Probing of primed and unprimed sites of calpains: Design, synthesis and evaluation of epoxysuccinyl-peptide derivatives as selective inhibitors

Article abstract of DOI:10.1016/j.ejmech.2014.05.058

Calpains are intracellular cysteine proteases with important physiological functions. Up- or downregulation of their expression can be responsible for several diseases, therefore specific calpain inhibitors may be considered as promising candidates for drug discovery. In this paper we describe the synthesis and characterization of a new class of inhibitors derived from the analysis of amino acid preferences in primed and unprimed sites of calpains by incorporation of l- or d-epoxysuccinyl group (Eps). Amino acids for replacement were chosen by considering the substrate preference of calpain 1 and 2 enzymes. The compounds were characterized by RP-HPLC, amino acid analysis and ESI-MS. Selectivity of the compounds was studied by using calpain 1 and 2; and cathepsin B. We have identified five calpain specific inhibitors with different extent of selectivity. Two of these also exhibited isoform selectivity. Compound NH ₂-Thr-Pro-Leu-(d-Eps)-Thr-Pro-Pro-Pro-Ser-NH₂ proved to be a calpain 2 enzyme inhibitor with at least 11.8-fold selectivity, while compound NH₂-Thr-Pro-Leu-(l-Eps)-Ser-Pro-Pro-Pro-Ser-NH₂ possesses calpain 1 enzyme inhibition with at least 4-fold selectivity. The results of molecular modeling calculations suggest that the orientation of the bound inhibitor in the substrate binding cleft is markedly dependent on the stereochemistry of the epoxysuccinyl group.

Full text of DOI:10.1016/j.ejmech.2014.05.058

European Journal of Medicinal Chemistry 82 (2014) 274e280
Contents lists available at ScienceDirect
European Journal of Medicinal Chemistry
journal homepage: http://www.elsevier.com/locate/ejmech
Original article
Probing of primed and unprimed sites of calpains: Design, synthesis
and evaluation of epoxysuccinyl-peptide derivatives as selective
inhibitors
a
b
c
a, d
ꢀ
ꢀ
ꢀ
ꢀ
Levente E. Dokus , Dora K. Menyhard , Agnes Tantos , Ferenc Hudecz
,
a, *
ꢀ
ꢀ ꢀ
Zoltan Banoczi
a
€
€
ꢀ
MTA-ELTE Research Group of Peptide Chemistry, Hungarian Academy of Sciences (HAS), Eotvos Lorand University (ELTE), Budapest, Hungary
b
€
€
ꢀ
Laboratory of Structural Chemistry and Biology and MTA-ELTE Protein Modeling Group, Eotvos Lorand University, Budapest, Hungary
c Institute of Enzymology, Research Centre for Natural Sciences of the Hungarian Academy of Sciences, Budapest, Hungary
d
€
€
Department of Organic Chemistry, Eotvos L. University, Budapest, Hungary
a r t i c l e i n f o
a b s t r a c t
Article history:
Calpains are intracellular cysteine proteases with important physiological functions. Up- or down-
regulation of their expression can be responsible for several diseases, therefore specific calpain inhibitors
may be considered as promising candidates for drug discovery. In this paper we describe the synthesis
and characterization of a new class of inhibitors derived from the analysis of amino acid preferences in
Received 20 January 2014
Received in revised form
19 May 2014
Accepted 23 May 2014
Available online 24 May 2014
primed and unprimed sites of calpains by incorporation of L- or D-epoxysuccinyl group (Eps). Amino acids
for replacement were chosen by considering the substrate preference of calpain 1 and 2 enzymes. The
compounds were characterized by RP-HPLC, amino acid analysis and ESI-MS. Selectivity of the com-
pounds was studied by using calpain 1 and 2; and cathepsin B. We have identified five calpain specific
inhibitors with different extent of selectivity. Two of these also exhibited isoform selectivity. Compound
Keywords:
Calpain
Epoxysuccinyl-peptide
Epoxysuccinic acid
Enzyme inhibitor
NH2-Thr-Pro-Leu-(
D-Eps)-Thr-Pro-Pro-Pro-Ser-NH2 proved to be a calpain 2 enzyme inhibitor with at
least 11.8-fold selectivity, while compound NH2-Thr-Pro-Leu-(
L-Eps)-Ser-Pro-Pro-Pro-Ser-NH2 possesses
calpain 1 enzyme inhibition with at least 4-fold selectivity. The results of molecular modeling calcula-
tions suggest that the orientation of the bound inhibitor in the substrate binding cleft is markedly
dependent on the stereochemistry of the epoxysuccinyl group.
© 2014 Elsevier Masson SAS. All rights reserved.
1. Introduction
Most of the peptide- and non-peptide-based calpain inhibitors
interact only with Sn positions of the active site [10], although
expanding the binding surface towards Sn sites may result in more
selective inhibitors. Design of inhibitors by studying the preference
of amino acids in certain positions of calpain substrates can be
The members of calpain family are intracellular Ca2þ-ion
dependent cysteine proteases [1,2] which are present in almost all
eukaryotes and also in some bacteria. The two ubiquitous forms are
0
the calpain 1 (
m-calpain) and calpain 2 (m-calpain). Calpains have
considered as a promising strategy. Tompa et al. analyzed
important role in various cellular activities (cell adhesion and
motility, signal transduction, exocytosis and regulation of gene
expression etc.) [3e7]. Over-activated calpains could be involved in
the pathogenesis of a wide range of disorders such as Alzheimer's,
Huntington's and Parkinson's diseases, multiple sclerosis, amyo-
trophic lateral sclerosis or ischemic and traumatic brain injury [8,9].
numerous cleavage sites in calpain substrates, and established a
matrix containing preferences of amino acids in positions P4eP7
[11]. Based on this matrix, a FRET calpain substrate with excellent
characteristics, and also its cell-penetrating derivative was suc-
cessfully prepared and applied to monitor intracellular calpain ac-
tivity [12,13].
0
Based on this substrate sequence, we have designed a group of
azaglycine (Agly) containing analogues as potential novel calpain
inhibitors [14]. In these compounds Lys in position P1 has been
changed to azaglycine.
* Corresponding author. MTA-ELTE Research Group of Peptide Chemistry, Hun-
€
€
garian Academy of Sciences, Eotvos L. University, Budapest 112, POB 32, H-1518,
Hungary.
The epoxysuccinyl group (Eps) can be used to design inhibitors
0
with different amino acids at Pn and Pn sites. Some synthetic and
ꢀ
ꢀ
E-mail address: banoczi@elte.hu (Z. Banoczi).
http://dx.doi.org/10.1016/j.ejmech.2014.05.058
0223-5234/© 2014 Elsevier Masson SAS. All rights reserved.

ꢀ
L.E. Dokus et al. / European Journal of Medicinal Chemistry 82 (2014) 274e280
275
natural peptide epoxides have already been described as proven
cysteine protease inhibitors [15].
washing with DMF (8 ꢀ 0.5 min). For coupling, amino acid de-
rivatives, DIC and HOBt dissolved in DMF were used in 3-fold molar
excess for the resin capacity. The reaction proceeded at RT for
60 min, then the resin was washed with DMF.
The epoxysuccinyl-pentapeptides were built up also on Rink-
amide MBHA resin (0.2 g, 0.52 mmol/g) by the same coupling
strategy. After the removal of the terminal Na-Fmoc protection, the
trans-epoxysuccinyl group was attached to the pentapeptide chain
The trans-L-epoxylsuccinyl-L-leucylamido-4-guanidino-butane
(E-64) was first published as an effective and selective cysteine
protease inhibitor possessing epoxy group [16]. This compound and
its derivatives are irreversible inactivators of cysteine proteases.
Different parts of E-64 were varied to improve its inhibitory effect
on calpains [17].
Using
positions P4eP2 was studied on calpain 1 and 2 inhibition [18]. As a
result of this systematic study a peptide analogue, WRH( -Eps)-OEt
was identified as a selective and irreversible calpain inhibitor.
Pfizer et al. attached different dipeptides to the free carboxylic
L
- or
D
-epoxysuccinyl analogues the effect of changes in
by L- or D-trans-epoxysuccinic acid using DIC and HOBt. The tri-
peptides as well as the epoxysuccinyl-pentapeptides were removed
from the resin by cleavage with 5 mL TFA containing 0.25 mL
distilled water. In case of Trp containing tripeptide segments the
cleavage was performed by TFA in the presence of scavengers (TFA,
water, thioanisole, EDT, crystalline phenol ¼ 5 mL, 0.25 mL, 0.25 mL,
0.125 mL, 0.375 g). The crude product was precipitated by dry
diethyl ether, dissolved in 10% acetic acid, freeze-dried and sepa-
rated by RP-HPLC. Yield of peptide synthesis was 35e40%.
D
group of Ep-460 (HO-(L)Eps-Leu-NH-[CH2]4-NH-Z; e.g. HO-AA1-
AA2-( )Eps-Leu-NH-[CH2]4-NH-Z), and studied the inhibitory ac-
L
tivity on calpain 1, cathepsin B and cathepsin L [19]. Ep-460 was
supposed to direct the dipeptide part to the primed position in the
substrate binding pocket. The amino acid preference was examined
0
0
using amino acid libraries of positions P1 and P2 .
2.2. Synthesis of epoxysuccinyl-nonapeptide analogues
Recently, Schiefer et al. reported E-64 analogues as potent calpain
1 selective inhibitors [20]. The capping group of E-64 (4-guanidino-
butane) at position P3/P4 and Leu at position P2 were changed. The
selectivity of these compounds was characterized using papain.
In the present paper we describe the design, synthesis and func-
tional characterization of novel potential peptide inhibitors with
epoxysuccinyl group of cysteine proteases calpain 1 and 2, and
cathepsin B. Molecular modeling was also performed to understand
The purified epoxysuccinyl-pentapeptides and the tripeptides
were coupled in solution (DMF). 10 mg (0.01 mmol) epoxysuccinyl-
pentapeptide was reacted with 1.2 equivalent tripeptide in the
presence of 1.2 equivalent DIC and HOBt. The reaction proceeded at
RT overnight, then the DMF was evaporated and the components of
the reaction mixture were separated by RP-HPLC. Yield of the
conjugation reactions was 28e35%.
the binding of inhibitors with L- or D-epoxysuccinyl group. The pre-
sented peptide analogues are based on the sequence of calpain sub-
strate peptide (TPLKSPPPSPR), containing different amino acids at
positions P and P0, derived from the preference matrix [11]. By incor-
2.3. RP-HPLC
Analytical RP-HPLC was performed on a Zorbax SB C18 column
poration of
L
- or
D
-epoxysuccinyl group, the amino acid preference at
(150 ꢀ 4.6 mm I.D.) with 5
mm silica (100 Å pore size) (Torrance, CA
positions P and P0 was scanned to identify novel calpain inhibitors.
USA) as a stationary phase. A linear gradient elution was devel-
oped: 0 min 0% B; 2 min 0% B; 22 min 90% B with eluent A (0.1% TFA
in water) and eluent B (0.1% TFA in acetonitrileewater (80:20, v/v)).
A flow rate of 1 mL/min was used at ambient temperature. Samples
2. Materials and methods
All amino acid derivatives were purchased from Bachem
(Bubendorf, Switzerland) and Reanal (Budapest, Hungary); whereas
DIEA, HOBt, DIC, TFA were FLUKA (Buchs, Switzerland) products.
Rink-amide MBHA resinwas from Iris Biotech GmbH (Marktredwitz,
were dissolved in eluent B, injection volume: 20
detected at
¼ 220 nm. The purification of the crude products was
carried out by RP-HPLC using Phenomenex Jupiter C18 column
m silica (300 Å pore size) (Torrance,
CA, USA), flow rate: 4 mL/min, room temperature. Linear gradient
elution was applied. Gradient I: 0 min 5% B, 5 min 5% B, 50 min 50%
B. Gradient II: 0 min 10% B, 5 min 10% B, 50 min 70% B.
mL. Peaks were
l
(250 ꢀ 10 mm I.D.) with 10
m
Germany). Racemic trans-
L/D
epoxysuccinic acid was obtained from
TCI (Tokyo, Japan). Solvents for synthesis and purification were
obtained from Molar Chemicals Kft (Budapest, Hungary). The fluo-
rescent substrates Suc-Leu-Tyr-AMC (catalog no. S 1153) and all
other chemicals used in biochemical experiments were purchased
from Sigma. The calpain 1 enzyme from human erythrocytes (EC
3.4.22.52) and cathepsin B from human liver (EC 3.4.22.1), as well as
the substrate III of cathepsin B (Z-Arg-Arg-AMC ꢀ 2 HCl) were ob-
tained from Calbiochem/Merck (Darmstadt, Germany). All buffers
were prepared with ion exchanged distilled water.
2.4. Amino acid analysis
The amino acid composition of the peptides was determined by
amino acid analysis performed on a Sykam Amino Acid S433H
analyser (Eresing, Germany). Prior to analysis, samples were hydro-
lyzed with 6 M HCl in sealed and evacuated tubes at 110 ꢁC for 24 h.
2.1. Synthesis of tripeptides and
pentapeptides
L
- or
D-epoxysuccinyl-
2.5. Mass spectrometry
The peptides and peptide analogues were identified by elec-
trospray ionization mass spectrometry (ESI-MS) on a Bruker Dal-
tonics Esquire 3000 Plus (Bremen, Germany) ion trap mass
The resolution of trans-L- and D-epoxysuccinic acid isomers us-
ing racemic starting material was performed as described by Tamai
et al. [21] without modification. Yield of resolution was 41% for the
spectrometer, operating in continuous sample injection at 4 mL/min
D
-enantiomer and 46% for the
L
-enantiomer. The tripeptide seg-
flow rate. Samples were dissolved in ACNewater (50:50 v/v%)
mixture containing 0.1 v/v% AcOH. Mass spectra were recorded in
positive ion mode in the m/z 50e2000 range.
ments were synthesized manually by solid phase peptide synthesis
on Rink-amide MBHA resin (0.2 g, 0.52 mmol/g) using Fmoc/tBu
strategy. The amino acid side-chain protecting groups were trityl
for Gln and 2,2,4,6,7-pentamethyldihydrobenzofuran-5-sulfonyl
for Arg. The side chain of Ser and Thr was protected with tert-
butyl group. The Na-Fmoc group was removed with 2% DBU in the
presence of 2% piperidine in DMF (2, 2, 5, 10 min) followed by
2.6. Enzyme purification
The 80-kDa large subunit and the 21-kDa truncated small sub-
unit of rat calpain 2 were expressed in Escherichia coli as described

ꢀ
L.E. Dokus et al. / European Journal of Medicinal Chemistry 82 (2014) 274e280
276
earlier [22] and purified with Akta Explorer (GE Healthcare) fast
protein liquid chromatography (FPLC) system, on a HisTrap HP
column using step elution. Purified calpain 2 was dialyzed against
10 volume of calpain buffer (10 mM HEPES, 150 mM NaCl, 1 mM
EDTA, pH ¼ 7.5). The enzyme preparation in solution was stored at
4 ꢁC until use.
mer calpain substrate sequence (TPLKSPPPSPR) [11]. In the group of
previously published inhibitory peptides, lysine at position P1 was
replaced with azaglycine [26]. Here we report on the design and
synthesis of a novel group of oligopeptide analogues containing
D-
or -trans-epoxysuccinyl-group ( -Eps, -Eps) (Table 1). In these 9-
L
D
L
mer compounds of TPLKSPPPS sequence the Lys residue at posi-
tion P1 was replaced by the epoxysuccinyl moiety.
2.7. Enzyme inhibition assay
The synthesis of epoxysuccinyl-nonapeptide analogues, as out-
lined in Scheme 1, was carried out by the combination of the sepa-
rate solid phase peptide synthesis of the two segments, and their
conjugation in solution. The N-terminal tripeptide amid0e (P4eP2) as
prepared on Rink-amide MBHA solid phase using Fmoc/tBu strategy.
The epoxysuccinyl group was attached to the N-terminal amino
Activity of the enzymes (calpain 1 and 2, cathepsin B) was
measured using a Synergy microplate reader, on a 96-well plate. In
0
case of the calpains, the reaction mixture contained 100
AMC (Sigma) substrate, 3 mM Ca2þ and the inhibitor at various
concentrations in 50 L of calpain buffer. The reaction was initiated
by adding and mixing rapidly 0.2 M calpain 2 or 0.5 M calpain 1.
The cathepsin B activity was measured in 50 L of 0.25 mM MES
M cathepsin B substrate III and
m
M Suc-LY-
well as the N-epoxysuccinyl-pentapeptide amide (P1 eP5 ) were
m
m
m
group of the pentapeptide on solid phase using L- or D-trans-epox-
m
ysuccinic acid. After removal from the resin, the purified free tri-
peptide amide and the purified epoxysuccinyl-pentapeptide amide
were conjugated in solution by in situ activation using DICeHOBt
coupling reagents. The epoxysuccinyl-nonapeptides were purified by
semipreparative RP-HPLC. It should be noted that no significant ring
opening was observed in case of epoxysuccinyl-pentapeptide ana-
logues during the cleavage from the resin by TFA in the presence of a
small amount of water. The homogeneity of the purified compounds
was characterized by analytical RP-HPLC (in Supplementary
information), while the amino acid composition and primary struc-
ture were verified by ESI-MS (Table 1) and amino acid analysis (in
Supplementary information).
buffer (pH ¼ 5.0) containing 100
m
inhibitor at various concentrations. The reaction was started by
adding and mixing rapidly 0.7 nM cathepsin B. Cathepsin B was
activated in 0.25 mM MES buffer containing 10 mM DTE. The enzyme
activity was followed by the change of fluorescence intensity
measured at excitation/emission wavelengths of
l
¼ 380/460 nm.
Three replicates of each experiment were performed. The kinetic
data were obtained from the pre-steady-state phase of the progress
curve as described previously, accepting the method of previous
authors for calculation of their epoxy-dipeptide inhibitors [19,23,24].
2.8. Molecular modeling
The common feature of this group of nonapeptide analogues is
that the P4eP2 segment (e.g. H-Leu-Pro-Thr-NH2) as well as the
pentapeptide unit (e.g. H-Ser-Pro-Pro-Pro-Ser-NH2) are attached
with their N-terminal amino group to the epoxysuccinyl group, so
all sequences have a CeN (epoxysuccinyl) NeC orientation in their
structure.
A model structure for the calpain 2 complex with inhibitor 2 and
11 was built. Starting structure was derived from the crystal
structure of the calpain 2ecalpastatin complex [25], using domains
II and III of the large catalytic unit, fitting both inhibitors to the
calpastatin active region (612e613) in two opposing orientations,
resulting in 4 different arrangements. The mutated catalytic residue
(Ser105) was re-mutated in silico to its wild type equivalent of Cys.
Monte Carlo Multiple Minimum (MCMM) searches were carried
out by involving the random variation (within the range of 0e180ꢁ)
of a randomly selected subset of all torsional angles of the inhibitor,
and the random translation (0e5 Å) and rotation (0e180ꢁ) of the
inhibitor within the enzyme matrix in a Monte Carlo step. The
perturbed structures were energy-minimized and the unique
structures were stored within a 40 kJ/mol energy window above
the global minimum. Calculations were carried out using the OPLS
2005 forcefield [26]. Solvent-effect was modeled by the GB/SA al-
gorithm (using water solvent). The first search of 3000 steps was
restrained, where the distance of the SG atom of Cys105 and C2 and
C3 of the epoxy-ring was held between 2.9 and 5.5 Å using a flat-
bottom constraint, while energy minimization was carried out on
all inhibitor atoms and the side chain atoms of all residues reaching
within 6 Å of the inhibitor. In a second calculation, 1000 steps of
constraint free MCMM search (Monte Carlo steps involving the
inhibitor, but energy minimization carried out on the entire
structure) was conducted on the lowest energy arrangement of
each restrained search. The unbound inhibitors were also subject to
10,000 steps constraint-free MCMM search to be able to estimate
the binding energy.
3.2. Inhibition of the activity of calpain 1 and 2
The inhibitory potential of epoxysuccinyl-peptides was studied
on calpain 1 and 2 using Suc-LY-AMC as substrate, and it was
Table 1
Chemical characteristics of epoxysuccinyl-nonapeptides.
Compound
Rta
(min.)
M (cal.) Mb
(meas.)
1. NH2-Thr-Pro-Leu-(
L
-Eps)-Ser-Pro-Pro-Pro-Ser-NH2 12.5 906.5 906.5
2. NH2-Thr-Trp-Leu-(
L-Eps)-Ser-Pro-Pro-Pro-Ser-NH2 12.6 995.8 995.6
3. NH2-Thr-Ser-Leu-(
4. NH2-Thr-Pro-Thr-(
L
-Eps)-Ser-Pro-Pro-Pro-Ser-NH2
-Eps)-Ser-Pro-Pro-Pro-Ser-NH2
10.3 884.4 884.7
9.7 882.5 882.4
10.7 892.7 892.6
L
5. NH2-Thr-Pro-Val-(
6. NH2-Thr-Pro-Leu-(
7. NH2-Thr-Pro-Leu-(
8. NH2-Thr-Pro-Leu-(
9. NH2-Thr-Pro-Leu-(
L-Eps)-Ser-Pro-Pro-Pro-Ser-NH2
L
L
L
L
-Eps)-Arg-Pro-Pro-Pro-Ser-NH2 11.1 976.0 976.7
-Eps)-Thr-Pro-Pro-Pro-Ser-NH2 11.0 921.8 921.5
-Eps)-Ser-Ser-Pro-Pro-Ser-NH2
-Eps)-Ser-Gln-Pro-Pro-Ser-NH2 10.8 937.5 937.5
-Eps)-Ser-Pro-Pro-Pro-Ser-NH2 10.9 906.5 906.7
12.0 896.5 896.5
10. NH2-Thr-Pro-Leu-(
11. NH2-Thr-Trp-Leu-(
12. NH2-Thr-Pro-Thr-(
13. NH2-Thr-Pro-Val-(
14. NH2-Thr-Ser-Leu-(
15. NH2-Thr-Pro-Leu-(
16. NH2-Thr-Pro-Leu-(
17. NH2-Thr-Pro-Leu-(
18. NH2-Thr-Pro-Leu-(
D
D
-Eps)-Ser-Pro-Pro-Pro-Ser-NH2 13.2 995.8 995.7
-Eps)-Ser-Pro-Pro-Pro-Ser-NH2 9.1 882.5 882.7
D-Eps)-Ser-Pro-Pro-Pro-Ser-NH2 10.2 892.5 892.5
D
D
-Eps)-Ser-Pro-Pro-Pro-Ser-NH2 10.2 884.4 884.7
D
-Eps)-Arg-Pro-Pro-Pro-Ser-NH2 11.1 976.0 975.9
-Eps)-Thr-Pro-Pro-Pro-Ser-NH2 12.3 921.8 921.8
-Eps)-Ser-Ser-Pro-Pro-Ser-NH2 11.9 896.5 896.5
-Eps)-Ser-Gln-Pro-Pro-Ser-NH2 10.7 937.5 937.5
D
D
D
3. Results
a
HPLC retention time; column: Zorbax SB C18 column (150 ꢀ 4.6 mm I.D.) with
m silica (100 Å pore size). Linear gradient elution: 0 min 0% B; 2 min 0% B; 22 min
5
m
3.1. Synthesis and chemical characterization of epoxysuccinyl-
peptide analogues
90% B with eluent A (0.1% TFA in water), and eluent B (0.1% TFA in acetonitrilee-
water (80:20, v/v)); flow rate: 1 mL/min, ambient temperature. Peaks were detected
at
l
¼ 220 nm.
b
We demonstrated earlier that oligopeptides with 9e11 amino
acid residues as calpain inhibitors can be identified based on an 11-
ESI-MS analysis was carried out on a Bruker Esquire 3000 plus (Germany). The
sample was dissolved in acetonitrileewater (50:50, v/v), 0.1% acetic acid.

ꢀ
L.E. Dokus et al. / European Journal of Medicinal Chemistry 82 (2014) 274e280
277
Scheme 1. Outline of the synthesis strategy of epoxysuccinyl-peptide analogue.
characterized by Ki values (Table 2). If there was no inhibition at
50 M concentration, the Ki value was not calculated.
Concerning the activity of epoxysuccinyl-peptides with
isomer on calpains, we found that compound 1 with the -trans-
epoxysuccinyl group had no effect on calpain 2 even at c ¼ 50 M,
but had moderate effect on calpain 1 (Ki ¼ 12.50 M). Substitution
of Pro at position P3 with Trp (2) or Ser (3) resulted in compounds
with inhibitory activity on both calpain 2 (Ki ¼ 4.05 and 14.3 M,
M, respectively).
In order to study the effect of stereochemical orientation related
m
to the presence of -epoxysuccinic group instead of the
D
L-isomer,
L
-Eps
nonapeptides with D-isomer were also prepared and examined.
L
Among this group of analogues only compounds 11 and 16 inhibi-
ted calpain 2 (Ki ¼ 21.43 and 4.24 M), and compounds 11, 14 and 15
inhibited calpain 1 (Ki ¼ 3.70, 30.84 and 26.33 M).
m
m
m
m
m
3.3. Selectivity: inhibition of cathepsin B activity
Compounds with - or -epoxysuccinyl-group were also studied
respectively) and calpain 1 (Ki ¼ 17.14 and 21.92
m
The effect of replacement of Leu residue at position P2 with Thr (4)
had not effect on the activity, while the presence of Val (5) resulted
in a peptide analogue with moderate inhibitory effect only on cal-
D
L
for their inhibitory effect on cathepsin B. In both groups of non-
apeptide analogues, we found compounds with ability to inhibit
the activity of cathepsin B. Epoxysuccinyl-peptides (2) and (7) with
pain
(Ki ¼ 17.86
2
(Ki
m
¼
29.23
m
M). Moderate activity on calpain
2
M) and also on calpain 1 (Ki ¼ 23.73
mM) was observed
L
-Eps (Ki ¼ 5.00 and 12.07
mM, respectively), and also compound 11
with M) were active on cathepsin B (Table 2).
after incubati0on of the enzymes with compound 6 with Arg residue
D
-Eps (Ki ¼ 4.45
m
at position P1 . Compound 7 with Thr at position P10 exhibited good,
but not excellent effect on calpain 1 (Ki ¼ 8.38
mM). Compound with
Ser or Thr at position P20 (8 or 9) did not exhibit any activity against
3.4. Molecular modeling
calpains.
Epoxysuccinyl inhibitors have been co-crystallized with calpain
1 (2nqg [18], 2nqi, 2nqg [18], 1tlo [27]), papain (1cvz [28]) and
cathepsin B (1ito [29]) previously. These crystal structures show the
covalently linked product of the cleavage e the epoxy-ring open
and relaxed into an open-chain conformation. However, no small
molecular complexes of calpain 2 have been, to date, determined.
The calculations carried out here depict the pre-cleavage state of
calpain 2 in complex with compound 2 (the most effective inhibitor
identified) and compound 11 (its markedly less effective, stereo-
chemical counterpart), in the rate limiting docked state, prior to the
nucleophilic attack of Cys.
Table 2
The inhibitory effect of epoxysuccinyl-peptides.
Compound
Ki (m
M) (SD)a
Calpain 2
Calpain 1
Cathepsin B
1
2.
3.
4.
5.
6.
7.
8.
>50b
4.05 (2.10)
14.30 (2.70)
>100
29.23 (5.00)
17.86 (2.45)
>50
>100
>100
>50
21.43 (1.07)
no data
>100
no data
>50
4.24 (1.85)
>100
>100
12.50 (3.54)
17.14 (3.29)
21.92 (1.29)
>100
>50
5.00 (2.06)
>50
>100
>100
>50
12.07 (3.02)
>100
>100
>50
4.45 (1.79)
no data
>100
no data
>100
>50
>100
>100
>50
23.73 (7.28)
8.38 (0.06)
>100
>100
>100
3.70 (0.60)
>100
>50
30.84 (1.05)
26.33 (1.44)
>50
Peptide NH2-Thr-Trp-Leu-( -Eps)-Ser-Pro-Pro-Pro-Ser-NH2 (2) is
L
oriented within the binding grove by 11 intermolecular H-bonds
(see Fig. 1): residues P2eP4 are linked to Lys61, Gly103, Glu193,
Ser196 of domain II of the enzyme, t0he epoxysuccinyl part is bound
by the amide nitrogen of Gly198, P1 Ser by the carbonyl oxygen of
Gly261 of domain II and the P50 Ser by Asp425 and Met426 of domain
III0e it is thus bound in a conformation where not the P2 Leu, but the
P1 Ser is immersed in the S2 pocket. This is not so surprising, since
the S2 pocket is lined by polar residues (Thr200, Thr201, Ser241,
Glu339), whereby it is far from being an ideal docking site for Leu,
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
>100
>100
0
and although the P1 Ser is too short to reach these residues, it
a
Data represent the mean S.D. value of triplicate experiments.
b
benefits from the polarized environment created by them. P2 Leu is
stabilized by favorable contacts from Leu102 of the enzyme.
>50, >100: the epoxysuccinyl-peptide did not inhibit the enzyme at 50 or
M concentration.
100
m

ꢀ
L.E. Dokus et al. / European Journal of Medicinal Chemistry 82 (2014) 274e280
278
orientation places the Leu residue in P2 position in the S2 pocket.
The estimated binding energy for this complex, however, is 31 kJ/
mol weaker than that of the calpain 2ecompound 2 complex.
Binding mode of both inhibitors is canonical in the sense that
the central, epoxysuccinyl part of the molecules is anchored in both
cases by an H-bond between the carbonyl oxygen of the inhibitor
and the backbone amide of Gly198. Gly198 (and the corresponding
Gly208 of calpain 1, Gly66 of papain and Gly74 of cathepsin B)
forms a similar H-bond with the carbonyl oxygen atom of the Leu at
position P2 of calpastatin (3df0 [30], ltl9 [27]), or the epoxysuccinyl
carbonyl oxygen of E64 (1tlo [27]), E64C (1ito [29]), WR18 (2nqg
[18]), WR13 (2nqi [18]) or also of CLIK148 [28].
4. Discussion and conclusions
Over-activated calpain enzymes are involved in several patho-
logical processes in cells. Identification of appropriate inhibitors
(effective and selective) may lead compounds for the treatment of
diseases associated with over-activation. Such inhibitors could also
be useful in studying the function of calpains even in normal cells.
Inhibitors capable to distinguish not only between different
cysteine proteases, but calpain isoforms could be even more valu-
able. Our approach to develop selective calpain inhibitors is based
on the analysis of the sequences of relevant protein substrates [11].
The most examined part of calpain inhibitors is the P4eP1
preference region [10]. Only few studies were reported on the
0
0
preference of the primed site (P4 eP1 ) of inhibitors. We have
selected the epoxysuccinyl functional group as a suitable moiety to
examine the effect of amino acids at primed and non-primed sites
on the inhibitory activity and selectivity. This group has an
important contribution to the inhibitory potency of a cysteine
protease selective inhibitor, E-64 [16]. Therefore the incorporation
of this functional moiety could result in compounds with selectivity
against cysteine proteases.
According to our knowledge, no reports were published so far
in which the effect of simultaneous substitutions at both non-
primed and primed binding sites were analyzed. We have
designed and synthesized compounds with amino acid sub-
stitutions considering the substrate preference matrix [11]. Amino
acids with the first, the second and the third highest score were
incorporated into the substrate sequence, TPLKSPPPSPR. Guided
by the results of our earlier study with a group of azaglycine
substituted substrate peptides exhibiting inhibitory properties
[14], we have prepared a limited set of the analogues for the
present investigations.
4.1. The effect of the replacement of Lys at position P1 by
and the amino acid substitution on inhibitory activity
L/D-Eps,
As our data suggest, the replacement of Lys residue at position
P1 with -Eps in the substrate sequence resulted in selective calpain
1 inhibition, while the analogue containing -Eps did not have any
L
D
activity. These findings are in agreement with the literature
describing that the stereochemistry of the epoxy-group could also
have an effect on the activity. The analysis of inhibitory activity of
Fig. 1. A. Binding of compound 2 (NH2-Thr-Trp-Leu-(L-Eps)-Ser-Pro-Pro-Pro-Ser-NH2)
(shown in blue with non-polar hydrogen atoms omitted for clarity) to the substrate
binding cleft of calpain 2 (orange surface). The corresponding segment of calpastatin is
also shown as a ribbon (purple). B. H-bond network formed by the binding of com-
pound 2. C. Overlay of compound 2 and compound 11 bound in opposing orientation.
(For interpretation of the references to color in this figure legend, the reader is referred
to the web version of this article.)
compounds with L- or D-Eps modified sequences showed difference
between their potency and selectivity against enzymes, although
compounds 2 and 11 with the same amino acid composition
possessed similar potency against calpain 1 and 2 as well as
cathepsin B. It seems that the binding mode of these compounds
makes inhibitory potency insensitive for the stereochemistry of the
The energetically most favorable conformation of compound 11
is of reversed orientation. While the epoxysuccinyl part of the
peptide analogue is stabilized by an H-bond from the backbone
amide group of Gly198, just as in case of compound 2, the reversed
epoxysuccinyl group. While compound 16 with
inhibitory activity exclusively against calpain 2, its analogue with
Eps, compound 7, inhibited both calpain 1 and cathepsin B, but not
calpain 2.
D-Eps had a marked
L-

ꢀ
L.E. Dokus et al. / European Journal of Medicinal Chemistry 82 (2014) 274e280
279
Our results confirmed the findings of Courrier et al. [18] that at
position P2 the Leu residue was preferred. Although its substitution
to Val (5) resulted in a moderate calpain 2 inhibitory activity,
without inhibiting the other enzymes. Thr (4) did not result in in-
hibition in this position.
inhibitors (1 and 16, respectively) are selective also over cathepsin
B (4- and 11.8-fold, respectively). It is interesting to note that
compounds with inhibitory activity against cathepsin B have also
inhibited calpain 1 (compounds 2, 7, 11).
Taken together, the results presented in this communication
demonstrate that it is possible to design novel calpain inhibitory
peptide analogues based on a substrate sequence by insertion of
epoxysuccinyl moiety at position P1 even into a long, 9-mer
sequence. It is also described that the inhibitory activity does not
depend only on the amino acid sequence, but also on the stereo-
chemical arrangement of the epoxysuccinyl group. This finding was
further supported by the results of the molecular modeling studies
suggesting that a change in the stereochemistry of the epox-
ysuccinyl group might reverse the orientation of the inhibitor
within the substrate binding groove of calpain 2. Calculations also
showed that specifically bound inhibitors could form a number of
well-positioned H-bonds with the protein matrix including both
side chain and main-chain contacts, originating from an extensive
surface of the protein.
The replacement of Pro in position P3 had dramatic effect on the
inhibitory potency. In the L-series, substitution with Trp or Ser,
compound 2 and 3 resulted in calpain 2 inhibitors. It should be
noted that compound 2 with Trp was found to be the most active
nonapeptide analogue. These changes at the same time decreased
the inhibitory activity towards the calpain 1 enzyme, and only the
Trp analogue (2) expressed cathepsin B inhibitory activity. Our
observation about the preference of Trp residue at this position is
also in harmony with findings in the literature [18]. Trp was the
most preferred amino acid next to the Arg at this position, and Pro
was one of the most unfavorable. In the D-series, the substitutions
had little effect on the inhibitory potency of the original sequence.
Only the presence of Trp residue at position P3 (11) and of Thr at
0
position P1 (16) resulted in inhibitory activity. The former com-
pound was active on all three enzymes (calpain 1 and 2, cathepsin
Furthermore, with compounds 1 and 16 we could demonstrate
that there is a possibility of developing selective calpain 1 and 2
inhibitors by the appropriate combination of epoxysuccinyl moiety,
simultaneous extension of peptide parts at both sides, and topo-
logical considerations. Further studies should be performed to
investigate the effect of these inhibitors on intracellular calpains.
Based on our preliminary studies, these inhibitors are not taken up
by cells, therefore in order to study their effect on intracellular
enzymes cell-penetrating conjugate derivatives will be prepared.
B), while the latter was selective on calpain 2. The appearance of
0
Thr residue at position P1 in the
L-Eps analogue (7) exhibited an
opposite effect. This compound inhibited calpain 1 and cathepsin B,
but not calpain 2.
4.2. Comparison of the sequences of inhibitory peptide analogues
with azaGly or L/D-Eps
When the sequence of the peptide TPLKSPPPS was modified by
incorporation of azaglycine instead of Lys, the compound inhibited
both calpain 1 and 2 [14]; while the presence of L/D-Eps in this
Author contributions
position resulted in no activity. These observations suggest that the
binding mode of two the structures can be different. While the
azaglycine modification preserved the manner of substrate binding,
Designed the experiments: ZB, AT, FH, DKM.
Performed the experiments: ZB, LED, AT, DKM.
Analyzed and interpreted the data: ZB, LED, AT, DKM, FH.
Wrote the paper: LED, ZB, AT, DKM, FH.
the incorporation of L- or D-Eps moiety had no such influence. There
are two main differences between the structures of molecules with
azaamino acid or with epoxysuccinyl group. Incorporation of an
azaamino acid does not alter the length of the peptide backbone,
while the presence of an epoxysuccinyl moiety increases the dis-
tance with an additional methylene group. The second (major)
difference is in the orientation of amino acid side chains originating
from the reverse order of amino acid residues. In case of azapep-
tides only N to C direction occurs in the peptide sequence. In sharp
contrast, in compounds with epoxysuccinyl group the orientation
of the amide bonds in the “N-terminal” (P4eP2 direction) part is
Revised critically the paper: all authors.
All authors read and approved the final version of the
manuscript.
Conflict of interest
All authors declare no conflict of interest.
Acknowledgments
reverse. These differences between the azaGly and
substituted analogues may result in markedly changed inhibitory
activity even in case of an “identical” peptide sequence.
L-Eps
This study was supported by grants: OTKA PD-83923, PD-
ꢀ ꢀ
101095, K 104385 and GVOP-3.2.1-2004-04-0352/3.0. Z. Banoczi
ꢀ
acknowledges the support of Bolyai Janos Scholarship of the Hun-
garian Academy of Sciences. The authors are thankful for the proof-
reading to Dr. Katalin Uray.
4.3. Comparison of selectivity of inhibitory peptide analogues with
L
/
D-Eps
Appendix A. Supplementary data
In order to study the selectivity of compounds described in this
and earlier published papers [14], the inhibitory activity was
studied not only on calpain 1 and 2, but also on cathepsin B. The
calpain 1 and 2 enzymes are major, ubiquitous isoforms of calpain
with similar substrate preferences [31]. In fact, as our data suggest,
we have identified a specific calpain 2 inhibitor, compound NH2-
Supplementary data related to this article can be found at http://
dx.doi.org/10.1016/j.ejmech.2014.05.058.
References
Thr-Pro-Leu-(
at least 11.8-fold selectivity over calpain 1. Another nonapeptide
analogue, NH2-Thr-Pro-Leu-( -Eps)-Ser-Pro-Pro-Pro-Ser-NH2 (1)
could inhibit calpain 1 with at least 4-fold selectivity.
Here we also describe that the incorporation of the epox-
ysuccinyl moiety could result in compounds with cathepsin B
D
-Eps)-Thr-Pro-Pro-Pro-Ser-NH2 (16) which exhibits
[1] H. Sorimachi, S. Ishiura, K. Suzuki, Structure and physiological function of
calpains, Biochem. J. 32 (1997) 721e732.
[2] H. Sorimachi, K. Suzuki, The structure of calpain, J. Biochem. (Tokyo) 129
(2001) 653e664.
[3] D.E. Goll, V. Thompson, F.H. Li, W. Wei, J. Cong, The calpain system, Physiol.
Rev. 83 (2003) 731e801.
[4] A. Glading, D.A. Lauffenburger, A. Wells, Cutting to the chase: calpain pro-
teases in cell motility, Trends Cell Biol. 12 (2002) 46e54.
L
selectivity. The above mentioned calpain
1 and 2 selective

ꢀ
L.E. Dokus et al. / European Journal of Medicinal Chemistry 82 (2014) 274e280
280
[5] K. Sato, S. Kawashima, Calpain function in the modulation of signal trans-
P. Petukhov, R.J. Gregory, Thatcher design, synthesis, and optimization of
novel epoxide incorporating peptidomimetics as selective calpain inhibitors,
J. Med. Chem. 56 (2013) 6054e6068.
duction molecules, Biol. Chem. 382 (2001) 743e751.
[6] T.W. Rutledge, S.W. Whiteheart, SNAP-23 is a target for calpain cleavage in
activated platelets, J. Biol. Chem. 277 (2002) 37009e37015.
[21] M. Tamai, C. Yokoo, M. Murata, K. Oguma, K. Sota, E. Sato, Y. Kanaoka, Efficient
[7] S.J. Franco, A. Huttenlocher, Regulating cell migration: calpains make the cut,
J. Cell Sci. 118 (2005) 3829e3838.
synthetic
method
for
ethyl
(þ)-(2S,3S)-3-[(S)-3-methyl-1-(3-
methylbutylcarbamoyl)butylcarbamoyl]-2-oxiranecarboxylate (EST),
a
new
[8] Y. Huang, K.K. Wang, The calpain family and human disease, Trends Mol. Med.
7 (2001) 355e362.
inhibitor of cysteine proteinases, Chem. Pharm. Bull. 35 (1987) 1098e1104.
[22] J.S. Elce, C. Hegadorn, S. Gauthier, J.W. Vince, P.L. Davies, Recombinant
calpain-II e improved expression systems and production of a 105A active-
site mutant for crystallography, Protein Eng. 8 (1995) 843e848.
[23] J.F. Morrison, The slow-binding and slow, tight-binding inhibition of enzyme-
catalyzed reactions, Trends Biochem. Sci. 7 (1982) 102e105.
[9] E.M. Saez, R. Ramirez-Lorca, J.F. Moron, A. Ruiz, The therapeutic potential of
the calpain family: new aspects, Drug Discov. Today 11 (2006) 917e923.
[10] O.I. Donkor, Calpain inhibitors: a survey of compounds reported in the patent
and scientific literature, Expert Opin. Ther. Pat. 21 (5) (2011) 601e636.
ꢀ
ꢀ ꢀ
[11] P. Tompa, P. Buzder-Lantos, A. Tantos, A. Farkas, A. Szilagyi, Z. Banoczi,
F. Hudecz, P. Friedrich, On the sequential determinants of calpain cleavage,
J. Biol. Chem. 279 (2004) 20775e20785.
[24] C.G. Knight, in: A.J. Barrett, G. Salvesen (Eds.), Proteinase Inhibitors, Elsevier,
Amsterdam, 1986, pp. 23e51.
[25] R.A. Hanna, R.L. Campbell, P.L. Davies, Calcium-bound structure of calpain and
its mechanism of inhibition by calpastatin, Nature 456 (2008) 409e412.
[26] J.L. Banks, H.S. Beard, Y. Cao, A.E. Cho, W. Damm, R. Farid, A.K. Felts,
T.A. Halgren, D.T. Mainz, J.R. Maple, R. Murphy, D.M. Philipp, M.P. Repasky,
L.Y. Zhang, B.J. Berne, R.A. Friesner, E. Gallicchio, R.M. Levy, Integrated
Modeling Program, Applied Chemical Theory (IMPACT), J. Comp. Chem. 26
(2005) 1752.
[27] T. Moldoveanu, R.L. Campbell, D. Cuerrier, P.L. Davies, Crystal structures of
calpain-E64 and -leupeptin inhibitor complexes reveal mobile loops gating
the active site, J. Mol. Biol. 343 (2004) 1313e1326.
ꢀ
ꢀ
[12] Z. Banoczi, A. Alexa, A. Farkas, P. Friedrich, F. Hudecz, Novel cell-penetrating
calpain substrate, Bioconjugate Chem. 19 (2008) 1375e1381.
ꢀ
ꢀ
ꢀ
ꢀ ꢀ
[13] I. Vilagi, D.S. Kiss, A. Farkas, S. Borbely, K. Tarnok, K. Halasy, Z. Banoczi,
F. Hudecz, F. Friedrich, Synthetic calpain activator boosts neuronal excitability
without extra Ca2þ, Mol. Cell Neurosci. 38 (2008) 629e636.
ꢀ
ꢀ
ꢀ
[14] Z. Banoczi, A. Tantos, A. Farkas, Z. Majer, E.L. Dokus, P. Tompa, F. Hudecz, New
m-calpain substrate-based azapeptide inhibitors, J. Pept. Sci. 19 (2013)
370e376.
[15] C.J. Powers, L.J. Asgian, D.O. Ekici, E.K. James, Irreversible inhibitors of serine,
cysteine, and threonine proteases, Chem. Rev. 102 (2002) 4639e4750.
[16] K.M.T. Hanada, M. Yamagishi, S. Ohmura, J. Sawada, I. Tanaka, Isolation and
characterization of E-64, a new thiol protease inhibitor, Agric. Biol. Chem. 42
(1978) 523e528.
[28] H. Tsuge, T. Nishimura, Y. Tada, T. Asao, D. Turk, V. Turk, N. Katunuma, Inhi-
bition mechanism of cathepsin L-specific inhibitors based on the crystal
structure of papain-CLIK148 complex, Biochem. Biophys. Res. Commun. 266
(1999) 411e416.
[17] C. Parkers, A.A. Kembhavi, J.A. Barrett, Calpain inhibition by peptide epoxides,
Biochem. J. 230 (1985) 509e516.
[18] D. Cuerrier, T. Moldoveanu, L.R. Campbell, J. Kelly, B. Yoruk, L.H.S. Verhelst,
D. Greenbaum, M. Bogyo, L.P. Davies, Development of calpain-specific inac-
tivators by screening of positional scanning epoxide libraries, J. Biol. Chem.
282 (2007) 9600e9611.
[19] M.J. Pfizer, I. Assfalg-Machleidt, W. Machleidt, L. Moroder, N. Schaschke,
Primed-site probing of papain-like cysteine proteases, Int. J. Pept. Res. Ther. 13
(2007) 93e104.
[20] T.I. Schiefer, S. Tapadar, V. Litosh, M. Siklos, R. Scism, T.G. Wijewickrama,
P.E. Chandrasena, V. Sinha, E. Tavassoli, M. Brunsteiner, M. Fa', O. Arancio,
[29] A. Yamamoto, T. Tomoo, K. Matsugi, T. Hara, Y. In, M. Murata, K. Kitamura,
T. Ishida, Structural basis for development of cathepsin B-specific
noncovalent-type inhibitor: crystal structure of cathepsin B-E64c complex,
Biochim. Biophys. Acta 1597 (2002) 244e251.
[30] T. Moldoveanu, K. Gehring, D.R. Green, Concerted multi-pronged attack by
calpastatin to occlude the catalytic cleft of heterodimeric calpains, Nature 456
(2008) 404e408.
[31] K. Takahashi, Calpain substrate specificity, in: R.L. Mellgren, T. Murachi (Eds.),
Intracellular Calcium-dependent Proteolysis, Fla, CRC Press Inc, Boca Raton,
1990, pp. 56e74.