Investigation into the P3 binding domain of m-calpain using photoswitchable diazo- and triazene-dipeptide aldehydes: New anticataract agents

Article abstract of DOI:10.1021/jm061455n

The photoswitchable N-terminal diazo and triazene-dipeptide aldehydes 8a-d, 10a,b, and 17a,b present predominantly as the (E)-isomer, which purportedly binds deep in the S₃ pocket of calpain. All compounds are potent inhibitors of m-calpain, wi

Full text of DOI:10.1021/jm061455n

2916
J. Med. Chem. 2007, 50, 2916-2920
Brief Articles
Investigation into the P3 Binding Domain of m-Calpain Using Photoswitchable Diazo- and
Triazene-dipeptide Aldehydes: New Anticataract Agents
Andrew D. Abell,*,† Matthew A. Jones,† Axel T. Neffe,†,‡ Steven G. Aitken,† Thomas P. Cain,† Richard J. Payne,†
Stephen B. McNabb,† James M. Coxon,† Blair G. Stuart,† David Pearson,† Hannah Y.-Y. Lee,§ and James D. Morton§
Department of Chemistry, UniVersity of Canterbury, PriVate Bag 4800, Christchurch, New Zealand, and Agriculture and Life Sciences
DiVision, Post Office Box 84, Lincoln UniVersity, Canterbury, New Zealand
ReceiVed December 20, 2006
The photoswitchable N-terminal diazo and triazene-dipeptide aldehydes 8a-d, 10a,b, and 17a,b present
predominantly as the (E)-isomer, which purportedly binds deep in the S3 pocket of calpain. All compounds
are potent inhibitors of m-calpain, with 8b being the most active (IC50 of 35 nM). The diazo-containing
inhibitors 8a, 8c, and 10a were irradiated at 340 nm to give a photostationary state enriched in the (Z)-
isomer, and in all cases, these were less active. The most water soluble triazene 17a (IC50 of 90 nM) retards
calpain-induced cataract formation in lens culture.
Introduction
Calpains are Ca2+-dependent cysteine proteases that are found
in many types of organisms, including mammals, plants, and
protozoa.1 The most common isoforms, µ- and m-calpain,
require µmol and mmol concentrations of calcium, respectively,
for activation in vitro.2 Both isoforms are heterodimers, consist-
ing of a large 80 kDa catalytically active subunit and a smaller
28 kDa regulatory subunit.3
The over-activation of calpain is central to a number of
medical conditions associated with cellular damage, including
traumatic brain injury, stroke, and cataract.4 In the case of
cataract, calpain catalyzes the breakdown of the major lens
proteins (crystallins), resulting in clouding of the lens and
ultimately blindness.5 Specific calpain inhibitors are thus
attractive therapeutic agents;6 however, their design has been
somewhat restricted by the, until recent, lack of available X-ray
data.7 Calpains are known to cleave a large number of
substrates,8 and something of their specificity is known. This
helps in inhibitor design. Calpains recognize unordered se-
quences located between globular domains of proteins, and the
preferred amino acids that neighbor the cleavage sites are
known.6
The S1, S2, and S3 binding sites of calpain that bind these
preferred sequences are located deep within a canyon-like
groove defined by the steep surfaces of domain I on one side,
and domain II on the other.9 This groove is tightest around the
S1 site (<7 Å across, approx 15 Å deep), with the S2 site being
much deeper and generally being occupied by branched
hydrophobic residues. The groove ends in a wider opening (>15
Å) at the S3 site, which is thought to be accommodated by an
N-terminal aryl group10 within dipeptide-based inhibitors, such
as SJA6017 1 (Figure 1), and a range of hydrophobic or basic
groups in the cases of tri- and larger sequences.11 These calpain
Figure 1. Calpain inhibitors SJA6017 1, leupeptin 2, and SNJ1715 3
for which X-ray structures of each bound in µ-calpain have been
determined.
inhibitors most likely bind in an extended â-strand conformation
that spans approx 15 Å of the complementary active site cleft,
but little work has been done to confirm this and, in particular,
to understand interactions in the S3 binding pocket.
An X-ray structure of leupeptin 2, with engineered µ-calpain,
shows this conformation to be defined by two key hydrogen
bonds between the NH and carbonyl groups of Leu (P2) and
Gly208 and also an additional hydrogen bond between the NH
of the P1-P2 amide bond and Gly271 (Figure 2).7a More recent
crystal structures, for example, calpain bound prodrug 3, confirm
the importance of the same series of hydrogen bonds.7b-c
In this paper we report the preparation of a series of (E)-
diazodipeptide aldehydes (amides 8a-d and sulfonamides
10a,b) that contain an N-terminal diazo group designed to extend
deep into the S3 binding pocket and a C-terminal aldehyde for
attachment to the active site cysteine. The N-terminal diazo
groups of (E)-8a,c and (E)-10a were photochemically isomerized
to give an alternative photostationary state in which the (Z)-
isomer predominates. These mixtures, enriched in either the (E)-
or (Z)-isomer, were assayed against m-calpain to further explore
and define the P3 binding pocket. This new class of N-terminal
group was extended to more water soluble triazene-dipeptide
* To whom correspondence should be addressed. Tel.: +64-3-364 2118.
Fax: +64-3-364 2110. E-mail: andrew.abell@canterbury.ac.nz.
† University of Canterbury.
‡ Current address: GKSS Research Centre, Institute of Polymer Research,
Kantstrasse 55, 14513 Teltow, Germany.
§ Lincoln University.
10.1021/jm061455n CCC: $37.00 © 2007 American Chemical Society
Published on Web 05/12/2007

Brief Articles
Journal of Medicinal Chemistry, 2007, Vol. 50, No. 12 2917
Table 1. Inhibition of m-Calpain by Different Photostationary States of
the Diazo and Triazene-dipeptide Aldehydes 8a-d, 10a,b, and 17a,b
IC50
(nM)
compounda
8a
8a (irradiated)
8b
E/Z
subb
X
Y
R1
4.8:1
1:4.3
5.4:1
4.9:1
1:4
5.4:1
3.3:1
1:3.7
7:1
4
4
4
3
3
3
4
4
4
4
4
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
CO
CO
CO
CO
CO
CO
SO2
SO2
SO2
SO2
SO2
Leu
Leu
Phe
Leu
Leu
Phe
Leu
Phe
Phe
Leu
Leu
45
175
35
Figure 2. Depiction of noncovalent interactions between calpain and
aldehyde-based inhibitors bound in an antiparallel â-strand conforma-
tion.9 Hydrogen bonds from the carbonyl group of Gly208, the NH group
of Gly208, and the carbonyl group of Gly271 are labeled A, B, and C,
respectively.
8c
75
8c (irradiated)
8d
10a
105
170
40
100
90
10a (irradiated)
10b
17a
17b
Ph
pip
morp
Scheme 1. Synthesis of 8a-d and 10a,ba
23:1
23:1
90
420
a The major isomer before irradiation is assigned as the thermodynami-
cally more stable (E)-isomer, based on literature precedence.12 b Substitution
position.
Scheme 2. Synthesis of the Triazenes 17a,ba
a Reagents and conditions: (a) leucinol or phenylalaninol, HATU,
DIPEA, DMF; (b) 4 M HCl/dioxane; (c) HATU, DIPEA, DMF, 3- or
4-phenyldiazobenzoic acid, respectively; (d) DMF, Et3N, 4-phenyldiaz-
obenzenesulfonyl chloride; (e) SO3-Pyr, DMSO, DCM, DIPEA.
a Reagents and conditions: (a) 4-nitrobenzene sulfonyl chloride, DMF,
Et3N; (b) PtO2, H2, MeOH/DCM; (c) NaNO2, 6 M HCl/MeOH, then
morpholine or piperidine; (d) NaOH, H2O/THF; (e) leucinol, HATU,
DIPEA, DMF; (f) SO3-Pyr, DMSO, DCM, DIPEA.
aldehydes 17a,b, with 17a being assayed in lens culture to
determine its ability to arrest the development of calpain-induced
cataract.
The major isomer in each case is assigned the thermodynami-
cally more stable (E)-isomer based on literature precedence.12
Results and Discussion
The sulfonamides 10a,b were synthesized (Scheme 1) by
separately treating 6a and 6b with 4-phenyldiazobenzenesulfonyl
chloride to give 9a and 9b in 60 and 50% yield, respectively.
Oxidation with SO3-pyridine gave aldehydes 10a,b, each in 70%
yield. Both the (E)- and (Z)-isomers were observed for 9a and
10a,b, with ratios (E/Z) of 6.5:1, 3.3:1, and 7:1, respectively.
However, little of the (Z)-isomer was observed for 9b, with a
ratio (E/Z) of >20:1.
Compounds 8a-d were prepared (Scheme 1) by coupling
Boc-valine 4 to leucinol and phenylalaninol to give 5a and 5b,
respectively. Deprotection of 5a,b with 4 M HCl/dioxane gave
6a,b, which were separately coupled to 3- and 4-phenylazoben-
zoic acid to give 7a-d in 29-67% yield. Subsequent oxidation
with SO3-pyridine gave the desired aldehydes 8a-d (52-89%).
The azobenzenes 7a-d and 8a-d had (E/Z) ratios of 3.7:1,
3.5:1, 13:1, and 9.2:1 for 7a-d and 4.8:1, 5.4:1, 4.9:1, and 5.4:1
For the synthesis of triazenes 17a,b (Scheme 2) the valine
methyl ester 11 was treated with 4-nitrobenzenesulfonyl chloride
1
for 8a-d, respectively, as determined by H NMR (Table 1).

2918 Journal of Medicinal Chemistry, 2007, Vol. 50, No. 12
Brief Articles
to give 12 in 73% yield. Reduction of the nitro group with PtO2
in a hydrogen atmosphere at 120 psia, with a mixture of
dichloromethane and methanol (1:1) as solvent, gave 13 in 81%
yield. A lower hydrogen pressure (20 psi), even with an
extended reaction time, resulted in a decreased yield of 13
(63%). The reduction was also carried out in ethyl acetate (20
psi), but this solvent resulted in an even lower yield of 13 (30%).
The amine 13, converted to its diazonium salt on treatment with
NaNO2 in a 1:1 mixture of 6 M aqueous HCl and methanol,
was treated with piperidine or morpholine to give the triazenes
14a,b in 49% and 65% yield, respectively. The methyl esters
of 14a,b were separately hydrolyzed with sodium hydroxide in
THF/water (1:1) to give 15a,b in 82% and 86% yields,
Figure 3. (a) â-Strand backbone conformations of (E)-8a, (E)-8b, (E)-
8c, (Z)-8c, (E)-10a, (E)-17a, and (E)-17b, resulting from the induced-
fit modeling. (b) Induced-fit overlay of the S3 subsite of calpain with
docked (E)-10a (blue) and (Z)-10a (red).
1
respectively. H NMR showed the (E/Z) ratios to be 25:1 and
18:1 for the piperidine series 14a-15a and 10.7:1 and 9.7:1
for the morpholine series 14b-15b. The carboxylic acids 15a,b
were coupled to leucinol to give 16a,b in 58% and 49%, with
an (E/Z) ratio of 25:1 and 24:1. A final oxidation gave the
desired aldehydes 17a,b in 55% and 59% yield, with an (E/Z)
ratio of 23:1 in each case (Table 1).
Samples of the dipeptidic aldehydes 8a-d, 10a,b, and 17a,b,
consisting of predominantly the (E)-isomer, were assayed against
m-calpain using a fluorescence-based assay13 to determine in
vitro potency, and the results are summarized in Table 1. The
initial (E)/(Z)-isomer mixtures for 8a, 8c, and 10a were
irradiated with ultraviolet light (500 W mercury arc lamp
through a UV filter with a narrow wavelength band centered at
340 nm) to give samples enriched in the (Z)-isomer [1:4.3, 1:4,
and 1:3.7 mixture of (E)- and (Z)-isomers, respectively, Table
1]. These new photostationary states of 8a, 8c, and 10a (Table
1) were also assayed against m-calpain to further assess the
available space in the S3 binding pocket, to which the diaz-
obenzene groups purportedly bind as revealed by molecular
modeling.14
µ-calpain construct are shown in Figure 3a. This is consistent
with published crystal structures of µ-calpain with inhibitor
bound.7 The docked poses for (Z)-8a, (E)-8d, (Z)-10a, and (E)-
10b (not shown in Figure 3a) also revealed a â-strand
conformation of the inhibitor but with only two of the three
hydrogen bonds apparent. The model also predicts the aldehyde
group in all cases to be in close proximity to the active site
cysteine, as required for mechanism-based inhibition.
The inhibitory activity of the isomeric mixtures of 8a, 8c,
and 10a decreased, on irradiation, by a factor of 3.9, 1.4, and
2.5, respectively. For example, the inhibitory activity of 8a
decreased from 45 nM for a sample containing 83% (E)-isomer
to 175 nM for a mixture containing 4-fold less of the (E)-isomer
(i.e., 19% (E)-isomer with 81% (Z)-isomer), see Table 1. The
(E)-isomers are significantly more active in each case,18
reflecting a better fit in the S3 binding pocket associated with
their geometry and dipole moment. In fact, modeling suggests
that the diazo group of (Z)-10a, unlike (E)-10a, appears not to
bind in the S3 binding pocket, but rather interacts with a
hydrophobic patch on the mobile loop that defines the calpain
active site (Figure 3b). This is consistent with the observed
higher potency of the photostationary state enriched in the (E)-
isomer (see Table 1 and previous discussion). A photoisomer-
izable N-terminal diazobenzene of this type provides both a
useful probe for mapping the size and geometry of the S3 binding
domain and evidence for enhancing interactions between this
pocket and a P3 aryl group of an inhibitor. An empirical
preference for the interaction of an aryl group with S3 has been
previously suggested,7a and our study supports this observation.
The in vivo potential of these highly potent inhibitors was
next determined by assessing their ability to retard calpain-
induced cell damage in sheep lenses cultured in Eagle’s minimal
essential medium. Inhibitor 17a was chosen for study because
of its superior water solubility, as compared to 8a-d and 10a,b,
and also its high in vitro potency, as compared to the other
triazene 17b. The triazene moiety of 17a is also amenable to
salt formation that makes it attractive as a potential drug
candidate.
The most active dipeptidic aldehyde mixture 8b had an IC50
value of 35 nM, which is significantly more potent than the
calpain inhibitor SJA6017 1 (IC50 ) 80 nM in our assay).15
The inhibitor mixtures 8a (45 nM), 8c (75 nM), 10a (40 nM),
and 17a (40 nM) are also particularly potent. Of all the sample
mixtures, only 8d (170 nM) and 17b (420 nM) are less potent
than 1 (Table 1).
The S1 pocket can accommodate both Leu and Phe at P1;
however, there does seem to be some preference for Leu at this
position; compare assay results for 8a,b, 8c,d, and 10a,b (Table
1). With regard to potency, there does not seem to be any
consistent preference for an amide or sulfonamide linker;
compare the activity of 8a-10a and 8b-10b (Table 1).
Inhibitors with 4-substitution of the N-terminal aryl group (8a,b)
are more potent than those with 3-substitution (8c,d).
The triazenes 17a,b are less potent than the diazo compounds
8a-d and 10a,b. The planar and extended aromatic nature of
the (E)-diazo group of 8a,b and 10a,b appears to allow for better
interaction with this flat S3 binding pocket, when compared to
the corresponding triazenes 17a,b. This is consistent with the
lower potency of the triazenes 17a,b (see Table 1). The lesser
potency of 17b may reflect the presence of an electronegative
oxygen at the N-terminus, which would be expected to disfavor
binding in the hydrophobic S3 pocket.
Inhibitor 17a (0.8 µΜ) was added to one lens from each pair
of sheep lenses in culture media. Two hours later calcium was
added to all lenses to activate the constituent calpains and hence
induce cataract formation. After 24 h, all lenses were photo-
graphed and the opacity graded; see Figure 4 for representative
examples. Lenses treated with calcium only (e.g., lens 1 in
Figure 4) clearly showed the opacity associated with cataract
formation; however, lenses treated with calcium in the presence
of 17a (e.g., lens 2) remained essentially clear, as revealed by
the reference grid placed behind each lens. The loss of
transparency was significantly reduced by 17a (p < 0.005) in
Induced fit modeling16 shows each µ-calpain docked inhibitor
adopts a â-strand conformation.17 The poses of (E)-8a, (E)-8b,
(E)-8c, (Z)-8c, (E)-10a, (E)-17a, and (E)-17b docked with a
a Abbreviations: DIPEA, N,N-diisopropylethylamine; HATU, O-(7-
azabenzotriazole-1-yl)-N,N,N′,N′-tetramethyluronium hexafluorophosphate;
psi, pounds of force per square inch.

Brief Articles
Journal of Medicinal Chemistry, 2007, Vol. 50, No. 12 2919
CDCl3) 0.87 (3H, d, J ) 6.7 Hz, CH3), 0.95 (3H, d, J ) 6.8 Hz,
CH3), 1.73 (6H, mc, 3 × CH2), 2.02-2.04 (1H, m, CH(CH3)2),
3.45 (3H, s, OCH3), 3.73 (1H, dd, J ) 9.9 Hz and J ) 5.1 Hz,
CHCH(CH3)2), 3.85 (4H, mc, 2 × CH2), 5.06 (1H, d, J ) 9.9 Hz,
NH), 7.48 (2H, app d, CHAr), 7.75 (2H, app d, CHAr). Selected
data for minor (Z)-isomer from mixture: 1H NMR (500 MHz,
(CD3)2SO) 7.55 (2H, app d, CHAr), 7.82 (2H, app d, CHAr). Data
from mixture: 13C NMR (75 MHz, (CD3)2SO) 18.3 (CH3), 18.8
(CH3), 23.6 (CH2pip), 30.5 (CH(CH3)2), 51.6 (OCH3), 61.5 (CHCH-
(CH3)2), 120.0 (CHAr), 127.8 (CHAr), 136.6 (CSO2), 153.3 (CNdN),
171.3 (CO2Me); Anal. Calcd for C17H26N4O4S: C, 53.38; H, 6.85;
N, 14.65%. Found: C, 53.66; H, 6.86; N, 14.52%.
(S)-3-Methyl-2-[4-(piperidin-1-ylazo)-benzenesulfonylamino]-
butyric Acid (15a). To a solution of ester 14a in THF/H2O ([3:1],
1 mL/10 mg) was added NaOH (2.5 equiv), and the solution was
heated to 65 °C with stirring overnight. Concentration in vacuo
gave a crude orange solid that was partitioned between ethyl acetate
(100 mL) and water (100 mL). The aqueous layer was acidified to
pH 2 with 1 M HCl and extracted with ethyl acetate (2×). The
combined ethyl acetate layers were washed with water and brine
and dried (MgSO4). Concentration in vacuo gave 15a as an orange
solid [inseparable mixture of isomers, 18(E)/1(Z); 1.61 g, 82%]:
mp 117-119 °C; HRMS (ES+) calcd for C16H25N4O4S ([M + H]+),
369.1596; found, 369.1613; νmax (KBr) 3312 (CO2H), 1701 (CO2H),
1355 and 1153 (SO2NH). Data for major (E)-isomer from mix-
ture: 1H NMR (500 MHz, (CD3)2SO) 0.77 (3H, d, J ) 6.9 Hz,
CH3), 0.80 (3H, d, J ) 6.9 Hz, CH3), 1.65 (6H, mc, 3 × CH2),
1.91-1.93 (1H, m, CH(CH3)2), 3.47 (1H, dd, J ) 8.8 Hz and J )
5.9 Hz, CHCH(CH3)2), 3.80 (4H, mc, 2 × CH2), 7.42 (2H, app d,
CHAr), 7.69 (2H, app d, CHAr), 7.89 (1H, d, J ) 9.3 Hz, NH),
12.45 (1H, br s, CO2H). Selected data for minor (Z)-isomer from
mixture: 1H NMR (500 MHz, (CD3)2SO) 7.59 (2H, app d, CHAr),
Figure 4. Calcium-induced cataract in sheep lenses. The scores
represent the average result using three lens pairs. Opacification scores
of 100 ) full opacity, whereas a score of 1 ) clear and transparent.
a paired t-test. It would thus appear that this class of inhibitor,
with an N-terminal group capable of binding deep into the S3
binding pocket, is active in vitro and in vivo (lens assay) and is
thus of interest as a potential drug candidate.
Conclusion
A series of diazo- and triazene-dipeptide aldehydes 8a-d,
10a,b, and 17a,b have been prepared predominantly as the (E)-
isomers. All of the inhibitor mixtures are highly potent against
m-calpain, with the most potent 8b [5.4(E)/1(Z)] having an IC50
value of 35 nM. Photoisomerism of the diazo inhibitors 8a-c
and 10a gave samples enriched in the (Z)-isomer that proved
to be significantly less active.
SAR data is presented, which suggests that an N-terminal
diazo group (8a-d) is favored over a triazene (17a,b).
Furthermore 4-substitution of the N-terminal diazo group (8a,b)
is favored over 3-substitution (8c,d). The triazene of 17a imparts
improved water solubility for in vivo studies and was shown to
arrest the development of calpain-induced cataract formation
in sheep lens culture.
7.75 (2H, app d, CH
Ar), 8.00 (1H, d, J ) 7.8 Hz, NH). Data from
mixture: 13C NMR (75 MHz, (CD3)2SO) 17.9 (CH3), 19.1 (CH3),
23.6 (CH2pip), 30.4 (CH(CH3)2), 61.3 (CHCH(CH3)2), 120.0
(CHAr), 127.8 (CHAr), 137.0 (CSO2), 153.2 (CNdN), 171.3 (CO2H);
Anal. Calcd for C16H24N4O4S: C, 52.16; H, 6.57; N, 15.21%.
Found: C, 52.34; H, 6.61; N, 14.92%.
Molecular modeling predicts that these compounds all bind
in an extended â-strand conformation, as defined by three key
hydrogen bonds to Gly208 and Gly271. This is consistent with
published crystal structures of µ-calpain with the inhibitor
bound.7 The reactive carbonyl in each case is located in close
proximity to the active site cysteine, as would be required for
mechanism-based inhibition. As a consequence, the N-terminal
group of the (E)-diazo dipeptide aldehydes 8a-d and 10a,b
extends deep into the S3 binding pocket.
Finally, we suggest that a calpain inhibitor, the activity of
which can be influenced by irradiation, offers some potential
as a means to control cataracts. The work presented here is a
first step toward this goal.
(S)-N-((S)-1-Hydroxymethyl-3-methyl-butyl)-3-methyl-2-[4-
(piperidin-1-ylazo)benzenesulfonylamino]-butyramide (16a). To
a stirred solution of 15a in dry DMF (1 mL/60 mmol) was added
leucinol (1.1 equiv), DIPEA (4 equiv), and HATU (1.1 equiv). The
solution was stirred at rt overnight, diluted with ethyl acetate (5:
1), washed with water (2×) and brine, and dried over MgSO4.
Concentration in vacuo gave a pale orange solid that was purified
by column chromatography to give 16a as a yellow solid [insepa-
rable mixture of isomers, 25(E)/1(Z); 1.13 g, 58%]: mp 178-
180 °C; Rf ) 0.51 (ethyl acetate/petroleum ether [1:1]); HRMS
(ES+) calcd for C22H38N5O4S ([M + H]+), 468.2644; found,
468.2656; νmax (KBr) 1643 (CdONH), 1320 and 1157 (SdO2NH),
1026 (CsOH). Data for major (E)-isomer from mixture: 1H NMR
(500 MHz, (CD3)2SO) 0.61 (3H, d, J ) 5.9 Hz, CH3), 0.70 (3H, d,
J ) 5.8 Hz, CH3), 0.76 (3H, d, J ) 6.8 Hz, CH3), 0.82 (3H, d, J
) 6.8 Hz, CH3), 0.96-0.98 (1H, m, CHHCH(CH3)2), 1.12-1.14
(2H, m, CH2CH(CH3)2 and CHHCH(CH3)2), 1.66 (6H, mc, 3 ×
CH2), 1.77-1.79 (1H, m, CH(CH3)2), 3.03-3.06 (1H, m, CH2OH),
3.14-3.19 (1H, m, CH2OH), 3.46-3.49 (1H, m, CHCH2OH), 3.80
(4H, mc, 2 × CH2), 4.50-4.53 (1H, m, CHCH(CH3)2), 7.39 (2H,
app d, CHAr), 7.47 (1H, d, J ) 8.3 Hz, NH), 7.68 (2H, app d, CHAr).
Selected data for minor (Z)-isomer from mixture: 1H NMR (500
MHz, (CD3)2SO) 7.57 (1H, d, J ) 7.1 Hz, NH), 7.79 (2H, app d,
CHAr). Data from mixture: 13C NMR (75 MHz, (CD3)2SO) 17.9
(CH3), 19.2 (CH3), 21.8 (CH2CH(CH3)2), 23.2 (CH3), 23.6 (CH2-
pip), 23.8 (CH3), 31.4 (CH(CH3)2), 38.3 (CH2CH(CH3)2), 48.6
(CHCH2OH), 61.3 (CHCH(CH3)2), 63.6 (CH2OH), 119.9 (CHAr),
127.7 (CHAr), 137.1 (CSO2), 153.0 (CNdN), 169.5 (CONH); Anal.
Calcd for C22H37N5O5S‚HCl: C, 52.42; H, 7.60; N, 13.89%.
Found: C, 52.26; H, 7.48; N, 14.05%.
Experimental Section
(S)-3-Methyl-2-[4-(piperidin-1-ylazo)-benzenesulfonylamino]-
butyric Acid Methyl Ester (14a). To a salt-ice-cooled solution of
13 in 6 M HCl/MeOH (50 mL, [5:1]) was added dropwise a solution
of 2.5 M NaNO2 (1.0 equiv) so that the temperature did not exceed
5 °C. The crude orange diazonium salt was cooled to 0 °C, a
solution of piperidine (1.0 equiv) in 1 M HCl (1.0 equiv) was added,
and the mixture was stirred for 30 min. To the resulting homoge-
neous solution was added solid Na2CO3 (caution!) to give a bright
orange precipitate. The precipitate was collected by filtration,
dissolved in ethyl acetate (50 mL), washed with water and brine,
and dried (MgSO4). Concentration in vacuo gave a bright orange
oil that was purified by column chromatography to give 14a as a
pale yellow solid [inseparable mixture of isomers, 25(E)/1(Z); 2.05
g, 49%]: mp 125-127 °C; Rf ) 0.65 (ethyl acetate/petroleum ether
[2:1]); HRMS (ES+) calcd for C17H27N4O4S ([M + H]+), 383.1771;
found, 383.1753; νmax (KBr) 1738 (CO2Me), 1346 and 1142 (SO2-
NH). Data for major (E)-isomer from mixture: 1H NMR (500 MHz,
(S)-N-((S)-1-Formyl-3-methyl-butyl)-3-methyl-2-[4-(piperidin-
1-ylazo)benzenesulfonylamino]-butyramide (17a). To a stirred
ice-cooled solution of alcohol 16a and 4 equiv of DIPEA in DCM/

2920 Journal of Medicinal Chemistry, 2007, Vol. 50, No. 12
Brief Articles
(6) Neffe, A. T.; Abell, A. D. Developments in the Design and Synthesis
of Calpain Inhibitors. Curr. Opin. Drug DiscoVery DeV. 2005, 8,
684-700.
DMSO (1:1, 1 mL/30 mmol) was added dropwise a solution of 4
equiv of sulfur trioxide pyridine complex in DMSO (1 mL/60
mmol). Stirring was continued in ice for 2 h at which time the
mixture was diluted with ethyl acetate (10:1), washed with aqueous
1 M HCl (2×), saturated aqueous NaHCO3 (2×), and brine (10
mL), and dried over MgSO4. Concentration in vacuo gave an orange
oil that was purified by column chromatography to give 17a as a
yellow solid [inseparable mixture of isomers, 23(E)/1(Z); 0.62 g,
55%]: mp 75-77 °C; Rf ) 0.51 (ethyl acetate/petroleum ether
[1:1]); HRMS (ES+) calcd for C22H36N5O4S ([M + H]+), 466.2488;
found, 466.2484; νmax (KBr) 1734 (CHO), 1659 (CONH), 1345
and 1110 (SdO2NH). Data for major (E)-isomer from mixture: 1H
NMR (500 MHz, (CD3)2SO) 0.69 (3H, d, J ) 5.9 Hz, CH3), 0.79
(3H, d, J ) 6.3 Hz, CH3), 0.82 (3H, d, J ) 6.8 Hz, CH3), 0.85
(3H, d, J ) 6.8 Hz, CH3), 1.13-1.18 (1H, m, CH2CH(CH3)2),
1.32-1.37 (2H, m, CH2CH(CH3)2), 1.66 (6H, mc, 3 × CH2), 1.82-
1.86 (1H, m, CH(CH3)2), 3.54-3.57 (1H, m, CHCHO), 3.81 (5H,
mc, 2 × CH2 and CHCH(CH3)2), 7.40 (2H, app d, CHAr), 7.70
(3H, mc, CHAr and SO2NH), 8.23 (1H, d, J ) 6.9 Hz, NH), 9.13
(1H, s, CHO). Selected data for minor (Z)-isomer from mixture:
1H NMR (500 MHz, (CD3)2SO) 7.52 (2H, app d, CHAr), 7.80 (2H,
mc, CHAr), 7.92 (1H, d, J ) 5.2 Hz, SO2NH), 8.25 (1H, d, J ) 6.9
Hz, NH). Data from mixture: 13C NMR (75 MHz, (CD3)2SO) 18.1
(CH3), 19.1 (CH3), 21.4 (CH2CH(CH3)2), 22.9 (CH3), 23.6 (CH2-
pip), 23.8 (CH3), 31.1 (CH(CH3)2), 36.3 (CH2CH(CH3)2), 56.6
(CHCHO), 61.1 (CHCH(CH3)2), 119.9 (CHAr), 127.8 (CHAr), 137.0
(CSO2), 153.1 (CNdN), 170.6 (CONH), 200.9 (CHO); Anal. Calcd
for C22H35N5O4S: C, 56.75; H, 7.58; N, 15.04%. Found: C, 56.40;
H, 7.73; N, 15.44%
(7) (a) Moldoveanu, T.; Campbell, R. L.; Cuerrier, D.; Davies, P. L.
Crystal Structures of Calpain-E64 and -Leupeptin Inhibitor Com-
plexes Reveal MobileLoops Gating the Active Site. J. Mol. Biol.
2004, 343, 1313-1326. (b) Cuerrier, D.; Moldoveanu, T.; J. Inoue,
Davies, P. L.; Campbell, R. L. Calpain Inhibition by R-Ketoamide
and Cyclic Hemiacetal Inhibitors Revealed by X-ray Crystallography.
Biochemistry 2006, 45, 7446-7452. (c) Li, Q.; Hanzlik, R. P.;
Weaver, R. F.; Schoenbrunn, E. Molecular Mode of Action of a
Covalently Inhibiting Peptidomimetic on the Human Calpain Protease
Core. Biochemistry 2006, 45, 701-708.
(8) Tompa, P.; Buzder-Lantos, P.; Tantos, A.; Farkas, A.; Szila´gyi, A.;
Ba´no´czi, Z.; Hudecz, F.; Friedrich, P. On the Sequential Determinants
of Calpain Cleavage. J. Biol. Chem. 2004, 279, 20775-20785.
(9) Note the use of Schechter-Berger nomenclature [Schechter, I.;
Berger, A. On the size of the active site in proteases. Biochem.
Biophys. Res. Commun. 1967, 27, 157]; the residues on the N-terminal
side of the peptide bond that is cleaved are denoted (in order) P1-
Pn, and those on the C-terminus are denoted P1′-Pn′. In turn, the
corresponding subsites on the enzyme are denoted Sn-Sn′.
(10) (a) Donkor, I. O. A survey of Calpain Inhibitors. Curr. Med. Chem.
2000, 7, 1171-1188. (b) Iqbal, M.; Messina, P. A.; Freed, B.; Das,
M; Chatterjee, S.; Yripathy, R.; Yao, M.; Josef, K. A.; Dembofsky,
B.; Dunn, D.; Griffith, E.; Siman, R.; Senadhi, S. E.; Biazzo, W.;
Bozyczko-Coyne, D.; Meyer, S. L.; Ator, M. A.; Bihovsky, R. Subsite
Requirements for Peptide Aldehyde Inhibitors of Human Calpain I.
Bioorg. Med. Chem. Lett. 1997, 7, 539-544.
(11) (a) Inoue, J.; Nakamura, M.; Cui, Y.-S.; Sakai, Y.; Sakai, O.; Hill,
J. R.; Wang, K. W. W.; Yuen, P.-W. Structure-Activity Relationship
Study and Drug Profile of N-(4-Fluorophenylsulfonyl)-L-valyl-L-
leucinal (SJA6017) as a Potent Calpain Inhibitor. J. Med. Chem. 2003,
46, 868-871. (b) Angelastro, M. R.; Mehdi, S.; Burkhart, J. P.; Peet,
N. P.; Bey, P. R-Diketone and R-Keto Ester Derivatives of N-
Protected Amino Acids and Peptides as Novel Inhibitors of Cysteine
and Serine Proteinases. J. Med. Chem. 1990, 33, 11-13. (c) Mehdi,
S.; Angelastro, M. R.; Wiseman, J. S.; Bey, P. Inhibition of the
Proteolysis of Rat Erythrocyte Membrane Proteins by a Synthetic
Inhibitor of Calpain. Biochem. Biophys. Res. Comm. 1988, 157,
1117-1123.
(12) (a) Westmark, P. R.; Kelly, J. P.; Smith, B. D. Photoregulation of
Enzyme Activity. Photochromic, Transition-State-Analog Inhibitors
of Cysteine and Serine Proteases. J. Am. Chem. Soc. 1993, 115,
3416-3419. (b) Harvey, A. J.; Abell, A. D. Azobenzene-Containing,
Peptidyl R-Ketoesters as Photobiological Switches of R-Chymot-
rypsin. Tetrahedron 2000, 56, 9763.
Acknowledgment. We thank Matthew Muir for the supply
of calpain, Janna Nikkel for assistance with the inhibition assay,
and Dr. Quentin MacDonald (QBit NZ) for assistance with
modeling. The research was supported by a University of
Canterbury Fellowship (A.T.N., S.G.A.) and grants from the
New Zealand Public Good Science and Technology Fund
(M.A.J., J.M.N.), the Tertiary Education Commission enterprise
scholarship (H.Y.-Y.L.), the Foundation for Research Science
and Technology (S.B.M.), Douglas Pharmaceuticals, and the
Auckland Centre for Molecular Biodiscovery (B.G.S.).
Supporting Information Available: Experimental details cor-
responding to the synthesis of the compounds described in this
paper, spectral data for all compounds, and assay and molecular
modeling methods. This material is available free of charge via
the Internet at http://pubs.acs.org.
(13) Thompson, V. F.; Saldana, S.; Cong, J.; Goll, D. E. A BODIPY
Fluorescent Microplate Assay for Measuring Activity of Calpains
and Other Proteases. Anal. Biochem. 2000, 279, 170-178.
(14) Assay of irradiated 8a, 8c, and 10a enriched in the (Z)-isomer were
carried out in dim lighting conditions, that is, glassware wrapped in
foil and with the lights turned off.
(15) The calpain inhibitor SJA6017 1 was chosen as a literature standard
to validate our assay procedure and for comparison of inhibitor
potency.
References
(1) Perrin, B. J.; Huttenlocher, A. Calpain. Int. J. Biochem. Cell Biol.
2002, 34, 722-725.
(2) Reverter, D.; Sorimachi, H.; Bode, W. The Structure of Calcium-
Free Human m-Calpain: Implications for Calcium Activation and
Function. Trends CardioVasc. Med. 2001, 11, 222-229.
(3) Goll, D. E.; Thompson, V. F.; Li, H.; Wei, W.; Cong, J. The Calpain
System. Physiol. ReV. 2003, 83, 731-801.
(16) Induced-fit docking using Glide and Prime: Glide, version 4.0;
Schro¨dinger, LLC: New York, 2005. Prime, version 1.5; Schro¨dinger,
LLC: New York, 2005.
(17) The X-ray crystal structure for rat µ-calpain was used for the docking
studies as the m-calpain X-ray crystal structure in an inactivated form.
Structures for sheep m- or µ-calpain have not been published;
however, sequence analysis indicates 94% homology between sheep
and rat m-calpain.
(4) Huang, Y.; Wang, K. K. W. The Calpain Family and Human Disease.
Trends Mol. Med. 2001, 7, 355-362.
(5) (a) David, L. L.; Wright, J. W.; Shearer, T. R. Calpain II Induced
Insolubilization of Lens beta-Crystallin Polypeptides may Induce
Cataract. Biochim. Biophys. Acta 1992, 1139, 210-216. (b) Rob-
ertson, L. J. G.; Morton, J. D.; Yamaguchi, M.; Bickerstaffe, R.;
Shearer, T. R.; Azuma, M. Calpain may Contribute to Hereditary
Cataract Formation in Sheep. InVest. Ophthalmol. Vision Sci. 2005,
46, 4634-4640.
(18) For example, the (E)-isomer of 8a is calculated to be 40-fold more
active than the (Z)-isomer.
JM061455N