Non-nucleoside inhibitors of the hepatitis C virus NS5B polymerase: Discovery and preliminary SAR of benzimidazole derivatives

Article abstract of DOI:10.1016/j.bmcl.2003.10.023

Benzimidazole 5-carboxamide derivatives from a combinatorial screening library were discovered as specific inhibitors of the NS5B polymerase of the hepatitis C virus (HCV). Initial hit-to-lead activities taking advantage of high-throughput parallel synthetic techniques, identified a 1,2-disubstituted benzimidazole 5-carboxylic acid scaffold as the minimum core for biological activity. Potent analogues in this series inhibit the polymerase at low micromolar concentrations and provide an attractive 'drug-like' lead structure for further optimization and the development of potential HCV therapeutics.

Full text of DOI:10.1016/j.bmcl.2003.10.023

Bioorganic & Medicinal Chemistry Letters 14 (2004) 119–124
Non-nucleoside inhibitors of the hepatitis C virus NS5B
polymerase: discovery and preliminary SAR of
benzimidazole derivatives
a,
a
a
a
a
¨
Pierre L. Beaulieu, * Michael Bos, Yves Bousquet, Gulrez Fazal, Jean Gauthier,
a
a
a
a
James Gillard, Sylvie Goulet, Steven LaPlante, Marc-Andre Poupart,
´
b
b
b
c
Sylvain Lefebvre, Ginette McKercher, Charles Pellerin, Volkhard Austel
and George Kukoljb
aDepartment of Chemistry, Boehringer Ingelheim (Canada) Ltd, Research and Development, 2100 Cunard Street,
Laval (Qu e´ bec) Canada H7S2G5
b
Biological Sciences, Boehringer Ingelheim (Canada) Ltd, Research and Development, 2100 Cunard Street,
Laval (Qu e´ bec) Canada H7S2G5
c
Department of Chemistry, Boehringer Ingelheim KG, Biberach, Germany
Received 5 August 2003; accepted 3 October 2003
Abstract—Benzimidazole 5-carboxamide derivatives from a combinatorial screening library were discovered as specific inhibitors of
the NS5B polymerase of the hepatitis C virus (HCV). Initial hit-to-lead activities taking advantage of high-throughput parallel
synthetic techniques, identified a 1,2-disubstituted benzimidazole 5-carboxylic acid scaffold as the minimum core for biological
activity. Potent analogues in this series inhibit the polymerase at low micromolar concentrations and provide an attractive ‘drug-
like’ lead structure for further optimization and the development of potential HCV therapeutics.
#
2003 Elsevier Ltd. All rights reserved.
The hepatitis C virus was discovered more than a dec-
1
two-thirds of the polyprotein consists of non-structural
(NS) proteins (NS2!NS5) whose enzymatic functions
ade ago, and the prevalence of the infection is now
estimated at 1–3% of the world population (ca. 170
7
are essential for viral replication. The NS5B segment
encodes a 65 kd RNA-dependent RNA polymerase
2
million people). Chronic HCV infections can lead to
fatal liver disorders such as cirrhosis and hepatocellular
carcinomas that eventually require organ transplant-
8
(RdRp). Because of the central role this enzyme plays
in the replication of HCV RNA, it has become a target
of choice for the screening and design of small molecule
inhibitors which, in principle, should interfere with viral
3
ation. Therapies for the treatment of chronic HCV
infection are currently limited to various forms of
interferon, in combination with the broad spectrum
antiviral ribavirin.4 A significant proportion of the
patient population does not respond to therapy and
there is an urgent demand for novel strategies addres-
5
replication.
Our sample collection was screened in an HCV poly-
merase assay using an N-terminally histidine-tagged
recombinant full-length NS5B protein from a 1b geno-
5
a
sing this clearly unmet medical need.
9
type. We identified a set of benzimidazole derivatives
from a combinatorial library (as exemplified by com-
HCV is a small, enveloped virus whose ꢀ9.6 Kb single-
1
0
strand (+)-sense-RNA genome encodes a polypeptide
of approximately 3000 amino acids. The C-terminal
pounds 1 and 2) as specific inhibitors of the poly-
11
merase (Fig. 1).
6
In order to confirm that inhibition was mediated by
specific interactions of the compounds with NS5B,
NMR differential line broadening (DLB) and
*
Corresponding author. Fax: +1-450-682-8434; e-mail: pbeaulieu@
lav.boehringer-ingelheim.com
0
960-894X/$ - see front matter # 2003 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2003.10.023

120
P. L. Beaulieu et al. / Bioorg. Med. Chem. Lett. 14 (2004) 119–124
transferred-NOESY experiments were performed with 2
1
inhibition resulted from specific interactions of the
ligands with the polymerase rather than through unde-
sired mechanisms such as binding to the RNA primer/
template substrate.
2
in the presence of a soluble form of the polymerase.
1
Figure 2A shows a representative H resonance of 2
(
H ) that broadens upon incremental additions of
14
NS5B, which would be expected for rapid and reversible
binding of a small ligand to a large macromolecule.
Figure 2B shows other resonances of 2 that are also
subject to DLB (shown in red on the trace and the
structure), indicating regions of the molecule that are
likely contacting the protein (‘epitope’ mapping). The
Based on the similar inhibitory potencies of 1 and 2 and
on the NMR data (vide supra), it was evident that the
basic left-hand side of 2 contributed very little to the
observed activities. Hence, we chose derivative 1 as a
starting point for optimization studies. Preliminary hit-
to-lead activities consisted of systematically truncating
the right-hand side amide substituent which eventually
lead us to a minimal core that is required for significant
biological activity.
1
2
observation of transferred-NOESY cross-peaks (data
not shown) provided further evidence that 2 binds to
HCV polymerase. The observation of DLB for the
resonances of the cyclohexyl substituent (H s), the C-2
1
1
aryl group and parts of the right-hand side of the inhi-
bitor (H atoms shown in red) suggested that these seg-
ments contact the protein in the bound state. In
As shown in Table 1, excision of a methylene group
from 1 resulted in a small increase in potency (3), yet
further truncation to anilide 4 was not tolerated.
Remarkably, amides 5 and 6 showed partial activity,
which was further improved in the case of the free car-
boxylic acid. Compound 7 had comparable potency to
benzylic amide 3 and was 3-fold more potent than the
original hit molecule 1. The feasibility of substituting
the neutral amide right-hand side of compounds such as
contrast, the left-hand portion of 2 (H1 ) lacks DLB
À7
and does not appear to interact with the protein, in
agreement with the SAR (vide infra). These results,
combined with the observed specificity of the inhibitors
for HCV NS5B indicated that the observed enzymatic
1
or 3 with a charged carboxylic acid group as in 7
without loss in potency, suggested that inhibitors may
exploit diverse interactions with the protein. Despite the
substantial size reduction and modifications in chemical
properties resulting from this change, inhibitor 7 main-
tained specificity for HCV polymerase as compared to
both the poliovirus and mammalian calf thymus (CT)
RNA polymerases.
Isomeric 4-, 6- and 7-carboxy-benzimidazole derivatives
had little or no inhibitory activity in the assay (results
not shown). Furthermore, the homologous phenylacetic
acid derivative (5-CH COOH) corresponding to 7 was
2
also inactive. This suggested that the carboxylic acid
function of 7 is implicated in a well-defined and specific
interaction with the protein. Whereas replacement of
the 5-carboxylic acid group of 7 by the isosteric tetra-
zole decreased potency 6-fold, removal of the 5-COOH
substituent or replacement by other functional groups
Figure 1. Original hits from a combinatorial screening library.
such as CN, CF resulted in a complete loss of activity
3
(results not shown). Compound 7, showing low mM
specific activity against NS5B (IC =4.3 mM) and
5
0
improved physicochemical properties relative to 1 and 3
solubility), became the starting point for further opti-
mization studies.
(
Initial benzimidazole hits from our screening campaign
0
1
that showed specificity for HCV polymerase all car-
1
ried a cyclohexyl substituent at the N -position and
hence, the importance of this position was explored
(
SAR carried out at this position are presented in Table
Table 2). Only a few select examples of the extensive
Figure 2. The structure of 2 is provided with color-coded numbering
that summarizes the DLB NMR data observed upon addition of
NS5B to a solution containing the inhibitor: red, black and green
correspond to DLB, lack of DLB and no observation possible due to
resonance overlap, respectively. A shows DLB for H14 upon addition
of incremental amounts of the polymerase (2/NS5B ratio from 54:1 to
1
2
. Removal of the N -substituent (compound 8) oblit-
erated inhibitory activity. Increasing ring size to a
seven-membered ring (9) was detrimental to activity,
but replacement with a cyclopentyl ring resulted only in
a 3-fold decrease in potency (10). Further ring-size
reduction or cleavage of the ring to produce acyclic
variants was unfavorable (11–13).
1
4:1). B provides a selected portion of the aliphatic region showing
DLB for the cyclohexane protons (H11) at an inhibitor/NS5B ratio of
7:1 (red trace).
2

P. L. Beaulieu et al. / Bioorg. Med. Chem. Lett. 14 (2004) 119–124
121
Table 1. Identification of the minimum core required for biological
activity in the NS5B polymerase enzymatic assay
Table 2. SAR at the 1-position of the benzimidazole scaffold
Compd
1R
HCV NS5B IC50
Poliovirus IC50
a
(
mM)
(mM)
Compd
X
HCV NS5B IC50
Poliovirus IC50
a
(
mM)
(mM)
7
8
4.3Æ1
>500
>500
H
1
3
4
12Æ3
7.1Æ1
>7b
>500
>500
9
49Æ19
12Æ4
1
1
1
1
1
1
0
1
2
3
4
5
6
>500
76Æ15
139Æ46
408Æ12
50Æ13
13Æ4
5
6
7
NHCH
3
28Æ6
27Æ8
4.3Æ1
NH
OH
2
>500
a
Values are means of duplicate experiments on two separate weigh-
ings. Actual compound concentrations determined by HPLC using
standards.
b
1
3–30% inhibition at maximum soluble concentration (7 mM).
Substitution of the cyclohexane ring with methyl groups
14 and 15) reflected rather tight steric requirements for
(
the 4-position (incorporation of an adamantane ring at
this position resulted in a>20-fold decrease in potency;
result not shown); however, the 2-position of the cyclo-
hexane ring appeared to be more permissive. Never-
theless, no net beneficial effect resulted from these
susbtitutions. Aromatization of the ring (16) also resul-
ted in a significant loss in potency. These findings and
NMR experiments discussed above confirm that the
cyclohexyl substituent is optimally interacting with a
very well defined lipophilic pocket on the polymerase
enzyme. Similar results were observed when this exercise
was performed with the neutral benzylic amide series
using compound 3 as a reference point, suggesting a
similar mode of binding at this position by both series
>500
1
360Æ14
a
Values are means of duplicate experiments on two separate weigh-
ings. Actual compound concentrations determined by HPLC using
standards.
extend beyond the first aromatic ring. Replacement of
the 2-pyridyl group in 7 with a neutral phenyl ring (19)
resulted in a 2-fold decrease in potency. Substitution of
this ring with a variety of single or multiple functional
groups (OH, OR, halo, alkyl, CN, CONHR, COOH,
NO , NHCOR, NR ) did not produce any inhibitors
(
results not shown).
The C-2 position on the benzimidazole scaffold, in con-
1
trast to the N -position, was very permissive. SAR pat-
2
2
terns were difficult to establish and the main feature that
was extracted from the examples presented in Tables 3
and 4 is that small heterocycles were preferred. More
than 100 alternatives were evaluated at the 2-position
and only select compounds are presented in the tables.
with IC s below 10 mM (results not shown).
5
0
ortho-Substituted aromatics were particularly detri-
mental to potency as shown for the example with tri-
methylated phenyl derivative 20. Presumably, ortho-
substituents can sterically influence positioning of the
1
The necessity for a substituent at C-2 was obvious, as
unsubstituted derivative 17 was approximately 40-fold
less potent. Incorporation of the basic entity found in
one of the initial hit compounds (2) to give 18 resulted
in a 3-fold loss in potency relative to the 2-pyridyl sub-
stitution (7). This was consistent with the DLB findings
for 2 discussed above, that suggested minimal interac-
tion with the protein for segments of inhibitors that
critical cyclohexane ring at the N -position within its
well-defined lipophilic pocket in the enzyme. Larger
bicyclic substituents such as benzofuran 21 (as well as
quinolines and indoles) were significantly less potent
relative to 7. Likewise, aliphatic substituents such as
cyclohexyl analogue 22 (and others: tert-butyl, cyclo-
propyl, isobutyl) resulted in a minimum 5-fold decrease
in potency. Vinylic substituents (e.g., 23) were slightly

122
P. L. Beaulieu et al. / Bioorg. Med. Chem. Lett. 14 (2004) 119–124
Table 3. SAR at the 2-position of the benzimidazole scaffold
Table 4. SAR at the 2-position of the benzimidazole scaffold: most
potent analogues
Compd
2R
HCV NS5B IC50 Poliovirus IC50 CT IC50
a
(
mM)
(mM)
(mM)
2R
Compd
HCV NS5B IC50
Poliovirus IC50
CT IC50
(mM)
a
(
mM)
(mM)
7
4.3Æ1
>500
7
4.3Æ1
>500
>250
>500
>500
>250
>500
>250
>500
>250
>250
>500
>500
>500
>500
>500
>500
>500
>500
>500
>500
>500
>500
1
7
8
H
186Æ76
13Æ3
1
26
27
28
29
28Æ14
17Æ6
1
2
2
2
2
2
9
0
1
2
3
4
5
8.6Æ2
120Æ29
58Æ24
20Æ9
250
>500
3.7Æ1.3
1.6Æ0.6
5.9Æ2
3
3
3
3
0
1
2
3
19Æ5
>500
10Æ4.5
1.8Æ0.4
6.3Æ4
62Æ35
44Æ9
2
a
Values are means of duplicate experiments on two separate weighings.
Actual compound concentrations determined by HPLC using standards.
34
3.4Æ1.5
4.6Æ0.2
3
5
better than their saturated counterparts, yet, potency
2
was decreased at least 5-fold compared to 7. Other sp -
hybridized groups such as ester 24 and ketone 25 were
also examined but conferred no benefit.
a
Values are means of duplicate experiments on two separate weigh-
ings. Actual compound concentrations determined by HPLC using
standards.
Particularly interesting compounds discovered to date
in this series are those depicted in Table 4, which incor-
porated small heterocycles at the 2-position of the ben-
zimidazole scaffold.
Bulky groups such as phenyl rings may likely influence
the conformation or positioning of the critical cyclo-
hexyl substituent on the neighboring N -position, due to
the presence of two ortho-hydrogens. Smaller five-mem-
bered aromatic heterocycles or the placement of a
heteroatom ortho to the point of attachment on six-
membered rings (as exemplified by compound 7 relative
to 26 and 27) may allow for partial relief of detrimental
steric interactions with the cyclohexyl ring and generally
led to more active compounds.
1
Replacement of the 2-pyridyl ring in 7 with small five-
membered heterocycles (e.g., furans, pyrroles, thio-
phenes, imidazoles, thiazoles) either increased potency
slightly (ca. 3-fold: 29 and 32) or had no effect. In gen-
eral, substitution at various positions of these small
heterocycles with alkyl, halogen, COOH or other
groups led to significant decreases in potency (some
examples are shown in Fig. 3). Though no clear SAR
trends were apparent for the C-2 position, we believe
that steric factors are an important determinant for
inhibitor activity.
Electronic factors do not appear to play a major role as
neutral (28–30 and 32–34) and more basic groups (e.g.,

P. L. Beaulieu et al. / Bioorg. Med. Chem. Lett. 14 (2004) 119–124
123
Scheme 1. Synthesis of 5-carboxybenzimidazole inhibitors. Reagents
and conditions: (a) EtOH, SOCl , reflux; (b) oxalyl chloride, cat.
then Wang resin, DIEA; (c) 1RNH
, DIEA, DMSO,
dihy-
2
DMF, CH
2
Cl
2
2
ꢁ
6
0–80 C; (d) H
2
(1 atm), 20% Pd(OH)
2
/C, MeOH; (e) SnCl
2
drate, water, DMF; (f) RCHO, chloranil or oxone1, DMF, water
2
2
then NaOH, water/MeOH then AcOH; (g) RCOOH, HATU, DIEA,
DMF or DMSO then TFA; (h) RNH , TBTU or HATU, DIEA,
DMF or DMSO.
Figure 3. Less potent C-2 heterocyclic substituents (IC50>30 mM).
2
7
, 26, 27, 31, and 35) are equally tolerated at this posi-
tion. As can be seen from the data presented in Table 4,
the compounds still retained their specificity for HCV
polymerase relative to the polio virus RdRp and, more
importantly, the mammalian DdRp II isolated from calf
thymus (CT).
Recently, cell lines that support high levels of self-repli-
cating sub-genomic HCV RNAs (HCV replicons) in
1
3
Huh-7 cells have become available. This viral replic-
ation surrogate cell-based assay is useful for evaluating
the potential for compounds to inhibit their target (in
this case NS5B polymerase) within a cellular environ-
ment. None of the selected compounds from the present
study (7, 29, 33, and 34) showed inhibition of HCV
RNA replication at non-toxic concentrations
(
EC >100 mM). Some of the factors that may explain
5
0
this lack of efficacy include: the compounds are not
reaching the target enzyme in the cell (lack of cellular
permeation); the mechanism by which they inhibit the
polymerase is not relevant in vivo; the compounds are
not potent enough. Further optimization of these spe-
cific HCV NS5B inhibitors will be required in order to
promote this class of compounds as potential HCV
therapeutics. Further optimization of benzimidazole-
based HCV polymerase inhibitors will be reported in
due course.
Scheme 2. Synthesis of 4-carboxybenzimidazole inhibitor. Reagents
and conditions: (a) Ac O, heat; (b) 90% HNO , Ac O, cat. concd
2
3
2
ꢁ
H
2
SO
4
, 0–5 C then rt then separate by flash chromatography; (c) 2 N
KOH, heat then 4 N HCl; (d) CH
2
Pd(OH) /C, MeOH; (f) cyclohexanone, BH pyridine complex,
2 2 2
N , ether; (e) H (1 atm), 20%
3.
1
MeOH, rt; (g) 2-pyridinecarboxaldehyde, oxone
ꢁ
, DMF–water, rt; (h)
KOH, MeOH, 60 C then HCl.
The 6-COOH and 7-COOH analogues were prepared
from commercially available 3-fluoro-4-nitrobenzoic
and 2-chloro-3-nitrobenzoic acids, respectively, using a
similar sequence to that described in Scheme 1.
A general procedure for the synthesis of the benzimida-
zole inhibitors is shown in Scheme 1. Since screening
hits originated from a combinatorial library, well-
established methodologies for solid-or solution p- hase,
parallel synthesis of 5-carboxy-benzimidazole deriva-
All inhibitors were purified by reversed-phase HPLC to
>90% homogeneity and isolated as TFA salts in most
cases. Compounds gave spectral data (electrospray MS
1
and H NMR) consistent with their assigned structures.
1
4
tives were used. For compounds prepared in solution
phase (P=Et), a newly developed protocol using
We have discovered a series of small 5-carboxy-
benzimidazole derivatives that specifically inhibit the
NS5B RNA-dependent RNA-polymerase of the hepati-
tis C virus with IC s in the low micromolar range.
1
14c
oxone as an oxidant was used,
particularly convenient when larger quantities of mate-
and found to be
rial were required. For rapid optimization of sub-
1
5
0
stituents at N and C-2, solid-phase synthesis (P=Wang
Preliminary SAR studies revealed strict requirements
1
resin) became the method of choice. The isomeric 4-
COOH benzimidazole derivative was prepared as shown
in Scheme 2.
for a small C –C carbocyclic substituent at the N -
6
5
position. At the 2-position of the benzimidazole scaf-
fold, small aromatic heterocycles were preferred.

124
P. L. Beaulieu et al. / Bioorg. Med. Chem. Lett. 14 (2004) 119–124
Although the moderate potency of these compounds
and their physicochemical properties may be limiting
their efficacy in a cell-based HCV RNA replication
assay, they do nevertheless, represent an attractive
starting point in our search for more potent molecules
and HCV therapeutics.
7. Kolykhalov, A. A.; Mihalik, K.; Feinstone, S. M.; Rice,
C. M. J. Virol. 2000, 74, 2046.
8
. (a) Behrens, S.-E.; Tomei, L.; De Francesco, R. EMBO J.
996, 15, 12. (b) Lohmann, V.; Korner, F.; Herian, H.;
Bartenschlager, R. J. Virol. 1997, 71, 8416.
1
¨
9
. Full length HCV NS5B was purified as a N-terminal
hexa-histidine fusion protein from baculovirus infected
insect cells. Compounds were tested for inhibitory activity
using a 12 nucleotide RNA oligo-uridylate primer mod-
Acknowledgements
0
ified with biotin at the free 5 C position and a com-
plementary poly-adenylate template of heterogeneous
length (1000–10,000 nucleotides). Polymerase activity was
We wish to acknowledge the contribution of the Boeh-
ringer Ingelheim Pharmaceuticals Inc. (Ridgefield, CT,
USA) screening team, and in particular Drs. Carol
Homon and Paul Kaplita, Gustavo Rodriguez and
George Rogers. The authors also thank Lukxmi Bala-
thasan and Leslie Magtanong for their support running
the polymerase enzymatic assays. We are grateful to
Norman Aubry, Colette Boucher, Sylvain Bordeleau,
Michael Little and Serge Valois for analytical support.
3
measured as incorporation of UMP-[5,6 H] into the
chain elongated from the oligo-U primer using SPA-bead
1
1a
capture and quantification (TopCount).
A more
detailed description of this assay will be published else
where (McKercher, G. et al.).
1
0. Compounds were tested against polio virus RNA-depen-
dent RNA polymerase and a mammalian DNA depen-
dent RNA polymerase II isolated from calf thymus in an
assay format similar to that described for HCV poly-
merase.
1. (a) Austel, V., Beaulieu, P. L., Fazal, G., Gillard, J.,
Kukolj, G. US patents 6,448,281 B1 and 6,479,508 B1,
9
,11a
1
References and notes
2
002, to Boehringer Ingelheim (Canada) Ltd. (b) A recent
patent application from Japan Tobacco Inc. also discloses
derivatives of benzimidazole 5-carboxylic acids as HCV
polymerase inhibitors: Hashimoto, H., Mizutani, K.,
Yishida, A. EP 1 162 196 A1, 2001.
1
. Choo, Q.-L.; Kuo, G.; Weiner, A. J.; Overby, L. R.;
Bradley, D. W.; Houghton, M. Science 1989, 244, 359.
. Wasley, A.; Alter, M. J. Semin. Liver Dis. 2000, 20, 1.
. Berenger, M.; Wright, T. L. Proc. Assoc. Am. Phys. 1998,
2
3
12. The buffer used for NMR experiments consisted of 20
mM Tris-d11 at pH 7.5, 10% glycerol-d , 2 mM DTT-d ,
1
10, 98.
. Di Bisceglie, A. M.; Hoofnagle, J. H. Hepatology 2002,
6 (Suppl. 1), S121.
. (a) Beaulieu, P. L.; Llina
8
10
4
5
1 mM EDTA-d , 300 mM NaCl, 15 mL DMSO-d and
1
2
6
3
2
10% D O (v/v). Data acquisition was performed at
`
s-Brunet, M. Curr. Med. Chem.
600 MHz and 300 K using methods similar to those
reported previously: LaPlante, S. R.; Cameron, D. R.;
Aubry, N.; LeFebvre, S.; Kukolj, G.; Maurice, R.; Thi-
beault, D.; Lamarre, D.; Llinas-Brunet, M. J. Biol. Chem.
1999, 274, 18618. A soluble form of the polymerase,
NS5BÁ21, lacking C-terminal 21 amino acids was used:
Ferrari, E.; Wright-Minogue, J.; Fang, J. W. S.; Baroudy,
B. M.; Lau, J. Y. N.; Hong, Z. J. Virol. 1999, 73, 1649.
Anti-Infective Agents 2002, 1, 163. (b) For recent dis-
closures of other non-nucleoside HCV NS5B polymerase
inhibitors, see: Dhanak, D.; Duffy, K. J.; Johnston, V. K.;
Lin-Goerke, J.; Darcy, M.; Shaw, A. N.; Gu, B.; Silver-
man, 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.; Macar-
ron, 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. (c) Chan, L.; Reddy, T. J.;
Proulx, M.; Das, S. K.; Pereira, O.; Wang, W.; Siddiqui,
A.; Yannopoulos, C. G.; Poisson, C.; Turcotte, N.;
Drouin, A.; Alaoui-Ismaili, M. H.; Bethell, R.; Hamel,
M.; L’Heureux, L.; Bilimoria, D.; Nguyen-Ba, N. J. Med.
Chem. 2003, 46, 1283.
¨
13. Lohmann, V.; Korner, F.; Koch, J. O.; Herian, U.;
Theilmann, L.; Bartenschlager, R. Science 1999, 285, 110.
14. (a) Grimmet, M. R.; In Comprehensive Heterocyclic
Chemistry; Katritzky, A. R., Rees, C. W., Eds.; Perga-
mon: Oxford, 1984; Vol. 5, p 457. (b) Mayer, J. P.; Lewis,
G. S.; McGee, C.; Bankaitis-Davis, D. Tetrahedron Lett.
1
1998, 39, 6655. (c) For the use of oxone in the synthesis
of benzimidazoles from aldehydes and 1,2-phenylenedia-
6
. Reed, K. E.; Rice, C. M. Curr. Top. Microbiol. Immunol.
´
mines see: Beaulieu, P. L.; Hache, B.; von Moos, E.
Synthesis 2003, 1683.
2000, 242, 55.