Aminothiazole inhibitors of HCV RNA polymerase

Article abstract of DOI:10.1016/j.bmcl.2004.10.024

Aminothiazole-based inhibitors designed for HCV polymerase display low micromolar potencies in biochemical assays. These compounds show a stringent preference for a cyclohexyl hydrophobe at the 2-amino position. The composition of these compounds suggests that they may be interacting at a recently discovered allosteric site on the polymerase.

Full text of DOI:10.1016/j.bmcl.2004.10.024

Bioorganic & Medicinal Chemistry Letters 15 (2005) 115–119
Aminothiazole inhibitors of HCV RNA polymerase
Gerald W. Shipps, Jr.,a,* Yongqi Deng,a Tong Wang,a Janeta Popovici-Muller,a
Patrick J. Curran,a Kristin E. Rosner,a Alan B. Cooper,b Viyyoor Girijavallabhan,b
Nancy Butkiewiczb and Michael Cableb
aNeoGenesis Pharmaceuticals Inc., Department of Chemistry, 840 Memorial Drive, Cambridge, MA 02139, USA
bSchering-Plough Research Institute, 2015 Galloping Hill Road, Kenilworth, NJ 07033, USA
Received 26 August 2004; revised 8 October 2004; accepted 8 October 2004
Available online 28 October 2004
Abstract—Aminothiazole-based inhibitors designed for HCV polymerase display low micromolar potencies in biochemical assays.
These compounds show a stringent preference for a cyclohexyl hydrophobe at the 2-amino position. The composition of these com-
pounds suggests that they may be interacting at a recently discovered allosteric site on the polymerase.
Ó 2004 Elsevier Ltd. All rights reserved.
The hepatitis C virus (HCV) chronically infects approxi-
mately 3% of the worldÕs population and is a leading
cause of liver transplantation in the United States.1
Because the virus often results in fibrosis and cirrhosis
of the liver over a time course that can last decades, it
is projected that the societal costs of this disease will
increase over the next 20years.2 Current treatment
options are limited, with interferon a-2b/ribavirin com-
binations having the most widespread application.
Unfortunately this combination does not always pro-
duce sustained responses across all viral genotypes,
particularly genotype 1. Therefore, the development of
new inhibitors targeting HCV becomes urgent.
within a narrow cleft on the proteinÕs surface in the
ÔthumbÕ domain.10
In this letter we describe a novel series of HCV RNA
polymerase inhibitors developed from an original hit
7a identified by screening our library using NeoGenesis
ALISTM (automated ligand identification system) techno-
logy. Compound 7a showed approximately 20lM activ-
ity in the biochemical assay and the initial optimization
of 7a resulted in modest potency improvement. After a
comparison with other inhibitors reported in the litera-
ture (1, 2) we decided to shift the cyclohexyl group to
the amide nitrogen placing the two hydrophobic groups
adjacent to each other (7b). Although the initial change
resulted in loss of activity, the subsequent replacement
of pyrazole with thiazole generated inhibitor 8 with an
IC50 of 7lM. Herein, we would like to report the syn-
thesis and optimization of a series of HCV polymerase
inhibitors based on aminothiazole 8. Several compounds
were optimized to low micromolar potencies in the bio-
chemical assay. The SAR trends of these compounds
clearly reveal that hydrophobicity at the 2-amino posi-
tion is important for inhibition in the biochemical assay.
HCV is a single stranded RNA virus of the Flaviviridae
family, whose genome encodes for a polyprotein consist-
ing of both structural and nonstructural proteins.3 Of
the six nonstructural targets the two that have been pur-
sued most vigorously are the NS3 serine protease and
the NS5B RNA dependent RNA polymerase. Known
polymerase inhibitors can be grouped into three catego-
ries (Fig. 1): generally hydrophobic heterocyclic com-
pounds, which contain an acidic functionality (1–3),4–6
ketoacids or their isosteres that likely target the active
site (4, 5)7,8 and a third series (6), which has been
recently disclosed by GSK.9 The first category of inhibi-
tors (i.e., 3) has recently been linked to an allosteric site
The general synthesis of the aminothiazole inhibitors is
illustrated in Scheme 1. The acylation step proceeded
with high efficiency in most cases, while the yields of
the Mitsunobu reactions were modest (10–50%). Alter-
nate synthetic schemes, starting with alkylation of the
aminothiazole, followed by acylation were less success-
ful. The final deprotection of the ethyl ester with LiOH
in THF generated the test compounds in 80–90% yields.11
Keywords: HCV; Aminothiazole.
*
Corresponding author. Tel.: +1 617 588 5122; fax: +1 617 868
1515; e-mail: shipps@neogenesis.com
0960-894X/$ - see front matter Ó 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2004.10.024

116
G. W. Shipps, Jr. et al. / Bioorg. Med. Chem. Lett. 15 (2005) 115–119
Cl
HO
O
Cl
O
N
O
H
N
N
N
O
HO
Cl
O
N
H
O
O
N
N
S
O
O
Cl
HO
3
2
1
4
O
O
O
S
Cl
Cl
S
OH
N
N
H
S
N
N
H
N
H
O
OH
Cl
H
N
OH
N
N
O
Cl
O
O
OH
O
O
OH OH
5
6
O
O
O
N
O
N
O
S
N
PhO
PhO
OH
O
OH
PhO
OH
N
N
HN
N
R
N
R = H or Me
7b
8
7a
Figure 1.
O
showed similar SAR trends. The removal of the cyclo-
hexyl group resulted in complete loss of activity (9,
21). Replacing the cyclohexane ring with smaller
branched alkyl groups such as isopropyl (10, 22) also
obliterated the inhibitory activity. While introduction
of a sulfide group into the ring caused a slight loss of
activity (11), the oxidation of sulfide into sulfone (12) re-
sulted in complete loss of activity (12). Substitution of
the cyclohexane ring with a methyl group at the 2-posi-
tion had modest impacts on the potency, as the cis was
slightly favored (14, 24) over the trans form (13, 23).
Substitution of cyclohexane at 4-position with a methyl
group had little impact on the activity in the 5-carboxy-
lic acid series (15, 16) but for the 4-carboxylic acid series,
twofold activity improvement was observed (25, 26).
Substitution with a much larger t-butyl group, particu-
larly in the cis-form, resulted in a twofold improvement
in potency in the 5-carboxylic acid series (18) and three-
fold in the 4-carboxylic acid series (28).
O
N
N
CO2Et
CO2Et
CO2Et
a
c
+
Ar
Ar
Ar
H2N
N
S
S
Cl
H
b
O
O
N
N
CO2H
Ar
N
R
N
R
S
S
Scheme 1. General synthetic sequence for HCV polymerase inhibitors.
Reagents and conditions: (a) DMAP, 10equiv pyridine, dichlorometh-
ane, 0°C–rt, 2h; (b) triphenylphosphine, diisopropyl azodicarboxylate,
ROH, THF, 0°C–rt, 4h; (c) 5equiv 1M LiOH in THF, rt, 12h.
To measure the efficacy of these compounds an SPA-
based RNA polymerase assay was performed using radio-
labeled GTP, a poly C oligo G template/primer and the
D-55 construct of NS5B according to a modified litera-
ture procedure.12 In some cases compounds were also
assayed using the D-21 (BK isolate) construct of the
enzyme and similar, albeit usually lower, IC50 values
were determined.13
In our initial optimization of 7, we found that 4-phen-
oxy or 4-benzoxy hydrophobic motifs are necessary for
activity. Attempts to replace the oxygen linker with
nitrogen or carbon caused a loss of activity. Directly
linking the distal ring to the phenyl group also resulted
in loss of activity. From this initial SAR, it became clear
that the 4-phenoxy and benzyloxy compounds were of
greater interest. Since both 4-, and 5-carboxylic acid ser-
ies showed similar SAR trends in the previous optimiza-
The importance of the cyclohexyl substituent at the N2-
position was explored in both the 4 and the 5 substituted
aminothiazole series and is highlighted in Table 1 using
data from the D-55 construct. Overall, both series

G. W. Shipps, Jr. et al. / Bioorg. Med. Chem. Lett. 15 (2005) 115–119
Table 1. Modification of cyclohexyl group
117
O
5
1
O
S
OH
4
N
R
2
N
3
PhO
HCV NS5B IC50 (lM)
R
H
Compd
5-Carboxylic acid
Compd
4-Carboxylic acid
10 1.0
8
7
0.3
20
9
>50
>50
21
22
>50
>50
10
11
12
13 0.5
S
O
O
S
>50
13
14
9.4 1.6
6.1 1.0
23
24
11 2.0
trans
cis
7.6 0.5
trans
15
16
17
18
5.0 1.0
6.6 1.4
6.0 0.3
3.4 0.4
25
26
27
28
5.4 0.8
5.7 0.6
4.7 0.9
2.8 0.3
cis
trans
cis
19
6.0 2.0
29
5.5 0.8
tion, the SAR exploration on the distal group was
mainly focused on the 5-carboxylic acid series. In an
effort to expedite the compound synthesis, advanced
intermediates containing a readily functionalized phenol
group were prepared. General alkylation and copper
arylation procedures were then used to synthesize ana-
logs (Scheme 2). A large number of substituted phenyl
boronic acids and benzyl bromides were used to react
with 30, and some examples are shown in Table 2. In
the phenoxy analogs, it was apparent that substitution
of the phenyl ring at 3 or 4-position led to more active
compounds compared to 8, with the 3-position being
favorable (33–35). This SAR pattern was conserved in
the benzyloxy series as well, where compound 43 had
approximately fourfold improved potency compared to
40. The improvement in potency with the increase of
hydrophobicity of the distal ring led us to speculate that
this distal group might be occupying a hydrophobic
pocket during its binding to the protein. Indeed, when
the distal phenyl group in 8 was replaced with a cyclo-
hexane ring (31), the inhibitory activity was retained.
However, when a more polar group (e.g., pyridine)
was used in the place of phenyl (41), its potency was re-
duced compared to the parent compound (40).
A preliminary evaluation of in vitro DMPK (drug
metabolism and pharmacokinetics) properties was per-
formed on a few selected compounds and favorable
results were obtained. For example, compound 43 had
147nM/s permeability in the CACO2 experiment and
was found to be 5.0lL/min/million clearance rate in
human hepatocytes. It was also found to be selective
against CYP’s and had an IC50 of over 30lM in both
CYP 3A4 and CYP 2D6 assays.

118
G. W. Shipps, Jr. et al. / Bioorg. Med. Chem. Lett. 15 (2005) 115–119
CO2H
CO2H
O
S
S
N
RO
N
N
CO2Et
CO2Et
O
O
S
S
a
N
N
BnO
HO
N
N
O
N
RO
30
Scheme 2. Synthetic routes used for discrete syntheses. Reagents and conditions: (a) BCl3, dichloromethane, À78°C; (b) copper(II) acetate, boronic
acid, pyridine, air, 48h; (c) cesium carbonate, benzyl bromides or chlorides, MeCN, 55°C, overnight; (d) 5equiv 1M LiOH in THF, rt, 12h.
Table 2. SAR at the distal hydrophobic group
O
O
RO
S
N
OH
N
R
Compd
HCV NS5B IC50 (lM)
R
Compd
HCV NS5B IC50 (lM)
8
7.0 0.3
40
6.5 1.0
31
32
33
5.5 0.4
4.2 0.8
2.8 0.2
41
42
43
15 1.0
7.0 1.0
1.4 0.3
N
S
Cl
Cl
F3C
F3C
34
35
36
37
4.4 1.0
3.1 0.4
8.5 1.6
2.6 0.2
44
45
46
47
5.1 0.5
4.0 1.0
3.9 0.2
6.5 1.0
CF3
MeO
F
OMe
F
F
Cl
Cl
MeO
38
39
3.5 0.5
5.2 0.7
48
49
8.5 1.6
3.2 0.5
OMe
MeO
MeO
F3CO
In conclusion, suitably substituted aminothiazoles have
been shown to be active against the NS5B RNA depend-
ent RNA Polymerase. The need for a branched aliphatic
hydrophobe on the 2-amino group was quite conserved
across all analogs. There was some variability in the ori-
entation of the distal aromatic hydrophobes, with a
slight preference for meta substitution. The position of
the carboxyl group was also variable with no significant
preference for either the 4 or 5 positions on the thiazole
ring. One explanation for this lack of preference may be
that the ring heteroatoms are not forming close or
highly directional contacts with NS5B. The composition

G. W. Shipps, Jr. et al. / Bioorg. Med. Chem. Lett. 15 (2005) 115–119
119
7. (a) Altamura, S.; Tomei, L.; Kocii, U.; Neunter, P. J.;
Summa, V. International Patent WO 0006529, 1999; (b)
Summa, V.; Petrocchi, A.; Pace, P.; Matassa, V. G.; De
Francesco, R.; Altamura, S.; Tomei, L.; Koch, U.;
Neuner, P. J. Med. Chem. 2004, 47, 14.
8. Gardelli, C.; Giuliano, C.; Harper, S.; Koch U.; Narjes,
F.; Ontoria, J.; Poma, M.; Ponzi, S.; Stansfield, I.; Summa,
V. International Patent WO 06246, 2002.
9. Dhanak, D.; Duffy, K.; Johnston, V. K.; Lin-Goerke, J.;
Darcy, M.; Shaw, A. N.; Gu, B.; Silverman, C.; Gates, A.
T.; Nonnemacher, M. R.; Earnshaw, D. L.; Casper, D. J.;
Kaura, A.; Baker, A.; Greenwood, C.; Gutshall, L. L.;
Maley, D.; DelVecchio, A.; Macarron, R.; Hofmann, G.
A.; Alnoah, Z.; Cheng, H.-Y.; Chan, G.; Khandekar, S.;
Keenan, R. M.; Sarisky, R. T. J. Biol. Chem. 2002, 277,
38322.
of these compounds suggests that they may be interact-
ing at the recently discovered allosteric site on the
polymerase. Select compounds have been analyzed in
DMPK studies and their profiles in these assays are
encouraging. This series is currently undergoing more
extensive evaluation and it is hoped that further efforts
on this novel HCV series will produce more efficacious
compounds.
Acknowledgements
The authors would like to acknowledge Dr. A. Siddiqui,
Dr. B. A. Malcolm, and Dr. B. M. Baroudy for insight-
ful discussions.
10. (a) Wang, M.; Ng, K. K.-S.; Cherney, M. M.; Chan, L.;
Yannopoulos, C. G.; Bedard, J.; Morin, N.; Nguyen-Ba,
N.; Alaoui-Ismaili, M. H. C.; Bethell, R. C.; James, M. N.
G. J. Biol. Chem. 2003, 278, 9489; (b) Love, R. A.; Parge,
H. E.; Yu, X.; Hickey, M. J.; Diehl, W.; Gao, J.; Wriggers,
H.; Ekker, A.; Wang, L.; Thomson, J. A.; Dragovich, P.
S.; Fuhrman, S. A. J. Virol. 2003, 77, 7575.
References and notes
1. (a) Choo, Q.-L.; Kuo, G.; Weiner, A. J.; Overby, L. R.;
Bradley, D. W.; Houghton, M. Science 1989, 244, 359; (b)
Memon, M. I.; Memon, M. A. J. Viral. Hepat. 2002, 9, 84.
2. (a) Hoofnagle, J. H. Hepatology 1997, 26, S15; (b) Wasley,
A.; Alter, M. J. Semin. Liver Dis. 2000, 20, 1.
11. All compounds in this study were characterized by 1H
NMR and MS. Purity was assessed by HPLC.
3. (a) Beaulieu, P. L.; Montse, L.-B. Curr. Med. Chem. Anti-
Infective Agents 2002, 1, 163; (b) Tan, S.-L.; Pause, A.;
Shi, Y.; Sonenber, N. Nature 2002, 1, 867; (c) Adachi, T.;
Ago, H.; Habuka, N.; Okuda, K.; Komatsu, M.; Ikeda, S.;
Yatsunami, K. Biochim. Biophys. Acta 2002, 1601, 39.
4. Hiromasa, H.; Kenji M.; Hitoshi Y. Japanese Patent JP
247550A, 2001.
5. Beaulien, P. L.; Bo¨s, M.; Bousquet, Y.; DeRoy, P.; Fazal,
G.; Gauthier, J.; Gillard, J.; Goulet, S.; Mckercher, G.;
Poupart, M.-A.; Valois, S.; Kukolj, G. Bioorg. Med.
Chem. Lett. 2004, 14, 967.
6. (a) Chan, L.; Reddy, T. J.; Proulx, M.; Das, S. K.; Pereira,
O.; Wang, W.; Siddiqui, A.; Yannopoulos, C.; Turcotte,
N.; Drouin, A.; Alaouii, H.; Bethel, R. C.; Lucile, H.;
Nguyen-Ba, N. J. Med. Chem. 2003, 46(8), 1283; (b)
Ismaili, H.; Moulay, A.; Cheng, Y.-X.; Lavallee, J.-F.;
Siddiqui, A.; Storer, R. International Patent WO 60315,
2001 and U.S. Patent 0,019,363, 2002.
12. Briefly, 50lL reaction mixture containing 20mM HEPES
(pH7.3), 7.5mM DTT, 20units/mL RNasIN, 0.5lg/mL
biotinylated oligoG12, 5lg/mL polyC, 0.5lM GTP, 1lCi/
mL [3H]-GTP, 10mM MgCl2, 60mM NaCl, 100lg/mL
BSA, and 6nM NS5B (D55) were incubated at room
temperature for 50min in 96-well plates with or without test
compounds. Assay was terminated by the addition of 50lL,
10mg/mL streptavidin-coated SPA beads supplemented
with 100mM EDTA, and the incorporation of labeled GTP
determined by a TopCount Scintillation Counter. IC50
values were calculated from single experiments using 11
serial 2-fold dilutions (0.05–50lM), and data were consid-
ered reliable only when the IC50 value of a positive internal
control was within standard deviation range.
13. Assay was identical to that described in (12) except 5lM
GTP, 20lCi/mL [3H]-GTP, and 50nM NS5B (D-21)
enzyme form were used in a 3h reaction at room
temperature.