Improved replicon cellular activity of non-nucleoside allosteric inhibitors of HCV NS5B polymerase: From benzimidazole to indole scaffolds

Article abstract of DOI:10.1016/j.bmcl.2006.07.074

Benzimidazole-based allosteric inhibitors of the hepatitis C virus (HCV) NS5B polymerase were diversified to a variety of topologically related scaffolds. Replacement of the polar benzimidazole core by lipophilic indoles led to inhibitors with improved potency in the cell-based subgenomic HCV replicon system. Transposing the indole scaffold into a previously described series of benzimidazole-tryptophan amides generated the most potent inhibitors of HCV RNA replication in cell culture reported to date in this series (EC₅₀ ～ 50 nM).

Full text of DOI:10.1016/j.bmcl.2006.07.074

Bioorganic & Medicinal Chemistry Letters 16 (2006) 4987–4993
Improved replicon cellular activity of non-nucleoside
allosteric inhibitors of HCV NS5B polymerase:
From benzimidazole to indole scaffolds
Pierre L. Beaulieu,* James Gillard, Darren Bykowski, Christian Brochu,
´
Nathalie Dansereau, Jean-Simon Duceppe, Bruno Hache, Araz Jakalian,
´
Lisette Lagace, Steven LaPlante, Ginette McKercher, Elaine Moreau,
´
Stephane Perreault, Timothy Stammers, Louise Thauvette,
Jeff Warrington and George Kukolj
Boehringer Ingelheim (Canada) Ltd., Research and Development, 2100 Cunard Street, Laval (Que´bec), Canada H7S 2G5
Received 8 June 2006; revised 14 July 2006; accepted 17 July 2006
Abstract—Benzimidazole-based allosteric inhibitors of the hepatitis C virus (HCV) NS5B polymerase were diversified to a variety of
topologically related scaffolds. Replacement of the polar benzimidazole core by lipophilic indoles led to inhibitors with improved
potency in the cell-based subgenomic HCV replicon system. Transposing the indole scaffold into a previously described series of
benzimidazole–tryptophan amides generated the most potent inhibitors of HCV RNA replication in cell culture reported to date
in this series (EC50 ꢀ 50 nM).
Ó 2006 Elsevier Ltd. All rights reserved.
Chronic infection by the hepatitis C virus afflicts an esti-
mated 2–3% of the world population and represents a
health issue of increasing concern. While it can remain
asymptomatic for decades, the infection may progress
to life-threatening liver diseases such as cirrhosis (20%
of cases) and hepatocellular cancers (2.5%). HCV infec-
tion is the major reason for liver transplantation in the
western world and causes more than 10,000 deaths
annually in the US alone.1 The last 10 years have wit-
nessed major advances in our understanding of this
pathogen from the identification of essential viral func-
tions,2 to the development of subgenomic replicons that
allow HCV RNA replication in Huh-7 cells,3 to replicat-
ing specific viral strains in the laboratory.4 As a result,
several drug candidates that target specific viral enzymes
have recently been described1b and clinical testing in
HCV patients is producing unprecedented results.5
Recently, we reported the discovery, preliminary optimi-
zation, and mechanism of action of non-nucleoside, allo-
steric inhibitors of the NS5B polymerase of the HCV
virus.6 These compounds are characterized by a benz-
imidazole scaffold harboring a critical cyclohexyl or
cyclopentyl ring at the N1-position, a small aromatic or
heteroaromatic ring at C-2 and a carboxylic acid function
at C-5 (e.g., compound 1, Fig. 1).6b While low micromolar
potency in a truncated genotype 1b HT-NS5BD21 enzy-
matic assay6a was attained, this class of inhibitors did
not inhibit HCV RNA replication in the cell-based repli-
con assay.7 Significant improvements in potency (58-fold)
were attained by elaborating the right-hand side of the
inhibitors through attachment of functionalized amino
acid residues6c (e.g., compound 2). Adjustments to the
physicochemical properties of the molecules by replace-
ment of the ionizable carboxylic acids eventually led to
the discovery of inhibitors (e.g., compound 3) active in cell
culture at low micromolar concentrations.8
At this stage of the optimization process, much of the
SAR surrounding the benzimidazole scaffold had been
exploited and further improvements in potency (enzy-
matic and cell-based) required modification of the benz-
imidazole core itself. In particular, we hypothesized that
Keywords: HCV NS5B polymerase; Hepatitis C virus; Allosteric inhi-
bitors; Replicon; Antiviral; Indoles.
*
Corresponding author. Tel.: +1 450 6824640; fax: +1 450 6828434;
e-mail: pbeaulieu@lav.boehringer-ingelheim.com
Undergraduate chemistry COOP students.
0960-894X/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2006.07.074

4988
P. L. Beaulieu et al. / Bioorg. Med. Chem. Lett. 16 (2006) 4987–4993
O
pyrroles 6F and pyrrazole 6G were prepared. The com-
pounds were >500-fold less potent than the corre-
sponding indole 6C. This dramatic drop in potency
may be due to the loss of critical hydrophobic interac-
tions between the indole benzene ring and the protein.
Since the carboxylic acid group interacts with a critical
basic protein residue,10 alterations to the pKa and
changes in the orientation of the carboxyl function in
6F and 6G may also contribute to the observed
decrease in potency.11
N
N
OH
OH
IC50 = 1.1 µM
1
2
O
EC50 = not active
H
O
N
O
N
N
N
H
O
O
IC50 = 0.019 µM
EC50 = not active
COOH
S
The scaffold study was extended to 3-cyclopentyl
derivatives (Table 2), which were previously shown to
have comparable potency to cyclohexyl analogs.6b
6-Indolecarboxylic acids (8 and 9) were more potent than
the reference benzimidazole 7, showing improved enzy-
matic potency (ꢀ100-fold) and cellular activity compara-
ble to the cyclohexyl derivatives (4B and 4C). As
previously noted, N-methylation did not affect potency
(8 vs 9) and 6-indolecarboxylic acids were superior (7-fold
in this case) to the 5-isomer (9 vs 10). Benzofuran 11 and
benzothiophene 12, which are topologically similar to
indoles (see Fig. 2 depicting 2-phenyl derivatives), had
reduced enzymatic (IC50 = 0.08 lM, a 4-fold drop com-
pared to unsubstituted indole 8) and replicon potency.
Figure 2 shows a superposition between key pharmaco-
phores (cyclopentyl ring and carboxylic acid function)
within the three systems, the largest deviation resulting
from the longer C–S bonds in the benzothiophene analog.
In the latter compound, a slight change in orientation of
the C-2 substituent is apparent but the effect is not reflect-
ed in the potency of the analogs. This is in agreement with
the observation that the C-2 position tolerates a variety of
structural modifications.6b
H
N
N
O
N
N
N
H
3
O
OH
IC50 = 0.22 µM
EC50 = 1.1 µM
Figure 1. Initial SAR in the benzimidazole series.
replacement of the polar benzimidazole system by more
lipophilic, topologically related variants may favor cel-
lular permeation and thereby, improve the modest cell-
based potency achieved so far in this series.
Retaining the critical pharmacophores of 4A, we
prepared a series of indole derivatives as illustrated in
Table 1. Replacement of the benzimidazole core of 4A
(calcd logP = 3.4) with the more lipophilic indole moie-
ty generated a compound (4B, calcd logP = 4.4) which
displayed a 37-fold improvement in enzymatic potency
(IC50 = 0.030 vs 1.1 lM).6a Furthermore, indole 4B
was active in the replicon (EC50 = 6.7 lM),7 presumably
reflecting the expected increase in lipophilicity and
improved cellular permeation.9 Encouraged by these
results, we proceeded with N-methylation of the indole
nitrogen, a newly accessible position for further elabora-
tion of the scaffold, achieving an additional 2-fold
improvement in enzymatic potency (4C) while retaining
similar replicon activity. The isomeric 5-indole carboxyl-
ic acid analog 4D was slightly less potent (2-fold). In an
attempt to improve cellular permeation by reducing ion-
ization of the carboxylic acid by modifying its pKa,
7-azaindole analogs 4E and 5E were prepared. Even
though enzymatic potency was slightly reduced (2- to
3-fold) compared to indole 4C, 7-aza analog 5E was
3-fold more potent in cell culture, reflecting the contri-
bution of the pKa/ionization of the carboxyl function
to cellular permeation and replicon activity.9c
Finally, oxidation of the thiophene ring (compound 13)
resulted in complete loss of activity. The presence of the
two polar oxygen atoms may interfere with stacking of
the scaffold within the hydrophobic enzyme pocket
and account for the loss of activity.10
As mentioned previously, the improved cellular potency
of azaindole 5E over the corresponding carbo-analog 5C
(See Table 1) suggested that the pKa of the carboxylic
acid function and the overall ionization state of the mol-
ecule may be contributors to cellular permeation and
cell culture activity. Previous work on benzimidazoles
(unpublished results) had shown that addition of a
methyl group was well tolerated in the 4-position of that
scaffold (7-position in 4C or 9) but was detrimental to
potency at the 6-position of 4A (5-position on indole
9). The effect on potency of introducing substituents at
the 7-position of the indole scaffold was thus examined,
with the intention of probing electronic and steric effects
on ionization of the nearby critical carboxylic acid func-
tion and replicon potency. We also hoped that introduc-
tion of functional groups on the indole scaffold would
lead to the discovery of unexploited interactions with
the protein and further improvements in overall potency
(see Table 3).
Evaluation of the scaffolds in a small combination ma-
trix with representative C-2 substituents (identified in
our previous work on benzimidazoles)6b confirmed the
advantage of the N-methyl-6-indolecarboxylic acid scaf-
fold in both enzymatic and cell culture assays (Table 1).
Notably, the 2-pyridyl analog 6C had comparable
replicon potency to the furyl derivative 4C despite a
5.5-fold decrease in intrinsic potency,
indication of improved cellular permeation.
a possible
With the aim of further reducing compound size and
probing the effect of conformational restriction,
Unfortunately, apart from the 7-hydroxy analog 14
which was 3 times less potent than unsubstituted indole

P. L. Beaulieu et al. / Bioorg. Med. Chem. Lett. 16 (2006) 4987–4993
4989
Table 1. Benzimidazole scaffold replacements (IC50 values determined using a truncated genotype 1b HT-NS5B-D21 construct)6a
R
=
N
O
5
4
6
a
a
a
a
a
a
Compound
IC50 (lM)
EC50 (lM)
IC50 (lM)
EC50 (lM)
IC50 (lM)
EC50 (lM)
O
N
OH
OH
R
N
1.1
NA
3.1
NA
>10
2.7
NA
A
O
H
N
R
0.030
6.7
0.05
0.23
>10
B
O
Me
N
R
OH
OH
OH
0.016
0.044
0.051
4.2
0.03
0.18
0.08
6.6
>10
2.2
0.09
3.9
C
O
R
N
>10
0.82
>10
D
O
Me
N
N
R
Not tested
E
O
H
N
OH
OH
R
R
>50
Not tested
Not tested
F
O
N
>100
N
G
a Values are means of duplicate experiments on two separate weightings. NA (not active).
9, all other derivatives were >13-fold less potent. The
observed trends were consistent with a steric rather than
electronic signature (cf. 15, 16 vs 17). The low potency of
diacid 18 may be due to an unfavorable intramolecular
interaction between the two carboxyl functions that
preclude the 6-COOH from engaging in a productive
interaction with the polymerase.
DNA-dependent RNA polymerase II (derived from
mammalian calf thymus as previously described)6a assay
and were found to be inactive (IC50 > 125–250 lM).
We and others9b have established the superiority of in-
dole scaffolds over the original benzimidazole-based
leads, with respect to intrinsic potency against the
NS5B enzyme. However, the presence of the carboxylic
acid function that is ionized at physiological pH in small
indole derivatives such as 4C results in relatively modest
Selected compounds (4B and 4C) were also evaluated in
a poliovirus RNA-dependent RNA polymerase and

4990
P. L. Beaulieu et al. / Bioorg. Med. Chem. Lett. 16 (2006) 4987–4993
Table 2. Scaffold replacements in 3-cyclopentyl analogs (IC50s deter-
mined using a genotype 1b HT-NS5B-D21 construct)6a
Table 3. Effect of substituents at the 7-position of 6-indolecarboxylic
acid derivatives. (IC50s determined using a genotype 1b HT-NS5B-D21
construct)6a
Compound
HT-NS5B-D21 Replicon
a a
IC50
,
(lM)
EC50
, (lM)
O
X
Me
N
O
O
O
O
O
O
O
OH
N
N
O
OH
OH
OH
OH
OH
OH
OH
7
8
2.2
NA
O
O
O
O
O
O
O
Compound
X
H
HT-NS5B-D21
Replicon
a
a
IC50
,
(lM)
EC50, (lM)
H
N
9
14
15
16
17
18
0.021
0.075
0.34
0.29
0.48
5.3
>10
19
OH
OMe
Cl
0.025
0.021
0.15
6.6
6.6
39
40
37
COOMe
COOH
22
Not tested
Me
N
a Values are means of duplicate experiments on two separate
weightings.
9
inhibition in the replicon assay (EC50 = 4 lM), despite
high levels of inhibition in enzymatic assays
(IC50 = 16 nM). Compound 19 (Fig. 3), the indole ver-
sion of amide 3,8 showed a 2-fold improvement in
potency in the enzymatic assay and was 6-fold less po-
tent than carboxylic acid 4C. However, the improved
cellular permeation resulting from removal of the car-
boxylic acid charge led to a dramatic increase in cellular
potency and 19 is the most potent inhibitor of HCV
RNA replication in the cell-based replicon that has been
reported to date from this class of non-nucleoside allo-
steric NS5B inhibitors. The simple replacement of a
nitrogen by a carbon atom in compound 3 resulted
in a 22-fold improvement in cell culture activity
(EC50 ꢀ 50 nM; the value was derived after correcting
for actual compound concentration in assay medium
as determined by HPLC analysis, due to the very poor
aqueous solubility of this compound). Efforts toward
further optimization of this series will be reported in
due course.
10
11
12
13
N
O
0.081
0.078
21
S
22
O
S
O
>100
Not tested
The synthesis of indole-6-carboxylic acid analogs is
shown in Scheme 1 for the cyclohexyl series:9,11 conden-
sation of commercially available 6-indolecarboxylic acid
20 with excess cyclohexanone under basic conditions fol-
lowed by hydrogenation produced the 3-cyclohexyl acid
intermediate 21. The carboxyl function was esterified to
give indole 22, or 21 was per-methylated to N-methylin-
dole ester 23. Alternatively, ester 22 was N-methylated
to 23 in a separate step. Bromination of the 2-position
using pyridinium perbromide gave bromoindoles 24 or
25 that served as substrates for a palladium-catalyzed
a Values are means of duplicate experiments on two separate weigh-
tings. NA (not active).
S
H
N
N
O
IC50 = 0.1 µM
N
N
EC50 = 0.05 µM
H
O
OH
19
Figure 2. Overlay of 3-cyclopentyl-2-phenyl-6-carboxy-indole (green),
benzofuran (red), and benzothiophene (blue) scaffolds.
Figure 3. HCV NS5B polymerase indole-based derivative with sub-
micromolar potency in the cell-based replicon.

P. L. Beaulieu et al. / Bioorg. Med. Chem. Lett. 16 (2006) 4987–4993
4991
H
I
COOMe
I
COOMe
COOH
b
a
N
H
N
COOH
a, b
d
73%
HN
> 100%
H2N
27
26
21
20
c
Me
N
H
COOMe
COOMe
COOMe
N
COOMe
c
TMS
Br
N
N
82%
e
23
22
29
28
f
g
H
N
Me
N
COOH
COOMe
COOMe
see Scheme 1
d
4D (93%)
5D (91%)
6D (80%)
R
Br
Br
N
24
h, i
1-3D
25
h, i
Scheme 2. Synthesis of indole 5-carboxylic acid derivatives. Reagents
and conditions: (a) cyclohexanone (6 equiv), MgSO4 (10.5 equiv),
NaBH(OAc)3 (3.5 equiv), AcOH, rt overnight; (b) TMS-CꢁCCH3
(2.8 equiv), LiCl (1 equiv), KOAc (2 equiv), Pd(OAc)2 (0.05 equiv),
DMF, 100 °C overnight; (c) Br2 (0.9 equiv), CH2Cl2, 0 °C, 30 min;
(d) For 6D: n-BuLi (1.05 equiv), THF, À78 °C, 45 min then B(OMe)3
(1.2 equiv), À70 °C, 1 h then Pd(OAc)2 (0.01 equiv), P(p-tolyl)3 (0.06
equiv), K2CO3 (2 equiv), 2-bromopyridine (1.2 equiv), MeOH, reflux,
18 h.
Me
N
H
COOH
COOH
N
R
R
4-6B
4-6C
Scheme 1. Synthesis of indole 6-carboxylic acid derivatives. Reagents
and conditions: (a) cyclohexanone (3 equiv), NaOMe (6.5 equiv),
MeOH, reflux for 36 h (>100%). For the 3-cyclopentyl analogs:
cyclopentanone (5 equiv slow addition), KOH (10 equiv), MeOH,
water, 75 °C, 18 h (88%). (b) 20% Pd(OH)2/C, H2 gas (55 w), MeOH/
THF 1:1 (100%). (c) MeI (1.1 equiv), K2CO3 (1.4 equiv), DMF, rt, 18 h
(90%). (d) MeI (2.5 equiv), K2CO3 (2.5 equiv), DMF, rt, 2 h; then NaH
(1.5 equiv), 0 °C to rt, 4 h. (e) NaH (1.3 equiv), MeI (1.5 equiv), DMF,
rt, 18 h (86%). (f) Br2 (1.3 equiv), NaOAc (1.3 equiv), i-PrOAc, 5 °C,
30 min (76%). (g) PyrBr3 (1.4 equiv), THF/CHCl3 1:1, 0 °C to rt, 2.5 h
(71%). (h) For R = 3-furyl (99%) or Ph (88%): 3-furylboronic acid or
phenylboronic acid (1.3 equiv), LiCl (2 equiv), Na2CO3 (2.5 equiv),
Pd(PPh3)4 (0.04 equiv), toluene/EtOH/water 1:1:1, reflux overnight.
For R = 2-pyridyl: n-BuLi (1.3 equiv), THF, À78 °C, 1.5 h, then
Bu3SnCl (1.4 equiv) (79%) followed by: 2-bromopyridine (1.3 equiv),
PPh3 (0.2 equiv), CuI (0.1 equiv), LiCl (2 equiv), Pd(PPh3)4 (0.05
equiv), DMF, 100 °C, overnight (71%). (i) NaOH then HCl.
followed by Stille couplings were employed as described
in Scheme 1 to provide the desired inhibitors 4–6D. A
similar sequence replacing cyclohexanone by cyclopen-
tanone provided inhibitor 10.
7-Azaindole analogs 4E and 5E were prepared following
the route depicted in Scheme 3: commercially available
7-azaindole was condensed with cyclohexanone and
hydrogenated as described in Scheme 1 to give 30. Oxida-
tion of the pyridine to the N-oxide directed subsequent
N-methylation to the indole nitrogen to form 31. Treat-
ment with TMSCN under basic conditions provided
nitrile 32. After conversion to ethyl ester 33 the desired
inhibitors were obtained following the usual bromina-
tion, cross-coupling, and saponification sequence.
Suzuki cross-coupling with 3-furyl and phenylboronic
acid to provide final inhibitors 4–5B and 4–5C following
saponification of the ester group. In the case of the 2-
pyridyl analog 6C, 2-bromoindole 25 was trans-metal-
lated at low temperature and the 2-lithioindole was
stannylated with chlorotributyltin to provide an acid-
sensitive 2-stannylindole intermediate. Stille coupling
with 2-bromopyridine provided the desired indole 6C
in good yield following deprotection of the ester func-
tion. For 6B, indole 22 was protected as the N-Boc
derivative and stannylated in the 2-position (LDA/n-
Bu3SnCl). Cross-coupling with 2-bromopyridine under
standard Stille conditions followed by removal of pro-
tecting groups provided inhibitor 6B. Cyclopentyl ana-
logs 8 and 9 were prepared in a similar fashion.
The benzofuran 11 and benzothiophene 12 analogs were
prepared as described in Scheme 4: Friedel–Craft acyla-
tion of 3-bromophenol with acid chloride 34 derived
O
+
H
N
N
H
N
N
N
a,b
Scheme 1
72%
N
57%
31
30
7-azaindole
N
CN
COOEt
N
N
N
c
d
4-5E
60%
63%
e, then
Scheme 1
32
33
The synthesis of isomeric 5-carboxyindole derivatives is
depicted in Scheme 2. Reductive amination of methyl
4-amino-3-iodobenzoate 26 with cyclohexanone provid-
ed iodoaniline 27. Palladium-catalyzed cross-coupling
with 1-trimethylsilylpropyne gave 2-trimethylsilylindole
28 which was brominated to provide 2-bromoindole
29. Suzuki cross-couplings and metallation–stannylation
Scheme 3. Synthesis of 7-azaindoles. Reagents and conditions: Scheme
1, steps a–b. (a) MCPBA (2 equiv), DME, rt, 2 h; (b) NaH (2 equiv),
MeI (1 equiv), DMF, rt, 3 h; (c) Et3N (2.5 equiv), TMSCN (5 equiv),
MeCN, reflux 19 h; (d) EtOH, HCl gas, 15 min at rt then reflux for
1.5 h; (e) PyrHBr3 (1.5 equiv), THF, 65 °C, 16 h (96%).

4992
P. L. Beaulieu et al. / Bioorg. Med. Chem. Lett. 16 (2006) 4987–4993
HO
Br
the protocols described in Scheme 1 or the Supplemen-
tary data section.13,14 The synthetic approaches to pyr-
role 6F and pyrrazole 6G are also provided in the
Supplementary data section.14
COOH
COCl
b
a
O
35
c, d
34
Me2N
S
O
Br
In conclusion, we have shown that replacement of the
polar benzimidazole scaffold of allosteric HCV NS5B
polymerase inhibitors by a more lipophilic indole core
provides significant potency improvements in a cell-
based replicon assay of HCV RNA replication. Combi-
nation of an indole-based scaffold with a lipophilic tryp-
tophan-derived right-hand side has led to the most
potent inhibitor in cell culture reported to date in this
series.
g
HOOC
O
Br
O
40
41
O
h, i
36
e
HS
Br
O
Br
O
37
j, d.
Br
f
HOOC
S
Acknowledgments
O
CN
O
We thank Serge Valois and Michael Little for analytical
support. Drs. Paul Anderson, Michael Bo¨s, Michael
Cordingley, and Daniel Lamarre are acknowledged for
their guidance and support of this work.
42
e, f
38
m
CN
S
CN
43
O
Supplementary data
Br
k, l
39
n, o
Supplementary data associated with this article can be
found, in the online version, at doi:10.1016/j.bmcl.2006.
07.074
Scheme 1
p
then
Scheme 1
12
COOEt
S
Br
11
44
13
References and notes
Scheme 4. Synthesis of benzofuran and benzothiophene analogs.
Reagents and conditions: (a) SOCl2, pyridine (cat.), reflux; (b) 3-
bromophenol (1 equiv), AlCl3 (1.5 equiv), 1,2-dichloroethane, reflux,
2 h (78%); (c) NaH (1.5 equiv), ethyl bromoacetate (1.5 equiv), DMF,
1 h; (d) NaOH, THF–water; (e) NaOAc (2 equiv), Ac2O, reflux 5 h
(80%); (f) CuCN (1.2 equiv), DMF reflux, 7 h (72%, 3 steps); (g)
DABCO (2 equiv), dimethylthiocarbamyl chloride (2 equiv), DMF, rt,
3 h (100%); (h) 180–190 °C, neat, 5 h; (i) KOH, MeOH, reflux, 1.5 h
(91%, 2 steps); (j) ethylbromoacetate (1.2 equiv), K2CO3 (3 equiv),
acetone, rt overnight (82%); (k) Br2 (4 equiv), AcOH, rt overnight; (l)
EtOH–concd H2SO4, reflux 1–2 days (58%, 2 steps); (m) NBS (3.2
equiv), DMF, rt overnight (53%); (n) cross-coupling: see conditions in
Scheme 1; (o) concd H2SO4/AcOH/water 5:4:2, reflux 1.5 h; (p) 30%
H2O2 (6 equiv), AcOH (68%), 60 °C, 24 h.
1. (a) Shepard, C. W.; Finelli, L.; Alter, M. J. Lancet Infect.
Dis. 2005, 5, 558; (b) Beaulieu, P. L.; Tsantrizos, Y. Curr.
Opin. Invest. Drugs 2004, 5, 838.
2. (a) Reed, K. E.; Rice, C. M. Curr. Top. Microbiol.
Immunol. 2000, 242, 55; (b) Kolykhalov, A. A.; Mihalik,
K.; Feinstone, S. M.; Rice, C. M. J. Virol. 2000, 74, 2046.
3. Lohmann, V.; Ko¨rner, F.; Koch, J. O.; Herian, U.;
Theilmann, L.; Bartenschlager, R. Science 1999, 285, 110.
4. (a) Wakita, T.; Pietschmann, T.; Kato, T.; Date, T.;
Miyamoto, M.; Zhao, Z.; Murthy, K.; Habermann, A.;
Kra¨usslich, H. G.; Mizokami, M.; Bartenschlager, R.;
Liang, T. J. Nat. Med. 2005, 11, 791; (b) Lindenbach, B.
D.; Evans, M. J.; Syder, A. J.; Wo¨lk, B.; Tellinghuisen, T.
L.; Liu, C. C.; Maruyama, T.; Hynes, R. O.; Burton, D.
R.; McKeating, J. A.; Rice, C. M. Science 2005, 309, 521.
5. (a) Hinrichsen, H.; Benhamou, Y.; Wedemeyer, H.;
Reiser, M.; Sentjens, R. E.; Calleja, J. L.; Forns, X.;
Erhardt, A.; Cro¨enlein, J.; Chaves, R.; Yong, C.-L.;
Nehmiz, G.; Steinmann, G. G. Gastroenterology 2004,
from cyclopentanecarboxylic acid gave ketone 35.
Alkylation with bromoacetate followed by saponifica-
tion (36) and ring closure gave 6-bromobenzofuran 37
which was subjected to cyanation to provide nitrile 38.
Bromination (39) followed by Suzuki cross-coupling
with 3-furylboronic acid and hydrolysis of the nitrile
to the carboxylic acid gave analog 11. Conversion of
phenol 35 to thiol 41 was effected through a thermal
rearrangement of O,S-thiocarbamate 40. Thiol 41 was
then elaborated (42–44) to bromobenzothiophene 44
that was converted to analog 12 in the usual manner.
Benzothiophene dioxide 13 was obtained by oxidation
of intermediate 44 using H2O2/AcOH,12 and further
elaboration as described in Scheme 1.
`
127, 1347; (b) Goudreau, N.; Llinas-Brunet, M. Expert
Opin. Invest. Drugs 2005, 14, 1129; (c) McHutchison, J.
G.; Bartenschlager, R.; Patel, K.; Pawlotsky, J.-M. J.
Hepatol. 2006, 44, 411.
6. (a) McKercher, G.; Beaulieu, P. L.; Lamarre, D.;
LaPlante, S.; Lefebvre, S.; Pellerin, C.; Thauvette, L.;
Kukolj, G. Nucleic Acids Res. 2004, 32, 422, The HT-
NS5BD21 polymerase scintillation proximity assay was
performed as described with the following modifications:
12 nM of the soluble truncated enzyme was used but the
reaction had 0.01% w/v BSA and lacked the previously
supplemented 0.33% dodecyl-b-D-maltoside, 0.01% IGE-
PAL.; (b) Beaulieu, P. L.; Bo¨s, M.; Bousquet, Y.; Fazal,
G.; Gauthier, J.; Gillard, J.; Goulet, S.; LaPlante, S.;
Substituted indoles 14–18 were prepared from the corre-
sponding 7-substituted-6-indolecarboxylic acids using

P. L. Beaulieu et al. / Bioorg. Med. Chem. Lett. 16 (2006) 4987–4993
4993
Poupart, M.-A.; Lefebvre, S.; McKercher, G.; Pellerin,
C.; Austel, V.; Kukolj, G. Bioorg. Med. Chem. Lett.
2004, 14, 119; (c) Beaulieu, 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; (d) Hirashi-
ma, S.; Suzuki, T.; Ishida, T.; Noji, S.; Yata, S.; Ando,
I.; Komatsu, M.; Ikeda, S.; Hashimoto, H. J. Med.
Chem. 2006, 49, 4721.
M.; Migliaccio, G.; Laufer, R.; De Francesco, R.; Rowley,
M.; Narjes, F. J. Med. Chem. 2005, 48, 1314; (c) Ontoria,
`
J. M.; Martın Hernando, J. I.; Malancona, S.; Attenni, B.;
Stansfield, I.; Conte, I.; Ercolani, C.; Habermann, J.;
Ponzi, S.; Di Filippo, M.; Koch, U.; Rowley, M.; Narjes,
F. Bioorg. Med. Chem. Lett. 2006, 16, 4026.
10. (a) Coulombe, R.; Beaulieu, P. L.; Jolicoeur, R.; Kukolj,
G.; LaPlante, S.; Poupart, M.-A. WO 04/099241, 2004;
See also (b) Di Marco, S.; Volpari, C.; Tomei, L.;
Altamura, S.; Harper, S.; Narjes, F.; Koch, U.; Rowley,
7. Protocols for the cell-based luciferase reporter HCV RNA
replicon assay used in this work can be found in:
Tsantrizos, Y. S.; Chabot, C.; Beaulieu, P. L.; Brochu,
C.; Poirier, M.; Stammers, T.; Thavonekham, B.; Ran-
court, J. WO 05/080388, 2005. Compounds were devoid of
cytotoxicity at inhibitory concentrations.
8. Beaulieu, P. L.; Bousquet, Y.; Gauthier, J.; Gillard, J.;
Marquis, M.; McKercher, G.; Pellerin, C.; Valois, S.;
Kukolj, G. J. Med. Chem. 2004, 47, 6884.
9. (a) Beaulieu, P. L.; Fazal, G.; Kukolj, G.; Jolicoeur, E.;
Gillard, J.; Poupart, M.-A.; Rancourt, J. WO 03/010140,
2003; Similar findings were recently reported by an other
group (b) Harper, S.; Pacini, B.; Avolio, S.; Di Filippo,
´
M.; De Francesco, R.; Migliaccio, G.; Carfı, A. J. Biol.
Chem. 2005, 280, 29765.
11. All final inhibitors were purified by C-18 reversed-phase
HPLC to >97% homogeneity and gave 1H NMR, 13C
NMR, and MS data consistent with their assigned
structure.
12. Mitra, R. B.; Tilak, B. D. J. Sci. Indust. Res. 1956, 15B, 573.
13. Gassman, P. G.; Van Bergen, T. J.; Gilbert, D. P.; Cue, B.
W. J. Am. Chem. Soc. 1974, 96, 5495.
14. Detailed experimental procedures and characterization for
selected key intermediates are provided in the supplemen-
tary section of the article.