P2-P1′ macrocyclization of P2 phenylglycine based HCV NS3 protease inhibitors using ring-closing metathesis

Article abstract of DOI:10.1016/j.bmc.2011.06.064

Macrocyclization is a commonly used strategy to preorganize HCV NS3 protease inhibitors in their bioactive conformation. Moreover, macrocyclization generally leads to greater stability and improved pharmacokinetic properties. In HCV NS3 protease inhibitor

Full text of DOI:10.1016/j.bmc.2011.06.064

Bioorganic & Medicinal Chemistry 19 (2011) 4917–4927
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
P2–P10 macrocyclization of P2 phenylglycine based HCV NS3 protease
inhibitors using ring-closing metathesis
Anna Lampa a, Angelica E. Ehrenberg b, Aparna Vema a, Eva Åkerblom a, Gunnar Lindeberg a,
U. Helena Danielson b, Anders Karlén a, Anja Sandström a,
⇑
a Department of Medicinal Chemistry, Organic Pharmaceutical Chemistry, Uppsala University, BMC, Box 574, SE-751 23 Uppsala, Sweden
b Department of Biochemistry and Organic Chemistry, Uppsala University, BMC, Box 576, SE-751 23 Uppsala, Sweden
a r t i c l e i n f o
a b s t r a c t
Article history:
Macrocyclization is a commonly used strategy to preorganize HCV NS3 protease inhibitors in their bio-
active conformation. Moreover, macrocyclization generally leads to greater stability and improved phar-
macokinetic properties. In HCV NS3 protease inhibitors, it has been shown to be beneficial to include a
vinylated phenylglycine in the P2 position in combination with alkenylic P10 substituents. A series of
14-, 15- and 16-membered macrocyclic HCV NS3 protease inhibitors with the linker connecting the P2
phenylglycine and the alkenylic P10 were synthesized by ring-closing metathesis, using both microwave
and conventional heating. Besides formation of the expected macrocycles in cis and trans configuration as
major products, both ring-contracted and double-bond migrated isomers were obtained, in particular
during formation of the smaller rings (14- and 15-membered rings). All inhibitors had Ki-values in the
nanomolar range, but only one inhibitor type was improved by rigidification. The loss in inhibitory effect
Received 27 April 2011
Revised 21 June 2011
Accepted 22 June 2011
Available online 28 June 2011
Keywords:
HCV
Protease inhibitors
Macrocyclization
Phenylglycine
Metathesis
can be attributed to a disruption of the beneficial
which proved to be especially deleterious for the
p
–
p
interaction between the P2 fragment and H57,
-phenylglycine epimers.
Ó 2011 Elsevier Ltd. All rights reserved.
D
1. Introduction
The structural proteins C, E1, E2 and p7 are needed to build the viral
capsid while the non-structural (NS) proteins NS2, NS3, NS4A, NS4B,
NS5A, and NS5B are involved in the viral replication machinery.4–7
Of the viral proteins, NS3 is one of the most extensively examined
targets for anti-HCV therapy.8 NS3 is a dual-activity enzyme of both
serine protease and helicase/NTPase activities. In combination with
its cofactor NS4A, NS3 is responsible for processing the cleavage
between NS3/4A, NS4A/4B, NS4B/5A, and NS5A/5B.9 Also, recent
data suggest that the NS3 protease interferes with the antiviral
defense signaling processes of the infected cell, thereby assisting a
more persistent HCV infection.10 Early studies showed that knock-
out of the NS3/4A protease activity by point mutations in an HCV
RNA transcript resulted in interruption of viral replication in vivo.11
Later, proof of concept was established in humans when administra-
tion of the HCV NS3 protease inhibitor BILN-2061 reduced HCV RNA
plasma levels significantly, something that has also been established
by ensuing inhibitors.12
It is estimated that 130–170 million people, or 2.2–3.0% of the
world’s population are infected by hepatitis C virus (HCV).1 HCV is
a serious health problem that has become the major reason for liver
transplantations in the western world.2 The standard of care consists
of the broad-spectrum antiviral agent ribavirin, in combination with
pegylated interferon-a, taken for 24–48 weeks. The treatment is
often associated with severe side effects and is only efficient for
45–50% of the patients, depending on the viral genotype.3 Therefore,
much effort has been dedicated to the important search for new and
more effective therapies, a fact, that is, reflected by the number of
drug candidates pending in clinical trials. The expansion of drug-
resistant strains of HCV might aggravate anti-HCV therapy and is
an issue to consider when developing new drugs.
HCV belongs to the Flaviviridae family and contains single
stranded (+) RNA. After entry of virus into the host cell, the
cytoplasm receives the viral RNA from which a viral polyprotein of
3000 amino acids is synthesized. The polyprotein is further
processed into four structural and six non-structural viral proteins.
A characteristic of the NS3 protease is that it is inhibited by its
N-terminal cleavage product,13,14 a property that has been exploited
in the development of product-based NS3 protease inhibitors. Such
inhibitors possess a C-terminal acidic functionality, typically a car-
boxylic acid or a bioisostere thereof, for example, an acylsulfona-
mide (see Fig. 1). Other HCV NS3 protease inhibitors are based on
a C-terminal electrophilic group, such as an
of forming a covalent, reversible bond with the catalytic serine.
a-ketoamide, capable
⇑
Corresponding author. Tel.: +46 18 471 4285; fax: +46 18 471 4474.
E-mail address: Anja.Sandstrom@orgfarm.uu.se (A. Sandström).
0968-0896/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2011.06.064

4918
A. Lampa et al. / Bioorg. Med. Chem. 19 (2011) 4917–4927
O
F
N
O
H
N
N
O
N
O
S
O
O
O
O
H
N
O
N
N
H
N
O
O
S
O
N
O
H
N
H
OH
O
O
O
N
H
Danoprevir
IC50 < 0.25 nM, EC50 = 0.6 nM
Ciluprevir
Ki = 40 nM, EC50 = 350 nM
O
N
N
S
N
O
O
O
O
NH
O
N
O
H
N
NH
O
O
O
S
O
O
S
O
O
HN
O
O
N
N
H
Vaniprevir
Ki = 0.05 nM, EC50 = 5nM
TMC435
Ki = 0.36 nM, EC50 = 7.8 nM
Figure 1. Macrocyclic product-based HCV NS3 protease inhibitors that have been, or are, in clinical trials: Ciluprevir,12 Danoprevir,25,26 TMC43527 and Vaniprevir.28,29
There are currently several HCV NS3 protease inhibitors in differ-
ent phases of clinical trials. Two of the electrophilic inhibitors, tela-
previr (Vertex Pharmaceuticals) and boceprevir (Merck & Co., Inc.),
have recently been approved by FDA.15,16 Four of the product based
inhibitors that have been or are at present, in clinical trials are out-
lined in Figure 1. Common features are a proline or a proline mimic
(TMC435) in the P2 position and a macrocycle linker connecting the
P3 and P1 side chains (Ciluprevir, Danoprevir, and TMC435) or con-
necting the P3 capping group and the P2 substituent (Vaniprevir).
Inhibitors based on peptide sequences often suffer from low bio-
availability and poor pharmacokinetic properties.17 Common strat-
egies to overcome these problems involve incorporation of
unnatural amino acids,18,19 peptide isosteres20 or macrocycles21,22
in the peptide sequence. If macrocyclization preorganizes the inhib-
itor in its bioactive conformation it can interact with the protease
with reduced loss of entropy and enhanced binding. Indeed, macro-
cyclization has proven to be a successful approach for several prod-
uct based HCV NS3 protease inhibitors.23,24
as when binding to the wild-type enzyme. In contrast, mutated
forms of protease create an imposed change in binding conforma-
tion for P2 proline based compounds, and thereby loss of inhibitory
potency.32 Yet, an X-ray structure of a phenylglycine-based inhib-
itor co-crystallized with the enzyme would be required to draw
accurate conclusions about the binding conformation. Although
phenylglycine-based inhibitors are promising alternatives to the
proline-based inhibitors, significant improvements in inhibitory
potencies are required.
An optimization strategy previously reported by us33 includes
the introduction of a vinyl in the 3-position of the phenylglycine
and an alkenylic P10 elongation (Compounds A, Fig. 2). Especially
the alkenylic P10 extensions proved to be beneficial for inhibitory
effects of the compounds. However, a positive effect of the vinylic
substitution was observed for the inhibitors based on
cine, which according to modeling could be due to a postulated
interaction between H57 and mainly the vinyl in this case
(Fig. 2).30,33
Docking further indicated proximity between the alkenylic P10
D-phenylgly-
p–p
We have previously reported on phenylglycine (Phg) as an
alternative P2 moiety in HCV NS3 protease inhibitors,30 invoking
an advantageous resistance profile31 and structure–activity rela-
tionships different from proline-based inhibitors. Some structural
moieties, such as the vinyl-ACCA in P1, that are optimized to suit
proline-based inhibitors are rarely beneficial for phenylglycine-
based inhibitors. These findings suggest an alternative binding
mode for phenylglycine-based inhibitors. For example, phenylgly-
linker and the P2 position (see Fig. 2), which inspired us to explore
the corresponding macrocyclization. Although the vinyl and the
P10 side chain are positioned in opposite directions in the model,
we were interested in investigating the outcome of a corresponding
macrocyclization. To our knowledge, similar types of P2–P10 macro-
cyclizations have not been explored thoroughly in HCV NS3 protease
inhibitors previously, besides that described in a very recent patent
application on proline-based inhibitors.34 In addition, we expected
that macrocyclization could help in validating the accuracy of the
model as well as probing the bioactive conformation of phenylgly-
cine-based inhibitors.
cine might interact with the catalytic H57 by
p–p
stacking.30 The
aromaticity of phenylglycine also allows for chemical modifica-
tions that can be used in optimization processes. Both modeling
and inhibition experiments suggest that phenylglycine-based
inhibitors have the possibility to retain the same conformation
when binding to A156T and D168V mutated forms of the protease
Herein the synthesis and biochemical evaluation of a series of
tripeptidic P2 phenylglycine-based HCV NS3 protease inhibitors

A. Lampa et al. / Bioorg. Med. Chem. 19 (2011) 4917–4927
4919
S2
S2
O
O
N
N
Compound A (L-Phg)
Compound A (D-Phg)
O
O
Ki = 63 nM
Ki = 48 nM
O
O
O
O
O
O
O
O
H
H
H
H
O
N
N
S
O
N
N
S
N
H
N
N
H
N
S1'
S1'
H
H
O
O
O
O
S3
S3
S1
S1
Figure 2. Above: Acyclic P2 phenylglycine-based inhibitors based on
L
-phenylglycine (to the left) and D-phenylglycine (to the right), with the binding pockets of the protease
indicated (S3–S10). Below: Docked poses of the compounds above in the full length NS3 protease binding site (PDB code: 1CU1). The protease residues are displayed as
sticks.33
take a larger part in the H57 interaction for the
indicated in red (i.e., 4.7 Å and 6.1 Å in the compounds based on
A
p
–p
interaction between H57 and the phenyl of the
L-phenylglycine (to the left) compound is suggested in the docking solution, whereas the vinyl is proposed to
D
-phenylglycine compound (to the right). The measured distances in Angstrom between the P10-linker and the P2 residue are
L
-phenylglycine and
D-phenylglycine, respectively).
encompassing a macrocycle spanning P10 and P2 is reported. The
macrocyclic inhibitors were synthesized using ring-closing
metathesis and include 14–16-membered macrocycles with
different aromatic or aliphatic moieties built into the linker. Several
products were formed during the metathesis reaction, both ring-
contracted and double-bond migrated products. The product
infra). It should be emphasized that several isomers were formed
during the ring-closure and their isolation was sometimes extre-
mely difficult, a fact that influenced the final yields significantly.
Only pure middle fractions were collected and only the inhibitors
that could be isolated in pure form were evaluated in the inhibition
assay.
pattern was further complicated by the presence of both
L
- and
Compound 1, containing an o-allyl benzenesulfonamide in the
P10 position, was mixed with Grubb’s 2nd catalyst and heated in
toluene to 85 °C. The starting material was fully consumed only
after 40 h of heating and with addition of more catalyst halfway
through the reaction. The Boc protecting group was removed using
D
-phenylglycine epimers. Although the product picture was very
complex, we managed to isolate several potential inhibitors that
provided us with a SAR and information about the bioactive
conformation.
HCl in dioxane and the product subsequently coupled to Boc-
L-tLeu
2. Results
using N-[(dimethylamino)-1H-1,2,3-triazolo-[4,5-b]pyridine-1-
yl-methylene]-N-methylmethaneaminium hexafluorophosphate
N-oxide (HATU) and N,N-diisopropylethylamine (DIEA) in DMF.
NMR analysis of the purified material revealed that the major
product was the expected 15-ring in cis configuration (1a). In addi-
tion, the 14-membered rings 1b (cis) and 1c (trans) were obtained.
Reduction of 1a with PtO2 in methanol under H2-atmosphere
yielded the saturated compound 1d.
From compound 2, with a 1-allylcyclopropane sulfonamide in
the P10 position, the desired 14-membered ring was produced in
very low yield using the standard conditions. However, micro-
wave-heating at 110 °C for 5 min in trifluorotoluene led to almost
full conversion of the starting material. It should be noted that the
Boc group remained intact despite the increase in temperature. No
2.1. Chemistry
Dipeptidic compounds 1–5 (Scheme 1), with vinylated phenyl-
glycine and various alkenylic elongations in P10 were all synthe-
sized as described previously. Due to racemization of the
phenylglycine during the Suzuki vinylation, compounds 1–5 were
mixtures of L-Phg and D
-Phg isomers in a 2:1 ratio.33 Ring-closing
metathesis of compounds 1–5 led to the isolated macrocyclic prod-
ucts 1a–d, 2a–b, 3a–f and 4a–d (Scheme 1). Not all of compounds
1–5 underwent ring-closure by conventional heating in toluene
using Grubb’s 2nd catalyst at 85 °C, so various ruthenium catalysts,
solvents, additives and temperatures were eventually used (vide

4920
A. Lampa et al. / Bioorg. Med. Chem. 19 (2011) 4917–4927
O
N
1a, R=phenyl, n=1, cis, 15.
d
O
1d, R=phenyl, n=1, saturated, 15.
, R=cyclopropyl, n=1, cis, 14.
2a
n
R
2b, R=cyclopropyl, n=1, cis, 14, D-Phg.
3a, R=CH2, n=2, cis, 15.
O
O
O
O
Ring closure
S
H
N
BocHN
3b
, R=CH2, n=2, trans, 15.
3c, R=CH2, n=2, cis, 15, D-Phg.
4a
N
H
N
H
O
O
, R=CH2, n=3, trans, 16.
4b, R=CH2, n=3, trans, 16, D-Phg.
4c
N
d
, R=CH2, n=3, cis, 16.
O
4d
O
, R=CH2, n=3, saturated, 16.
N
Double-bond
migration
RCM (a), deprotection (b),
coupling (c)
O
O
O
H
N
O
S
BocHN
N
R
n
R
H
O
3d, R=CH2, n=1, cis:trans 1:2, 15.
O
O
O
O
S
H
N
BocHN
1, R=o-allylbenzene
2, R=1-allylcyclopropane
N
H
N
H
O
3
, R=pentene
4, R=hexene
5, R=o-vinylbenzene
Double-bond
migration and
ring-contraction
3f, R=CH2, trans, 14.
O
N
O
n
Ring contraction
1b
, R=phenyl, n=0, cis, 14.
R
1c, R=phenyl, n=0, trans, 14.
3e, R=CH2, n=1, cis, 14.
O
O
O
O
S
H
N
BocHN
N
H
N
H
O
Scheme 1. Reagents and conditions: (a) Grubb’s 2nd catalyst/Hoveyda-Grubb’s 2nd catalyst, p-benzoquinone, toluene/trifluorotoluene/1,2-dichloroethane, 85 °C/110 °C/
130 °C, MW (1 (29%), 2 (37%), 3 (44%), 4 (86%), 5 (0%)); (b) 4.0 M HCl in dioxane (100%); (c) Boc-t-Leu, HATU, DIEA, DMF, rt (1a (39%), 1b (4.5%), 1c (5.9%), 2a (27%), 2b (7.8%),
3a (6.3%), 3b (3.3%), 3c (2.8%), 3d (3.9%), 3e (3.6%), 3f (3.8%), 4a (8.4%), 4b (4.1%), 4c (2.5%)); (d) PtO2, H2, MeOH (1d (50%), 4d (13.5%)).
ring-contraction was observed in this case. After subsequent HATU
mediated coupling with Boc- -tLeu, the - and -phenylglycine iso-
mers (2a and 2b, respectively) with cis configuration could be iso-
lated by RP-HPLC.
The same microwave method was used on compound 3, bearing a
4-pentene sulfonamide. After ring-closure followed by coupling
formation of more ring-contracted and double-bond migrated prod-
ucts. Consequently, the ring-closing metathesis reaction of the next
compound (4), with a 5-hexene sulfonamide, was carried out using
three different methods. The reaction products were analyzed by
chiral HPLC, LC–MS and NMR. Method 1, using Grubb’s 2nd catalyst
in toluene and conventional heating at 85 °C, was compared toMeth-
od 2, using Grubb’s 2nd catalyst in trifluorotoluene and heating at
110 °C by microwave irradiation, and Method 3, using Hoveyda-
Grubb’s 2nd catalyst in dichloroethane, p-benzoquinone, and micro-
wave irradiation at 130 °C. p-Benzoquinone has previously been
shown to reduce the amount of undesired ring-contracted and
double-bond migrated isomers during ring-closing metathesis.35
However, in this case no double-bond migration occurred and
contracted rings were formed in only 5–8% of the total conversion
yield (Table 1). The presence of p-benzoquinone did not reduce the
amount of ring-contracted product. It seems, therefore, that the for-
mation of different isomers is more dependent on the substrate.
Thus, the expected products (16-membered rings) from the diaste-
reomeric starting material 4 (based on the enantiomeric pair of
L
L
D
with Boc-
(3a, 3c) and trans (3b) products of the expected 15-memberd rings,
in both the - (3a and 3b) and - (3c) phenylglycine configuration
L-tLeu several products could be isolated. Thus, both cis
L
D
were formed. Migration of the double-bond resulted in 3d (15-ring)
in a cis/trans ratio of 1:2. Ring-contraction was also observed (3e,
14-ring), as well as ring-contraction with double-bond migration
(3f, 14-ring).
The results above showed that ring-closure of 1, 2 and 3 required
different methods. For example, 2 did not cyclize using conventional
heating, while it was almost fully converted using the microwave
method. For compound 3 microwave-heating resulted in ring-
contraction and double-bond migration besides formation of the ex-
pected product. Althoughthe problems with double-bond migration
and ring-contraction described above paradoxically led to isolation
of several potential inhibitors, we wanted to examine what impact
the different methods had on the outcome of the reaction and if
the increased temperature and microwave irradiation resulted in
L- and D-phenylglycine) are formed in ca. 92–95% yield and the
ring-contracted products in a total yield of ca. 5–8%. It might be
worth mentioning that the Boc protecting group remained intact
in all cases despite which method was used.

A. Lampa et al. / Bioorg. Med. Chem. 19 (2011) 4917–4927
4921
Table 1
Products formed using three different methods for ring-closing metathesis of compound 4
O
O
O
O
O
O
N
N
N
N
N
N
O
O
O
O
O
O
+
+
+
+
RCM
O
O
O
O
O
O
O
O
H
N
H
N
H
N
H
N
H
N
H
N
S
S
O
O
S
O
O
S
O
O
S
O
O
S O
O
BocHN
N
BocHN
N
H
BocHN
N
H
BocHN
N
H
BocHN
N
H
BocHN
N
H
H
O
O
O
O
O
O
4
trans, L-Phg,
trans, D-Phg,
cis, L-Phg,
cis, D-Phg,
Ring contraction
precursor to 4a
precursor to 4b
precursor to 4c
final product not isolated
Method 1
Method 2
Method 3
58.4%
54.9%
51.7%
23.1%
20.2%
20.9%
7.7%
13.6%
15.1%
5.3%
4.2%
4.3%
5.1%
7.1%
7.9%
Method 1: Grubb’s 2nd catalyst, toluene, 85 °C, 6 h. Method 2: Grubb’s 2nd catalyst, trifluorotoluene, MW, 110 °C, 5 min. Method 3: Hoveyda-Grubb’s 2nd catalyst, dichlo-
roethane, p-benzoquinone, MW, 130 °C, 5 min.
After deprotection of the crude material obtained from ring-
closing metathesis of and coupling with Boc- -tLeu the
corresponding final products 4a with trans configuration (16-ring,
-Phg isomer), 4b with trans configuration (16-ring, -Phg isomer)
and 4c with cis configuration (16-ring, -Phg isomer) could be iso-
35 nM. Interestingly, the configuration of the phenylglycine has a
small impact on the inhibitory effect. Molecular modeling suggests
4
L
that the vinyl substituent of the
have a favorable interaction with the catalytic H57 that can
compensate for the lost interaction with the aromatic side
chain of phenylglycine visualized for the
-epimers (Fig. 2).33
D-phenylglycine-based inhibitors
L
D
p–p
L
p–p
lated in pure form by preparative chiral HPLC. However, the satu-
L
rated 16-ring 4d was obtained by reduction of a mixed fraction of
the L-phenylglycine isomer products 4a and 4c, using PtO2 in meth-
anol under H2-atmosphere.
Molecular modeling also inspired us to exploit the proximity of
the P10 sulfonamide substituents and the P2 residue to develop
macrocyclic HCV NS3 protease inhibitors with a linker connecting
the P2 and P10 side chains (Fig. 2). Although the vinyl on the P2
phenylglycine and the alkene substituent in P10 were pointing in
opposite directions in the model they could be utilized for cycliza-
tion via ring-closing metathesis. We hoped to find compounds
with enhanced binding and advantageous pharmacokinetic prop-
erties and to gather information about the bioactive conformation,
especially the vinyl/phenylglycine interaction with H57, that might
strengthen the suggested hypothesis. Thus, phenylglycine-based
P2-P10 macrocyclic HCV NS3 protease inhibitors of different ring-
sizes (14–16 atoms) and with various sulfonamide linkers were
synthesized using ring-closing metathesis (Scheme 1 and Tables
2 and 3). Based on the distance between the two connection points
in the model (Fig. 2) and docking of compounds of different ring
sizes (e.g., Fig. 3, vide infra), while considering both ring strain
and bulkiness, 14–16-membered rings appeared as reasonable
candidates for this initial study. Besides the target substances,
some side-products caused by ring-contraction and/or double-
bond isomerization could be isolated and evaluated as inhibitors.
Ring-closure of compound 2 was not possible under conven-
tional metathesis conditions (toluene, Grubbs 2nd catalyst,
85 °C). However, running the reaction in trifluorotoluene in a
microwave synthesizer yielded macrocyclic products in only
5 min. This rapid method was then applied for cyclization of com-
pound 3. Since small amounts of macrocyclic byproducts were
formed by ring-contraction and double-bond migration, a phe-
nomenon that have been observed earlier for ring-closing metath-
esis,38–40 we wanted to examine what impact different metathesis
methods had on the ring-closure of 4. The standard method using
Grubb’s 2nd catalyst and conventional heating was compared with
microwave-heating under similar conditions and also with micro-
wave-heating using Hoveyda-Grubb’s 2nd catalyst in the presence
of p-benzoquinone, a ruthenium hydride scavenger that has pro-
ven successful in suppressing double-bond migration in other
cases.35 However, it turned out that in the cyclization of 4 only
the expected products and small amounts of ring-contracted
products (5–8%) were formed regardless of the method used. No
method was successful in ring-closing the o-vinylbenzene sulfon-
For the cyclization of 5 with an o-vinylbenzene sulfonamide
aiming at a 14-membered ring, several methods were tried. Heat-
ing of 5 with Grubb’s 2nd catalyst in toluene 85 °C resulted in very
small amounts of desired product (14-ring), a small amount of
polymer and mostly starting material remained after 77 h. Change
of method, to microwave irradiation at 110 °C for 5 min with
Grubb’s 2nd catalyst in trifluorotoluene gave the same unsatisfying
result, as did change of catalyst to Hoveyda-Grubb’s 2nd catalyst
and microwave irradiation at 120 °C for 10 min in trifluorotoluene.
A last attempt to ring-close compound 5 was made with [1,3-bis(2-
methylphenyl)-2-imidazolidinylidene]dichloro-(benzylidene)(tri-
cyclophosphine)ruthenium(II), a catalyst designed for ring-closing
metathesis of sterically hindered olefins, heated in trifluorotoluene
by microwave irradiation at 130 °C for 1 h. Yet again, no formation
of product could be seen. Thereafter, no further attempts in ring-
closing 5 were made.
2.2. Biochemical evaluation
Ki-Values for compounds 1a–4d were determined as described
previously36 in an inhibition assay using the full length NS3 protein
expressed from mRNA of genotype 1a and a peptide fragment of 16
amino acids corresponding to the activating region of NS4a (Tables
2 and 3). Compounds 1a, 2a and 3b were also evaluated for EC50
-
values in a replicon system using the Huh-7 cell line containing
subgenomic HCV RNA genotype 1b replicon with firefly lucifer-
ase.37 Only compound 3b with an EC50-value of 6
lM measured
an EC50-value lower than 10 lM.
3. Discussion
It has previously been shown that a vinyl moiety on the P2 phen-
ylglycine in combination with alkenylic prime side-substituents
improve the activity of P2 phenylglycine-based HCV NS3 protease
inhibitors ( Fig. 2). The alkenylic P10 substituent seems to reach
the space between K136 and Q41, thereby maximizing the hydro-
phobic contact giving the best inhibitor of this type a Ki-value of

4922
A. Lampa et al. / Bioorg. Med. Chem. 19 (2011) 4917–4927
Table 2
Cyclized inhibitors. Enzyme inhibition constants (Ki-values) measured with wild-type full-length NS3, genotype 1a
O
N
O
R
O
S
O
O
H
N
O
BocHN
NH
N
H
O
Compound
R
Ring size
Double-bond configuration
Phg configuration
Ki SD (nM)
1a
15
14
cis
L
190 50
1b
cis
L
170 40
1c
1d
2a
14
15
14
trans
L
L
L
150 40
250 40
110 18
Saturated
cis
2b
3a
3b
3c
14
15
15
15
cis
D
L
870 110
390 25
240 22
760 100
cis
trans
cis
L
D
L
3e
14
cis
137 18
4a
4b
4c
16
16
16
16
trans
L
940 120
710 70
320 40
570 75
trans
D
L
cis
4d
Saturated
L
SD, standard deviation.

A. Lampa et al. / Bioorg. Med. Chem. 19 (2011) 4917–4927
4923
Table 3
Cyclized, double-bond migrated, inhibitors. Enzyme inhibition constants (Ki-values) measured on wild-type enzyme, genotype 1a
O
N
O
R
O
O
O
O
S
H
N
BocHN
N
H
N
H
O
Compound
R
Ring size
15
Double-bond configuration
Phg configuration
Ki SD (nM)
3d
cis:trans 1:2
L
470 50
222 24
3f
14
trans
L
SD, standard deviation.
Figure 3. Compound 4a (green, left) and compound 3e (green, right) docked in the full-length NS3 protease binding site (PDB code: 1CU1). The protease residues are
displayed as sticks.
amide compound 5 to yield the 14-membered ring with an aryl-
containing linker. Interestingly, a 14-membered ring could be ob-
tained as a ring-contracted byproduct from the cyclization of the
o-allylbenzenesulfonamide analog 1, forming compounds 1b and
1c.
inhibitory effect than their acyclic counterparts.33 A slight prefer-
ence was seen for the 14-membered macrocycles (compounds
1b, 1c, 2a, and 3e) over 15- (compounds 1a, d and 3a–d) and 16-
membered macrocycles (compounds 4a, 4b, 4c, and 4d). Among
the 14-membered macrocycles the linker type did not have a large
impact on the inhibitory effect. Compound 1b (Ki = 170 nM) with
an aryl linker was slightly less potent than compound 3e
(Ki = 137 nM) with an alkyl linker and compound 2a (Ki = 110 nM)
with a cyclopropyl linker. Comparing the 14-membered macrocy-
cles with aryl linkers in cis configuration (1b, Ki = 170 nM) and
trans configuration (1c, Ki = 150 nM) reveals that the double-bond
configuration is insignificant for the magnitude of the inhibitory ef-
fect. Neither does double-bond conjugation with the phenylglycine
show any large impact on inhibitory effect (compare conjugated
compound 3e, Ki = 137 nM, with non conjugated compound 3f,
Ki = 222 nM). However, the phenylglycine configuration has a sig-
nificant influence on the Ki-value. The comparison of the 14-mem-
bered macrocycle compounds with cyclopropyl linkers shows an
Taken together, despite the similarities of substrates 1, 2, 3, 4
and 5 they required different conditions and yielded different out-
comes upon macrocyclization. Formation of the 14-membered
macrocycles needed increased temperature, and microwave irradi-
ation for substrate 2 and was not possible for substrate 5. A possi-
ble explanation to the disability of 5 to undergo ring-closing
metathesis might be interference with formation of a metallacyc-
lobutane complex of the styrene which then chelates to the sulfo-
nyl oxygen.41,42 The 15-membered macrocycles could be formed
under conventional heating but formed some ring-contracted (sub-
strates 1 and 3) and double-bond migrated products (substrate 3).
Formation of the 16-membered macrocycles was uncomplicated
and resulted in product formation with only minor amounts of
ring-contracted products irrespective of what method was used.
This indicates that the outcome of the reaction depends more on
the substrate used than the method.
eight-fold preference for the
Ki = 110 nM vs -isomer 2b, Ki = 870 nM). This result supports the
view that the vinyl has an important interaction with H57,
L-configuration (L-isomer 2a,
D
p–p
All inhibitors in the macrocyclic series had Ki-values in the
upper nanomolar range, with the best inhibitor possessing a Ki-
value of 110 nM. All macrocyclic inhibitors except 2a had lower
disrupted upon macrocyclization. Notably, the cyclopropyl moiety
which is a commonly employed P10 substituent in HCV NS3 prote-
ase inhibitors (see Fig. 1) is present in the best inhibitor of this

4924
A. Lampa et al. / Bioorg. Med. Chem. 19 (2011) 4917–4927
series (2a). In the acyclic series, on the other hand, the cyclopropyl
containing inhibitors were the least potent.33
performed on aluminium sheets precoated with silica gel 60 F254
(0.2 mm). Chromatographic spots were visualized using UV-
detection and/or 2% ninhydrin in ethanol solution followed by
heating. Column chromatography was performed using silica gel
As was the case for the 14-membered macrocycles, the nature
of the different linkers had little impact on the inhibitory activity
of the 15-membered macrocycles. Moreover, the inhibitory po-
tency is not much affected by double-bond configuration or satura-
tion of the double-bond. The phenylglycine configuration shows
the largest influence on the inhibitory effect also in this case. Again,
60 (40–63
Finnigan ThermoQuest AQA system equipped with a C18 (Onyx
Monolithic C18 (50 ꢀ 4.6 mm)) or a C4 (Hichrom ACE C4 (5 m,
lm). Analytical HPLC–MS was performed on a Gilson-
l
50 ꢀ 4.6 mm)) column using MeCN/H2O (0.05% HCOOH) or MeOH/
H2O (0.05% HCOOH) with UV (254 nm) and MS (ESI) detection or
on a Gilson Thermo-Finnigan Surveyor MSQ system using MeCN/
H2O (0.05% HCOOH) with ELSD detection, or on a manual system
equipped with a C8 (Zorbax SB-C8 (4.8 ꢀ 50 mm)) column using
UV (220 or 230 nm) detection. Preparative HPLC–MS was performed
on a Gilson-Finnigan ThermoQuest AQA system equipped with a C8
the L-isomer 3a (Ki = 390 nM) is more potent than the D-isomer 3c
(Ki = 760 nM). Interestingly, for the 16-membered macrocycles
with straight alkyl linkers, which generally were found to be less
efficient inhibitors, the double-bond configuration is of impor-
tance. Comparison of the trans compound 4a (Ki = 940 nM) and
the cis-isomer 4c (Ki = 320 nM), shows a threefold preference for
the cis configuration. These results imply that a conjugated dou-
ble-bond pointing out from the phenylglycine in a trans fashion
is not accommodated by the protease for the 16-membered macro-
cycles, a view, that is, supported by modeling of the 16-membered
trans inhibitor 4a. From modeling of the 14-membered cis inhibitor
3e (Ki = 137 nM) it seems that the shorter linker tightens the mac-
rocycle and takes up less space. This can explain why the 16-mem-
bered macrocycle with cis conformation (4c) is preferred over the
trans compound (4a).
(Zorbax SB-C8 (5
l
m, 150 ꢀ 21.2 mm)) column using MeCN/H2O
(0.05% HCOOH) as the mobile phase with UV (254 nm) and MS
(ESI) detection or a manual system equipped with a C8 (Zorbax
SB-C8 (21.2 ꢀ 150 mm)) column using UV (220 or 230 nm) detec-
tion and MeCN/H2O (0.05% HCOOH) or MeCN/H2O (0.09% TFA) or
MeCN/H2O with 25 mM NH4OAc, pH 6.3 as the mobile phase. Chiral
analysis and separation was performed on a HPLC system equipped
with a Reprosil CHIRAL-NR-R (250 ꢀ 4.6 mm, 8
lm) column or a
m). Purity determinations
Reprosil CHIRAL-NR-R (250 ꢀ 20 mm, 8
l
To summarize, our previous assumption that
tween H57 and the P2 phenyl (in the acyclic -isomers) or the P2
vinyl (in the acyclic -isomers) is of importance is strengthened
p–p
stacking be-
were done by RP-HPLC using the following conditions (UV detection
L
at 254 nm): System 1 (Onyx Monolithic C18, 50 ꢀ 4.6 mm, MeCN/
D
H2O with 0.05% HCOOH) and system 2 (Hichrom ACE C4, 5 lm
in this study. Although all the macrocyclic inhibitors possess
Ki-values in the nanomolar range, they are generally less potent
than their acyclic precursors.33 Thus, it seems that upon macrocyc-
lization, the P2 phenyl is forced to twist out of its binding mode
perpendicular to the backbone, parallel to H57 into a conformation
more parallel to the backbone. Thereby, the beneficial interactions
are disrupted.
50 ꢀ 4.6 mm, MeCN/H2O with 0.05% HCOOH). Microwave reactions
were carried out in a SmithSyntheSizer™ or in an Initiator™ single-
mode microwave cavity producing controlled irradiation at
2450 MHz. NMR spectra were recorded on a Varian Mercury plus
spectrometer (1H at 399.8 MHz, 13C at 100.5 MHz) at ambient tem-
perature. Chemical shifts (d) are reported in ppm, indirectly refer-
enced to tetrametylsilane (TMS) via the solvent signal (1H: CHCl3 d
7.26, CD2HOD d 3.31; 13C: CDCl3 d 77.16, CD3OD d 49.00). Exact
molecular masses were determined on Micromass Q-Tof2 mass
spectrometer equipped with an electrospray ion source.
The inhibitory effects of the macrocyclic
D-isomers are 4–16
times lower than for their acyclic counterparts. For the
L-isomers
the reduction is less pronounced (0.25–9 times). The 14-membered
macrocycle incorporating a cyclopropyl linker is, in fact, more po-
tent than its linear precursor. These results support our view that
the H57-vinyl interaction, which might be lost in a macrocyclic
5.2. General procedure for synthesis of compounds 1a–c, 2a–b,
3a–e and 4a–d
inhibitor, is more important for the D-isomers. It could be expected
that a larger, more flexible linker would allow the vinyl phenylgly-
cine-based compounds to regain the stacking interaction with H57.
However, for the 16-membered macrocycles other factors, such as
the bulk accompanied by the linker seem to be the limitation.
The acyclic starting materials 1–4 were subjected to ring-clos-
ing metathesis as described in detail below. The Boc protecting
group was removed by treating the crude cyclized product with
4.0 M HCl in 1,4-dioxane (10 ml/mmol), until the starting material
could no longer be visualized by LC–MS. The HCl salt recovered
after evaporation of the solvent was used in the ensuing coupling
without further purification. The hydrochloride was mixed with
Boc-t-Leu, HATU, DIEA and DMF and stirred until the starting
material disappeared (LC–MS). After addition of ethyl acetate, the
organic phase was washed with 0.1 M sodium acetate, pH 4, and
evaporated. Purification and separation of isomers was accom-
plished by preparative HPLC (sometimes repeatedly). Yields of
purified final products are calculated from the amount of crude cy-
clized starting material.
4. Conclusion
Ring-closing metathesis was used to synthesize tripeptidic P2
phenylglycine-based HCV NS3 protease inhibitors with a macrocy-
cle spanning the P2 and P10 residues. Inhibitors of varying ring size
and different linkers were evaluated. Even though the different
starting materials are similar, they require different modes of heat-
ing for ring-closure. During the ring-closing metathesis, side reac-
tions leading to ring-contraction and double-bond migration were
observed and seemed to depend on substrate structure rather than
cyclization method. The ring closed inhibitors generally had lower
potency than their acyclic counterparts, indicating that the benefi-
cial interactions with H57 were disrupted upon macrocyclization,
5.2.1. Compounds 1a, 1b and 1c
Compound 1 (65.3 mg, 0.081 mmol) and Grubb’s 2nd catalyst
(9.30 mg, 0.011 mmol) was heated in dry toluene at 85 °C. After
6 h, more catalyst (7.90 mg, 0.009 mmol) was added. After 20 h of
heating, a third portion of catalyst (6.00 mg, 0.007 mmol) was
added and the reaction heated for another 20 h. The solvent was
evaporated and the crude product purified on silica gel (CH2Cl2–
MeOH 93:7). After N-deprotection, the cyclic material (18.0 mg,
0.0232 mmol) was coupled with Boc-t-Leu (10.7 mg, 0.0463 mmol),
especially for the D-phenylglycine epimers.
5. Experimental section
5.1. Chemistry
Reagents and solvents were obtained commercially and used
without further purification. Thin layer chromatography (TLC) was
HATU (21.2 mg, 0.0557 mmol), and DIEA (30.9 lL, 0.181 mmol) in

A. Lampa et al. / Bioorg. Med. Chem. 19 (2011) 4917–4927
4925
DMF (1.6 ml). Three consecutive purifications by preparative HPLC
(MeCN/H2O (0.1% TFA)) yielded 1a (8.1 mg, 39%), 1b (0.92 mg, 4.5%)
and 1c (1.2 mg, 5.9%) as white solids. In a separate experiment, after
extensive purification of the crude ring-closed product from 1, the
cis and trans configurations of the precursors to 1b and 1c, could
be established by NMR analysis, although the assignment was not
possible in the final products. Compound 1a: 1H NMR (CD3OD
(TFA salt)): d 8.59 (d, J = 9.3 Hz, 1H), 8.09 (dd, J = 1.6, 7.9 Hz, 1H),
7.74–7.70 (m, 2H), 7.68–7.55 (m, 6H), 7.42–7.38 (m, 2H), 7.33 (m,
1H), 7.29 (d, J = 7.7 Hz, 1H), 6.83 (d, J = 11.4 Hz, 1H), 6.75 (s, 1H),
6.06 (ddd, J = 7.9, 7.9, 11.5 Hz, 1H), 5.45 (s, 1H), 4.32–4.23 (m,
2H), 4.15 (m, 1H), 4.11 (s, 3H), 3.95 (s, 1H), 1.75–1.54 (m, 2H),
1.44–1.39 (m, 2H), 1.42 (s, 9H), 0.97 (s, 9H), 0.95 (t, J = 7.4 Hz,
3H). 13C NMR (CD3OD (TFA salt)): d 172.9, 172.3, 172.0, 168.4,
166.7, 158.7, 157.8, 151.6, 145.3, 140.0, 138.9, 137.5, 135.1, 133.7,
133.5, 132.8, 132.2, 131.6, 131.3, 130.8, 130.8, 129.9, 129.7, 127.8,
126.8, 125.8, 123.2, 122.4, 115.5, 102.8, 101.6, 80.6, 63.5, 57.7,
57.0, 55.7, 35.4, 33.8, 31.5, 28.7, 27.1, 20.0, 14.0. HRMS calcd for
25 mM NH4OAc, pH 6.3) two isomers, 2a (6.1 mg, 27%) and 2b
(1.5 mg, 7.8%), could be isolated as white solids. For solubility rea-
sons, the products were later transformed to TFA-salts. Compound
2a: 1H NMR (CD3OD (TFA salt)): d 8.55 (d, J = 9.2 Hz, 1H), 7.92–7.89
(m, 2H), 7.68 (d, J = 7.6 Hz, 1H), 7.65 (s, 1H), 7.63–7.56 (m, 5H), 7.43
(d, J = 8.3 Hz, 1H), 6.81 (s, 1H), 6.61 (d, J = 11.6 Hz, 1H), 5.55 (s, 1H),
5.48 (ddd, J = 7.4, 8.7, 11.6 Hz, 1H), 4.43 (t, J = 7.7 Hz, 1H), 4.11 (s,
1H), 3.98 (s, 1H), 3.19–3.11 (m, 1H), 2.92 (dd, J = 7.4, 14.3 Hz, 1H),
1.77–1.57 (m, 2H), 1.48–1.22 (m, 4H), 1.43 (s, 9H), 1.00 (s, 9H),
0.99 (t, J = 7.5 Hz, 3H), 0.93–0.86 (m, 2H). 13C NMR (CD3OD (TFA
salt)): d 173.2, 172.9, 172.0, 169.0, 167.1, 158.2, 157.8, 151.4,
144.3, 138.0, 134.0, 132.7, 132.5, 131.4, 130.9, 130.8, 129.9, 129.8,
128.5, 126.0, 123.2, 122.7, 115.4, 102.9, 100.8, 80.7, 63.6, 57.2,
57.1, 55.2, 41.9, 33.4, 28.7, 27.5, 27.1, 20.0, 14.0, 10.8, 8.7. HRMS
calcd for C46H55N5O9S [M+H]+ 854.3799, found: 854.3790. HPLC
purity (system 1: 97%, system 2: 89%). Compound 2b: 1H NMR
(CD3OD (TFA salt)) d 8.57 (d, J = 10.0 Hz, 1H), 7.83–7.79 (m, 2H),
7.72 (m, 1H), 7.68–7.62 (m, 2H), 7.61–7.56 (m, 4H), 7.46 (d,
J = 8.9 Hz, 1H), 6.85 (s, 1H), 6.53 (d, J = 11.9 Hz, 1H), 5.62 (s, 1H),
5.56 (ddd, J = 5.5, 10.3, 11.9 Hz, 1H), 4.40 (t, J = 7.8, 1H), 4.11 (s,
3H), 3.78 (s, 1H), 3.22 (m, 1H), 2.51 (m, 1H), 1.95–1.84 (m, 2H),
1.65 (m, 1H), 1.55–1.36 (m, 3H), 1.48 (s, 9H), 1.04–0.99 (m, 2H),
1.00 (s, 9H), 0.98 (t, J = 7.2 Hz, 3H). HRMS calcd for C46H55N5O9S
[M+H]+ 854.3799, found: 854.3793. HPLC purity (system 1: 96%,
system 2: 97%).
C
49H55N5O9S [M+H]+ 890.3799, found: 890.3807. HPLC purity (sys-
tem 1: 93%, system 2: 96%). Compound 1b: 1H NMR (CD3OD (TFA
salt)): d 8.56 (d, J = 9.2 Hz, 1H), 8.09–8.05 (m, 2H), 7.91 (m, 1H),
7.79 (dd, J = 1.6, 7.2 Hz, 2H), 7.73 (d, J = 7.7 Hz, 1H), 7.61–7.54 (m,
5H), 7.52–7.43 (m, 2H) 7.41–7.31 (m, 3H) 6.77 (s, 1H), 5.92 (s,
1H), 4.60 (m, 1H), 4.07 (s, 3H), 4.05 (s, 1H), 1.84 (m, 1H), 1.73 (m,
1H), 1.54–1.47 (m, 2H), 1.45 (s, 9H), 1.04 (s, 9H), 0.98 (t,
J = 7.3 Hz, 3H). HRMS calcd for C48H53N5O9S [M+H]+ 876.3642,
found: 876.3637. HPLC purity (system 1: 91%, system 2: 95%). Com-
pound 1c: 1H NMR (CD3OD(TFA-salt)): d 8.60 (d, J = 9.2 Hz, 1H),
8.10–8.05 (m, 2H), 7.82–7.78 (m, 3H), 7.66–7.49 (m, 8H), 7.47–
7.36 (m, 3H), 6.76 (s, 1H), 5.91 (s, 1H), 4.67 (m, 1H), 4.08 (s, 3H),
4.02 (s, 1H), 1.84 (m, 1H), 1.73 (m, 1H), 1.54–1.47 (m, 2H), 1.45 (s,
9H), 1.03 (s, 9H), 0.99 (t, J = Hz, 3H). HRMS calcd for C48H53N5O9S
[M+H]+ 876.3642, found: 876.3634. HPLC purity (system 1: 99%,
system 2: 75%).
5.2.4. Compounds 3a, 3b, 3c, 3d, 3e and 3f
Compound
3 (105 mg, 0.139 mmol), Grubb’s 2nd catalyst
(30.0 mg, 0.0353 mmol) and trifluorotoluene was heated by micro-
wave irradiation at 110 °C for 5 min. After filtration of the reaction
mixture, the solvent was evaporated. The crude product was puri-
fied on silica (CH2Cl2–EtOH(95%) 96:4). The material obtained after
N-deprotection (42.5 mg, 0.0606 mmol) was coupled with Boc-t-
Leu (28.0 mg, 0.121 mmol), HATU (55.2 mg, 0.145 mmol), and DIEA
(80.8 lL, 0.473 mmol) in DMF (2 mL). Purification by preparative
5.2.2. Compound 1d
HPLC (MeCN/H2O with 25 mM NH4OAc, pH 6.3) yielded 3a
(3.66 mg, 6.3%), 3d (2.26 mg, 3.9%), 3f (1.95 mg, 3.8%) as white sol-
ids. Compounds 3a and 3d were transformed to TFA salts for solu-
bility reasons. By further purification using preparative HPLC
(MeCN/H2O (0.05% HCOOH)) compounds 3b (1.92 mg, 3.3%), 3c
(1.42 mg, 2.8%) and 3e (1.80 mg, 3.6%) were recovered as white sol-
ids. Compound 3b was transformed to TFA salt to improve solubil-
ity. Compound 3a: 1H NMR (CD3OD (TFA salt): d 8.56 (d, J = 9.1 Hz,
1H), 7.81 (dd, J = 1.7, 7.0 Hz, 2H), 7.69 (d, J = 7.3 Hz, 1H), 7.66–7.56
(m, 5H), 7.54 (d, J = 2.0, 1H), 7.43 (d, J = 8.6 Hz, 1H), 6.81 (s, 1H),
6.45 (d, J = 11.4 Hz, 1H), 5.68 (ddd, J = 6.9, 7.9, 11.4 Hz, 1H), 5.59
(s, 1H), 4.47 (dd, J = 6.7, 8.4 Hz, 1H), 4.11 (s, 3H), 3.97 (s, 1H),
3.33–3.12 (m, 2H), 2.29–2.22 (m, 2H), 1.85–1.60 (m, 4H), 1.43 (s,
9H), 1.49–1.35 (m, 2H), 1.00 (s, 9H), 0.98 (t, J = 7.3 Hz, 3H). 13C
NMR (CD3OD (TFA salt)): d 173.8, 173.0, 171.9, 169.0, 167.1,
158.2, 157.8, 151.0, 144.2, 137.9, 136.1, 134.0, 132.8, 131.8,
131.5, 131.3, 131.0, 129.7, 126.1, 124.5, 123.1, 122.7, 115.4,
102.9, 100.7, 80.7, 63.6, 57.2, 57.1, 54.8, 53.3, 35.4, 34.4, 28.7,
28.4, 27.1, 24.9, 20.0, 13.9. HRMS calcd for C45H55N5O9S [M+H]+
842.3799, found: 842.3772. HPLC purity (system 1: 95%, system
2: 98%). Compound 3b: 1H NMR (CD3OD (TFA salt)): 8.61 (d,
J = 10.1 Hz, 1H), 7.85–7.81 (m, 2H), 7.73–7.58 (m, 5H), 7.56 (dd,
J = 2.4, 8.4 Hz, 1H), 7.40 (s, 1H), 7.34 (d, J = 8.4 Hz, 1H), 6.76 (s,
1H), 6.44 (d, J = 15.7 Hz, 1H), 6.21–6.09 (m, 1H), 5.81 (s, 1H), 4.80
(m, 1H), 4.11 (s, 3H), 3.98 (s, 1H), 3.68 (m, 1H), 3.29 (m, 1H),
3.20 (m, 1H), 2.45 (m, 1H), 2.16 (m, 1H), 1.80–1.59 (m, 2H), 1.44
(s, 9H), 1.44–1.41 (m, 2H), 1.02 (s, 9H), 1.01 (t, J = 9.4 Hz, 3H).
HRMS calcd for C45H55N5O9S [M+H]+ 842.3799, found: 842.3814.
HPLC purity (system 1: 96%, system 2: 96%). Compound 3c: 1H
NMR (CD3OD): d 8.27 (d, J = 9.1 Hz, 1H), 7.80–7.74 (m, 2H), 7.51–
7.44 (m, 5H), 7.40 (s, 1H), 7.28 (d, J = 8.2 Hz, 1H), 7.24 (dd, J = 2.5,
Compound 1a (4.98 mg, 0.00560 mmol) and PtO2 (0.75 mg,
0.00330 mmol) were stirred in methanol (2 mL) under H2-pres-
sure. More PtO2 (0.75 mg, 0.00330 mmol) was added after 4 hours.
The solution was filtered and evaporated after 23 h. Purification by
preparative HPLC (MeCN/H2O (0.1% TFA)) yielded 1d (2.8 mg, 50%)
as a white solid. 1H NMR (CD3OD (TFA salt)): d 8.54 (d, J = 9.4 Hz,
1H), 8.04 (dd, J = 1.4, 8.1 Hz, 1H), 7.91–7.87 (m, 2H), 7.70–7.56
(m, 7H), 7.54 (dd, J = 2.4, 9.3 Hz, 1H), 7.48 (d, J = 7.8 Hz, 1H),
7.43–7.30 (m, 2H), 7.20 (s, 1H), 5.49 (s, 1H), 4.34 (dd, J = 5.4,
9.2 Hz, 1H), 4.09 (s, 3H), 4.01 (s, 1H), 3.07 (m, 1H), 2.96–2.74 (m,
3H), 2.07–1.94 (m, 2H), 1.86–1.74 (m, 1H), 1.60 (m, 1H), 1.44–
1.38 (m, 2H), 1.41 (s, 9H), 1.01 (s, 9H), 0.94 (t, J = 7.3 Hz, 3H). 13C
NMR (CD3OD (TFA salt)): d 173.1, 172.5, 171.5, 168.2, 166.6,
158.8, 157.8, 152.8, 142.8, 138.0, 136.7, 135.3, 135.2, 133.9,
133.5, 132.4, 132.3, 131.9, 130.6, 129.8, 129.7, 129.6, 127.4,
125.8, 122.7, 122.4, 115.9, 103.4, 101.5, 80.6, 63.5, 58.1, 57.0,
55.0, 35.4, 33.6, 32.0, 30.4, 29.7, 28.7, 27.1, 19.9, 13.8. HRMS calcd
for C49H57N5O9S [M+H]+ 892.3955, found: 892.3969. HPLC purity
(system 1: 95%, system 2: 98%).
5.2.3. Compounds 2a and 2b
Compound 2 (54.5 mg, 0.0708 mmol) was mixed with Grubb’s
2nd catalyst (15.8 mg, 0.0403 mmol) and dry trifluorotoluene
(23 mL) in a microwave vial and heated by microwave irradiation
at 110 °C for 5 min. After filtration, the solvent was evaporated.
The crude product was purified on silica (CH2Cl2–MeOH 95:5). After
N-deprotection, the cyclic intermediate (19.3 mg, 0.0261 mmol)
was coupled with Boc-t-Leu (12.1 mg, 0.0521 mmol), HATU
(23.8 mg, 0.0626 mmol), and DIEA (34.8
(1 mL). By purification using preparative HPLC (MeCN/H2O with
lL, 0.204 mmol) in DMF

4926
A. Lampa et al. / Bioorg. Med. Chem. 19 (2011) 4917–4927
9.1 Hz, 1H), 6.61 (s, 1H), 6.41 (d, J = 11.6 Hz, 1H), 5.69 (s, 1H), 5.64
(m, 1H), 4.45 (t, J = 7.4 Hz, 1H), 4.00 (s, 3H), 3.84 (s, 1H), 3.50–3.12
(m, 2H), 2.45–2.17 (m, 2H), 1.96–1.70 (m, 4H), 1.47–1.41 (m, 2H),
1.44 (s, 9H), 1.03 (s, 9H), 0.97 (t, J = 7.3 Hz, 3H). HRMS calcd for
1H), 6.49 (d, J = 15.7 Hz, 1H), 6.29 (m, 1H), 5.75 (s, 1H), 4.60 (m,
1H), 4.02 (m, 1H), 4.01 (s, 3H), 2.25 (m, 1H), 2.15 (m, 1H), 1.92–
1.81 (m, 2H), 1.79–1.58 (m, 4H), 1.55–1.38 (m, 4H), 1.43 (s, 9H),
1.02 (s, 9H), 0.97 (t, J = 7.3 Hz, 3H). 13C NMR (CD3OD)174.2,
173.1, 172.4, 164.5, 164.0, 160.9, 157.8, 156.8, 151.2, 138.3,
136.6, 135.1, 132.7, 131.0, 130.5, 129.9, 128.8, 127.0, 125.3,
124.2, 123.2, 120.4, 115.8, 106.8, 102.0, 80.7, 63.7, 56.9, 56.3,
54.5, 53.9, 35.4, 35.1, 31.3, 28.8, 27.7, 27.2, 22.3, 20.2, 13.9. HRMS
calcd for C46H57N5O9S [M+H]+ 856.3955, found: 856.3964. HPLC
purity (system 1: 98%, system 2: 99%). Compound 4b: 1H NMR
C
45H55N5O9S [M+H]+ 842.3799, found: 842.3780. HPLC purity (sys-
tem 1: 89%, system 2: 90%). Compound 3d: 1H NMR (CD3OD (TFA
salt), major isomer reported): d 8.59 (d, J = 9.3 Hz, 1H), 7.87–7.80
(m, 2H), 7.74–7.53 (m, 6H), 7.37 (d, J = 8.2 Hz, 1H), 6.86 (s, 1H),
5.56 (s, 1H), 5.53 (m, 1H), 5.40 (ddd, J = 6.2, 7.6, 15.2 Hz, 1H),
4.46 (dd, J = 6.7, 8.2 Hz, 1H), 4.11 (s, 3H), 3.98 (s, 1H), 3.94 (dd,
J = 7.3, 15.0 Hz, 1H), 3.83 (dd, J = 6.5, 15.0 Hz, 1H), 2.85 (m, 1H),
2.75 (m, 1H), 2.60 (m, 1H), 2.47 (m, 1H), 1.81–1.61 (m, 2H),
1.45–1.40 (m, 2H), 1.42 (s, 9H), 1.01 (s, 9H), 0.98 (t, J = 7.4 Hz,
+
(CD3OD (NH4 salt)) d 8.30 (d, J = 9.2 Hz, 1H), 7.79–7.74 (m, 2H),
7.61 (m, 1H), 7.49 (d, J = 2.5 Hz, 1H), 7.48–7.42 (m, 3H), 7.39 (dd,
J = 2.3, 8.2 Hz, 1H), 7.29 (dd, J = 2.5, 9.2 Hz, 1H), 7.22 (d, 8.3 Hz,
1H), 6.64 (s, 1H), 6.50 (d, J = 15.8 Hz, 1H), 6.45 (m, 1H), 5.51 (s,
1H), 4.74 (m, 1H), 4.01 (s, 3H), 3.88 (s, 1H), 3.74 (m, 1H), 3.28
(m, 1H), 2.28–2.08 (m, 2H), 1.94–1.73 (m, 4H), 1.72–1.55 (m,
2H), 1.50 (m, 1H), 1.43 (s, 9H), 1.30 (m, 1H), 1.02 (s, 9H), 0.97 (t,
3H). HRMS calcd for
C
45H55N5O9S [M+H]+ 842.3799, found:
842.3815. HPLC purity (system 1: 98%, system 2: 99%). Compound
3e: 1H NMR (CD3OD): d 8.27 (d, J = 9.2 Hz, 1H), 7.87–7.84 (m, 2H),
7.66 (s, 1H), 7.56 (dd, J = 2.3, 8.3 Hz, 1H), 7.47 (d, J = 2.6 Hz, 1H),
7.49–7.40 (m, 3H), 7.27 (dd, J = 2.5, 9.2 Hz, 1H), 7.26 (d, J = 8.3 Hz,
1H), 6.69 (s, 1H), 6.55 (d, J = 11.8 Hz, 1H), 5.69 (m, 1H), 5.56 (s,
1H), 4.44 (t, J = 7.6 Hz, 1H), 4.02 (s, 1H), 4.00 (s, 3H), 3.27 (m,
1H), 3.14 (m, 1H), 2.90 (m, 1H), 2.47 (m, 1H), 1.82–1.61 (m, 2H),
1.52–1.35 (m, 2H), 1.43 (s, 9H), 1.02 (s, 9H), 0.99 (t, J = 7.4 Hz,
J = 7.3 Hz, 3H). 13C NMR (CD3OD (NH4 salt): d 175.2, 173.8, 1
+
72.9, 164.0, 163.7, 161.0, 158.1, 152.5, 151.6, 140.7, 136.5, 134.9,
132.9, 131.3, 130.8, 129.9, 128.7, 127.9, 124.8, 124.0, 123.6,
120.2, 115.8, 107.5, 101.9, 80.8, 64.5, 59.1, 56.1, 54.4, 53.7, 35.7,
34.9, 30.7, 28.8, 28.3, 27.2, 22.4, 20.0, 14.0. HRMS calcd for
3H). HRMS calcd for
C
44H53N5O9S [M+H]+ 828.3642, found:
C
46H57N5O9S [M+H]+ 856.3955, found: 856.3964. HPLC purity (sys-
tem 1: 95%, system 2: 97%). Compound 4c: 1H NMR (CD3OD (NH4
salt)) d 8.28 (d, J = 9.4 Hz, 1H), 7.84–7.79 (m, 2H), 7.65 (br s, 1H),
7.56 (dd, J = 2.3, 8.3 Hz, 1H), 7.52–7.40 (m, 4H), 7.30–7.25 (m,
2H), 6.60 (s, 1H), 6.34 (d, J = 11.5 Hz, 1H), 5.60–5.52 (m, 2H), 4.35
(dd, J = 6.4, 8.7 Hz, 1H), 4.01–3.98 (m, 4H), 2.19–2.10 (m, 3H),
1.91–1.79 (m, 1H), 1.77–1.63 (m, 3H), 1.60–1.47 (m, 3H), 1.43 (s,
9H), 1.29 (m, 1H), 1.18 (m, 1H), 1.00 (s, 9H), 0.99 (t, J = 7.4 Hz,
+
828.3668. HPLC purity (system 1: 97%, system 2: 63%). Compound
3f: 1H NMR (CD3OD (NH4+ salt)) d 8.2 (d, J = 9.2 Hz, 1H), 7.78–7.75
(m, 2H), 7.50–7.45 (m, 5H), 7.32 (d, J = 2.4 Hz, 1H), 7.27 (dd, J = 2.5,
9.2 Hz, 1H), 7.24 (d, J = 8.4, 1H), 6.64 (s, 1H), 5.85 (ddd, J = 5.9, 7.4,
15.4 Hz, 1H), 5.75 (s, 1H), 5.51 (m, 1H), 4.55 (dd, J = 6.9, 8.2 Hz, 1H),
4.04 (s, 1H), 4.00 (s, 3H), 4.15–3.96 (m, 2H), 3.47–3.24 (m, 2H), 1.79
(m, 1H), 1.67 (m, 1H), 1.43 (s, 9H), 1.49–1.34 (m, 2H), 1.03 (s, 9H),
0.97 (t, J = 7.3 Hz, 3H). HRMS calcd for C44H53N5O9S [M+H]+
828.3642, found: 828.3660. HPLC purity (system 1: 98%, system
2: 99%).
3H). 13C NMR (CD3OD (NH4 salt)) d 174.5, 172.9, 172.1, 164.4,
+
163.7, 160.5, 152.5, 152.0, 150.8, 140.0, 136.2, 135.8, 132.2,
131.1, 131.0, 130.2, 129.3, 128.5, 124.3, 124.1, 123.1, 120.1,
115.7, 107.1, 101.5, 80.7, 63.5, 57.4, 56.2, 55.0, 52.1, 35.5, 34.5,
28.9, 28.7, 27.5, 27.2, 24.6, 20.2, 13.9. HRMS calcd for C46H57N5O9S
[M+H]+ 856.3955, found: 856.3943. HPLC purity (system 1: 97%,
system 2: 99%).
5.2.5. Compounds 4a, 4b and 4c
Starting material 4 was ring-closed using three different meth-
ods. Method 1: 4 (25.0 mg, 0.0300 mmol) and Grubbs 2nd catalyst
(2.50 mg, 0.00300 mmol) were stirred in dry toluene (60 mL) in a
round bottomed flask under N2-atmosphere and heated at 85 °C.
The solvent was evaporated after 6 h of heating and the crude
product was purified on silica (CH2Cl2–MeOH 7%). Method 2: 4
(25.0 mg, 0.0300 mmol) and Grubb’s 2nd catalyst (2.50 mg,
0.00300 mmol) were dissolved in dry trifluorotoluene (20 mL) in
a microwave vessel. The vessel was sealed, evacuated and flushed
with N2 before heating by microwave irradiation at 110 °C for
5 min. The solvent was evaporated and the crude product was
purified on silica (CH2Cl2–MeOH 7%). Method 3: 4 (25.0 mg,
5.2.6. Compound 4d
The combined products 4a and 4c (5.90 mg, 0.00689 mmol)
were dissolved in MeOH (3 mL) and stirred together with PtO2
(0.930 mg, 0.00413 mmol) in an H2-atmosphere until the starting
material was no longer visible by LC–MS. The solution was filtered
and evaporated. Purification by RP-HPLC (MeCN–H2O (0.1% TFA))
yielded 4d (0.8 mg, 13.5%).1H NMR (CD3OD (TFA salt)) d 8.55 (d,
J = 8.7 Hz, 1H), 7.82 (dd, J = 7.8, 1.3 Hz, 2H), 7.70–7.53 (m, 7H),
7.31 (d, J = 8.7 Hz), 6.86 (s, 1H), 5.62 (s, 1H), 4.43 (m, 1H), 4.10 (s,
3H), 4.02 (s, 1H), 3.08–2.93 (m, 2H), 1.84 (m, 1H), 1.77–1.67 (m,
2H), 1.61–1.46 (m, 5H), 1.42 (s, 9H), 1.36–1.28 (m, 3H), 1.16–
1.08 (m, 3H), 1.01 (s, 9H), 0.98 (t, J = 7.4 Hz, 3H). HRMS calcd for
0.0300 mmol),
Hoveyda-Grubbs
2nd
catalyst
(0.900 mg,
0.00150 mmol) and p-benzoquinone (0.320 mg, 0.00300 mmol)
were dissolved in dichloroethane (20 mL) in a microwave vessel.
The vessel was sealed, evacuated and flushed with N2 before heat-
ing by microwave irradiation at 140 °C for 5 min. The solvent was
evaporated and the crude product purified on silica (CH2Cl2–MeOH
7%) before the outcomes of the methods were evaluated. The prod-
ucts obtained from using the different methods were combined
(57.0 mg, 0.0770 mmol) and deprotected according to the general
C
46H59N5O9S [M+H]+ 858.4033, found: 858.4118. HPLC purity (sys-
tem 1: 95%, system 2: 99%).
5.3. Enzyme inhibition
The protease activity of the full-length HCV NS3 protein (prote-
ase–helicase/NTPase) from genotype 1a was measured using a
FRET-assay as previously described.36,43 In short, 1nM enzyme
was incubated for 10 min at 30 °C in 50 mM HEPES, pH 7.5,
procedure. The resulting hydrochloride was mixed with Boc-
(53.0 mg, 0.231 mmol), HATU (105 mg, 0.277 mmol) and DIEA
(110 L, 0.631 mmol) in DMF (1.8 mL). After purification using pre-
L-tLeu
l
parative chiral HPLC (i-PrOH–i-hexane 30:70) followed by RP-HPLC
(MeCN/H2O 0.05% HCOOH) or (MeCN/H2O with 25 mM NH4OAc,
pH 6.3), compounds 4a (5.6 mg, 8.4%), 4b (2.7 mg, 4.1%) and 4c
(1.7 mg, 2.5%) were isolated as white solids. Compound 4a: 1H
NMR (CD3OD (NH4 salt)) d 8.33 (d, J = 9.3 Hz, 1H), 7.79–7.76 (m,
2H), 7.54 (d, J = 2.4 Hz, 1H), 7.50 (d, J = 2.4 Hz, 1H), 7.49–7.44 (m,
4H), 7.31 (dd, J = 2.4, 9.0 Hz, 1H), 7.21 (d, J = 8.3 Hz, 1H), 6.65 (s,
10 mM DTT, 40% glycerol, 0.1% n-octyl-b-
with 25 of the peptide cofactor 2K-NS4A (KKGSVVIV-
GRIVLSGK), and inhibitor. The reaction was started by the addition
of 0.5 substrate (Ac-DED(Edans)EEAbu [COO]ASK(Dabcyl)-
NH2) obtained from AnaSpec Inc. (San Jose, USA). Non-linear
regression analysis of the data was made using Grafit 5.0.13 (Erith-
acus software limited).
D-glucoside, 3.3% DMSO
l
M
l
M
w
+

A. Lampa et al. / Bioorg. Med. Chem. 19 (2011) 4917–4927
4927
11. Kolykhalov, A. A.; Mihalik, K.; Feinstone, S. M.; Rice, C. M. J. Virol. 2000, 74,
2046.
5.4. Computational methodology
12. Lamarre, D.; Anderson, P. C.; Bailey, M.; Beaulieu, P.; Bolger, G.; Bonneau, P.;
Bos, M.; Cameron, D. R.; Cartier, M.; Cordingley, M. G.; Faucher, A. M.;
Goudreau, N.; Kawai, 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.; Weldon, S. M.; Yong, C. L.; Llinas-
Brunet, M. Nature 2003, 426, 186.
13. Llinas-Brunet, M.; Bailey, M.; Fazal, G.; Goulet, S.; Halmos, T.; Laplante, S.;
Maurice, R.; Poirier, M.; Poupart, M. A.; Thibeault, D.; Wernic, D.; Lamarre, D.
Bioorg. Med. Chem. Lett. 1998, 8, 1713.
14. Steinkuhler, C.; Biasiol, G.; Brunetti, M.; Urbani, A.; Koch, U.; Cortese, R.; Pessi,
A.; De Francesco, R. Biochemistry 1998, 37, 8899.
15. U. S. Food and Drug Administration http://www.fda.gov/NewsEvents/
Newsroom/PressAnnouncements/ucm255390.htm (Accessed June 2011).
16. Vertex Pharmaceuticals homepage: http://www.vrtx.com/current-projects/
approved-medicines/incivek.html (Accessed June 2011).
All the ligands were built in maestro and geometry optimization
is carried out using OPLS-2005 force field. The crystal structure of
HCV bifunctional protease helicase (1CU1) is used in the study.
This NS3 protein complex consists of the C-terminal helicase do-
main and the N-terminal protease domain with a covalently linked
NS4A cofactor. The protease active site is occupied by the substrate
(NS3 C-terminus which is part of the helicase domain). The active
site was made available for the docking studies by deleting these
terminal residues (624–631) occupying the active site.
All the crystallographic waters were removed and the protein
was prepared for docking using the protein preparation tool imple-
mented in the Schrödinger Suite 2008.44 The generated structures
were subjected to restrained minimization using the OPLS-2005
force field with ‘normal’ BatchMin cutoffs (7.0 Å VDW; 12.0 Å
ELE). To account for the conformational changes in the protein, In-
17. Ripka, A. S.; Rich, D. H. Curr. Opin. Chem. Biol. 1998, 2, 441.
18. Marshall, G. R. Tetrahedron 1993, 49, 3547.
19. Trabocchi, A.; Cini, N.; Menchi, G.; Guarna, A. Tetrahedron Lett. 2003, 44, 3489.
20. Jones, R. C. F.; Ward, G. J. Tetrahedron Lett. 1988, 29, 3853.
21. Marsault, E.; Peterson, M. L. J. Med. Chem. 2011, 54, 1961.
22. Vlieghe, P.; Lisowski, V.; Martinez, J.; Khrestchatisky, M. Drug Discovery Today
2010, 15, 40.
duced fit docking is carried out using FLO (also called QXP).45
A
truncated protein structure that included only amino acids within
9 Å from the residues 624–631 (C-terminus) was used in the dock-
ing studies using FLO. Protein flexibility is accounted for by allow-
ing crucial amino acid residues in the binding pocket to move
freely up to 0.2 Å. Movement larger than 0.2 Å was penalized by
20.0 kJ/mol/Å.2 Flexible residues occupying the active site viz., R
155, Q 526 and K136 were given full conformational freedom. Con-
straints were applied to the inhibitor to restrict its translation
away from the active site. For each inhibitor, 10 unique binding
poses were generated using 2000 Monte Carlo perturbation cycles.
These poses were further subjected to 20 steps of simulated
annealing followed by energy minimization. Each cycle involves
400 rapid Monte Carlo steps generating unique conformations
within an energy window of 50 kJ/mol/Å and an RMSD >0.5 Å. Each
of the complexes was submitted to a 3 fs dynamics at 600 K after
3000 fs of equilibration steps. The maximum movement of an atom
in any single step was limited to 0.1 Å. Hydrogen vibrations were
damped by assigning an atomic weight of 10. Among the 10 best
poses generated, the most plausible binding mode conformation
for each compound was selected based on visual inspection.
23. Venkatraman, S.; Njoroge, F. G. Curr. Top. Med. Chem. 2007, 7, 1290.
24. Tsantrizos, Y. S. Biopolymers 2004, 76, 309.
25. Condroski, K. R. Z. H.; Seiwert, S.D.; Ballard, J. A.; Bernat, B. A.; Brandhuber, B. J.;
Andrews, S. W.; Josey, J. A.; Blatt, L. M. Structure-Based Design of Novel
Isoindoline Inhibitors of HCV NS3/4A Protease and Binding Mode Analysis of
ITMN-191 by X-ray Crystallography: Digestive Disease Week, May 20–25, Los
Angeles, 2006, Poster #T1794;.
26. Mo, H. M.; Bae, A.; Sun, S. C.; Qi, X. P.; Chen, X. W.; Ku, K. R.; Worth, A.; Wong, K.
A.; Harris, J.; Miller, M. D. Antimicrob. Agents Chemother. 2010, 54, 5288.
27. Raboisson, P.; de Kock, H.; Rosenquist, Å.; Nilsson, M.; Salvador-Oden, L.; Lin,
T.-I.; Roue, N.; Ivanov, V.; Wähling, H.; Wickström, K.; Hamelink, E.; Edlund,
M.; Vrang, L.; Vendeville, S.; Van de Vreken, W.; McGowan, D.; Tahri, A.; Hu, L.;
Boutton, C.; Lenz, O.; Delouvroy, F.; Pille, G.; Surleraux, D.; Wigerinck, P.;
Samuelsson, B.; Simmen, K. Bioorg. Med. Chem. Lett. 2008, 18, 4853.
28. Liverton, N. J.; Carroll, S. S.; DiMuzio, J.; Fandozzi, C.; Graham, D. J.; Hazuda, D.;
Holloway, M. K.; Ludmerer, S. W.; McCauley, J. A.; McIntyre, C. J.; Olsen, D. B.;
Rudd, M. T.; Stahlhut, M.; Vacca, J. P. Antimicrob. Agents Chemother. 2010, 54,
305.
29. McCauley, J. A.; McIntyre, C. J.; Rudd, M. T.; Nguyen, K. T.; Romano, J. J.;
Butcher, J. W.; Gilbert, K. F.; Bush, K. J.; Holloway, M. K.; Swestock, J.; Wan, B. L.;
Carroll, S. S.; DiMuzio, J. M.; Graham, D. J.; Ludmerer, S. W.; Mao, S. S.; Stahlhut,
M. W.; Fandozzi, C. M.; Trainor, N.; Olsen, D. B.; Vacca, J. P.; Liverton, N. J. J. Med.
Chem. 2010, 53, 2443.
30. Örtqvist, P.; Peterson, S. D.; Åkerblom, E.; Gossas, T.; Sabnis, Y. A.; Fransson, R.;
Lindeberg, G.; Danielson, U. H.; Karlen, A.; Sandström, A. Bioorg. Med. Chem.
2007, 15, 1448.
31. Dahl, G.; Sandström, A.; Åkerblom, E.; Danielson, U. H. Antivir. Ther. 2007, 12,
733.
32. Örtqvist, P.; Vema, A.; Ehrenberg, A. E.; Dahl, G.; Rönn, R.; Åkerblom, E.; Karlen,
A.; Danielson, U. H.; Sandström, A. Antivir. Ther. 2010, 15, 841.
33. Lampa, A.; Ehrenberg Angelica, E.; Gustafsson Sofia, S.; Vema, A.; Åkerblom, E.;
Lindeberg, G.; Karlen, A.; Danielson, U. H.; Sandström, A. Bioorg. Med. Chem.
2010, 18, 5413.
34. Wang, A. X.; Scola, P. M. WO 2010/077783, 2010.
35. Hong, S. H.; Sanders, D. P.; Lee, C. W.; Grubbs, R. H. J. Am. Chem. Soc. 2005, 127,
17160.
Acknowledgments
We gratefully acknowledge support from Knut and Alice Wal-
lenberg’s foundation. We also thank Dr Aleh Yahorau, Department
of Pharmaceutical Biosciences, Uppsala University, for help with
HRMS analyzes and Medivir AB, Huddinge, Sweden for EC50-deter-
minations and financial support.
36. Poliakov, A.; Hubatsch, I.; Shuman, C. F.; Stenberg, G.; Danielson, U. H. Protein
Expr. Purif. 2002, 25, 363.
37. Lohmann, V.; Korner, F.; Koch, J. O.; Herian, U.; Theilmann, L.; Bartenschlager,
R. Science 1999, 285, 110.
38. Courchay, F. C.; Sworen, J. C.; Ghiviriga, I.; Abboud, K. A.; Wagener, K. B.
Organometallics 2006, 25, 6074.
39. Schmidt, B. Eur. J. Org. Chem. 2003, 816.
40. Fuerstner, A.; Thiel, O. R.; Ackermann, L.; Schanz, H.-J.; Nolan, S. P. J. Org. Chem.
2000, 65, 2204.
References and notes
1. Lavanchy, D. Liver Int. 2009, 29, 74.
2. Brown, R. S. Nature 2005, 436, 973.
3. Feld, J. J.; Hoofnagle, J. H. Nature 2005, 436, 967.
4. Takamizawa, A.; Mori, C.; Fuke, I.; Manabe, S.; Murakami, S.; Fujita, J.; Onishi,
E.; Andoh, T.; Yoshida, I.; Okayama, H. J. Virol. 1991, 65, 1105.
5. Lindenbach, B. D.; Rice, C. M. Nature 2005, 436, 933.
6. Grakoui, A.; Wychowski, C.; Lin, C.; Feinstone, S. M.; Rice, C. M. J. Virol. 1993, 67,
1385.
41. Feldman, J.; Murdzek, J. S.; Davis, W. M.; Schrock, R. R. Organometallics 1989, 8,
2260.
42. Fu, G. C.; Grubbs, R. H. J. Am. Chem. Soc. 1992, 114, 7324.
43. Johansson, A.; Poliakov, A.; Åkerblom, E.; Lindeberg, G.; Winiwarter, S.;
Samuelsson, B.; Danielson, U. H.; Hallberg, A. Bioorg. Med. Chem. 2002, 10, 3915.
44. Maestro. v.8.5; Schrodinger Inc.: 101 SW Main Street, Suite 1300, Portland, OR,
97204.
7. Rosenberg, S. J. Mol. Biol. 2001, 313, 451.
8. Rönn, R.; Sandström, A. Curr. Top. Med. Chem. 2008, 8, 533.
9. Failla, C.; Tomei, L.; De Francesco, R. J. Virol. 1994, 68, 3753.
10. Gale, M.; Foy, E. M. Nature 2005, 436, 939.
45. McMartin, C.; Bohacek, R. S. J. Comput. Aided Mol. Des. 1997, 11, 333.