6-Acylamino-2-[(ethylsulfonyl)oxy]-1H-isoindole-1,3-diones mechanism-based inhibitors of human leukocyte elastase and cathepsin G: Effect of chirality in the 6-acylamino substituent on inhibitory potency and selectivity

Article abstract of DOI:10.1016/S0968-0896(00)00281-9

Inhibition of human leukocyte elastase(HLE) by a series of 6-acylamino-2-[(ethylsulfonyl)oxy)]-1H-isoindole-1,3-diones was determined and compared to their inhibition of ChT, PPE, and Cat G. The best inhibitor of the series was 6-((1′S)-camphanyl)amino-2-[(ethylsulfonyl)oxy]-1H-isoindole-1,3-dione 5b, with a k_obs/[I]=11,000M^-1 s^-1. This study revealed that HLE shows a preference for the S stereochemistry and tolerates hydrophobic substituents in the S_n′ binding sites. Molecular modeling of noncovalent HLE-inhibitor complexes was used as a tool to investigate our binding model. Buffer stability assays reveal that these compounds are susceptible to hydrolysis at physiological pH. Copyright

Full text of DOI:10.1016/S0968-0896(00)00281-9

Bioorganic & Medicinal Chemistry 9 (2001) 637±645
6
-Acylamino-2-[(ethylsulfonyl)oxy]-1H-isoindole-1,3-diones
Mechanism-based Inhibitors of Human Leukocyte Elastase
and Cathepsin G: E€ect of Chirality in the 6-Acylamino
Substituent on Inhibitory Potency and Selectivity
Lisa M. Vagnoni, Michael Gronostaj and John E. Kerrigan*
Department of Pharmaceutical Chemistry, College of Pharmacy, Rutgers University,
1
60 Frelinghuysen Road, Piscataway, NJ08854, USA
Received 23 August 2000;accepted 13 October 2000
AbstractÐInhibition of human leukocyte elastase(HLE) by a series of 6-acylamino-2-[(ethylsulfonyl)oxy)]-1H-isoindole-1,3-diones
0
was determined and compared to their inhibition of ChT, PPE, and Cat G. The best inhibitor of the series was 6-((1 S)-campha-
À1 À1
nyl)amino-2-[(ethylsulfonyl) oxy]-1H-isoindole-1,3-dione 5b, with a kobs/[I]=11,000 M
s
. This study revealed that HLE shows a
0
preference for the S stereochemistry and tolerates hydrophobic substituents in the Sn binding sites. Molecular modeling of non-
covalent HLE±inhibitor complexes was used as a tool to investigate our binding model. Bu€er stability assays reveal that these
compounds are susceptible to hydrolysis at physiological pH. # 2001 Elsevier Science Ltd. All rights reserved.
Introduction
inhibitors of HLE.11 Enzyme catalyzed hydrolysis of the
inhibitor exposes a reactive functionality which cova-
lently and irreversibly binds to the enzyme. A multitude
of diverse mechanism-based inhibitors have been repor-
ted in the literature. Among these are the 3-alkoxy-7-
Human leukocyte elastase (HLE) is a serine protease
1
produced by polymorphonuclear leukocytes. The pri-
mary role of HLE is the degradation of microorganisms
ingested by leukocytes during phagocytosis. In the pro-
cess, HLE degrades connective tissue proteins like elas-
1
2,13
amino-4-chloroisocoumarins,
1
the saccharin deriva-
15,16
4
tives, the N-((alkylsulfonyl)oxy)succinimides,
17
and
the b-lactams. More recently, researchers have repor-
ted on the coumarinic derivatives, thieno-oxazin-4-
2
tin in pathological states. Extensive elastolysis by HLE
1
8
is primarily prevented by an endogenous inhibitor, a1-
antitrypsin, also known as a1-PI. Depressed levels of
active a1-PI, caused by a genetic de®ciency or cigarette
smoking, can lead to excessive proteolysis of elastin and
result in the disease pulmonary emphysema.3 It is
believed that cigarette smoke causes the oxidation of the
reactive methionine residue of a1-PI, thereby rendering
1
9
20À22
ones, the 1,2,5-thiadiazolidin-3-one-1,1-dioxides,
2
3,24
and the N-((alkylsulfonyl)oxy) phthalimides.
It has
been hypothesized by us and others that the mechanism
of action of the N-sulfonyloxy phthalimides involves the
formation of an imidazole-N-carboxamide covalent com-
plex much like their N-sulfonyloxy succinimide counter-
23,25,26
4À6
it a less e€ective inhibitor of HLE.
parts.
This is achieved through the initial acylation
of the catalytic serine residue followed by the release of a
latent isocyanate. The isocyanate reacts with a catalytic
histidine residue forming an imidazole-N-carboxamide.
One of the current treatments for patients su€ering
from emphysema is high cost a1-PI replacement therapy
7
(
$20,000±$30,000 per patient per year). Recently, much
attention has focused on the synthesis of cost-e€ective,
low molecular weight inhibitors of HLE that may incur
therapeutic utility. Mechanism-based inhibitors
have been explored extensively as potent and speci®c
Our proposed binding model is shown in Figure 1. The
rationale for this assumption is based on the structural
similarity observed between the 6-amino-2-[(ethylsulfo-
nyl)oxy]isoindole-1,3-diones and 7-amino-4-chloro-3-
ethoxyisocoumarins. The binding model of isocoumarin
mechanism-based inhibitors of elastase has been estab-
lished by X-ray crystallography of PPE-isocoumarin
8À10
*
6
Corresponding author. Tel.: +1-732-445-5763;fax: +1-732-445-
312;e-mail: jkerriga@cop.rutgers.edu
2
7À29
and further supported by molecular
complexes
0968-0896/01/$ - see front matter # 2001 Elsevier Science Ltd. All rights reserved.
PII: S0968-0896(00)00281-9

638
L.M. Vagnoni et al. / Bioorg. Med. Chem. 9 (2001) 637±645
diones with bulkier hydrophobic substituents at the 6-
position and the in vitro activity of the derivatives
towards HLE, bovine a-chymotrypsin (ChT), porcine
pancreatic elastase (PPE), and human neutrophil
cathepsin G (Cat G).
Results
Synthesis
Figure 1. Proposed binding model.
The sequence used for the preparation of the 6-acylamino-
2-[(ethylsulfonyl)oxy)]-1H-isoindole-1,3-diones is out-
lined in Scheme 1. The commercially available 4-nitro-
phthalic anhydride was reacted with O-benzylhydroxyl-
amine hydrochloride to form compound 2, which was
subsequently reduced via a transfer hydrogenation method
to give intermediate 3. Sulphonylation of 3 produced the
starting material used for all subsequent reactions. Reac-
tion of 4 with the appropriate acid chloride a€orded the
series of inhibitors discussed in this paper. Compounds,
which were not available as acid chlorides, were re¯uxed
with thionyl chloride in THF to form the intermediate
acid chloride and immediately combined with 4 to form
the corresponding inhibitor.
modeling studies.30,31 In Figure 2, ¯exible alignment of
the two structures using the FlexAlign program in
MOE (Chemical Computing Group, Inc.) shows a
remarkable similarity between the isocoumarins and the
isoindole-1,3-diones. The proposed binding model in
Figure 1 is based upon the isocoumarin model. Kerri-
gan et al. have demonstrated that ¯exible non-bulky
hydrophobic substituents at the 6-position in tandem
with small aliphatic (methyl or ethyl) groupings in the
1
2
selectivity of 6-(substituted)amino-2-[(alkylsulfonyl)oxy]
-(alkylsulfonyl)oxy substituent increase potency and
isoindole-1,3-dione inhibitors of HLE over ChT.2
4,25
In
a similar fashion, hydrophobic groupings in the 7-posi-
tion of isocoumarins increases potency and selectivity of
In vitro assays
1
3,28
these compounds for HLE.
It was observed that
The reaction of an irreversible inhibitor with enzyme is
generally accepted to proceed through a noncovalent
complex (EÀI*) followed by formation of the covalent
complex (EÀI in eq (1)). The inhibitory activity of
compounds 5a±5g toward HLE, ChT, PPE, and Cat G
changing the R group from a methyl to an ethyl group
increased selectivity up to 4-fold for inhibitors with non-
bulky ¯exible chains in the 6-position. These studies
0
support the possibility that the 6-substituent (R ) might
0
32
be binding in the hydrophobic Sn subsites of HLE. As
was determined using the progress curves method. The
pseudo ®rst order rate constant (kobs) was determined
an extension of earlier studies, exploration of the e€ect
of bulky substituents possessing chirality on inhibitor
potency and selectivity for HLE is addressed in this
paper. Herein, we report the synthesis of a series of 6
3
3
1
using eq (3). The solver in Excel (Microsoft Corp.)
was used to simultaneously solve for two unknowns,
A
measured by monitoring a decrease in absorbance at
and kobs, by least squares ®t to eq (3). Inhibition was
1
-
acylamino - 2 - [(ethylsulfonyl)oxy) - 1H - isoindole - 1,3 -
Figure 2. Overlay of aminoisocoumarin with aminoisoindole-1,3-dione.

L.M. Vagnoni et al. / Bioorg. Med. Chem. 9 (2001) 637±645
639
ꢀ
Scheme 1. Reagents: (a) PhCH
ꢁ
2
ONH
2
HCl, TEA, toluene, re¯ux;(b) 5% Pd/C, cyclohexene, EtOH/THF (1/1), re¯ux;(c) ethanesulphonyl chloride,
, EtOAc.
NaHCO
3
, H
2
O, 0 C;(d) RCOCl, TEA, THF -or- NaHCO
3
4
10 nm due to the release of p-nitroaniline upon
Discussion
enzyme-catalyzed hydrolysis of the p-nitroanilide sub-
strate. The results in the form of kobs/[I] values (see eq (2)
are summarized in Table 1.
Inhibition of elastase
Compound 5b was the best inhibitor of HLE in this
series. Compound 5b was a 4-fold better inhibitor of
HLE over ChT and 11-fold more selective for HLE over
Cat G. Compound 5c was the next best inhibitor, which
was only slightly less potent than 5b and very selective
for HLE over the other three enzymes. For purposes of
clarity, we will refer to the chiral center directly adjacent
to the amide linkage of the side chain to the phthalimide
ring when making reference to `R' and `S'. Assays
involving those compounds possessing chirality in their
side chains demonstrated that HLE has a preference for
S stereochemistry over R for chiral centers adjacent to
the amide linkage as can be seen from the increase in
inhibition by 5b over 5a and 5d over 5e. This is con-
k3
Ã
ÀÀ*
E ‡ I )ÀÀ E À I ÀÀ! E À I
ꢀ1†
Ki
k3
kobs=‰IŠ ˆ
ꢀ2†
ꢀ3†
Ki ‡ ‰IŠ
Acalc ˆ A1 À A1eÀꢀkobst†
A =®nal absorbance and Acalc=absorbance, calcu-
1
lated at time t.
1
3
sistent with other ®ndings in the literature. In stark
contrast, chymotrypsin exhibits the opposite pattern in
this series of compounds showing a preference for the R
stereochemistry. Surprisingly, the tosyl-l-Val and tosyl-
L-Phe substituted derivatives interacted poorly with
HLE compared to their inhibition of ChT. In contrast,
Kerrigan et al. found that the tosyl-l-Phe derivative of
Bu€er stability studies
Compounds 4, 5b and 5f were selected for bu€er stabi-
lity assays. All compounds were found to be stable
under the bu€er (pH 5.4) assay conditions used in the
mobile phase of the HPLC assay. Each compound was
incubated in either Tris or HEPES bu€er, and their
hydrolysis rates monitored via HPLC. Half-lives were
calculated based on observations made at ®ve di€erent
time intervals. Plots of ln[(A ÀA )/(A ÀA )] versus time,
the methylsulfonyloxy phthalimides inhibited HLE very
À1 À1 24
well (kobs/[I]=160,000 M
extension of the alkylsulfonyloxy chain strongly in¯u-
s ). This suggests that
t
f
o
f
0
the 6-position.
where A and A are the peak areas of the remaining
ences the Sn binding interactions of the substituent at
t
f
non-hydrolyzed sulfonyloxy phthalimide at times t and
the ®nal reading, were used to determine the rate of
hydrolysis. Half-life values were calculated using t
Three representative compounds were selected for buf-
fer stability study. Derivative 4 was the most stable of
the three in both HEPES and Tris. Acylation of the
1
/2
À1
=0.693/k where k (s ) is obtained from the slope
(Table 2).

640
L.M. Vagnoni et al. / Bioorg. Med. Chem. 9 (2001) 637±645
Table 1. Inhibitory activity of derivatives 5a±g toward human
leukocyte elastase, chymotrypsin, porcine pancreatic elastase and
cathepsin G
6-amino function in the parent compound 4 appears to
increase the susceptibility of compounds 5b and 5f to
hydrolytic decomposition. These compounds have two
potential sites for hydrolytic attack due to the asym-
metric nature of the 6-substituted-isoindole-1,3-dione
ring (Scheme 2). Opening of the ring exposes two pos-
sible intermediate isocyanates. Hydrolysis of the iso-
cyanate gives two possible carboxamic acids, which
upon loss of carbon dioxide will decompose to the
anthranilic acids. An example of an HPLC trace (Fig. 3)
of the progress of the hydrolysis reaction of compound
À1 À1 a,b
k
obs/[I] (M
s
)
5f in Tris bu€er reveals the formation of two major
Compound
HLEc
ChTd
PPEe
Cat Gf
product peaks, one at 6.8 min and one at 7.4 min. The
peak at 10.5 min is the non-hydrolyzed starting mate-
rial. This observation supports the suggested duality of
binding modes. Additional work is under way to deter-
mine the exact nature of the structure of these two
major products and their kinetics of formation and will
be published elsewhere.
5
5
5
5
5
5
5
a
b
c
d
e
f
8100
11,000
10,000
7500
3600
6600
17,000
2800
3500
3600
7800
270
1100
3100
950
2300
170
1900
950
2000
g
ND
ND
ND
2200
g
g
14,000
18,000
g
9000
610
a
ꢁ
0
.1 M HEPES, 0.5 M NaCl, pH 7.5 at 25 C and <6% DMSO.
b
All data was collected in duplicate with an average relative standard
deviation of Æ14%.
Modeling
c
The substrate for HLE was MeO-Suc-Ala-Ala-Pro-Val-pNA ([E]=
The interaction energies of the noncovalent HLE-inhi-
bitor complexes (EÀI* in eq (1)) of the two sets of
enantiomers are reported in Table 3. The interaction
3
1 nM, [S]=95±100 mM).
d
The substrate for ChT was Suc-Ala-Ala-Pro-Phe-pNA ([E]=100±
1
58 nM, [S]=80 mM).
e
energy (E ) is the sum of all nonbonding interactions in
int
The substrate for PPE was Suc-Ala-Ala-Ala-pNA ([E]=200±400 nM,
the complex (i.e., van der Waals, Evdw and electrostatic,
Eelec in eq. (4)). Interaction energies (E ) are not bind-
[
S]=225 mM).
The substrate for Cat G was the same as that for ChT ([E]=42.5 nM,
f
int
[S]=200 mM).
g
ND, not determined.
ing free energies (i.e., ÁGbind);however, they can be
used to rank complexes and have been demonstrated to
3
4
show moderate correlation with experiment. We will
refer to the chiral center directly adjacent to the amide
linkage of the side chain to the phthalimide ring when
making reference to `R' and `S'. In both sets of enan-
tiomers, the S stereochemistry gave the lower interac-
tion energy. In support of these results, the experimental
results indicate that the S enantiomer is the better inhi-
bitor (see Table 1). One reason for this preference for
the S enantiomer is the nature of the binding of the
Table 2. Bu€er stability data for selected compoundsa
t
1/2 (min)
Compound
HEPESb
Trisc
4
8.9
2.1
6.1
6.1
4.0
6.8
5
5
b
f
0
chiral group in the Sn sites. In Figure 4 is a picture of
the AMBER89 energy minimized noncovalent complex
a
2
2
All log plots had R > 0.89 with average R =0.95.
ꢁ
.1 M HEPES, 0.5 M NaCl, pH 7.5 at 25 C and <6% DMSO.
b
of S enantiomer 5b with HLE showing the relative
0
0
c0.05 M Tris, 0.15 M NaCl, pH 7.5 at 25 C and <6% DMSO.
ꢁ
position of key residues (PHE 41 and LEU 35 in S and
n
Scheme 2. Possible bu€er hydrolysis outcomes.

L.M. Vagnoni et al. / Bioorg. Med. Chem. 9 (2001) 637±645
641
Figure 3. HPLC trace of 5f in Tris bu€er after 5 min of incubation.
Table 3. Interaction energies (Eint) of noncovalent HLE±inhibitor
complexes
E
int (kcal/mol)
AMBER89b
Compound
MMFF94a
5
5
5
5
5
5
a R=(1R)-camphanyl
b R=(1S)-camphanyl
d R=(1S,2R,5S)-menthyl
e R=(1R,2S,5R)-menthyl
d `inverted'
À763
À818
À927
À857
À992
À1008
À5560
À5610
e `inverted'
a
ꢀ
Ê
Gradient convergence at 0.05 kcal/(mol A).
b
ꢀ
Ê
Gradient convergence at 0.001 kcal/(mol A).
Figure 5. Electrostatic potential mapped to electron density in active
site region of compound 5b noncovalent complex with HLE.
0
some ring strain was noted in the `R' enantiomer com-
pointing out and away from the S pocket. In addition,
n
plex possibly due to unfavorable steric interactions in S
and S . We investigated the `inverted' binding model
1
0
see Fig. 6) for the menthyl derivatives 5d and 5e. The R
n
(
enantiomer is the lower energy complex in the `inverted'
model. The inverted model gave results that were in
con¯ict with experimental observations. The energy gap
(
ÁE ) between 5d and 5e was 16 kcal/mol for the
int
inverted model compared to a ÁEint of 70kcal/mol for
our proposed binding model (Fig. 1). These results do not
necessarily rule out the inverted binding model;however,
they do further support our binding hypothesis.
Figure 4. Model of compound 5b noncovalent complex with HLE.
VAL 216 in S ) relative to the inhibitor. The electro-
1
static potential surface map of the complex of `S' enan-
tiomer 5b with HLE in Figure 5 shows the lactone
carbonyl oxygen of the camphanyl group making a
favorable electrostatic interaction with the enzyme
backbone along the peptide bond of CYS 58 and VAL
Eint ˆ Evdw ‡ Eelec
ꢀ4†
Conclusion
5
9. In comparison, the electrostatic surface potential
Substitution in the 6-position of 6-acylamino-2-[(ethylsul-
fonyl)oxy]-1H-isoindole-1,3-diones is an e€ective method
of improving selectivity of these inhibitors for HLE.
Compounds 5b and 5c were both selective inhibitors of
HLE over ChT, PPE, and Cat G. These data demonstrate
map of the `R' enantiomer 5a (not shown) has all the
same features as the `S' enantiomer 5b except for the
favorable contact between the enzyme and the lactone.
For the `R' enantiomer, the lactone ring oxygens are

642
L.M. Vagnoni et al. / Bioorg. Med. Chem. 9 (2001) 637±645
2
-N-(Benzyl)oxy-5-nitroisoindole-1,3-dione (2). Melting
ꢁ
25
ꢁ
1
point (EtOH) 185±187.5 C. Lit. 188±190 C. H NMR
DMSO) d 5.20 (s, 2H), 7.45 (m, 5H), 8.15 (d, J=8.1Hz,
(
1H), 8.48 (d, J=1.8 Hz, 1H) 8.63 (dd, J=8.2, 2.0 Hz, 1H).
6-Amino-2-N-hydroxy-1H-isoindole-1,3-dione (3). Melt-
ing point (EtOH) 270±273 C Lit. 268 C. H NMR
ꢁ
25
ꢁ
1
Figure 6. `Inverted' binding model.
(DMSO) d 6.5 (s, 2H), 6.78 (d, J=8.1 Hz, 1H), 6.90 (s,
1H), 7.45 (d, J=8.3 Hz, 1H), 10.4 (s, 1 H).
that, while hydrophobic groups may increase selectivity
0
6-Amino-2-[(ethylsulfonyl)oxy)]-1H-isoindole-1,3-dione (4).
Compound 3 (1.96 g, 11.0 mmol) was taken up in H O
(70 mL) and cooled to 0 C. To this solution was added
sodium bicarbonate (924 mg, 11.0 mmol) and ethane-
sulphonyl chloride (1.0 mL, 11.0 mmol). The reaction
was allowed to stir overnight and the resulting pre-
for HLE;the combination of groups in P1 and P2 can
have dire e€ects on potency. HLE does not tolerate the
combination of bulky hydrophobic groups in both the
2
ꢁ
0
P1 (ethylsulfonyloxy group) and P2 regions of the
inhibitor. The incorporation of an additional carbon in
the alkylsulfonyloxy chain results in a loss of potency.
Extension of this chain may lead to the formation of a
higher energy complex in derivatives with bulky hydro-
phobic ring substituents perhaps through unfavorable
cipitate was diluted in H O and collected. Recrystalli-
2
zation in EtOH a€orded a yellow powder (1.83 g, 62%):
mp 185±186 C; H NMR (DMSO) d 1.49 (t, J=7.2 Hz,
ꢁ
3H), 3.75 (q, J=7.3 Hz, 2H), 6.80 (s, 2 H), 6.9 (d, J=
1
0
13
steric interactions in both the S and S sites and hence
1
8.3 Hz, 1H), 7.0 (s, 1H), 7.6 (d, J=8.3 Hz, 1H);
C
n
less potent inhibition. The bu€er stability assay indi-
cates that these inhibitors will not be suitable for intra-
venous or oral administration. However, these
compounds might be suitable for use in an inhaler.
NMR (DMSO) d 8.3, 46.5, 107.9, 112.6, 118.0, 126.6,
131.3, 156.2, 162.2, 163.1;IR (KBr) 3476 (NH) 3389,
À1
+
1733 (CˆO) cm ;EIMS m/z 269.8 (M ). Anal. calcd
for C H N O S: C, 44.44;H, 3.73;N, 10.38. Found:
2
1
0
10
5
C, 44.50;H, 3.78;N, 10.41.
0
6
-((1 R)-Camphanyl)amino-2-[(ethylsulfonyl)oxy]-1H-iso-
Experimental
Chemistry
indole-1,3-dione (5a). (1R)-(+)-Camphanic acid (500mg,
2.3 mmol) in thionyl chloride (4 mL, 52 mmol) was stir-
red at re¯ux for 1.5 h. Thionyl chloride was removed in
vacuo and the resulting acid chloride was immediately
taken up in THF (5 mL) followed by addition of 4
(618 mg, 2.3 mmol) then triethylamine (0.4 mL, 2.8
All reactions were carried out under an atmosphere of
dry nitrogen. All melting points were recorded on a
Mel-Temp II apparatus and are uncorrected. HPLC
data was recorded on a Shimadzu Class VP HPLC
using a YMC ODS-AQ (C-18) 120 A 5 mm particle size
4.6 Â 150 mm analytical column with the UV detector
set at 254 nm. The solvent mixtures used were acetoni-
ꢁ
mmol) at 0 C. The mixture was gradually warmed to
Ê
room temperature and allowed to stir overnight. The
mixture was diluted in ethyl acetate (40 mL), washed
with 10% HCl (16 mL), saturated sodium bicarbonate
trile: 25 mM NH OAc pH 5.4 bu€er (30:70) as Solvent
4
(2Â8 mL), brine (16 mL), dried (MgSO ) and con-
4
A and acetonitrile (100%) as Solvent B. The gradient
was ramped from 100% Solvent A to 100% Solvent B
in a linear fashion over a 10-min time period followed
by 4 min with 100% Solvent B with a ¯ow rate of
centrated. The resulting precipitate was recrystallized in
methyl tert-butyl ether in cyclohexanes, yielding a beige
ꢁ
1
powder (870 mg, 89%): mp 167±168 C; H NMR
(DMSO) d 0.90 (s, 3H), 1.10 (m, 7H), 1.49 (t, J=7.3 Hz,
3H) 1.62 (m, 1H) 2.02 (m, 2H), 3.81 (q, J=7.3 Hz, 2H),
7.93 (d, J=8.4 Hz, 1H), 8.25 (dd, J=8.2, 2.0 Hz, 1H), 8.41
(d, J=2 Hz, 1H), 10.45 (s, 1H). IR (KBr) 3323 (NH),
1
1
.00 mL/min. IR data was obtained on a Perkin±Elmer
1
3
600 series FTIR spectrometer. Proton and C spectra
were recorded on a Varian Gemini 200 FTNMR spec-
trometer. Optical rotations were determined on a Per-
kin±Elmer 241 polarimeter using a micro-cell with a
À1
+
2974, 1743 (CˆO) cm . EIMS m/z 450.2 (M ). Anal.
calcd for C H N O S: C, 53.33;H, 4.92;N, 6.22. Found:
2
C, 53.57;H, 5.11;N, 6.20;[ a]D +18.8 (c 4, CHCl ).
0
22
2
8
2
5
ꢁ
1
decimeter path length. Elemental analyses were per-
3
formed by Atlantic Microlab, Inc., Norcross, GA. Mass
spectrometric analysis was performed by the Depart-
ment of Chemistry, Washington University, St. Louis,
0
6-((1 S)-Camphanyl)amino-2-[(ethylsulfonyl)oxy]-1H-iso-
indole-1,3-dione (5b). This compound was prepared
from (1S)-(À)-camphanic acid in the same manner as
the procedure described for 5a. The crude precipitate
was recrystallized from methyl tert-butyl ether in cyclo-
hexane giving a beige powder (200 mg, 48%): mp 168.9±
3
5
MO. Flash chromatography was carried out using 32±
Ê
3
6 mm particle size silica gel (pore diameter, 60 A). Thin-
layer chromatography was performed using Allied-
Signal silica plates and visualized by UV light, iodine
vapor or phosphomolybdic acid spray. Tetrahydrofuran
ꢁ
1
169.0 C; H NMR (DMSO) d 0.90 (s, 3H), 1.05 (m,
7H), 1.49 (t, J=7.3 Hz, 3H), 1.60 (m, 1H), 2.00 (m, 2H),
(
THF) was distilled from sodium benzophenone ketyl.
3.75 (q, J=7.3 Hz, 2H), 7.95 (d, J=8.5 Hz, 1H), 8.25
(dd, J=8.4, 1.5 Hz, 1H), 8.40 (s, 1H), 10.51 (s, 1H);
NMR (DMSO) d 8.3, 9.8, 16.5, 16.7, 28.6, 30.3, 40.4,
54.3, 54.9, 92.0, 115.7, 123.3, 125.5, 126.4, 129.9, 144.5,
161.8, 162.1, 166.8, 178.1. IR (KBr) 3324 (NH), 2965,
ꢁ
13
Tosyl-L-valine. Melting point (aq EtOH) 146.2±147.9 C
C
36
ꢁ
.90 (m, 1H), 2.35 (s, 3H), 4.44 (m, 2H), 7.33 (d, J=8.1 Hz,
1
Lit. 147 C. H NMR (DMSO) d 0.78 (t, J=6.6Hz, 6H),
1
2H), 7.65 (d, J=8.2Hz, 2H), 7.9 (d, J=9.3 Hz, 1H).

L.M. Vagnoni et al. / Bioorg. Med. Chem. 9 (2001) 637±645
643
À1
+
1
C H N O S: C, 53.33;H, 4.92;N, 6.22. Found: C,
749 (CˆO) cm . EIMS m/z 450 (M ). Anal calcd for
a€orded a crude oil. The crude oil was puri®ed by col-
umn chromatography (ethyl acetate/hexanes, 1:1) giving
a yellow powder after removal of solvent (140 mg,
2
0
22
2
8
2
5
ꢁ
5
3.57;H, 5.09;N, 6.16;[ a]D À19.5 (c 2, CHCl ).
3
ꢁ
1
31%): mp 177.6±180.1 C dec. H NMR (DMSO) d 0.85
6
-(Benzoyl)amino-2-[(ethylsulfonyl)oxy]-1H-isoindole-1,
-dione (5c). Triethylamine (0.2 mL, 1.1 mmol) was added
(dd, J=6.6, 5.5 Hz, 6H), 1.55 (t, J=7.3 Hz, 3H), 1.89 (m,
1H), 2.15 (s, 3H) 3.60 (t, J=8.1 Hz, 1H), 3.78 (q,
J=7.2 Hz, 2H), 7.18 (d, J=8.1 Hz, 2H), 7.60 (s, 1H),
7.62 (d, J=8.1 Hz, 2H), 7.9 (m, 2H), 8.23 (d, J=10.0 Hz,
1H), 10.55 (s, 1H); C NMR (DMSO) d 8.3, 18.8, 19.1,
20.0, 30.9, 46.5, 63.1, 113.9, 122.4, 124.7, 125.7, 126.9,
3
to a THF (6 mL) solution of compound 4 (250 mg,
0
0
.9 mmol) and phenylacetyl chloride (0.1 mL, 1.0 mmol) at
C. The mixture was allowed to stir at room temperature
ꢁ
overnight. The work up employed was identical to that
13
used for 5b. Recrystallization in ethanol yielded a light-
ꢁ
129.4, 129.9, 138.2, 142.8, 144.7, 162.1, 170.3. IR (KBr)
À1
1
yellow powder (120 mg, 35%): mp 180.9±181.6 C; H
NMR (DMSO) d 1.45 (t, J=7.3 Hz, 3H), 3.75 (m, 4H),
3251 (NH), 2963, 1748 (CˆO) cm . FABMS m/z 524.0
(M+H). Anal. calcd for C H N O S : C, 50.47;H,
8 2
2
2
25
3
7
.31 (m, 5H), 7.92 (m, 2H), 8.25 (s, 1H), 10.89 (s, 1H);
4.81;N, 8.03. Found: C, 50.34;H, 4.74;N, 7.91.
1
3
C NMR (DMSO) d 8.2, 43.6, 46.4, 173.6, 122.2, 124.4,
25.8, 127.0, 128.7, 129.5, 130.3, 135.4, 145.8, 161.8, 162.1,
1
1
6-(N-Tosyl-L-phenylalanyl)amino-2-[(ethylsulfonyl)oxy]-
1H-isoindole-1,3-dione (5g). Triethylamine (0.1 mL, 0.9
mmol) was added to a THF (10 mL) solution of com-
pound 4 (200 mg, 0.74 mmol) and N-p-toluenesulpho-
À1
70.6. IR (KBr) 3372 (NH), 3058, 1739 (CˆO)cm
.
+
EIMS m/z 388 (M ). Anal calcd for C H N O S: C,
6
1
8
16
2
5
5.66;H, 4.15;N, 7.21. Found: C, 55.63;H, 4.20;N, 7.12.
ꢁ
nyl-l-phenylalanyl chloride (304 mg, 0.9 mmol) at 0 C.
0
0
0
6
-((1 S,2 R,5 S)-Menthyloxycarbonyl)amino-2-[(ethylsul-
The mixture was warmed to room temperature and
stirred overnight. The work up employed was identical
to that used for 5b. The crude residue was recrystallized
in EtOH a€ording a dark orange powder (148 mg,
fonyl)oxy]-1H-isoindole-1,3-dione (5d). Compound 4
250 mg, 0.93 mmol) was combined with (+)-menthyl
(
chloroformate (0.2 mL, 1.0 mmol) as described for the
preparation of 5b. This preparation a€orded an oil, which
was puri®ed by column chromatography (ethyl acetate/
hexanes, 1:1). The isolated oil was dried in vacuo giving
ꢁ
1
35%);mp 189.9±191 C dec. H NMR (CDCl ) d 1.66
3
(t, J=7.3 Hz, 3H), 2.43 (s, 3H), 3.06 (m, 2H), 3.65 (q,
J=7.3 Hz, 2H), 4.05 (m, 1H), 5.18 (d, J=6.2 Hz, 1H),
ꢁ
a crystalline compound (220 mg, 53%): mp 242 C dec.
1
6.95 (m, 2H), 7.2 (m, 5H), 7.55 (d, J=8.1 Hz, 2H), 7.84
13
H NMR (DMSO) d 0.7±1.15 (m, 12H), 1.4 (m, 2H), 1.5
t, J=7.2 Hz, 3H), 1.65 (m, 2H), 1.95 (m, 2H) 3.85 (q,
(m, 2H) 8.24 (s, 1H), 8.56 (s, 1H); C NMR (CDCl ) d
3
(
J=7.2 Hz, 2H), 4.62 (dt, J=6.6, 4.1 Hz, 1H), 7.88 (m,
8.4, 21.7, 37.9, 48.2, 58.7, 74.2, 97.3, 115.4, 125.3, 125.7,
127.3, 127.7, 129.3, 121.3, 130.2, 134.7, 143.9, 144.7,
161.5, 169.3. FABMS m/z 572 (M+H). HRFABMS
calcd for C H N O S : 572.1161. Found: 572.1137.
2
8
7
1
H), 8.09 (s, 1H), 10.5 (s, 1 H); 13C NMR (DMSO) d
.3, 16.5, 20.8, 22.2, 23.2, 26.0, 31.2, 33.9, 46.5, 47.1,
2
6
26
3
8 2
4.8, 112.5, 112.2, 123.3, 125.9, 130.4, 146.4, 153.3,
À1
62.3. IR (KBr) 3364 (NH), 2953, 1748 (CˆO) cm
.
Biological assays
HRFABMS calcd for C H N O S: 453.1696. Found:
2
2
1
28
7
4
6
+
53.1697. Anal calcd for C H N O S: C, 55.74;H,
2
In vitro assays were carried out in plastic cuvettes on a
Shimadzu UV±vis 2101PC scanning spectrophotometer
equipped with a thermo jacketed cell holder. Substrate
concentrations for enzyme assays were taken from Naka-
1
28
2
7
2
5
.24;N, 6.19. Found: C, 55.70;H, 6.29;N, 6.08;[
34.7 (c 2, CHCl ).
3
a]
D
ꢁ
0
fonyl)oxy]-1H-isoindole-1,3-dione (5e). Compound 4
0
0
37
6
-((1 R,2 S,5 R)-Menthyloxycarbonyl)amino-2-[(ethylsul-
jima and Powers. All inhibitor and substrate stock
solutions were prepared in DMSO. Stock solutions of
HLE and Cat G were prepared in 0.05 M NaOAc,
0.15 M NaCl bu€er, pH 5.5;and stock solutions of PPE
and ChT were prepared in 1 mM HCl. HEPES [4-(2-
hydroxyethyl)-1-piperazineethane]sulfonic acid bu€er
(0.1 M HEPES, 0.5 M NaCl, pH 7.5) was used in all in
(
250 mg, 0.9 mmol) was combined with (À)-menthyl
chloroformate (0.2 mL, 1.0 mmol) as described for the
preparation of 5d. This preparation a€orded the oppo-
site enantiomer, which was puri®ed by column chroma-
tography (ethyl acetate/hexanes, 1:1). The oil obtained
1
ꢁ
was dried in vacuo (160 mg, 38%). H NMR (DMSO) d
vitro enzyme assays. Assays were run at 25 C and initi-
0
.7±1.2 (m, 12H), 1.4 (m, 2H), 1.5 (t, J=7.3 Hz, 3H);
ated by addition of enzyme. The pseudo ®rst order inhi-
bition rates were monitored by following the decrease in
absorbance at 410 nm of the p-nitroaniline group
1
(
(
.65 (d, 2H), 1.98 (m, 2H), 3.75 (q, J=7.3 Hz, 2H), 4.62
dt, J=6.5, 4.0 Hz, 1H), 7.88 (m, 2H), 8.10 (s, 1H), 10.5
1
3
À1 À1
s, 1H); C NMR (DMSO) d 8.3, 16.4, 20.8, 22.2, 23.2,
(e410=8800 M
cm ) released from the substrate by
2
1
3
6.0, 31.2, 33.9, 46.4, 47.1, 74.8, 107.4, 112.5, 121.2,
23.3, 125.9, 130.4, 146.3, 153.3, 161.9, 162.3. IR (KBr)
364 (NH), 2953, 1748 (CˆO) cm . EIMS m/z (%) 188
enzymatic hydrolysis. Human Leukocyte Elastase and
Cathepsin G were obtained from Athens Research and
Technology, Inc., Athens, GA. Porcine Pancreatic Elas-
tase and Chymotrypsin were obtained from Sigma-
Aldrich, St. Louis, MO. All substrates were obtained
from Bachem Bioscience, King of Prussia, PA.
À1
(100), 163 (43), 315 (26). Anal. calcd for C H N O S:
21 28 2 7
C, 55.74;H, 6.24;N, 6.19. Found: C, 55.84;H, 6.33;N,
6
2
5
ꢁ
.11;[ a] À45.5 (c 2, CHCl ).
d
3
6
-(Tosyl-L-valinyl)amino-2-[(ethylsulfonyl)oxy]-1H-iso-
Molecular modelling
indole-1,3-dione (5f). Compound 4 (250 mg, 0.9 mmol)
was combined with tosyl-l-valine (1.0 g, 3.7 mmol) as
described for the preparation of 5a. This preparation
All computational work was performed on an
1
OCTANE (Silicon Graphics, Inc.) system or a Pentium

644
L.M. Vagnoni et al. / Bioorg. Med. Chem. 9 (2001) 637±645
III1 (Intel, Inc.) 933 MHz personal computer. The small
molecule inhibitors were built and geometry optimized
4. Powers, J. C.;Bengali, Z. H. Am. Rev. Respir. Dis. 1986,
134, 1097.
5. Johnson, D.;Travis, J. J. Biol. Chem. 1978, 253, 7142.
3
8
using the Merck Molecular Force Field (MMFF94) in
1
6. Beatty, K.;Matteson, N.;Travis, J.
Physiol. Chem. 1984, 365, 731.
Hoppe-Seyler's Z.
PCSpartan Pro (Wavefunction, Inc.). For those com-
pounds used in the AMBER89 calculations, single-point
ab initio calculations were performed on the resulting
7
4
8
4
9
. Hay, J. W.;Robin, E. D. Am. J. Public Health 1991, 81,
27.
. Leung, D.;Abbenante, G.;Fairlie, D. J. Med. Chem. 2000,
3, 305.
. Bernstein, P. R.;Edwards, P. D.;Williams, J. C. In Pro-
3
9,40
small molecule structures using the STO-3G*
4
basis
The CHelpG electrostatic
1
set in Gaussian 98W.
potential ®tting procedure in Gaussian was used to
compute charges on the individual atoms of each inhi-
bitor after single point ab initio calculation.42 For the
MMFF94 calculations, MMFF94 charges were used on
both the inhibitor and the protein. The ethylsulfonyl
portion of the energy-minimized inhibitors was super-
imposed on the side chain of the valine residue of the
MeOSuc-Ala-Ala-Pro-Val-chloromethyl ketone inhi-
bitor in the crystal structure (PDB;1PPG). Molecular
mechanics calculations using the Merck Molecular Force
Field (MMFF94) on the noncovalent enzyme-inhibitor
gress in Medicinal Chemistry;Ellis, G. W., Luscombe, D. K.,
Eds.;Elsevier: Amsterdam, 1994;Vol. 31, pp 59±120.
10. Edwards, P. D.;Bernstein, P. R. Med. Res. Rev. 1994, 14,
1
1
27.
1. Powers, J. C.;Harper, J. W.
Proteinase Inhibitors;
Elsevier: Amsterdam, 1986.
1
1
2. Harper, J. W.;Powers, J. C. Biochemistry 1985, 24, 7200.
3. Kerrigan, J. E.;Oleksyszyn, J.;Kam, C.;Selzler, J.;Powers,
J. C. J. Med. Chem. 1995, 38, 544.
4. Groutas, W. C.;Chong, L. S.;Venkataraman, R.;Kuang,
1
R.;Epp, J. B.;Houser-Archield, N.;Huang, H.;Hoidal, J. R.
Bioorg. Med. Chem. 1996, 4, 1393.
15. Groutas, W. C.;Brubaker, M. J.;Chong, L. S.;Venka-
taraman, R.;Huang, H.;Epp, J. B.;Kuang, R.;Hoidal, J. R.
J. Med. Chem. 1995, 38, 1571.
1
complexes were performed in Sybyl (Tripos, Inc.) on
the OCTANE. Molecular mechanics calculations on
enzyme-inhibitor complexes employing the Kollman all-
4
3
atom implementation of AMBER89 was performed
using MOE (Chemical Computing Group, Inc.) on the
personal computer.
1
6. Groutas, W. C.;Brubaker, M. J.;Venkataraman, R.;
Stanga, M. A. Arch. Biochem. Biophys. 1992, 297, 144.
7. Wilmouth, R. C.;Westwood, N. J.;Anderson, K.;
1
Brownlee, W.;Claridge, T. D. W.;Clifton, I. J.;Pritchard, G.
J.;Aplin, R. T.;Scho®eld, C. J. Biochemistry 1998, 37, 17506.
Ê
The re®ned 2.3 A crystal structure of HLE (PDB;
1
the noncovalent enzyme inhibitor complexes.
PPG) was used as a starting point for construction of
All
18. Pochet, L.;Doucet, C.;Dive, G.;Wouters, J.;Masereel,
B.;Reboud-Ravaux, M.;Pirotte, B. Bioorg. Med. Chem. 2000,
44
crystallographic water atoms and peripheral sugar resi-
dues were removed. The valine chloromethyl ketone
inhibitor was extracted and used for initial positioning
of the inhibitors. Hydrogen atoms were added and
charges on the individual residues were loaded from the
8, 1489.
19. Gutschow, M.;Neumann, U. J. Med. Chem. 1998, 41,
1729.
2
0. Groutas, W. C.;Kuang, R.;Ruan, S.;Epp, J. B.;Venka-
taraman, R.;Truong, T. M. Bioorg. Med. Chem. 1998, 6, 661.
21. Kuang, R.;Venkataraman, R.;Ruan, S.;Groutas, W.
Bioorg. Med. Chem. Lett. 1998, 8, 539.
4
3
Kollman all-atom charge set in Sybyl. The Kollman
all-atom charge set in AMBER89 is derived from ab
2
2. Kuang, R.;Epp, J. B.;Ruan, S.;Chong, L.;Venkatara-
4
3
initio STO-3G calculations. Inhibitors were pre-posi-
tioned in the active site of the enzyme followed by par-
man, R.;Tu, J.;He, S.;Truong, T.;Groutas, W. Bioorg. Med.
Chem. 2000, 8, 1005.
Ê
tial minimization of the entire complex. A 6 A region
23. Neumann, U.;Gu
21561.
È
tschow, M. J. Biol. Chem. 1994, 269,
about the inhibitor was de®ned as the cavity. The inhi-
bitors/cavity region was minimized to the gradient con-
vergence criteria speci®ed in the text. All graphical
24. Kerrigan, J. E.;Walters, M. C.;Forrester, K. J.;Crowder,
J. B.;Christopher, L. J. Bioorg. Med. Chem. Lett. 2000, 10, 27.
25. Kerrigan, J. E.;Shirley, J. J. Bioorg. Med. Chem. Lett.
1
displays were generated using Sculpt , version 3.0
(
1
996, 6, 451.
6. Groutas, W. C.;Giri, P. K.;Crowley, J. P.;Castrisos, J.
MDL Information Systems, Inc.).
2
C.;Brubaker, M. J. Biochem. Biophys. Res. Comm. 1986, 141,
7
2
1
2
41.
Acknowledgements
7. Bode, W.;Meyer, E.;Powers, J. C. Biochemistry 1989, 28,
951.
8. Hernandez, M. A.;Powers, J. C.;Glinski, J.;Oleksyszyn,
The authors wish to thank the New Jersey Thoracic
Society of the American Lung Association and the Col-
lege of Pharmacy, Rutgers University for their generous
support. The authors also wish to thank the journal
referees for their input and helpful suggestions.
J.;Vijayalakshmi, J.;Meyer, E. F. J. Med. Chem. 1992, 35,
1121.
29. Meyer, E. F.;Presta, L. G.;Radhakrishnan, R. J. Am.
Chem. Soc. 1985, 107, 4091.
3
0. Plaskon, R. R.;Kam, C.;Burgess, E. M.;Powers, J. C.;
Suddath, F. L. Proteins 1992, 13, 141.
1. Plaskon, R. R.;Kam, C.;Kerrigan, J. E.;Burgess, E. M.;
Powers, J. C.;Suddath, F. L. Arch. Biochem. Biophys. 1993,
3
References and Notes
3
3
00, 588.
2. Tian, W.;Tsou, C. Biochemistry 1982, 21, 1028.
1
1. Dewald, B.;Rindler-Ludwig, R.;Bretz, U.;Baggiolini, M.
J. Exp. Med. 1975, 141, 709.
33. Billo, E. Excel for Chemists: A Comprehensive Guide;
Wiley-VCH: New York, 1997.
2. Jano€, A.;Scherer, J. J. Exp. Med. 1968, 128, 1137.
3. Weinbaum, G.;Giles, R.;Krell, R. Ann. NY. Acad. Sci.
1991, 624, 45.
34. Holloway, M. K.;Wai, J. M.;Halgren, T. A.;Fitzgerald,
P. M. D.;Vacca, J. P.;Dorsey, B. D.;Levin, R. B.;Thomp-
son, W. J.;Chen, L. J.;deSolms, S. J.;Gan, N.;Ghosh, A.

L.M. Vagnoni et al. / Bioorg. Med. Chem. 9 (2001) 637±645
K.;Giuliani, E. A.;Graham, S. L.;Guare, J. P.;Hungate, R.
W.;Lyle, T. A.;Sanders, W. M.;Tucker, T. J.;Wiggins, M.;
Wiscount, C. M.;Woltersdorf, O. W.;Young, S. D.;Darke, P.
L.;Zugay, J. A. J. Med. Chem. 1995, 38, 305.
645
Burant, J.;Dapprich, S.;Millam, J.;Daniels, A.;Kudin, K.;
Strain, M.;Farkas, O.;Tomasi, J.;Barone, V.;Cossi, M.;
Cammi, R.;Mennucci, B.;Pomelli, C.;Adamo, C.;Cli€ord,
S.;Ochterski, J.;Petersson, G.;Ayala, P.;Cui, Q.;Mor-
okuma, K.;Malick, D.;Rabuck, A.;Raghavachari, K.;
Foresman, J.;Cioslowski, J.;Ortiz, J.;Baboul, A.;Stefanov,
B.;Liu, G.;Liashenko, A.;Piskorz, P.;Komaromi, I.;Gom-
perts, R.;Martin, R.;Fox, D.;Keith, T.;Al-Laham, M.;Peng,
3
2
3
1
3
5. Still, W. C.;Kahn, M.;Mitra, A. J. Org. Chem. 1978, 43,
923.
6. McChesney, E.;Swann, W. J. Am. Chem. Soc. 1937, 59,
116.
7. Nakajima, K.;Powers, J. C.;Ashe, B. M.;Zimmerman,
C.;Nanayakkara, A.;Challacombe, M.;Gill, P.;Johnson, B.;
Chen, W.;Wong, M.;Andres, J.;Gonzalez, C.;Head-Gordon,
M.;Replogle, E.;Pople, J. Gaussian, revision A.9;Gaussian,
Inc.: Pittsburgh, PA, 1998.
42. Breneman, C.;Wiberg, K. J. Comp. Chem. 1990, 11, 361.
43. Weiner, S. J.;Kollman, P. A.;Case, D. A.;Singh, U. C.;
Ghio, C.;Alagona, G.;Profeta, S.;Weiner, P. J. Am. Chem.
Soc. 1984, 106, 765.
M. J. Biol. Chem. 1979, 254, 4027.
3
3
2
4
8. Halgren, T. J. Comp. Chem. 1996, 17, 490.
9. Hehre, W.;Stewart, R.;Pople, J. J. Chem. Phys. 1969, 51,
657.
0. Hehre, W.;Radom, L.;Schleyer, P.;Pople, J.
Ab Initio
Molecular Orbital Theory;Wiley: New York, 1986.
1. Frisch, M.;Trucks, G.;Schlegel, H.;Scuseria, G.;Robb, M.;
Cheeseman, J.;Zakrzewski, V.;Montgomery, J.;Stratmann, R.;
4
44. Wei, A.;Mayr, I.;Bode, W. FEBS Lett. 1988, 234, 367.