Synthesis, antiviral activity, and conformational studies of a P3 aza-peptide analog of a potent macrocyclic tripeptide HCV protease inhibitor

Article abstract of DOI:10.1016/j.bmcl.2008.02.053

BILN 2061 is a macrocyclic tripeptide inhibitor of hepatitis C virus NS3-4A protease that has shown efficacy in the clinic for treating patients infected with HCV. We have synthesized a P3 aza-peptide analog of a potent macrocyclic tripeptide inhibitor closely related to BILN 2061. This aza-derivative was found to be >2 orders of magnitude less active than the parent macrocycle in both isolated enzyme (HCV NS3-4A) and HCV subgenomic replicon assays. NMR studies of P3 aza-peptides revealed these compounds adopt a β-turn conformation stabilized by an intramolecular H-bonding interaction. Molecular models of these structures indicate a d-like configuration of the P3 aza-residue. Thus, the configurationally undefined nature at P3 in the aza-peptide allows the compound to adopt an H-bond stabilized conformation that is substantially different from that necessary for tight binding to the active site of HCV NS3 protease.

Full text of DOI:10.1016/j.bmcl.2008.02.053

Available online at www.sciencedirect.com
Bioorganic & Medicinal Chemistry Letters 18 (2008) 2745–2750
Synthesis, antiviral activity, and conformational studies of a P3
aza-peptide analog of a potent macrocyclic tripeptide HCV
protease inhibitor
John T. Randolph,* Xiaolin Zhang, Peggy P. Huang, Larry L. Klein, Kevin A. Kurtz,
Alex K. Konstantinidis, Wenping He, Warren M. Kati and Dale J. Kempf
Global Pharmaceutical Research and Development, Antiviral Research, 200 Abbott Park Road, Abbott Park, IL 60064, USA
Received 27 November 2007; revised 15 February 2008; accepted 21 February 2008
Available online 26 February 2008
Abstract—BILN 2061 is a macrocyclic tripeptide inhibitor of hepatitis C virus NS3-4A protease that has shown efficacy in the clinic
for treating patients infected with HCV. We have synthesized a P3 aza-peptide analog of a potent macrocyclic tripeptide inhibitor
closely related to BILN 2061. This aza-derivative was found to be >2 orders of magnitude less active than the parent macrocycle in
both isolated enzyme (HCV NS3-4A) and HCV subgenomic replicon assays. NMR studies of P3 aza-peptides revealed these com-
pounds adopt a b-turn conformation stabilized by an intramolecular H-bonding interaction. Molecular models of these structures
indicate a D-like configuration of the P3 aza-residue. Thus, the configurationally undefined nature at P3 in the aza-peptide allows the
compound to adopt an H-bond stabilized conformation that is substantially different from that necessary for tight binding to the
active site of HCV NS3 protease.
Ó 2008 Published by Elsevier Ltd.
Hepatitis C virus (HCV) infection is a chronic disease
affecting 170 million people worldwide, according to
current estimates. It is the primary cause of chronic liver
disease and death due to liver disease in the United
States today, a problem that is expected to grow with
the aging population in the coming decades.1 The cur-
rent standard of care for treating patients with HCV
infection is co-administration of pegylated interferon-a
and ribavirin.2 This therapy is not well tolerated by
many patients during the long duration of treatment
as the result of numerous side effects associated with
its use. Furthermore, the sustained viral response
(SVR) is <50% for patients with genotype 1a and 1b
virus, which make up the majority of infections in the
United States and throughout much of Europe and
Asia. Thus, there is a growing need for improved thera-
pies to combat HCV infection, particularly those effec-
tive in treating genotype 1 patients.3
NS3 is a dual function enzyme incorporating both a ser-
ine protease and RNA helicase domain that only be-
comes fully functional as a serine protease in a
heterodimeric complex with NS4A. This enzyme com-
plex, which is responsible for four of the five polyprotein
cleavages necessary for viral replication, has been vali-
dated as a target for treating HCV infection through
successful clinical studies using peptidic inhibitors. Ini-
tial clinical success was demonstrated using BILN
2061 (1), a 15-membered ring macrocyclic tripeptide
that produced an impressive reduction in viral RNA lev-
els in patients infected with HCV genotype 1 virus fol-
lowing oral administration.5 This compound is a
potent inhibitor of HCV NS3-4A protease in vitro, dis-
playing single-digit nanomolar activity in both enzyme
(IC50 = 3.0 nM) and replicon (EC50 = 1.2 nM) assays.6
Our interest in these compounds led us to investigate the
properties of a P3 aza-peptide analog of macrocyclic
compounds such as 1 (Fig. 1). The use of aza-peptide
compounds in the treatment of HIV infection offers
some evidence for the feasibility of the approach.7 It
was hoped that the macrocyclic core would reduce flex-
ibility such that the aza-analog would adopt a similar
conformation to 1. Furthermore, a P3 aza-analog might
provide some advantage from the standpoint of synthe-
Intensive research has focused on identifying small mol-
ecule inhibitors of the HCV NS3-4A serine protease.4
Keywords: HCV protease inhibitor; Azapeptide; Macrocycle.
*
Corresponding author. Tel.: +1 847 970 9861; fax: +1 847 938
2756; e-mail: john.randolph@abbott.com
0960-894X/$ - see front matter Ó 2008 Published by Elsevier Ltd.
doi:10.1016/j.bmcl.2008.02.053

2746
J. T. Randolph et al. / Bioorg. Med. Chem. Lett. 18 (2008) 2745–2750
MeO
N
Ar
MeO
N
Ph
O
O
H
N
H
N
P1
CO2H
CO2H
N
N
P2
H
H
P3
O
N
O
N
O
O
H
H
R
O
N
O
O
O
S
N
1
3
Ar =
Ar =
R =
N
H
2
R =
Figure 1. Macrocyclic tripeptide inhibitors of HCV NS3-4A protease. 1 = BILN 2061.
sis since it would avoid the chiral synthesis of the com-
plex heptenylglycine P3 component in 1.8 In the effort,
we selected the structurally similar macrocycle 2 as the
parent compound for modification in order to simplify
the problem from the point of synthesis. Thus, we tar-
geted P3 aza-peptide macrocycle 3 for study.
macrocyclization was prepared as shown in Scheme 1.
Esterification of heptenoic acid 4 followed by reduction
with diisobutylaluminum hydride gave aldehyde 5.
Reaction of 5 with tert-butylcarbazate in refluxing 2-
propanol afforded the hydrazone intermediate, which
was reduced by the action of sodium cyanoborohydride
in the presence of p-toluenesulfonic acid to give heptenyl
analog 6. Acylation of 6 with pentafluorophenyl chloro-
formate provided the carbazate derivative 7, which re-
acted with the known dipeptide 8,11 to give RCM
precursor tripeptide 9.
Our strategy for the synthesis of 3 was analogous to that
reported for the synthesis of 1 and 2, which utilizes a
late-stage ring-closing metathesis (RCM) reaction to
form the macrocycle.9,10 The di-olefin substrate for this
1) BOCNHNH2
2-PrOH, reflux
1) EtOH, HCl, reflux
PFPOCOCl
CO2H
CHO
NHNHBOC
2) DIBAL, DCM
(85%)
2) NaCNBH3
TsOH, MeOH
(46%)
THF, NMM, 0oC
(>95%)
4
4
5
5
6
4
ArHO
ArHO
H
N
CO2Et
N
H
O
H
CO2Et
N
F5
H
O
H
N
N
H
BOC
N
O
O
N
O
8
H
N
O
DMF, 50oC
(73%)
O
5
5
7
9
MeO
N
Ph
Ar =
Scheme 1. Synthesis of P3 aza-peptide.

J. T. Randolph et al. / Bioorg. Med. Chem. Lett. 18 (2008) 2745–2750
2747
MeO
N
O
Ph
MeO
N
Ph
Hoveyda's catalyst (10 mol%)
toluene
O
H
N
H
N
CO2Et
CO2R
N
N
H
H
O
N
O
O
N
H
O
H
N
O
N
O
O
O
n
9 (n = 5)
11 (n = 3)
10 R = Et
3 R = H
LiOH
aq. THF
O
Cl
Cl
Ru
P
3
Hoveyda's Catalyst
Scheme 2. Macrocyclization of aza-peptide using RCM.
Macrocyclization was performed by RCM as shown in
Scheme 2. Initially, 9 was subjected to conditions re-
ported and used for the synthesis of 2 (Hoveyda’s cata-
lyst, 40 °C in CH2Cl2). This resulted in only 50%
macrocyclization after 3 days, giving a 4:1 ratio of cis:-
trans olefin products. When the temperature was raised
to 50 °C using toluene as solvent, only a trace amount of
9 remained after 3 days. The resulting macrocyclic prod-
uct was obtained in a 62% yield as a 3:2 ratio of cis:trans
olefin isomers. These isomers were readily separated by
column chromatography on silica gel to give macrocy-
clic tripeptide 10. Saponification of the ester gave the
target inhibitor 3. The unexpectedly sluggish reactivity
of 9 in the RCM reaction, and the resulting poor
cis:trans ratio of products are likely the result of confor-
mational bias in this intermediate. NMR studies of 9 re-
vealed a b-turn conformation for this compound
resulting from intramolecular H-bonding between the
tert-butylcarbamate carbonyl oxygen and the proton
on the P1-P2 amide nitrogen (Fig. 2). Evidence for this
intramolecular H-bond was obtained by standard meth-
ods comparing chemical shifts for the NH protons in
CDCl3 and DMSO-d6.12 The difference in chemical
shifts observed in the two solvents for the P3 NH (Ha)
was 3.1 ppm, the strong solvent effect indicating the
availability of this proton to interact with solvent. In
the case of the P1 NH (Hb), the observed solvent effect
on chemical shift is almost negligible, with a chemical
shift difference of only 0.2 ppm between the two
solvents. This small solvent effect is typical of a proton
involved in a stable intramolecular H-bonding interac-
tion.13 Furthermore, a positive NOE observed between
the tert-butyl group of the N-terminal BOC and protons
on both the cyclopropane ring and the olefin of the vinyl
cyclopropane (VCP) group is further support for the
proposed b-turn conformation. Finally, the lack of an
NOE between the P1 NH (Hb) and the proton at C-1
of the P2 proline (Hc) indicates compound 9 does not
adopt an extended conformation typical for tripeptide
HCV PIs.14 All of these NMR findings point to a b-turn
conformation for 9. A model of 9 showing the b-turn
conformation is shown in Figure 3. The conformational
constraints imparted by intramolecular H-bonding in 9
would be expected to affect reactivity during RCM,
which is consistent with our experimental findings.
Molecular modeling was used to examine the extent to
which intramolecular H-bonding could affect RCM in
these systems. A model of compound 9 in the b-turn
conformation revealed that enough flexibility remained
such that RCM could still occur (i.e., the two olefins
could still be brought close enough to one another such
that bonding would not be unexpected). However, these
models made clear the added difficulty of RCM in a b-
turn constrained system, thus explaining the need for
higher temperature and longer reaction time to achieve
macrocyclization of 9. This modeling work indicated
that the intramolecular H-bond is made possible by
the P3 aza group adopting a D-like configuration (this
center is inverted relative to the L-amino acids in the
parent tripeptide inhibitors).15 In order to determine
the extent to which the b-turn conformation might affect
ArO
Hc
O
O
N
O
N
Hb
O
N
DMSO-d6
5
CDCl3
O
Ha
Hb
δ 9.71
δ 6.64
H
N
Ha
δ 7.85
δ 8.09
NOE
O
Figure 2. NMR evidence for b-turn conformation of 9.

2748
J. T. Randolph et al. / Bioorg. Med. Chem. Lett. 18 (2008) 2745–2750
in the biochemical assay against NS3-4A protease (J4)
was in the low-micromolar range (IC50 = 3.8 lM). Activ-
ity in the HCV subgenomic replicon assay was further
attenuated 10-fold (EC50 = 31 lM). Activity data re-
ported for parent macrocyclic inhibitor 2 are NS3-4A
protease IC50 = 0.011 lM and replicon EC50 = 0.077
lM.6 Thus, the P3 aza-modification resulted in a loss in
activity in both the enzyme inhibition and replicon assays
of 350- and 400-fold, respectively. This large loss in po-
tency for 3 is likely the result of a significant difference
in overall conformation relative to 2.
Figure 3. Molecular model of b-turn conformation of 9 showing the
close proximity of tert-butyl ester and vinylcyclopropyl groups.21
reactivity, pentenyl analog 11 was prepared using the
methods described for the synthesis of 9. Molecular
modeling studies of 11 suggested that, in the b-turn
form, RCM would be highly difficult if not impossible
due to conformational constraints preventing cycliza-
tion of the intermediate ruthenium alkylidene. Subject-
ing 11 to RCM conditions used to cyclize 9 (50 °C for
3 days in toluene) resulted in no observed macrocycliza-
tion. Raising the temperature to 60 °C resulted primarily
in slow decomposition of 11. After 2 days, a cyclic prod-
uct was obtained in low yield (7% isolated) that was
identified as the trans-olefin macrocycle (the cis-olefin
was not obtained).
Figure 5. Molecular model of 10 showing that an intramolecular H-
bond can exist in the 16-membered ring macrocycle.21
Compound 3 was tested to determine the effect of the P3
aza-peptide modification on in vitro potency.16 Activity
MeO
N
Ph
MeO
N
O
Ph
O
H
N
H
N
CO2H
CO2H
N
N
H
H
O
N
O
N
O
O
H
H
N
O
N
O
O
O
R
12 R =
13 R =
IC50 = 3.5 μM
3
IC50 = 3.8 μM
EC50 = 31 μM
IC50 = 2.1 μM
EC50 = 53 μM
Figure 4. In vitro activity of P3 aza-peptide analogs.

J. T. Randolph et al. / Bioorg. Med. Chem. Lett. 18 (2008) 2745–2750
2749
Attempts to determine the conformation of 3 by NMR
were complicated due to the complexity of the spectra,
especially in CDCl3.17 For this reason, we were unable
to establish the presence of an intramolecular H-bond
using the method described for the study of 9. However,
two non-cyclic P3 aza-peptide compounds 12 and 13
(Fig. 4) gave similar biochemical activity to 3 and were
determined by NMR study to adopt a b-turn conforma-
tion.18 While acyclic azapeptides such as 9, 12, and 13
adopt the b-turn structure, it was unknown whether
the additional constraints imparted by the macrocyclic
structure of 3 would destabilize the intramolecular H-
bond, resulting in a different conformation in solution.
We were able to observe evidence for an intramolecular
H-bond between the P1 NH and tert-butylcarbamate
carbonyl oxygen in compound 10, the ethyl ester of
3.19 The chemical shift for P3 NH (Ha) in 10 differs by
2.65 ppm between spectra in CDCl3 and DMSO-d6,
while the P1 NH (Hb) proton is only shifted 0.60 ppm
from 7.70 ppm in CDCl3 to 8.30 ppm in DMSO-d6.
The small solvent effect on Hb chemical shift is an evi-
dence that it is involved in an intramolecular H-bond,
indicating 10 adopts a b-turn conformation. The some-
what larger solvent effect on Hb in macrocycle 10 in
comparison to the solvent effect on Hb in 9 may indicate
a weaker H-bonding interaction in the case of 10, most
likely due to conformational constraints in macrocycle
10 not present in acyclic peptide 9. A weak NOE ob-
served between the P1 NH and the Ha proton at P2 in
10 suggests the overall conformation of this compound
differs from that of 9, where no such NOE is observed.
A model of 10 showing a conformation stabilized by
an intramolecular H-bond is shown in Figure 5.
Table 1 shows a comparison of NMR data collected for
P3 aza-peptide analogs 9, 10, 12 (ethyl ester), and 13 (ethyl
ester) and macrocyclic tripeptides 1 and 2 (ethyl ester).20
The difference in chemical shift between CDCl3 and
DMSO-d6 for the P1–P2 amide NH (Hb) for each of the
aza-analogs ranges from 0.1 to 0.6 ppm, indicating
involvement of this proton in a strong intramolecular
H-bonding interaction in each case. Such is not the case
for macrocycle 2, where the chemical shift difference was
measured to be 1.69 ppm. The strong NOE observed be-
tween Hb and Hc for macrocycles 1 and 2 is absent from
NOE spectra for the aza-peptides, with the exception of
a weak NOE observed in the case of 10. Likewise, the
NOE observed between the BOC tert-butyl and the
VCP group (both Hd and one of the olefinic protons on
the terminal carbon of the vinyl group) for aza-peptides
9, 12, and 13 is absent in the case of 1 and 2. The NMR
data for P3 aza-peptides consistently support a b-turn
Table 1. Comparison of NMR data for P3 aza-peptides and macrocyclic tripeptide HCV PIs
Compound ID
12 (ethyl ester)
13 (ethyl ester)
9
OAr
OAr
OAr
Ha
Ha
Ha
O
O
O
O
Hc
Hc
Hc
N
N
Hb
N
N
Hb
N
N
Hb
O
N
N
O
N
N
N
N
CO Et
CO Et
2
Structure
CO Et
2
2
O
O
O
O
O
O
5
Hd
Hd
Hd
Evidence for b turn
dHb (CDCl3)
Yes
Yes
Yes
7.82 ppm
7.99 ppm
6.57 ppm
9.55 ppm
No
8.09 ppm
8.20 ppm
6.51 ppm
9.51 ppm
No
7.85 ppm
8.09 ppm
6.64 ppm
9.71 ppm
No
dHb (DMSO-d6)
dHa (CDCl3)
dHa (DMSO-d6)
NOE: Hb-Hc (CDCl3)
NOE: Hb-Hc (DMSO-d6)
NOE: BOC-VCP (CDCl3)
NOE: BOC-VCP (DMSO-d6)
No
Yes
No
Yes
No
Yes
Yes
Yes
Yes
10
2 (ethyl ester)
1
OAr
OAr
OAr
Ha
Ha
Ha
O
O
Hc
Hc
O
Hc
N
Hb
N
Hb
N
Hb
N
N
N
N
N
Structure
N
N
O
O
CO Et
CO Et
2
2
O
CO
H
2
O
O
O
O
O
O
Hd
Hd
Hd
Evidence for b turn
dHb (CDCl3)
Yes
No
No
NA
7.70 ppm
8.30 ppm
6.60 ppm
9.25 ppm
NA
7.03 ppm
8.65 ppm
5.30 ppm
6.99 ppm
Strong
Strong
No
dHb (DMSO-d6)
dHa (CDCl3)
dHa (DMSO-d6)
NOE: Hb-Hc (CDCl3)
NOE: Hb-Hc (DMSO-d6)
NOE: BOC-VCP (CDCl3)
NOE: BOC-VCP (DMSO-d6)
8.48 ppm
NA
7.00 ppm
NA
Weak
NA
NA
Strong
NA
No
No

2750
J. T. Randolph et al. / Bioorg. Med. Chem. Lett. 18 (2008) 2745–2750
reau, N.; Kawal, S. H.; Kukolj, G.; Lagace, L.; LaPlante,
S. R.; Narjes, H.; Poupart, M.-A.; Rancourt, J.; Sentjens,
R. E.; St. George, R.; Simoneau, B.; Steinmann, G.;
Thibeault, D.; Tsantrizos, Y. S.; Welden, S. M.; Yong, C.-
L.; Llinas-Brunet, M. Nature 2003, 426, 186.
conformation that is substantially different from the con-
formation of macrocyclic HCV PIs such as 1 and 2. Thus,
the poor activity of P3 aza-peptide analogs, including
macrocycle 3, s likely the result of conformational con-
straints due to intramolecular H-bonding preventing the
compounds from adopting the necessary extended con-
formation for tight binding to the active site of NS3
protease.
6. Llinas-Brunet, M.; Bailey, M. D.; Bolger, G.; Brochu,
C.; Faucher, A.-M.; Ferland, J. M.; Garneau, M.;
Ghiro, E.; Gorys, V.; Grand-Maitre, C.; Halmos, T.;
Lapeyre-Paquette, N.; Liard, F.; Poirer, M.; Rheaume,
M.; Tsantrizos, Y.; Lamarre, D. J. Med. Chem. 2004,
47, 1605.
In conclusion, we have modified a potent macrocyclic tri-
peptide inhibitor of HCV protease (2) by replacing the P3
alpha-carbon with nitrogen, thereby producing a P3 aza-
macrocyclic peptide analog. This structural modification
resulted in a large loss in inhibitory activity in vitro, as
aza-analog 3 was orders of magnitude less active than 2
in both the biochemical and replicon assays. It is believed
that the loss in potency is due to conformational con-
straints on these aza-analogs preventing an extended con-
formation required for tight binding to the protease active
site. Solution NMR studies of P3 aza-tripeptide HCV
protease inhibitors indicate these compounds adopt a b-
turn structure resulting from intramolecular H-bonding
between the P1 NH and N-terminal tert-butylcarbamate
at P3. Molecular modeling of an aza-peptide analog in
the b-turn conformation indicate a D-like configuration
at P3. Thus, it is believed that the increased flexibility at
the aza-center of 3 relative to the absolute configuration
of the P3 L-amino acid in 2, allowed 3 to adopt an overall
conformation that was stabilized by an intramolecular H-
bond.
7. Atazanavir is an aza-peptide HIV PI. For a recent review,
see: Perez-Elias, M. J. Expert Opin. Pharmacother. 2007, 8,
689.
8. Faucher, A.-M.; Bailey, M. D.; Beaulieu, P. L.; Brochu,
C.; Duceppe, J.-S.; Ferland, J.-M.; Ghiro, E.; Gorys, V.;
Halmos, T.; Kawai, S. H.; Poirer, M.; Simoneau, B.;
Tantrizos, Y. S.; Llinas-Brunet, M. Org. Lett. 2004, 6,
2901.
9. Tsantrizos, Y. S.; Ferland, J.-M.; McClory, A.; Poirier, M.;
Farina, V.; Yee, N. K.; Wang, X. J.; Haddad, N.; Wei, X.;
Xu, J.; Zhang, L. J. Organometal. Chem. 2006, 691, 5163.
10. For a recent review on olefin metathesis, see: Schrock, R. R.;
Hoveyda, A. H. Angew. Chem. Int. Ed. Engl. 2003, 42, 4592.
11. Tsantrizos, Y. S.; Cameron, D.; Faucher, A. -M.; Ghiro,
E.; Goudreau, N.; Halmos, T.; Llinas-Brunet, M. U.S.
Patent 6,608,027 B1, 2003.
12. Andre, F.; Vicherat, A.; Boussard, G.; Aubrey, A.;
Marraud, M. J. Peptide Res. 1997, 50, 372.
13. Andre, F.; Marraud, M.; Boussard, G. Tetr. Lett. 1996,
37, 183.
14. We have observed a strong NOE between the P1 NH and
P2 Ha in NMR spectra of non-aza tripeptide HCV PIs,
including 1 and 2. This NOE is consistent with an
extended conformation for the peptide portion of these
compounds.
Acknowledgment
15. Our model is consistent with X-ray crystal studies of aza-
Asx-Pro dipeptides that reveal a b-turn structure in which
the aza-residue exhibits a D-like chirality.13
The authors thank Kent Stewart for his help in generat-
ing molecular models of compounds 9 and 10.
16. For assay procedures, see: Lu, L.; Pilot-Matias, T. J.;
Stewart, K. D.; Randolph, J. T.; Pithawalla, R.; He, W.;
Huang, P. P.; Klein, L. L.; Mo, H.; Molla, A. Antimicrob.
Agents Chemother. 2004, 48, 2260.
Supplementary data
17. The NMR spectra for 3 in both CDCl3 and DMSO-d6
were characterized by broad signals at room temperature.
It was necessary to heat the DMSO-d6 sample to 60 °C to
obtain useful NMR data for analysis.
18. Huang, P. P.; Randolph, J. T.; Flentge, C.; Klein, L. L.;
Kurtz, K.; Konstantinidis, A.; Kati, W.; Zhang, X.; Stoll,
V. S.; Kempf, D. J. 232nd ACS National Meeting, San
Francisco, CA, Sept, 2006; MEDI-176.
Experimental procedures for the synthesis of compound
3, and detailed NMR data, including 2D spectra in both
CDCl3 and DMSO-d6 for compound 9. Supplementary
data associated with this article can be found, in the on-
line version, at doi:10.1016/j.bmcl.2008.02.053.
19. The P1 NH (Hb) and P3 NH (Ha) in 10 are clearly
observed and assigned by 2D NMR in DMSO-d6 at
60 °C, but are too broad to be assigned directly in
CDCl3. The five points titration at 0%, 5%, 10%, 20%,
and 50% DMSO in CDCl3, including variable temper-
ature (from 30o C to 80 °C) at each titration point, was
used to determine the chemical shift of P1 NH (Hb) and
P3 NH (Ha) in CDCl3.
20. Data for ethyl ester analogs are reported due to the
relative ease of obtaining quality NMR data in CDCl3
for these compounds compared to the corresponding
carboxylic acids. The NMR of 1 in CDCl3 was not
obtained.
References and notes
1. Kim, W. R. Hepatology 2002, 36, s30.
2. Cornberg, M.; Wedemeyer, H.; Manns, M. P. Curr.
Gastroenterol. Rep. 2002, 4, 23.
3. For a recent review on HCV antivirals, see: (a) Ni, Z.-J.;
Wagman, S. Curr. Opin. Drug Disc. Dev. 2004, 7, 446; Also,
see: (b) Bhopale, G. M.; Nanda, R. K. Hepatol. Res. 2005,
32, 146.
4. For a recent review of NS3-4A protease inhibitors, see: (a)
Thomson, J. A.; Perni, R. B. Curr. Opin. Drug Disc. Dev.
2006, 9, 1367; Also, see: (b) Sheldon, J.; Barreiro, P.;
Vincent, V. Expert Opin. Invest. Drugs 2007, 16, 1171.
5. Lamarre, D.; Anderson, P. C.; Bailey, M.; Beaulieu, P.;
Bolger, G.; Bonneau, P.; Boes, M.; Cameron, D. R.;
Cartier, M.; Cordingley, M. G.; Faucher, A.-M.; Goud-
21. The structures corresponding to compounds 9 (Fig. 3) and
10 (Fig. 5) were energy minimized using the CFF force
field within Insight software (Accelrys, San Diego).