ator,
M. A.; Chatterjee, S. Bioorg. Med. Chem. Lett. 1998, 8, 2647.
9. Peet, N. P.; Kim, H.-O.; Marquart, A. L.; Angelastro,
M. R.; Nieduzak, T. R.; White, J. N.; Friedrich, D.; Flynn,
G. A.; Webster, M. E.; Vaz, R. J.; Linnik, M. D.; Kochl, J. R.;
Mehdi, S.; Bey, P.; Emary, B.; Hwang, K.-K. Bioorg. Med.
Chem. Lett. 1999, 9, 2365.
Table 1. Diastereomeric ratios, Ki (mM) valuesa and selectivity ratios
for inhibition of porcine calpain I and human liver cathepsin B by 4a–d
and 5
Compd
Calpain I
Cathepsin B
SRb
DRc LL:DL
4a
4b
4c
4d
5
0.25
0.08
0.07
0.54
2.75
12.1
–
5.26
4
11
151
–
10
–
100:0
81:19d
86:14
100:0
0.020.07
aValues are means of triplicate experiments.
bSR is selectivity ratio to the nearest whole number. It was determined
by dividing the Ki for cathepsin B inhibition by the Ki for calpain I
inhibition.
cDR is diastereomeric ratio (P1:P2).
dDR for 4b is that for the LD:DD isomers not the LL:DL isomers.
10. Donkor, I. O.; Zheng, X.; Miller, D. D. Bioorg. Med.
Chem. Lett. 2000, 10, 2497.
11. Donkor, I. O.; Zheng, X.; Han, J.; Lacy, C.; Miller, D. D.
Bioorg. Med. Chem. Lett. 2001, 11, 1753.
Cathepsin B, like calpain belongs to the papain super-
family of cysteine proteases and is known to be inhib-
ited by peptidyl aldehydes and several calpain
inhibitors.5 We therefore assayed the compounds
against cathepsin B to determined selectivity for calpain
I. Generally, the compounds were better inhibitors of
caplain I compared to cathepsin B. Compound 4b,
which showed 150-fold selectivity for calpain I over
cathepsin B was the most selective member of the series
(Table 1). The size of the P2 residue influenced the cal-
pain I inhibitory potency of the compounds. The ki
values for inhibition of calpain I ranged from 0.07 mM
to 0.54 mM. Compounds with a five-membered ring at
the P2-position were the most potent members of the
series. Thus, 4b with a d-proline residue at the P2-posi-
tion and 4c with a l-thiazolidine-4-carboxylic acid resi-
due at this position were better inhibitors of calpain I
compared to 4a and 4d, which had d-azetidine-2-car-
boxylic acid and l-pipecolic acid as the P2 residue,
respectively. The equipotency of 4b and 4c is in agree-
ment with the observation that calpain can accom-
modate d- and l-proline at the P2 position.8 The
conformational behavior of the proline mimetics may
account for the observed difference in the potency of the
compounds.15 Compounds 4b and 4c, which were the
most potent calpain I inhibitors of the series were less
12. l-Azetidine-2-carbozylic acid (1a), l-thiazolidine-4-car-
boxylic acid (1c), and dl-pipecolic acid (1d) were purchased
from Aldrich. d-Proline (1b) was obtained from Advanced
ChemTech.
13. The calpain assay consisted of 50 mM Tris HCl (pH 7.4),
50 mM NaCl, 10 mM dithiothreitol, 1 mM EDTA, 1 mM
EGTA, 0.2mM or 1.0 mM Suc-Leu-Tyr-AMC, 2 mg porcine
erythrocyte calpain I, varying inhibitor concentrations in
DMSO (2%) and 5 mM CaCl2 in a final volume of 250 mL.
Assays were initiated by addition of CaCl2 and the increase in
fluorescence (lex=370 nm, lem=440 nm) was monitored at
amibient temperature. Ki values were by Dixon plots16 of tri-
plicate assays with correlation coefficient >0.95.
14. Serine protease inhibition assays: Each 200 mL reaction
mixture contained buffer (20 mM Tris pH 7.4, 150 mM
NaCl, 2mM CaCl 2, 0.1% BSA), inhibitor (0.1–100 mM),
enzyme (40 nM trypsin, or 40 nM a-thrombin, or 10 nm
FXa), and substrate (100 mM Bzl-Phe-Val-Arg-pNA for
trypsin and a-thrombin assays, or 100 mM Bzl-Ile-Glu-Gly-
Arg-pNA for FXa assay). The reaction mixture was incu-
bated at 25 ꢀC for 5 min prior to the addition of the substrate.
After substrate addition, the hydrolytic rates were read for 2
min at 412nm.
15. Toniolo, C.; Int, J. Peptide Protein Res. 1990, 35, 287.
16. Dixon, M. Biochem. J. 1972, 129, 197.