Design and synthesis of potent, non-peptide inhibitors of HCV NS3 protease

Article abstract of DOI:10.1016/S0960-894X(03)00032-5

Starting from a hexapeptide boronic acid lead, 3-amino bicyclic pyrazinones as novel β-sheet dipeptide mimetics have been designed and synthesized. Side-chain manipulation of this scaffold generated a series of potent, nonpeptidic inhibitors of HCV NS3 protease.

Full text of DOI:10.1016/S0960-894X(03)00032-5

Bioorganic & Medicinal Chemistry Letters 13 (2003) 1157–1160
Design and Synthesis of Potent, Non-peptide Inhibitors of
HCV NS3 Protease
Xiaojun Zhang,* Aaron C. Schmitt, Wen Jiang, Zelda Wasserman and Carl P. Decicco
Discovery Chemistry, Bristol-Myers Squibb, Route 141 and Henry Clay Road, Wilmington, DE19880, USA
Received 22 September 2002; accepted 12 December 2002
Abstract—Starting from a hexapeptide boronic acid lead, 3-amino bicyclic pyrazinones as novel b-sheet dipeptide mimetics have
been designed and synthesized. Side-chain manipulation of this scaffold generated a series of potent, nonpeptidic inhibitors of HCV
NS3 protease.
# 2003 Elsevier Science Ltd. All rights reserved.
Hepatitis C virus (HCV) infection is a major cause of
chronic liver disease leading to cirrhosis, hepatocellular
carcinoma or liver failure.1 Current therapies involve
treatment with interferon-a, either alone or in combi-
nation with ribavirin; however, poor efficacy and severe
side effects are frequent.2 There is thus an urgent need
to discover inhibitors of HCV-specific proteins for the
development of anti-HCV drugs. One of the most
intensively studied and best understood targets is the
NS3 serine protease.2 Recent work has demonstrated
that NS3 protease activity is required for HCV replica-
tion in chimpanzee.3 A low-molecular-weight inhibitor
of NS3 protease thus would be therapeutically useful
against HCV replication and infection.
domain reveal a substrate-binding pocket that is shal-
low and relatively solvent exposed.5,6 These features
underscore the difficulties of designing small molecule
inhibitors.7 Many of the reported inhibitors are peptides
derived from the C-terminal NS3 cleavage product.8À12
Potent peptide-based inhibitors were obtained when the
C-terminal carboxylates are activated as ‘serine traps’
such as a-ketoacid, ester and amide.13À15 Replacement
of the penultimate amino acid with an a-amino boronic
acid also gave potent inhibitors via a reversible, cova-
lent interaction of the boron with the serine hydroxyl
group.16,17 Starting from a hexapeptide boronic acid
lead, we report herein the design and synthesis of a ser-
ies of potent non-peptidic inhibitors of NS3 protease.
Many efforts have been made to identify potent inhibitors
of HCV NS3.4 X-ray crystal structures of the protease
Hexapeptide boronic acid 1 (Fig. 1) was shown by
Kettner to be very potent against NS3 with Ki=4 nM in
Figure 1. From hexapeptide inhibitor 1 to a non-peptide inhibitor 3.
*Corresponding author. Fax: +1-302-467-6850; e-mail: xiaojun.zhang@bms.com
0960-894X/03/$ - see front matter # 2003 Elsevier Science Ltd. All rights reserved.
doi:10.1016/S0960-894X(03)00032-5

1158
X. Zhang et al. / Bioorg. Med. Chem. Lett. 13 (2003) 1157–1160
Scheme 1. (a) LHMDS, THF, À78 ꢀC, R3X, R4=H, 50–75%; (b) LiBEt3H, THF, À78 ꢀC, then MeOH, H+(cat), >80%; (c) TMSCN, BF3 OEt2,
CH2Cl2, À78 ꢀC, then TFA; (d) (COCl2)2, toluene, 90 ꢀC, 35–60% for three steps; (e) R2NH2, EtOAc, 80 ꢀC, >90%; (f) LiOH, THF/H2O, then
.
.
NH2CH(allyl)B(O2C10H16) HCl, PyAOP, DIEA, DMF, 50–65%.
the enzyme assay.18 Considering the likely unfavorable
metabolic stability of structure 1, our goal was to design
a scaffold to significantly reduce the peptidic character
of inhibitor 1. Based on the assumption that proteases
typically bind their substrate in an extended conforma-
tion, we looked for b-sheet mimetics that have been
successfully used in other serine protease inhibitors. One
such mimetic, 3-amino pyrazinone, has been reported
effective against thrombin.19 It was postulated that the
CO and NH of the pyrazinone ring could form a pair of
b-sheet H-bonds with the enzyme backbone. We thus
replaced Val-Pro in 1 with a chloropyrazinone and the
Ac-Asp-Glu-Val of 1 with a non-peptide side chain R2
to give compound 2. A library of about 100 compounds
was prepared, but the binding affinities were moderate
(1–50 mM). However, studies of SAR suggested that a
bulky R1 (e.g., tert-butyl, neo-pentyl) increased the
binding affinity to low or sub mM level. This is likely a
(R3) at the 8-position increased the potency 18-fold,
giving 3b with IC50=0.26 mM. para-Substitution on the
benzyl with methyl, methoxy, or phenyl (3c, 3e, and 3g)
further increased the potency to about 50 nM, but meta-
substitution (3d and 3f) decreased the binding affinity.
Phenylpropyl (3h) and naphthylpropyl (3i) substitution
increased potency, with naphthylpropyl proving the best
(IC50=20 nM). Replacing the P1 allyl in 3i with an
ethyl gave an equally potent compound 3j. Compounds
3i and 3j represent a significant step towards completely
non-peptide NS3 inhibitors.
The enzyme is sensitive to the stereochemistry at the
8-position of 3. Though 3i is potent with IC50=20 nM,
its diastereomer 3l is 25-fold less potent. Similarly, 3k is
35-fold less potent than its diastereomer 3h.
We have also investigated the SAR of the amino cap-
ping group R2 (Table 2). Aliphatic chains of different
sizes (3m–s) had little effect on the potency, suggesting
that the R2 groupdid not strongly interact with the
enzyme. However, electron-withdrawing aromatics (3i
and 3w–y) slightly (ca. 5-fold) increased the binding
affinity with the enzyme. To gain more insight of the
SAR, we have generated a computer model illustrating
1
conformational effect due to the 1,2-relationshipof R
with the side chain. A crowded R1 would retard the
rotation of the N–CH2 bond, producing a relatively
rigid conformation. We reasoned that by a further con-
straint using a bicyclic pyrazinone as in compound 3,
more rigid conformation could be generated, creating
more potent compounds. Our results demonstrate that
this is indeed the case.
Table 1. SAR of the substituents at the 8-position of compound 3
Synthesis of 3 (Scheme 1) started with a stereoselective
alkylation of pyroglutamate 4 to 5.20,21 Chemo-selective
reduction of the ring carbonyl with LiBEt3H in 5 to a
hemiaminal, followed by solvent exchange with MeOH
gave 6. TMSCN under Lewis acid conditions converted
6 to aminonitrile 7 through an N-acyl iminium inter-
mediate. Cyclization of 7 with oxaly chloride smoothly
afforded dichloropyrazinone 8, which reacted with a
wide variety of primary amines to afford 3-amino bicyc-
lic pyrazinone 9 in high yields. Hydrolysis of 9 and cou-
Compd
R3
R4
R5
IC50, mMa
3a
3b
3c
3d
3e
3f
3g
3h
3i
H
C6H5CH2
H
H
H
H
H
H
H
H
Allyl
Allyl
Allyl
Allyl
Allyl
Allyl
Allyl
Allyl
Allyl
Et
4.6
0.26
0.06
0.40
0.05
0.26
0.06
0.06
0.02
0.02
2.0
.
pling with NH2CH(allyl)B(C10H16O2) HCl gave target 3
p-CH3C6H4CH2
m-CH3C6H4CH2
p-CH3OC6H4CH2
m-CH3OC6H4CH2
p-PhC6H4CH2
C6H5(CH2)3
2-naph(CH2)3
2-naph(CH2)3
H
in good yield. It should be noted that the boronic acid was
protected as a pinanediol ester to simplify the chemistry,
the active species is the acid which should be in fast equi-
librium with the ester under the assay condition. We made
the discrete free boronic acids of 3h and 3i (Table 1),
which displayed identical IC50 with the corresponding
pinanediol esters as expected. Compound 3h was shown
to be active-site directed, reversible inhibitor.
H
H
3j
3k
3l
C6H5(CH2)3
2-naph(CH2)3
Allyl
Allyl
H
0.51
aAll in vitro assays were performed as described previously;13 standard
deviation for the enzyme assays were typically Æ50% of the means or
less.
The parent pyrazinone 3a (R3=R4=H, Table 1) is a low
mM inhibitor with IC50=4.6 mM. A benzyl substitution

X. Zhang et al. / Bioorg. Med. Chem. Lett. 13 (2003) 1157–1160
1159
the NH off the pyrazinone ring, making it a better
H-bond donor, thus increasing binding affinity. The
phenylpropyl seems to cover a salt bridge formed by
Asp-Arg of the enzyme. However, the low solubility of
3h prevented it from being further evaluated in a func-
tional assay. Modification of the physical properties is
desired to further progress this series of inhibitors.
Table 2. SAR of the side chain R2 in compound 3
In summary, we have designed a 3-amino bicyclic pyra-
zinone scaffold that effectively replaces a dipeptide seg-
ment in a hexapeptide lead. By further manipulation of
the side chains of the scaffold, a series of potent non-
peptidic inhibitors of HCV NS3 was obtained. Because
of the correct alignment of the CO and NH, coupled
with rigidity of the bicyclic pyrazinone ring, we believe
this scaffold could be effective as a general b-sheet
mimetic that will find use in peptide chemistry and pro-
tease inhibitor design.
Compd
R2
IC50, mM
3m
3n
3o
3p
3q
3r
H
Me
Et
Pr
n-Bu
i-Bu
0.19
0.32
0.23
0.18
0.16
0.17
0.20
0.24
0.11
0.06
0.09
0.06
0.04
0.02
3s
3t
C6H11CH2
C6H5CH2
3,4-HOC6H3CH2
m-MeOC6H4CH2
2-Picolyl
m-FC6H4CH2
m-NO2C6H4CH2
m-CF3C6H4CH2
3u
3v
3w
3x
3y
3i
Acknowledgements
We thank Lorraine Gorey-Feret and James L. Meek
for measurement of IC50, Charles Kettner for helpful
discussions on aminoboronic acids as serine protease
inhibitors.
References and Notes
1. Cohen, J. Science 1999, 285, 26.
2. Kwong, A.; Kim, J.; Rao, G.; Lipovsek, D.; Raybuck, S.
Antiviral Res. 1998, 40, 1.
3. Kolykhalov, A.; Mihalik, K.; Feinstone, S.; Rice, C. J.
Virol. 2000, 74, 2046.
4. Walker, M. Drug Discov. Today 1999, 4, 518.
5. Love, R.; Parge, H. E.; Wickersham, J. A.; Hostomsky, Z.;
Habuka, N.; Moomaw, E. W.; Adachi, T.; Hostomska, Z. Cell
1996, 87, 331.
6. Kim, J. L.; Morgenstern, K. A.; Griffith, J. P.; Dwyer,
M. D.; Thomson, J. A.; Murcko, M. A.; Lin, C.; Caron, P. R.
Structure 1998, 6, 89.
7. Love, R.; Parge, H. E.; Wickersham, J. A.; Hostomsky, Z.;
Habuka, N.; Moomaw, E. W.; Adachi, T.; Margosiak, S.;
Dagostino, E.; Hostomska, Z. Clin. Diagn. Virol. 1998, 10,
151.
Figure 2. Inhibitor 3h modeled into the active site of HCV-NS3.
8. Ingallinella, P.; Bianchi, E.; Ingenito, R.; Koch, U.; Stein-
kuehler, C.; Altamura, S.; Pessi, A. Biochemistry 2000, 39,
12898.
9. Llinas-Brunet, M.; Bailey, M.; Deziel, R.; Fazal, G.; Gorys,
V.; Goulet, S.; Halmos, T.; Maurice, R.; Poirier, M.; Poupart,
M.; Rancourt, J.; Thibeault, D.; Wernic, D.; Lamarre, D.
Bioorg. Med. Chem. Lett. 1998, 8, 2719.
the important features that maybe involved in binding
of 3h in the active site. The computer model was based
on the structure of HCV protease, PDB file 1jxp.22
Because of the flexibility of R2 and R3 of 3h, multiple
calculations having different starting conformations
were carried out. Other than covalent attachment of the
ligand to the catalytic serine, the initial conformations
were generated at random and had little overlap. After
minimization and short molecular dynamics simulation
runs, using the method of Luty et al.,23 a predominant
binding mode emerged and is illustrated in Figure 2. In
this model, the R2 grouppoints away from the enzyme
surface into the solvent, explaining the flat SAR of R2.
Inhibitor 3h makes three H-bond interactions with the
enzyme backbone, a pair from the pyrazinone CO and
NH, the third from the P1 amide NH. We believe the
electron-withdrawing R2 groups increase the acidity of
10. Pessi, A. J. Peptide Sci. 2001, 7, 2.
11. Steinkuhler, C.; Biasiol, G.; Brunetti, M.; Urbani, A.;
Koch, U.; Cortese, R.; Pessi, A.; Francesco, R. Biochemistry
1998, 37, 8899.
12. Llinas-Brunet, M.; Bailey, M.; Fazal, G.; Ghiro, E.;
Gorys, V.; Golet, S.; Halmos, T.; Maurice, R.; Poirier, M.;
Poupart, M.-A.; Rancourt, J.; Thibeault, D.; Wernic, D.;
Lamarrer, D. Bioorg. Med. Chem. Lett. 2000, 10, 2267.
13. Han, W.; Hu, Z.; Jiang, X.; Decicco, C. Bioorg. Med.
Chem. Lett. 2000, 10, 711.
14. Narjes, F.; Brunetti, M.; Colarusso, S.; Gerlach, B.; Koch,
U.; Biasiol, G.; Fattori, D.; De Francesco, R.; Matassa, V.;
Steinkuehler, C. Biochemistry 2000, 37, 1849.

1160
X. Zhang et al. / Bioorg. Med. Chem. Lett. 13 (2003) 1157–1160
15. Marco, S.; Rizzi, M.; Volpari, C.; Walsh, M.; Narjes, F.;
Colarusso, S.; Francesco, R.; Matassa, V.; Sollazzo, M. J.
Biological Chem. 2000, 275, 7152.
16. Dunsdon, R.; Greening, J.; Jones, P.; Jordan, S.; Wilson,
F. Bioorg. Med. Chem. Lett. 2000, 10, 1577.
17. Attwood, M.; Bennett, J.; Campbell, A.; Canning, G.; Carr,
M.; Conway, E.; Dunsdon, R.; Greening, J.; Jones, P.; Kay, P.;
Handa, B.; Hurst, D.; Jennings, N.; Jordan, S.; Keech, E.;
O’Brien, M.; Overton, H.; King-Underwood, J.; Raynham, T.;
Stenson, K.; Wilkinson, C.; Wilkinson, T.; Wilson, F. Anti-
viral Chem. Chemother. 1999, 10, 259.
I.; Chen, Z.; Cook, J. J.; Cooper, C. M.; Gardell, S. J.; Hare,
T. R.; Krueger, J. A.; Lewis, S. D.; Lin, J. H.; Lucas, B. J.; Lyle,
E. A.; Lynch, J. J.; Stranieri, M. T.; Vastag, K.; Yan, Y.; Shafer,
J. A.; Vacca, J. P. J. Med. Chem. 1998, 41, 4466.
20. Ezquerra, J.; Pedregal, C.; Rubio, A.; Valenciano, J.;
Navio, J. L. G.; Alvarez-Builla, J.; Vaquero, J. J. Tetrahedron
Lett. 1993, 34, 6317.
21. Ezquerra, J.; Pedregal, C.; Rubio, A. J. Org. Chem. 1994,
59, 4327.
22. Yan, Y.; Li, Y.; Munshi, S.; Sardana, V.; Cole, J. L.;
Sardana, M.; Steinkuehler, C.; Tomei, L.; DeFrancesco, R.;
Kuo, L. C.; Chen, Z. Protein Sci. 1998, 7, 837.
23. Luty, B. A.; Wasserman, Z. R.; Stouten, P. F. W.; Hodge,
C. N.; Zacharias, M.; McMammon, J. A. J. Computational
Chem. 1995, 16, 454.
18. Kettner, C. Dupont Pharmaceuticals Company. Unpub-
lished data.
19. Sanderson, P. E. J.; Lyle, T. A.; Cutrona, K. J.; Dyer, D. L.;
Dorsey, B. D.; McDonough, C. M.; Naylor-Olsen, A. M.; Chen,