Highly potent and selective peptide-based inhibitors of the hepatitis C virus serine protease: Towards smaller inhibitors

Article abstract of DOI:10.1016/S0960-894X(00)00465-0

Structure-activity studies on a hexapeptide N-terminal cleavage product of a dodecamer substrate led to the identification of very potent and highly specific inhibitors of the HCV NS3 protease/NS4A cofactor peptide complex. The largest increase in potency

Full text of DOI:10.1016/S0960-894X(00)00465-0

Bioorganic & Medicinal Chemistry Letters 10 (2000) 2267±2270
Highly Potent and Selective Peptide-Based Inhibitors of the
Hepatitis C Virus Serine Protease: Towards Smaller Inhibitors
Montse Llinas-Brunet,* Murray Bailey, Gulrez Fazal, Elise Ghiro, Vida Gorys,
Sylvie Goulet, Ted Halmos, Roger Maurice, Martin Poirier, Marc-Andre Poupart,
Jean Rancourt, Diane Thibeault, Dominik Wernic and Daniel Lamarre
Boehringer Ingelheim (Canada) Ltd, Research and Development, 2100 Cunard Street, Laval, Quebec, Canada H7S 2G5
Received 9 June 2000; accepted 28 July 2000
AbstractÐStructure±activity studies on a hexapeptide N-terminal cleavage product of a dodecamer substrate led to the identi®ca-
tion of very potent and highly speci®c inhibitors of the HCV NS3 protease/NS4A cofactor peptide complex. The largest increase in
potency was accomplished by the introduction of a (4R)-naphthalen-1-yl-4-methoxy substituent to the P2 proline. N-Terminal
truncation resulted in tetrapeptides containing a C-terminal carboxylic acid, which exhibited low micromolar activity against the
HCV serine protease. # 2000 Elsevier Science Ltd. All rights reserved.
Hepatitis C virus (HCV) infection is an important cause
of chronic liver disease leading to cirrhosis and end-
stage liver disease in humans. Liver failure from chronic
hepatitis C is now the leading indication for liver trans-
plantation in the United States. Over 150 million people
world-wide are persistently infected with HCV and the
number of deaths attributable to chronic infection is likely
to rise dramatically over the next decade.1 Current thera-
pies for HCV disease involve treatment with interferon-a,
either alone or in combination with the purine nucleoside
analogue ribavirin. However, poor ecacy and relapse
after cessation of therapy are frequent.2 Therefore,
hepatitis C is an important emerging infectious disease
that warrants signi®cant e€orts in the development of
novel and more e€ective treatment strategies.
inhibited by N-terminal cleavage products of peptide
substrates. Hexapeptide 1, the N-terminal cleavage pro-
duct of a peptide substrate corresponding to the NS5A/
5B cleavage site, is an inhibitor of the enzyme (Table 1).6
Capping the N-terminus of 1 with an acetyl group and
replacing the cysteine at P1 with norvaline produced
chemically stable hexapeptide 2, which exhibited mod-
erate potency. Introduction of a d-aspartic acid at the
P5 position increased inhibitor potency by ꢀ10-fold
(compound 3).6,8 In contrast to what is observed with
other serine proteases, peptide sequences containing
electrophilic carbonyls such as tri¯uoromethyketone 4
did not o€er an overall advantage over the correspond-
ing carboxylic acid.8 The carboxylic acid functionality
at the C-terminus contributes considerably to potency
and imparts great speci®city to these peptide-based
inhibitors of the HCV serine protease. In this Letter,
we describe structure±activity studies on this series of
C-terminal carboxylic acid inhibitors. These studies led
to a substantial increase in potency allowing for a
reduction in the peptidic nature of these compounds.
One of the most intensively studied targets for antiviral
therapy against HCV is the serine protease of the NS3
protein.3 The NS3 protease domain, which constitutes
the amino-terminal third of the NS3 protein, is a chy-
motrypsin/trypsin-like serine protease responsible for the
proteolytic cleavage of the nonstructural NS3, NS4A,
NS4B, NS5A, and NS5B proteins. The protease activity is
enhanced by the NS4A protein, which acts as a cofactor
essential to polyprotein maturation.4 Recent work by
Rice et al. has demonstrated that NS3 protease activity
is required for HCV replication in the chimpanzee.5
Table 1. Hexapeptides inhibitors of NS3-4Apep protease
Compound
Sequenceb
IC50, mMa
1
2
3
4
DDIVPC-OH
71
150
17
We,6 and others,7 have shown that the NS3 protease
complexed with NS4A cofactor peptide (NS3-4Apep) is
Ac-DDIVP-Nva-OH
Ac-DdIVP-Nva-OH
Ac-DdIVP-Nva-CF3
22
a
*Corresponding author. Tel.: +1-450-682-4640; fax: +1-450-682-4189;
e-mail: mllinas@lav.boehringer-ingelheim.com
Values are the mean of at least four determinations.
d=d-aspartic acid; Nva=Norvaline.
b
0960-894X/00/$ - see front matter # 2000 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(00)00465-0

Á
M. Llinas-Brunet et al. / Bioorg. Med. Chem. Lett. 10 (2000) 2267±2270
2268
Table 2. SAR at the P2 proline
Table 3. Optimization of 4-R-Substituent in proline
Compound 2 IC50=150mM
Compound 5 IC50=7 mM
8 IC50=10 mM
9 IC50=60 mM
10 IC50=3.3 mM
5 IC50=7 mM
6 IC50=93 mM
7 IC50=206 mM
Results and Discussion
11 (o) IC50=6.8 mM
12 (m) IC50=3.5 mM
13 (p) IC50=2.5 mM
14 IC50=0.71 mM
15 IC50=0.39 mM
Structure±activity relationship (SAR) studies on hexa-
peptide 2 revealed that the introduction of a benzyloxy
group with the (R)-stereochemistry at the 4-position of
the P2 proline (compound 5, Table 2) resulted in a 21-
fold increase in potency.9 The importance of the benzyl
ring and its stereochemistry is evidenced by compounds
6 and 7. (4R)-Hydroxyproline analogue 6 and (4S)-
benzyloxyproline derivative 7 have activities similar to
the unsubstituted analogue 2.
found to be a better substitution than the d-aspartic
acid (compound 17 vs 3, Table 1).
Additivity of bene®cial substitutions
The oxygen in benzyloxy derivative 5 does not sig-
ni®cantly contribute to the potency since the carbon
analogue 8 is equipotent (Table 3). In order to study the
optimal distance between the phenyl and proline rings,
compounds 9 and 10 were synthesized. A shorter chain
length, as represented by compound 9, is detrimental,
but a longer chain length (compound 10) slightly
increases the inhibitor potency.
Combining the optimal substitutions at each position in
the hexapeptide resulted in very potent and competitive
inhibitors of the enzyme. Compound 18 (Fig. 1), which
contains a (4R)-naphthen-2-yl- methoxyproline at P2, a
cyclohexylglycine at P411 and a d-glutamic acid at P5,
has an IC50 of 0.027 mM. Kinetic studies showed that
compound 18 is a competitive inhibitor of the NS3-
4Apep protease with a Ki=0.003 mM.9 In addition,
We have previously reported NMR and computational
chemistry studies indicating that compound 5 binds the
protease in a well-de®ned extended conformation.10 In
addition, a di€erential line broadening NMR study
showed that the benzyloxy substituent, although quite
¯exible, is in direct contact with the protease.10 There-
fore, in an attempt to further increase the inhibitory
potency, substituents were introduced onto the phenyl
ring of compound 5 (Table 3). A methyl group at either
the ortho, meta, or para position of the phenyl ring
(compounds 11, 12, and 13) was well tolerated and did
not produce a signi®cant change in the potency of these
inhibitors. However, a substantial increase in potency
was observed by the addition of an additional phenyl
ring (compounds 14 and 15). The naphthen-1-ylmeth-
oxy derivative 15 was 18-fold more potent than the
benzyloxy analogue 5 and ꢀ400-fold more potent than
analogue 2, giving rise to inhibitors of the NS3-4Apep
serine protease with submicromolar potency.
Table 4. Optimization of compound 2
Compound
IC50
(mM)
2
150
16
17
54
7.5
Table 4 presents further optimizations of compound 2.
1-Amino-cyclopropylcarboxylic acid was identi®ed as a
good replacement for the P1 norvaline (compound 16).
In addition, d-glutamic acid at the P5 position was

Á
M. Llinas-Brunet et al. / Bioorg. Med. Chem. Lett. 10 (2000) 2267±2270
2269
Previous NMR studies10 showed that compound 5
binds to the protease in a well-de®ned extended con-
formation from P1 to P4 with little to no contact
observed between the P5 and P6 side chains and the
protein. This suggested that N-terminal truncation
could be possible. Table 5 shows the e€ect on inhibitory
potency upon N-terminal truncation.
Replacing the Ac-Asp-d-Glu- residues in compound 19
with a 4-carboxybutanoyl moiety a€orded compound
20 that is ꢀ70-fold less potent. The 4-carboxybutanoyl
capping group in 20 did not contribute much to inhi-
bitor potency as shown by the acetyl tetrapeptide deri-
vative 21. Compound 21, with an IC50 of 3.5 mM,
represents a low micromolar, monocharged inhibitor of
the protease. These tetrapeptides are also highly speci®c
as demonstrated by the lack of activity against HLE
(IC50>600 mM), a serine protease with related substrate
recognition. The NMR-derived bound conformation of
one such HCV serine protease inhibitor is reported in
the next Letter.
Figure 1.
SAR directed to improve the potency of this new series
is currently underway.
compound 18 is very speci®c for the HCV serine pro-
tease. As shown in Figure 1, compound 18 does not
signi®cantly inhibit other serine or cysteine proteases.
Chemistry
Compound 19 (Table 5), which contains a 1-amino-
cyclopropylcarboxylic acid at P1 in addition to other
optimal substitutions, is also a very potent inhibitor of
the enzyme.
The synthesis of inhibitors was carried out by standard
solid-phase or solution-phase peptide chemistry.12,13
Various 4-substituted prolines were prepared as
follows:
Size reduction
Having identi®ed low nanomolar NS3-4Apep protease
inhibitors, we decided to explore reducing their size.
a. 4-Arylmethoxyproline derivatives required for the
synthesis of compounds 5, 7, and 11±15 were
prepared from the sodium alkoxide of N-Boc-
(4R)-hydroxyproline and various aryl bromides
according to the procedure of Smith.14
Table 5. Size reduction
b. The synthesis of (4R)-benzyl and (4R)-(3-phenyl)-
propylproline derivatives (for compounds 9 and
10, respectively) was achieved via alkylation of
pyroglutamate ester 22 according to the procedure
of Ezquerra et al. (Scheme 1).15 The lithium eno-
late of protected pyroglutamic ester could be ste-
reoselectively alkylated with reactive electrophiles
without epimerization of the a-center. The resulting
lactams 23 could then be reduced chemoselectively
with lithium triethylborohydride followed by fur-
ther reduction with triethylsilane/borontri¯uoride
etherate to yield (4R)-substitued prolines 24.16
Compound
IC50
(mM)
19
0.013
c. The 4-phenethylproline derivative incorporated
into compound 9 was prepared as shown in
Scheme 2. Starting from allyl derivative 24c,
osmylation of the double bond followed by oxi-
dative cleavage of the resulting diol 25 with
sodium periodinate, gave the aldehyde 26 in
quantitative yield. A Grignard reaction with phe-
nylmagnesium bromide gave the expected benzylic
alcohol 27 in 57% yield. Hydrogenation over pal-
ladium on charcoal resulted in hydrogenolysis of
both the benzyl alcohol and the benzyl ester to
yield the desired (4R)-phenethylproline 28.
20
0.9
21
3.5

Á
M. Llinas-Brunet et al. / Bioorg. Med. Chem. Lett. 10 (2000) 2267±2270
2270
Acknowledgements
We thank P. Bonneau, C. Plou€e and H. Li for speci®-
city assays, the analytical chemistry department for
assistance and S. Kawai for proof reading the manu-
script. We also thank P. Anderson, M. Bos, and M.
Cordingley for encouragement and support.
References and Notes
1. Houghton, M. In Virology, 4th ed.; Fields, B. N., Knipe, P.
M., Howley, P. M., Eds.; Lippincott-Raven: Philadelphia
1998; pp 1035±1035.
2. Moussalli, J.; Opolon, P.; Poynard, T. J. Viral Hep. 1998,
5, 73.
Scheme 1.
3. For recent reviews see: (a) Bartenschlager, R. J. Viral Hep.
1999, 6, 165, (b) De Francesco, R.; Steinkuhler, C. Hepatitis C
Viruses. 2000, 242, 149 and references therein.
4. (a) Bartenschlager, R.; Lohmann, V.; Wilkinson, T.; Koch,
J. O. J. Virol. 1995, 69, 7519. (b) Failla, C.; Tomei, L.; De
Francesco, R. J. Virol. 1995, 69, 1769. (c) Lin, C.; Thomson,
J. A.; Rice, C. M. J. Virol. 1995, 69, 4373.
5. Kolykalov, A. A.; Mihalik, K.; Feinstone, S. M.; Rice, C.
M. J. Virol. 2000, 74, 2046.
6. Llinas-Brunet, M.; Bailey, M.; Fazal, G.; Goulet, S.; Halmos,
T.; Laplante, S.; Maurice, R.; Poirier, M.; Poupart, M.-A.; Thi-
beault, D.; Wernic, D.; Lamarre, D. Bioorg. Med. Chem. Lett.
1998, 8, 1713.
7. Steinkuhler, C.; Biasiol, G.; Brunette, M.; Urbani, A.;
Koch, U.; Cortese, R.; Pessi, A.; De Francesco, R. Biochem-
istry 1998, 37, 8899.
Scheme 2.
8. Llinas-Brunet, M.; Bailey, M.; Deziel, R.; Fazal, G.; Gorys,
V.; Goulet, S.; Halmos, T.; Maurice, R.; Poirier, M.; Poupart,
M-A.; Rancourt, J.; Thibeault, D.; Wernic, D.; Lamarre, D.
Bioorg. Med. Chem. Lett. 1998, 8, 2719.
9. The protease strain and assay conditions used to measure
the IC50 of the inhibitors are described in ref 6. All the IC50
values are an average of at least four measurements.
10. LaPlante, S.; Cameron, D. R.; Aubry, N.; Lefebvre, S.;
Kukolj, G.; Maurice, R.; Thibeault, D.; Lamarre, D.; Llinas-
Brunet, M. J. Biol. Chem. 1999, 274, 18618.
11. Cyclohexylglycine at P4 systematically produces inhibitors
that are three- to ®ve-fold more potent than those containing
an isoleucine (data not shown).
12. Bodanszky, M. Peptide Chemistry, 2nd rev. ed.; Spinger-
Verlag: Berlin, 1993.
13. Stewart, J. M.; Young, J. D. In Solid Phase Peptide
Synthesis, 2nd ed.; Gross, E., Meienhofer, J., Udenfriend, S.,
Ed.; Pierce Chemical Co: Rockford, 1984.
14. Smith, E. M.; Swiss, G. F.; Neustadt, B. R.; Gold, E. H.;
Sommer, J. A.; Brown, A. D.; Chiu, P. J. S.; Moran, R.;
Sybertz, E. J.; Baum, T. J. Med. Chem. 1988, 31, 875.
15. Ezquerra, J.; Pedregal, C.; Rubio, A.; Yruretagoyena, B.;
Escribano, A.; Sanchez-Ferrando, F. Tetrahedron 1993, 38, 8665.
16. Pedregal, C.; Ezquerra, J.; Escribano, A.; Carreno, M. C.;
Garcia Ruano, J. L. Tetrahedron Lett. 1994, 13, 2053.
Summary
Structure±activity studies on hexapeptide 2 led to the
identi®cation of very potent, speci®c and competitive
inhibitors of the HCV serine protease. The largest
increase in potency (ꢀ400-fold) was obtained by the
introduction of a (4R)-naphthen-1-ylmethoxy group to
the P2 proline. The combination of (4R)-naphthen-1-
ylmethoxy proline at P2, d-glutamic acid at P5, cyclo-
hexyl glycine at P4 and 1-amino-cyclopropyl carboxylic
acid at P1 generated compound 19, a low nanomolar
inhibitor with an IC50 of 13 nM. These compounds are
competitive inhibitors of the enzyme as demonstrated
for compound 18 with a Ki of 3 nM. The large increase
in potency has allowed a reduction in the size and pep-
tidic nature of these compounds. A new series of potent
and speci®c tetrapeptide inhibitors of the NS3-4Apep
protease has been introduced with compound 21 which
displays an IC50 of 3.5 mM.