Synthesis and SAR of piperazinyl-N-phenylbenzamides as inhibitors of hepatitis C virus RNA replication in cell culture

Article abstract of DOI:10.1016/j.bmcl.2009.01.066

The RNA replication machinery of HCV is a multi-subunit membrane-associated complex. NS5A has emerged as an active component of HCV replicase, possibly involved in regulation of viral replication and resistance to the antiviral effect of interferon. We report here substituted piperazinyl-N-(aryl)benzamides as potent inhibitors of HCV replication exerted via modulation of the dimerization of NS5A.

Full text of DOI:10.1016/j.bmcl.2009.01.066

Bioorganic & Medicinal Chemistry Letters 19 (2009) 1779–1783
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Synthesis and SAR of piperazinyl-N-phenylbenzamides as inhibitors
of hepatitis C virus RNA replication in cell culture
a
a
a
b
a
Immacolata Conte a, , Claudio Giuliano , Caterina Ercolani , Frank Narjes , Uwe Koch , Michael Rowley ,
*
Sergio Altamura c, Raffaele De Francesco d, Petra Neddermann e, Giovanni Migliaccio c, Ian Stansfield a
a Department of Medicinal Chemistry, Istituto di Ricerche di Biologia Molecolare ‘‘P. Angeletti” S.p.A., Merck Research Laboratories Rome, 00040 Pomezia, Rome, Italy
b Department of Computational Science, Istituto di Ricerche di Biologia Molecolare ‘‘P. Angeletti” S.p.A., Merck Research Laboratories Rome, 00040 Pomezia, Rome, Italy
c Department of Pharmacology, Istituto di Ricerche di Biologia Molecolare ‘‘P. Angeletti” S.p.A., Merck Research Laboratories Rome, 00040 Pomezia, Rome, Italy
d Department of Molecular Biology and Genomics, Istituto Nazionale Genetica Molecolare INGM, 20122 Milano,Italy
e Department of Biochemistry and Cell Biology, Istituto di Ricerche di Biologia Molecolare ‘‘P. Angeletti” S.p.A., Merck Research Laboratories Rome, 00040 Pomezia, Rome, Italy
a r t i c l e i n f o
a b s t r a c t
Article history:
The RNA replication machinery of HCV is a multi-subunit membrane–associated complex. NS5A has
emerged as an active component of HCV replicase, possibly involved in regulation of viral replication
and resistance to the antiviral effect of interferon. We report here substituted piperazinyl-N-(aryl)benz-
amides as potent inhibitors of HCV replication exerted via modulation of the dimerization of NS5A.
Ó 2009 Elsevier Ltd. All rights reserved.
Received 14 November 2008
Revised 20 January 2009
Accepted 22 January 2009
Available online 27 January 2009
This paper is dedicated to the memory of
Dr. Giovanni Migliaccio—an inspirational
scientist and sorely missed colleague.
Keywords:
Hepatitis C virus
Modulation of the dimerization of NS5A
Substituted piperazinyl-N-(aryl)benzamides
Hepatitis C virus (HCV) is the most frequent cause of chronic
viral hepatitis, with about 3% of the world’s population infected.1
In the majority of acutely infected individuals, HCV evades the im-
mune response and establishes a chronic infection associated with
liver cirrhosis and in some cases hepatocellular carcinoma.2 There
is currently no broadly effective therapy, making the development
of HCV-specific antiviral agents an urgent need.3,4
HCV is a small, enveloped, single stranded positive RNA virus in
the Flaviviridae family. The genome is approximately 10,000 nucle-
otides and encodes a single polyprotein of about 3000 amino acids.
This polyprotein comprises the structural (C, E1 and E2) and non-
structural (NS2, NS3, NS4A, NS4B, NS5A and NS5B) proteins that
are required for replication and packaging of the viral genomic
RNA. The RNA replication machine of HCV is a multi-subunit mem-
brane-associated complex. The nonstructural protein NS5A is an ac-
tive component of HCV replicase5,6, possibly involved in regulation
of viral replication and resistance to the antiviral effect of inter-
feron.7,8 NS5A is a large phosphoprotein (56–58 kDa) organized into
domain. Mutations disrupting either the membrane anchor9 or zinc
binding10 of NS5A have been shown to be lethal for RNA replication.
Thus, NS5A is a potential target for antiviral therapy.
Whilst HCV replicative enzymes, such as NS3 protease and
NS5B polymerase, have been extensively explored, efforts to target
NS5A have largely been hampered by an incomplete understanding
of the role of this multifunctional protein in replication—and a
resultant lack of a screening protocol for enzyme inhibition.
A few years ago, the development of subgenomic replicons that
replicate in a human hepatoma cell line provided a system to test
or screen compounds that inhibit genome replication11,12, and thus
make it possible to identify structural classes that disrupt the func-
tion of any of the viral proteins involved in replication—including
those with no known enzymic activity (such as NS5A).
In this report, we describe a class of piperazinyl-N-(aryl)benz-
amides (identified from screening the in-house compound collec-
tion) as potent inhibitors of HCV replication in such a surrogate
cell-based assay (Fig. 1). Resistant replicon mutants raised to this
class show mutations that map to domain IA of NS5A—implicating
modulation of NS5A function as a potential MOA for this class.
The compounds described herein were assessed for their ability
to inhibit replication of subgenomic HCV RNA, measured in HUH-7
cells using a modification of the procedure of Bartenschlager.13
three domains, including an amphipathic a-helix at its amino termi-
nus that promotes membrane association and a zinc binding
* Corresponding author. Tel.: +39 06 91093285; fax: +39 06 91093654.
E-mail address: immacolata_conte@merck.com (I. Conte).
0960-894X/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2009.01.066

1780
I. Conte et al. / Bioorg. Med. Chem. Lett. 19 (2009) 1779–1783
Table 1
linker
head
Head/tail group SAR: cell-based activity
R'
N
R '
N
N
H
N
N
H
N
tail
R/Ar
Ar
O
O
Figure 1. HCV replication inhibitors: (1) piperazinyl-N-(aryl)benzamides.
Compound
R1
Ar
EC50
>10
(l
M)a
Unless otherwise stated, cell-based data (EC50) were measured in
the presence of 10% fetal calf serum.
O
O
O
O
O
O
O
O
O
6
O
O
O
O
O
O
O
O
O
The piperazinyl-N-(aryl)benzamide scaffold proved readily
amenable to SAR in the head and tail regions. Thus, compounds re-
ported in Table 1 were prepared from commercially available
methyl 4-aminobenzoate, as outlined in Scheme 1. Heating methyl
4-aminobenzoate at reflux with bis(2-chloroethyl)amine hydro-
chloride in n-butanol for 7 days afforded arylpiperazine 3 which
could be readily isolated by filtration. Following capping of the sec-
ondary amine with a BOC group, base mediated hydrolysis of the
ester functionality afforded the free carboxylic acid 5. Amide bond
formation was then mediated either by standard peptide coupling
reagents, or by reaction of the carboxylic acid with a 1:1 mixture of
thionyl chloride/benzotriazole14 in dichloromethane. Filtration to
remove benzotriazole hydrochloride then afforded a stock solution
of the acid chloride, which could be readily aliquoted and reacted
in parallel with a set of amines—facilitating SAR in the tail region.
Final products were typically isolated either via automated direct
phase column chromatography or reverse phase HPLC.
In the instance where Ar was para-hydroxyaniline 7, protection
of the phenolic OH as an acetate ester and removal of the BOC
group gave 9 as a convenient common intermediate for SAR in
the head region. Urethanes, ureas and amides were prepared
through reaction of the secondary amine with, respectively, appro-
priate chloroformates, isocyanates, and acid chlorides, in the pres-
ence of an organic base. Excess reagent was scavenged, permitting
the O-acetate protected intermediates to be isolated by simple fil-
tration and evaporation of volatiles. Base mediated hydrolysis of
the acetate protecting group followed by acidification yielded the
target molecules, which frequently could be isolated via simple fil-
tration and washing with aq HCl or alternatively by automated re-
verse phase HPLC (Scheme 2).
Compounds with modifications to arylpiperazine core were pre-
pared according the Scheme 3. The amide fragment 28 using methyl
4-bromobenzoate, a commercially available starting material. At
first the methyl ester was hydrolysed with aq NaOH and the result-
ing free acid was coupled with 4-aminophenol in DMF, using PyBOP
as coupling agent in the presence of N-methylmorpholine. Protec-
tion of the hydroxyl group was affected with TBDMSCl in the pres-
ence of imidazole. This fragment was then used in the palladium
mediated N-arylation of amines. Following work-up and chroma-
tography, the desired O-de-silylated material was isolated directly,
without the need for a separate deprotection step to give 29 or 30.
Clearly, for inhibitors of replication in cell culture, the absence
of a biochemical assay complicates the interpretation of SAR. For
instance, a structural change might have an impact on either
intrinsic potency on an enzyme target or features pertinent to
the assay context or indeed both.
With this in mind, Table 1 illustrates our findings. In the tail re-
gion, aromatic functionality was preferred over aliphatic (entries 7
and 11). Substitution of the aromatic moiety was critical, with
para-hydroxy 7 proving particularly efficacious. Replacing the OH
with electron withdrawing (isoelectronic F 12, CN 13) or donating
(amide 14, amine 15) or pyridine N-oxide isostere 16 all proved
deleterious. Increasing the bulk of the substituent (e.g., meta-Br
7
0.16
>10
>10
5.6
OH
OH
F
11
12
13
14
15
16
17
CN
O
>10
>10
>10
0.58
N
H
NMe2
N
O
Br
O
18
19
0.69
0.14
Ph
O
O
O
O
O
O
OH
20
21
22
23
24
25
9.6
OH
OH
OH
OH
OH
OH
OH
O
0.55
4.1
Ph
O
O
O
O
0.47
2.7
N
H
O
Ph
N
H
O
19.6
0.78
N
H
O
26
17, acetophenone 18) brought activity back into the sub-
lM range,
a
EC50 values are the mean from at least 2 experiments.
with vinylphenol 19 proving equipotent to 7.

I. Conte et al. / Bioorg. Med. Chem. Lett. 19 (2009) 1779–1783
1781
O
HN
O N
a
b
H2N
N
N
CO2Me
CO2Me
CO2Me
2
3
4
O
O
O
N
(d+e) or f
O
N
c
N
N
R
N
CO2H
Ar
5
O
6 and 7
Scheme 1. Synthesis-SAR in the tail region. Reagents and conditions: (a) bis(2-chlroethyl)amine hydrochloride salt (1.0 equiv), n-butanol, reflux, 7 days; (b) (Boc)2O
(1.1 equiv), Et3 N (1.2 equiv), DMF, rt; (c) 1 N NaOH (aq): EtOH (1: 2), 60 °C; (d) SOCl2-benzotriazole 1.5 M stock solution (1.05 equiv), DCM, rt; (e) Ar/R-NH2 (1 equiv), NMM
(1.5 equiv), DMF, 40 °C; (f) PyBOP (1.5 equiv), NMM (3.0 equiv), 4-amino-phenol (1.1 equiv), DMF, rt.
O
.TFA
H
O
N
N
a
b
N
N
H
N
H
N
7
O
O
OAc
R1:
OAc
O
8
9
R1
N
c, d, e, f
O
O
N
R/Ar
H
N
R/Ar O
R/Ar N
H
O
10a
10b
10c
OH
10
Scheme 2. Synthesis-SAR in the head region. Reagents and conditions: (a) Ac2O (2.0 equiv), Et3 N (2.2 equiv), DCM, rt; (b) DCM: TFA (1:1), rt; (c) chloroformate or acychloride,
or isocianide, or sulfonyl chloride (2.0 equiv), Et3 N (3.5 equiv), THF, rt; (d) PS-trisNH2 resin scavenging; (e) 1 N NaOH (aq) (10.0 equiv), THF; (f) 1 N HCl (aq).
In the head region, carbamates, ureas and amides all afforded
sub- M inhibitors (21, 23 and 26). Bulky substituents proved best
(e.g., 7 and 23), with alkyl preferred over benzyl and aryl across all
the serine protease, ATPase or helicase activities associated with
the NS3-NS4A protein using either the full length NS3-4A complex
or the protease and helicase domains. Similarly, this compound did
l
3 classes (see compounds 7 and 21–26) (see Table 1).
not inhibit the activity of the
DC55 and DC21 forms of the NS5B
For the most part, modifications to the arylpiperazine core
proved incompatible with maintaining cell-based activity (Table
2). Deletion of either of the piperazine nitrogens, 31 (made as
in the Scheme 1 starting from commercially available N-Boc-4-
(4-carboxyphenyl) piperidine) and 29 resulted in a 10-fold loss
of activity with respect to 7. Opening of the piperazine ring
30, positional isomer 32 (made starting from methyl 3-amino-
benzoate) and simple N-methylation of the anilide 33 all led to
structures essentially devoid of activity.
To investigate the mechanism by which this chemotype inhibits
HCV replication, 7 was selected as a representative example for
further profiling. The activity of this compound on the viral en-
zymes was evaluated in cell-free assays performed with the puri-
polymerase. Lastly, the compound did not show any measurable
activity in a cell-based assay for the NS3-4A protease, ruling out
the possibility that the compound was indeed an inhibitor of this
enzyme but required metabolic activation by host cells. These re-
sults strongly suggest that the activity of 7 on replicon replication
is not exerted through inhibition of the viral enzymes for which
cell-free assays are available.
In the attempt to identify potential molecular target(s) for this
class, using escalating concentrations of 7 replicons resistant to 7
were selected and characterized. Several drug-resistant cell clones
were isolated, phenotypically similar to parental cells and express-
ing comparable levels of viral RNA and proteins but exhibiting
reduced susceptibility to 7 and its active analogues. These clones
fied proteins. At concentrations up to 10
lM, 7 did not inhibit
were still sensitive to inhibition by interferon-a as well as
R2
Br
N
R1
a, b, c
H
N
d
H
N
O
O
Br
O
O
29
30
OTBDMS
OH
28
Scheme 3. Synthesis-SAR in the head region. Reagents and conditions: (a) NaOH (1.1 equiv), THF, rt; (b) 4-aminophenol (1.1 equiv), pyBOP (1.5 equiv), NMM (3 equiv), DMF,
rt; (c) TBDMSCl (1.2 equiv), imidazole (2.5 equiv), DMF, rt; (d) Cs2CO3 (1.4 equiv), Pd2(dba)3 (0.02 equiv), BINAP (0.03 equiv), R1R2NH (1.2 equiv), toluene, rt.

1782
I. Conte et al. / Bioorg. Med. Chem. Lett. 19 (2009) 1779–1783
Table 2
Modifications to arylpiperazine core: cell-based activity
Compound
Structure
EC50 (l
M)a
O
O
N
N
29
1.6
H
N
O
O
OH
OH
O
O
NH
30
31
32
>10
H
N
O
O
N
1.7
H
N
O
OH
OH
O
O
N
N
O
>10
>10
N
N
N
H
O
O
33
Me
N
O
OH
a
EC50 values are the mean from at least 2 experiments.
Figure 2. Resistance mutations mapped in the crystal structure of NS5A: red A92 V,
orange Y93H and R157 W, (a) binding model of 7 and NS5A monomer and (b)
binding model of 7 with NS5A dimer.
inhibitors of the NS5B polymerase and NS3/4A protease, demon-
strating that resistance was specific for compound 7.
Analysis revealed that the mutations responsible for resistance
were located in the NS5A coding region. More specifically, repli-
cons containing three different single mutations in the NS5A pro-
tein (A92 V, Y93H and R157 W) were clearly resistant to
compound 7.15 As expected, the replicons carrying these mutations
were cross-resistant to active analogues in the piperazinyl-N-
arylbenzamide class, but not to other replication inhibitors, includ-
ing NS5B polymerase and NS3/4A protease inhibitors.
Taken together, these results point towards the inhibitory activ-
ity of this structural class being exerted by directly or indirectly
interfering with the function(s) of NS5A. In this regard, it is interest-
ing to note that mapping the resistance mutations (A92 V, Y93H and
R157 W) in the crystal structure of NS5A (Fig. 2) domain IA homo-di-
mer, shows them to lie at the interface of the two subunits. Thus, if
one assumes a biological relevance for the dimerization process,
then one might hypothesize that these analogs could exert their
anti-HCV activity via modulation of the dimerization of NS5A.
Modelling studies, docking 7 to the NS5A domain IA homo-di-
mer, show two potential binding modes (Fig. 2). Binding to one
subunit of the dimer, as in Figure 2(a), could perturb dimerization,
whilst binding (Fig. 2(b)) across the dimer interface could exert a
stabilizing effect. In this regard, it is tempting to interpret recent
work in the patent literature16, describing potent symmetrical
and pseudosymmetrical replication inhibitors (with NS5A as their
putative target) as favouring hypothesis (b), whereby this class
might bind across the dimerization interface.
the replicon in diverse structural classes, 7 and 20 in the 100 nM
range. Resistant replicon mutants raised to this class show muta-
tions in domain IA of NS5A – implicating modulation of NS5A func-
tion as a potential MOA for this class. These mutations map to the
homo-dimer interface in the crystal structure of domain IA of NS5A
and suggests a possible biological relevance for the dimerization
process of the protein.
Acknowledgments
We thank Giacomo Paonessa and Laura Pacini for enzymology;
Nadia Gennari and Mauro Cerretani for biological testing, and Fran-
cesca Naimo for RP-HPLC purifications. This work was funded in
part by a grant from the MIUR.
References
1. WHO. Hepatitis C Fact Sheet No. 164. Rev. October 2000.
2. Wong, T.; Lee, S. S. CMAJ 2006, 174, 649.
3. Hayashi, N.; Takehara, T. J. Gastroenterol. 2006, 41, 17.
4. Davis, G. L.; Lindsay, K. L. Lancet Infect. Dis. 2005, 5, 24.
5. Rice, C. M.; Lindenbach, B. D. Nature 2005, 436, 933.
6. Rice, C. M.; Marcotrigiano, J.; Tellinghuisen, T. L. Nature 2005, 435, 374.
7. Gale, M.; Katze, M. G. Pharmacol. Ther. 1998, 78, 29.
8. Gale, M. J., Jr.; Korth, M. J.; Tang, N. M.; Seng-Lai, Tan; Hopkins, D. A.; Dever, T.
E.; Polyak, S. J.; Gretch, D. R.; Katze, M. G. Virology 1997, 230, 217.
9. Seng-Lai, Tan; De Francesco, R. Virology 2007, 364, 1.
We have identified a novel class of potent inhibitors of HCV rep-
lication. SAR in head and tail regions afforded sub-lM activity in

I. Conte et al. / Bioorg. Med. Chem. Lett. 19 (2009) 1779–1783
1783
10. Tellinghuisen, T. L.; Marcotrigiano, J.; Gorbalenya, A. E.; Rice, C. M. J. Biol. Chem.
2004, 279, 48576.
13. Lohmann, V.; Korner, F.; Koch, J.; Henan, U.; Theilmann, L.; Bartenschlager, R.
Science 1999, 285, 110.
11. Lyle, F. R. U.S. Patent 6, 973,257, 1995; Chem. Abstr. 1995, 123, 2870.
12. Prasad, A.; Jackson, P. Abstracts of Papers, Part 2, 212th National Meeting of the
American Chemical Society, Orlando, FL, Aug 25–29, 1996; American Chemical
Society: Washington, DC, 1996; PMSE 189.
14. Sachin, S. C.; Krishnacharya, G. A. Synlett 1999, 11, 1763.
15. De Francesco et al., in preparation.
16. Serrano-Wu et al. Patent WO 2006/133326.