Design and synthesis of peptide-based carboxylic acid-containing transition-state inhibitors of human neutrophil elastase.

Article abstract of DOI:10.1016/S0960-894X(01)00797-1

In our search for a new agent, human neutrophil elastase (HNE) inhibitor, for the treatment of acute respiratory failure, we rationally designed and synthesized a series of peptide-based carboxylic acid-containing transition-state inhibitors. The presence

Full text of DOI:10.1016/S0960-894X(01)00797-1

Bioorganic & Medicinal Chemistry Letters 12 (2002) 551–555
Design and Synthesis of Peptide-Based Carboxylic
Acid-Containing Transition-State Inhibitors of
Human Neutrophil Elastase
Fuminori Sato,a,* Yasunao Inoue,a Tomoki Omodani,a Kiyomi Imano,b
Hiroshi Okazaki,b Tadashi Takemurab and Masanobu Komiyab
aDepartment of Chemistry II, Discovery Research Laboratories, Dainippon Pharmaceutical Co., Ltd.
Enoki 33-94, Suita, Osaka 564-0053, Japan
bDepartment of Pharmacology III, Discovery Research Laboratories, Dainippon Pharmaceutical Co., Ltd.
Enoki 33-94, Suita, Osaka 564-0053, Japan
Received 29August 2001; accepted 26 November 2001
Abstract—In our search for a new agent, human neutrophil elastase (HNE) inhibitor, for the treatment of acute respiratory failure,
we rationally designed and synthesized a series of peptide-based carboxylic acid-containing transition-state inhibitors. The presence
of valyl moiety is found to be essential for potent in vitro inhibitory activity and also prevention of an undesirable toxicity. Of these,
compound 9m has the most potent in vivo effect on HNE-induced lung hemorrhage in hamsters. # 2002 Elsevier Science Ltd. All
rights reserved.
Human neutrophil elastase (HNE, EC 3.4.21.37), a
neutral serine protease with broad substrate specificity,
is a major constituent in the azurophilic granules of
human neutrophils. Under normal conditions, the pro-
teolytic activity of the HNE is controlled by some
endogenous inhibitors such as a1-protease inhibitor, a2-
macrogloblin, secretroy leukocyte protease inhibitor,
and so on. An imbalance between the HNE and endo-
genous inhibitors has been suggested as a major caus-
ative factor in the acute respiratory distress syndrome or
the acute lung injury, which is a severe life-threatening
sequel to sepsis and other conditions.1 It is characterized
by massive infiltration of neutrophils into the lungs and
other organs, where the neutrophils cause tissue injury
by releasing the HNE and some toxic substances.
Although numerous HNE inhibitors including peptidic
and nonpeptidic inhibitors have been investigated, no
agent for clinical use has so far been developed.2 For the
treatment of the acute destructive diseases, the admin-
istration of a drug by injection is preferred. Therefore,
the high solubility and stability in aqueous solution are
essential physio-chemical properties for the inhibitor.
During our search for a new agent for the treatment of
such acute disorders, we have reviewed HNE inhibitors
reported previously.3 The known compounds are classi-
fied into two types based on their inhibitory mechanism:
(1) acyl-enzyme inhibitor and (2) transition-state inhi-
bitor (Fig. 1). In general, the in vitro inhibitory activities
Figure 1. Inhibitory mechanisms of acyl-enzyme inhibitor and transition-state inhibitor.
*Corresponding author. Tel.: +81-6-6337-7040; fax: +81-6-6338-7656; e-mail: fuminori-sato@dainippon-pharm.co.jp
0960-894X/02/$ - see front matter # 2002 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(01)00797-1

552
F. Sato et al. / Bioorg. Med. Chem. Lett. 12 (2002) 551–555
Scheme 1. (a) (i) HCl, EtOH, CHCl3; (ii) 2,6-dihydroxyaniline, Et3N, EtOH; (b) (i) TBSCl, DMAP, Et3N, CH2Cl2; (ii) DMP, CH2Cl2; (c) (i) TBAF,
THF; (ii) R-Br or R-I, NaH, DMF or RCOCl, Et3N, CH2Cl2..
state inhibitors, such as peptide-based trifluoromethyl
ketone, reversibly react with the active-site serine-195 of
HNE to form a tetrahedral configuration. The carbonyl
moiety (–COCF3) of trifluoromethyl ketone inhibitor is
chemically more stable than that (–O–CO–R) of acyl-
enzyme inhibitor.
In 1992, Edwards and co-workers at Zeneca disclosed
the crystal structure of the complex of substrate-based
inhibitor having a-ketoheterocyclic moiety and porcine
pancreatic elastase, demonstrating the importance of
the structure of the binding sites P1, P2, and P3 side
chains in the inhibitors (Fig. 2).4 This led us to intro-
duce substituent (OR) on benzoxazole ring, by which a
better fitting to S10 subsite of HNE could be achieved.
The compounds that lack of the amino acid moiety (Val
or Val-Pro in Fig. 2), may have some potent activities
with the presence of fitting substituent (OR) and also
expected to enhance the metabolic stability.
Figure 2. The inhibitory interaction between
benzoxazole and porcine pancreatic elastase.4
a peptidyl a-keto-
of the acyl-enzyme inhibitors are relatively high. How-
ever, it would be difficult to develop them to a water
stable form with a low toxicity, due to their high reactivity
toward nucleophiles. On the other hand, the transition-
Table 1. Substituent effect on HNE inhibitory activity
Compda
–OR
IC50 (mM)b
>10
Compda
–OR
IC50 (mM)b
4a
–H
4h
7.8
4b
4c
4d
4e
4f
–OH
>10
0.45
>10
>10
>10
>10
4i
4j
>10
>10
>10
>10
>10
>10
4k
4l
4m
4n
4g
aAll compounds are greater than 90% of diastereomer with the S configuration at the stereogenic center at a to the ketone carbony group.
bInhibition of HNE-catalyzed hydrolysis of the synthetic substrate Suc-Ala-Pro-Ala-MCA.6

F. Sato et al. / Bioorg. Med. Chem. Lett. 12 (2002) 551–555
553
Chemistry
benzoxazole ring. Among them, pivaloyl ester (4c) and
isopentyl ether (4h) groups were found to increase the
activity in comparison with unsubstituted compound
4a. Unsaturation (4i), migration of methyl group (4j),
and exchange of secondary carbon atom with nitrogen
atom (4k) of the isopropyl group in compound 4h lost
the activity at dose of 10 mM. Aryl alkyl ethers (4d–f)
are also not effective substituents. Furthermore, the
introduction of hydrophilic moieties (4b, 4m, and 4n)
resulted in loss the activity. These results suggest that at
the active site of the enzyme, there may be present a
lipophilic pocket that accommodates the substituents at
the 4-position of the benzoxazole ring.
The simplest unsubsituted benzoxazole derivative (4a,
Table 1) was prepared by the reported method.4 The
synthetic routes for the O-substituted benzoxazole deri-
vatives (4b–n) are summarized in Scheme 1. Thus, 4-
hydroxybenzoxazole derivative (2) derived from 1 was
prepared by Pinner condensation5 between 2-amino-
resorcinol with cyanohydrin. Phenolic hydroxy group
was selectively protected with tert-butyldimethylsilyl
group (TBS), and then oxidation of secondary hydroxy
group with Dess–Martin periodinane (DMP) gave 3.
After removing of TBS group, esterification or ether-
ification were performed by the usual manner to afford
4b–n. Scheme 2 shows the preparation of carboxylic
acid derivatives (9a–n). Deprotection of the Z-group in
3 provided 5, which was coupled with Z-Pro–OH or
Z-Val-Pro–OH to afford 6a or 6b. After removing the
Z-group, the terminal-ester group was introduced by the
condensation with half-esters to give 7a and 7b. After
oxidation of secondary alcohol of 7a and 7b with DMP,
transformation of TBSO group to ester or ether group
was then performed in a same manner as the methods
described in the Scheme 1. Finally, treatment with TFA
gave the desired compounds (9a–n).
Effect of amino acid residue
Several advanced derivatives were prepared in which
amino acid residues have incorporated (Table 2). Phen-
oxyacetic acid moiety was also introduced to increase
water-solubility, which did not affect the parent activity
(4c vs 9a). Although an incorporation of Pro-residue (9b
and 9c) did not increase in the in vitro activity, the in
vivo activities of these series are improved in some
extent at a high dose (100 mg/kg). The presence of Val
residue dramatically increased in the in vitro activity
(9d). The importance of the Val residue in this position
for the in vivo activity was in accordance with the pre-
viously reported results for peptide trifluoromethyl
ketone inhibitors.8 However, this modification did not
increase the in vivo potency; this indicates that the
decrease of the chain length in the peptide possibly
increase the metabolic stability. However, undesirable
toxicity (muscular stiffness in hamsters) was observed
when 9b and 9c were administrated intravenously at a
Results and Discussion
Substituent of benzoxazole ring
Effects of the substituents on the benzoxazole ring were
evaluated (Table 1). Various ester- and ether-based
substituents were incorporated into the 4-position of the
Table 2. Effect of amino acid residue on HNE inhibitory activity
Compda
–OR
–X–
IC50 (mM)b
% inhibition after iv bolus administrationc
100 mg/kg
30 mg/kg
9a
9b
9c
9d
9e
—
0.13
74
13
–Pro–
0.23
96
66
–P9ro–
0.299
0.00899
0.0076
ꢀ16
–Val-Pro9–
–Val-Pro–
73
98
56
aAll compounds are greater than 90% of diastereomer with the S configuration at the stereogenic center at a to the ketone carbony group.
bInhibition of HNE-catalyzed hydrolysis of the synthetic substrate Suc-Ala-Pro-Ala-MCA.6
cInhibition of HNE-induced lung hemorrhage in hamster.7

554
F. Sato et al. / Bioorg. Med. Chem. Lett. 12 (2002) 551–555
Scheme 2. (a) H2, Pd(OH)2, AcOEt; (b) Z-Pro-OH or Z-Val-Pro-OH, WSC, Py; (c) (i) H2, Pd(OH)2, AcOEt; (ii) tBuOOC-Y-COOH, WSC, Py; (d)
(i) DMP, CH2Cl2; (ii) TBAF, THF; (e) (i) R-Br or R-I, NaH, DMF or RCOCl, Et3N, CH2Cl2; (ii) TFA, CH2Cl2.
Table 3. Effect of terminal acid residue on HNE inhibitory activity
Compda
–G
IC50 (mM)b
% inhibition after iv bolus administrationc
100 mg/kg
30mg/kg
10 mg/kg
9f
0.066
0.072
0.28
61
9g
9h
9i
NDd
39
48
86
60
75
99
0.0072
0.0035
0.0074
0.013
0.033
90
99
37
30
15
12
46
9j
9k
9l
9m
aAll compounds are greater than 90% of diastereomer with the S configuration at the stereogenic center at a to the ketone carbony group.
bInhibition of HNE-catalyzed hydrolysis of the synthetic substrate Suc-Ala-Pro-Ala-MCA.6
cInhibition of HNE-induced lung hemorrhage in hamster.7
dDue to the low solubility, inhibitory activity was not determined.
Effect of terminal acid residue
dose of 100 mg/kg, but such behavior was not observed
for 9d. This toxicity seems to be caused from the lack of
Val residue. This observation directed us to fix on the X
as Val-Pro. As the introduction of the pivaloyl sub-
stituent was not effective for the Val-Pro type (9d vs 9e),
we then focused on the modification of the terminal
acidic moiety (Table 3).
Table 3 shows that the introduction of an aromatic ring
in the terminal acidic moiety is preferred for the in vitro
inhibitory activity. The substituent-position of car-
boxylic acid moiety of benzene ring was not crucial for
the activities (9e in Table 2 vs 9i in Table 3). Replacing

F. Sato et al. / Bioorg. Med. Chem. Lett. 12 (2002) 551–555
555
the oxygen atom of 9e with sulfur and introduction of
methyl group at acetic acid group for increasing the
bulkiness as in 9j resulted in similar potencies in both in
vitro and in vivo activities. The in vitro inhibitory
activities of the ester (9l) and the amide (9m) derivatives
were less potent than that of 9e. However, the in vivo
activity of 9m was the best in this series; HNE-induced
lung hemorrhage in hamsters was almost completely
inhibited at a dose of 30 mg/kg. The presence of amide
bond would increase a chemical stability or have better
pharmacokinetic properties than other synthesized
compounds. For further evaluation, both pharmacolo-
gical and synthetic studies on this amide compound 9m
and its derivatives are in progress.
2. (a) Metz, W. A.; Peet, N. P. Exp. Opin. Ther. Pat. 1999, 7,
851. (b) Skile, J. W.; Jeng, A. J. Exp. Opin. Ther. Pat. 1999, 7,
869.
3. (a) Edwards, P. D.; Berstein, P. R. Med. Res. Rev. 1994, 14,
127. (b) Berstein, P. R.; Edwards, P. D. Prog. Med. Chem.
1994, 31, 59. (c) Edwards, P. D.; Veale, C. A. Exp. Opin. Ther.
Patents 1997, 7, 17.
4. Edwards, P. D.; Meyer, E. F.; Vijayalakshmi, J.; Tuthill,
P. A.; Andisik, D. A.; Gomes, B.; Strimpler, A. J. Am. Chem.
Soc. 1992, 114, 1854.
5. Neilson, D. G. In Chemistry of The Amidines and Imidates;
Patai, S., Ed.; John Wiley & Sons: New York, 1975; Vol. 1,
p 385.
6. Fletcher, D. S.; Osinga, D. G.; Hand, K. M.; Dellea, P. S.;
Ashe, B. M.; Mumford, R. A.; Davies, P.; Hagmann, W.;
Finke, P. E.; Doherty, J. B.; Bonney, R. J. Am. Rev. Respir.
Dis. 1990, 141, 672.
7. Test compound was administrated into jugular vein 5 min
before an intratracheal instillation of HNE (25 units). One
hour after the induction of lung injury, BALF was collected
and hemorrhage in BALF was estimated by measuring of
absorbance at 414 nm.
8. Stein, R. L.; Strimpler, A. M.; Edwards, P. D.; Lewis, J. J.;
Mauger, R. C.; Schwartz, J. A.; Stein, M. M.; Trainor, D. A.;
Wildonger, R. A.; Zottola, M. A. Biochemistry 1987, 26, 2682.
References and Notes
1. (a) Weissman, G.; Smolen, J. E.; Korchak, H. M. N. Engl.
J. Med. 1980, 33, 7. (b) Weiland, J. E.; Davis, W. B.; Holter,
J. F.; Mohammed, J. R.; Dorinsky, P. M.; Gadek, J. E. Am.
Rev. Respir. Dis. 1986, 133, 218.