SAR studies on a series of N-benzyl-4-heteroaryl-1-(phenylsulfonyl) piperazine-2-carboxamides: Potent inhibitors of the polymerase enzyme (NS5B) of the hepatitis C virus

Article abstract of DOI:10.1016/j.bmcl.2011.03.008

Described herein is the initial optimization of (+/-) N-benzyl-4- heteroaryl-1-(phenylsulfonyl)piperazine-2-carboxamide (1), a hit discovered in a high throughput screen run against the NS5B polymerase enzyme of the hepatitis C virus. This effort resulted in the identification of (S)-N-sec-butyl-6-((R)-3- (4-(trifluoromethoxy)benzylcarbamoyl)-4-(4-(trifluoromethoxy)phenylsulfonyl) piperazin-1-yl)pyridazine-3-carboxamide (2), that displayed potent replicon activities against HCV genotypes 1b and 1a (EC₅₀ 1b/1a = 7/89 nM).

Full text of DOI:10.1016/j.bmcl.2011.03.008

Bioorganic & Medicinal Chemistry Letters 21 (2011) 3142–3147
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
SAR studies on a series of N-benzyl-4-heteroaryl-1-(phenylsulfonyl)piperazine-
2-carboxamides: Potent inhibitors of the polymerase enzyme (NS5B)
of the hepatitis C virus
Robert G. Gentles a, , Min Ding a, Xiaofan Zheng a, Louis Chupak a, Michael A. Poss a, Brett R. Beno a,
⇑
Lenore Pelosi c, Mengping Liu c, Julie Lemm c, Ying-Kai Wang c, Susan Roberts c, Min Gao c, John Kadow b
a Bristol Myers Squibb, Chemical and Protein Technologies, Research and Development, 5 Research Parkway, Wallingford, CT 06492, United States
b Bristol-Myers Squibb, Molecular Sciences and Candidate Optimization, Research and Development, 5 Research Parkway, Wallingford, CT 06492, United States
c Bristol Myers Squibb, Department of Virology, Research and Development, 5 Research Parkway, Wallingford, CT 06492, United States
a r t i c l e i n f o
a b s t r a c t
Article history:
Described herein is the initial optimization of (+/ꢀ) N-benzyl-4-heteroaryl-1-(phenylsulfonyl)piperazine-
2-carboxamide (1), a hit discovered in a high throughput screen run against the NS5B polymerase enzyme
of the hepatitis C virus. This effort resulted in the identification of (S)-N-sec-butyl-6-((R)-3-(4-(trifluoro-
methoxy)benzylcarbamoyl)-4-(4-(trifluoromethoxy)phenylsulfonyl)piperazin-1-yl)pyridazine-3-carbox-
amide (2), that displayed potent replicon activities against HCV genotypes 1b and 1a (EC50 1b/1a =
7/89 nM).
Received 7 January 2011
Revised 26 February 2011
Accepted 2 March 2011
Available online 6 March 2011
Keywords:
HCV
Ó 2011 Elsevier Ltd. All rights reserved.
NS5B
Allosteric inhibitors
SAR
Hepatitis C virus (HCV) infection is a pressing medical problem:
4 million cases have been reported in the United States and
approximately 180 million people are infected globally.1 Patients
with the disease frequently develop a persistent infection that re-
sults in variable hepatic necro-inflammation. Those patients pre-
senting with high inflammatory activity are at significant risk of
rapid disease progression leading to the development of fibrosis
and cirrhosis, the sequelae of which include portal hypertension,
liver failure and hepatocellular carcinoma.2 The current optimal
treatment involves extended administration (48 weeks) of pegylat-
ed interferon and ribavirin (Peg-IFN/RBV).3 This regime is associ-
ated with a poor response rate with only 40–50% of patients
infected with genotype 1 HCV able to achieve a sustained viral re-
sponse (SVR).4 In addition, concomitant severe flu-like side-effects
and anemia associated with this treatment lead to significant pa-
tient discontinuation rates. This has stimulated a number of com-
panies to attempt to identify direct acting antivirals (DAA’s) that
specifically target HCV.5
been reported,7 and a number of these are currently being clini-
cally evaluated, see Figure 1.8
In this Letter, we report on our exploratory work on the optimi-
zation of compound 1, a hit identified in an NS5B high throughput
screen (HTS).9 Preliminary structure–activity relationship (SAR)
investigations quickly led to the identification of the potent piper-
azine based NS5B inhibitor 2, as described below.10
The chemistry used in the preparation of analogs of 1 is shown
in Scheme 1, and primarily employs standard protecting group
manipulations and routine acylation and sulfonylation protocols.11
However, in order to accommodate the greatest array of reagents
in the N-arylation procedure (step f), a microwave mediated SNAr
protocol was adopted that differed slightly from previously re-
ported methodology.12 For the syntheses of N4-heteroaryl carbox-
amide derivatives such as 2, and 39–44, the pendant heteroaryl
groups were introduced onto the piperazine ring as the related het-
eroaryl esters that were subsequently hydrolyzed and condensed
with amines to provide the desired amides. In the case of com-
pound 2, (S)-butan-2-amine was used to furnish the targeted pyr-
idazine carboxamide.
All compounds discussed below were tested as racemates, un-
less otherwise stated, and in those instances, specific enantiomers
were isolated using chiral reverse-phase HPLC. In a limited number
of cases (e.g., Table 1, entries 10 and 11), specific enantiomers were
prepared by beginning the synthesis with the R-enantiomer of the
protected piperazine,13 as shown in Scheme 1.
HCV non-structural protein 5B (NS5B) is an RNA-dependent
RNA polymerase that is essential for viral replication.6 Accordingly,
it has long been considered an attractive target for drug develop-
ment, and several active-site and allosteric NS5B inhibitors have
⇑
Corresponding author.
E-mail address: robert.gentles@bms.com (R.G. Gentles).
0960-894X/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2011.03.008

R. G. Gentles et al. / Bioorg. Med. Chem. Lett. 21 (2011) 3142–3147
3143
Figure 1. Recently reported examples of active-site [PS-6130 and MK-0608] and allosteric [VX-759 and PF-868554] inhibitors of NS5B in clinical development.
Scheme 1. Reagents and conditions: (a) NaOH, 1:1, MeOH, THF; (b) HOAt, DCM, R1-NH2; (c) 1:2 TFA/DCM rt 30 min; (d) R2-SO2Cl, pyridine, DCM, rt; (e) 10% Pd/C, MeOH,
DCM; (f) n-butanol, 175 °C, Ar1X; (g) NaOH, THF/MeOH/H2O; (h) TBTU, DIPEA, R0 NH2.
2
The screening hit 1 can be viewed as a tri-functionalized piper-
azine with pendant moieties projecting along three vectors. Our
choice of the order of exploration of these groups was influenced
be the ease of chemistry. Correspondingly, we chose to sequen-
tially optimize the carboxamide, sulfonamide are N-aryl moieties
using the intermediates shown in Figure 3.
Compounds 3 and 5 can be accessed using variations of the
reaction sequence shown in Scheme 1. Intermediate 4 was pre-
pared using the chemistry depicted in Scheme 2.
As stated above, our initial investigations involved the prepara-
tion of a diverse set of amides derived from intermediate 3 using a
series of aliphatic, benzylic and heteroaromatic-methylamines. A
representative set of the product carboxamides is shown in Table
1, together with their associated inhibition data generated from
HCV 1b and 1a replicon assays.14 From these results and additional
data on related analogs (not reported), the following SAR could be
discerned. Firstly, removal of the methoxy group from 1, as shown
in compound 6 is associated with significant loss of activity.
Methylation at the benzylic position of the carboxamide, as exem-
plified in analog 7, results in the generation of inactive compounds.
This is also the case when the N–H moiety of the amide group is
alkylated, as observed with derivative 8. However, substitution at
the four-position of the benzylic aromatic moiety is generally well
tolerated. In relation to 1b activity, chloro 13, methyl 9 and meth-
oxy 1 derivatives are approximately equipotent, with the trifluoro-
methoxy analog 12 being slightly more active. A similar analysis of
the 1a activities showed 13, 9 and 1 to have similar activity, but
less than that of the trifluoromethoxy derivative 12, suggesting

3144
R. G. Gentles et al. / Bioorg. Med. Chem. Lett. 21 (2011) 3142–3147
Table 1
Replicon EC50 data on 4-(2,6-dimethoxypyrimidin-4-yl)-1-(4-ethylphenylsulfonyl)piperazine-2-carboxamides
a
Compound
Carboxamide N-substituent(s) R,R0
1b Replicon EC50
(lM)
1a Replicon EC50 (lM)
6
7
8
9
1
10
11
12
13
14
15
16
17
18
19
20
Phenylmethyl
2-Phenethyl (diastereomeric mixture)
N-Methyl, N-benzyl
8.8
na
na
na
>50
20.5
23
6.2
14.1
9.2
30.8
>50
>50
12.4
14.6
47.0
na
>50
1.7
2.2
0.9
10.1
0.5
2.1
3.7
2.4
1.2
0.9
1.2
na
4-Methylphenylmethyl
4-Methyoxyphenylmethyl
4-Methyoxyphenylmethyl R isomer
4-Methyoxyphenylmethyl S isomer
4-Trifluoromethoxyphenylmethyl
4-Chlorophenylmethyl
2-Chlorophenylmethyl
2,4-Dichlorophenylmethyl
2,4-Dimethoxyphenylmethyl
3,4-Dimethoxyphenylmethyl
4-(Thiophen2-yl)phenylmethyl
N-Cyclopropylmethyl
(3,5-Dimethyl-1-phenyl-1H-pyrazol-4-yl)methyl
2.1
20.8
a
Values are means of three experiments, and standard deviations are <40% of mean (na = not active, or therapeutic index <10).
Scheme 2. Reagents and conditions: (a) DIPEA, HATU, p-trifluoromethoxybenzylamine, DMF, rt; (b) piperidine/DMF (1:5), rt; (c) 6-chloro-2,4-dimethoxypyrimidine, DIPEA, n
BuOH, 180 °C, microwave; (d) TFA, DCM, rt.
the latter to be a preferred substituent. Analogs containing func-
tionality at the 2-position of the aryl group as in 14, or the di-
substituted analogs 15 and 16, generally did not lose 1b activity,
but were less active against the 1a genotype than the original lead
structure 1. Disubstitution at positions 3- and 4- also did not lead
to any significant enhancement of activity as exemplified by com-
pound 17. In a related exercise, all attempts to replace the aryl
moiety of the benzylic group with an aliphatic substituent resulted
in inactive compounds, as typified by analog 19. More productive,
was the replacement of the phenyl group in 1 with a variety of het-
erocycles, which although they did not enhance the activity of the
series, could introduce significant structural diversity into the
chemotype at little expense to potency, as observed with deriva-
tive 20.
On conclusion of this exercise we next used the intermediate 4
that contains one of the preferred carboxamide functionalities
identified above to examine the SAR of the pendant sulfonamide
moiety. Initial attempts to delete or replace the sulfonamide

R. G. Gentles et al. / Bioorg. Med. Chem. Lett. 21 (2011) 3142–3147
3145
linkage with a number of alternative moieties (alkyl, carboxamide,
carbamate, and urea) proved unsuccessful, as all activity at both
genotypes was lost. Subsequently, the sulfonamide derivatives
shown in Table 2 were prepared and the most salient aspects of
the SAR observed are discussed below.
Deletion of the ethyl group in 1 as shown in analog 21, resulted
in a significant loss of both 1b and 1a activities, as did substitution
at the ortho-position of the arylsulfonamide, as observed in deriv-
ative 23. Replacement of the phenyl group with an aliphatic moi-
ety, as shown in analog 22, was also unsuccessful. Substitution at
Table 2
Replicon EC50 data on 4-(2,6-dimethoxypyrimidin-4-yl)-N-(4-methoxy)benzyl)piperazine-1-sulfonamides
Sulfonyl substituent Ar2
1b Replicon EC50
(lM)
1a Replicon EC50 (lM)
a
Compound
21
22
23
24
25
26
27
28
29
30
31
32
33
Phenyl
Cyclohexyl
o-Fluorophenyl
m-Methylphenyl
m-Trifluoromethoxyphenyl
m-Trifluoromethylphenyl
m-(4-Fluorophenyl)phenyl
p-Trifluoromethoxyphenyl
p-Trifluoromethylphenyl
p-Phenylphenyl
p-Methoxyphenyl
p-Fluorophenyl
3,4-Dimethoxyphenyl
20.3
na
na
na
na
na
41
58
53
na
8.1
13
56
64
na
13
1.4
5.3
6.2
na
0.7
1.2
2.4
5.9
8.1
1.1
a
Values are means of three experiments, and standard deviations are <40% of mean (na = not active, or therapeutic index <10).
Table 3
Replicon EC50 data on 4-heteroaryll-N-(4-(trifluoromethoxy) benzyl)-1-(4-(trifluoromethoxy)phenylsulfonyl)piperazine-2-carboxamides
Ar1 substituent
1b Replicon EC50
(l
M)
1a Replicon EC50 (lM)
a
a
Compound
34
35
36
37
38
39
40
41*
42*
43*
2*
2,4-Dimethoxypyrimidin-4-yl
6-Carboxylic acid-pyridazin-3-yl
5-Carboxylic acid-pyrazin-2-yl
Methyl-6-carboxylatepyridazin-3-yl
Methyl-5-carboxylatepyrazin-2-yl
N-Methyl-6-carboxamidepyridazin-3-yl
N-Methyl-5-carboxyamidepyrazin-2-yl
N-Methyl-6-carboxamidepyridazin-3-yl
N-Cyclopropyl-6-carboxamidepyridazin-3-yl
N-Isopropyl-6-carboxamidepyridazin-3-yl
(R)-N-sec-butyl-6-carboxamidepyridazin-3-yl
N-tert-pentyl-6-carboxamidepyridazin-3-yl
0.4
na
na
2.0
na
na
3.9
0.38
0.14
0.02
0.22
0.007
0.009
0.008
0.007
0.043
1.6
0.33
2.23
0.22
0.169
0.139
0.087
0.46
44
a
Values are means of three experiments, and standard deviations are <40% of mean (na = not active, or therapeutic index <10).
Compounds that have an asterisk have the R absolute configuration at the 2-position of the piperazine ring in addition to the
designated stereochemistry indicated in table.
*

3146
R. G. Gentles et al. / Bioorg. Med. Chem. Lett. 21 (2011) 3142–3147
other vectors off the aromatic group is generally tolerated within
certain size constraints (see 27). However, again the preferred sub-
stituents identified were the p-substituted fluoroalkyl 29 and tri-
fluoromethoxy 28 groups. These observations are not likely
attributable to electronic effects, as 33, an electron rich system,
is equipotent with 28 and 29, both of which are moderately elec-
tron deficient.
Following these initial optimization exercises, we next used
intermediate 5 to explore the SAR at the four-position of the piper-
azine. Several compound arrays were prepared using a variety of
linking chemistries. However, direct heteroarylation of N4 of the
pirperazine was clearly preferred, and from an extensive set of het-
erocycles evaluated, carboxyl- and carboxamide-substituted pyrid-
azines and pyrazines clearly displayed the greatest potency, as
shown in Table 3.
Although the two acids 35 and 36 displayed no significant activ-
ity in the replicon assay, we attributed this to poor cellular pene-
tration, as both compounds displayed excellent activity in the
enzyme assay (data not shown). Correspondingly, we prepared
the esters 37 and 38, and were gratified to see significant activity
with both compounds. More interesting, was the methyl carbox-
amide 39 that displayed more than a 10-fold improvement in po-
tency against both genotypes relative to carboxylate 37.
Unfortunately, the related pyrazine carboxamide 40 did not show
a similar enhancement in activity, being approximately equipotent
to the related ester 38. We attributed this to the relative disposi-
tion of a nitrogen atom (⁄) in the preferred conformation of the
pyridazine and pyrazine carboxamides, as shown in Figure 4. This
will be discussed this in greater detail in a future manuscript.
Compound 39 was subsequently resolved, and the more active
(R)-enantiomer 41 displayed the expected enhancement in activ-
ity, with EC50’s of 7 nM and 220 nM against genotypes 1b and 1a,
respectively. A subsequent homologation exercise focused on the
introduction of small alkyl substituents onto the nitrogen atom
of the secondary carboxamide. The final products from these syn-
theses were fractionated using chiral reverse-phase HPLC, resulting
in the isolation of specific enantiomers 42 and 43, which displayed
similar activities to 41. Derivative 2, prepared by condensing
(S)-butan-2-amine with the appropriate racemic pyridazine
carboxamide, and isolation by preparative HPLC, displayed a mod-
erate enhancement of activity against genotype 1a, and was the
only analog within the series to display an EC50 of <100 nM against
this genotype. Initial attempts to extend this homologation further
resulted in a moderate reduction in activity, as seen with 44. How-
ever, it is apparent that further work around the aryl carboxamide
moiety of the chemotype is clearly merited.
Table 4
All of the compounds discussed above were screened against
NS5B genotype 1b and 1a enzymes, with activities that were gen-
erally consistant with those observed in the replicon system, as
typified by the data presented in Table 4 and Figure 2. In addition,
the compounds were assayed in a replicon system possessing a
Y452H mutation in the NS5B protein that is located within the
palm domain of the enzyme. All of the examples discussed above
displayed significantly less activity in the mutant replicon system,
NS5B enzymatic activity of compounds 1 and 2 against genotypes 1b and 1a and a
mutant Y452H replicon9
Compound
Wild type 1b
Wild type 1a
enzyme IC50
Mutant Y452H
replicon EC50 (lM)
enzyme IC50
(
lM)
(l
M)
1
2
0.3
0.013
12.5
19
0.4
Values are means of three experiments, and standard deviations are <40% of mean.
Figure 2. Screening hit 1 and optimized NS5B inhibitor, (S)-N-sec-butyl-6-((R)-3-(4-(trifluoromethoxy)benzylcarbamoyl)-4-(4-(trifluoromethoxy) phenylsulfonyl)piperazin-
1-yl)pyridazine-3-carboxamide (2).
Figure 3. The intermediates 3 and 5 can all be accessed using the chemical procedures shown in Scheme 1. For intermediate 4, see Scheme 2.

R. G. Gentles et al. / Bioorg. Med. Chem. Lett. 21 (2011) 3142–3147
3147
Figure 4. Preferred conformation of structurally related carboxamide substituted pyrazines and pyridazines.
6. Vo, N. V.; Tuler, J. R.; Lai, M. M. C. Biochemistry 2004, 43, 10579.
7. Beaulieu, P. L. Expert Opin. Ther. Patents 2009, 19, 145.
8. (a) Legrand-Abravanel, F.; Nicot, F.; Izopet, J. Expert Opin. Invest. Drugs 2010, 19,
963; (b) Enomoto, M.; Tamori, A.; Kawada, N. Rev. Recent Clin. Trials 2009, 4,
179.
9. An NS5B enzyme-based screen of the BMS Compound Library was conducted.
Inhibition of a genotype 1b enzyme (Con-1 with a C-terminal deletion of 18
amino acids) was detected as a decrease in 3H-UTP incorporation using a SPA
format (polyA template/biotin-labeled dT primer). Confirmed hits were tested
for activity and cytotoxicity in an HCV replicon assay.14
10. During the course of our work the following patent reported this compound
class to be active against HCV: Abe, H.; Tanaka, M.; Sugimoto, K.; Suma, A.;
Yokota, M.; Shiozaki, M.; Iio, K.; Ueyama, K.; Motoda, D.; Noguchi, T.; Adachi,
T.; Tsuruha, J.; Doi, S. PCTInt. Appl. WO 2007119889; Chem. Abstr. 2007, 120,
4767.
suggesting that these compounds bind in, or proximal to the palm
domain. This is discussed in detail in a related manuscript.15
In conclusion, we have identified a novel class of piperazine
NS5B inhibitors using high throughput screening. Subsequent
exploratory lead optimization led to a rapid improvement in po-
tency against HCV genotypes 1b (>300-fold) and 1a (>200-fold).
Key aspects of the pharmacophore have been clearly identified,
as have potential paths for the further advancement of this series.
Acknowledgments
The authors would like to thank Dr. Carl Bergstrom for his help
in reviewing this manuscript.
11. Breitenbucher, J. G.; Johnson, C. R.; Haight, M.; Phelan, J. C. Tetrahedron Lett.
1998, 1295.
12. Xiang, J.; Wan, Z. K.; Li, H. Q.; Ipek, M.; Binnun, E.; Nunez, J.; Chen, L.; McKew, J.
C.; Mansour, T. S.; Xu, X.; Suri, V.; Tam, M.; Xing, Y.; Li, X.; Hahm, S.; Tobin, J.;
Saiah, E. J. Med. Chem. 2008, 51, 4068.
13. Neya, M.; Yamazaki, H.; Kayakiri, N.; Sato, K.; Oku, T. PCT Int. Appl. WO
9827069, 1998.
14. Hudyma, T. W.; Zheng, X.; He, F.; Ding, M.; Bergstrom, C. P.; Hewawasam, P.;
Martin, S. W.; Gentles, R. G. PCT Int. Appl. WO 2006020082, 2006.
15. Gentles, R. G.; Sheriff, S.; Beno, B. R.; Wan, C.; Kish, K.; Ding, M.; Zheng, X.;
Chupak, L.; Poss, M. A.; Witmer, M. R.; Morin, M.; Wang, Y. K.; Rigat, R.; Lemm,
J.; Voss, S.; Liu, M.; Pelosi, P.; Roberts, S. B.; Gao, M.; Kadow, J. F. Bioorg. Med.
Chem. Lett., in press.
References and notes
1. (a) Shepard, C. W.; Finelli, L.; Alter, M. J. Lancet Infect. Dis. 2005, 5, 558; (b)
http://www.who.int/accine_research/diseases/viral_cancers/en/index2.html.
2. (a) Lauer, G. M.; Martin, K. A. N. Eng. J. Med. 2001, 345, 41; (b) Shackel, N. A.;
Patel, K.; McHutchison, J. Genomic Pers. Med. 2009, 2, 1375.
3. Zeuzem, S. Nat. Clin. Pract. Gastroenterol. Hepatol. 2008, 5, 610.
4. Field, J. J.; Hoofnagle, J. H. Nature 2005, 436, 967.
5. Li, H.; Shi, S. T. Future Med. Chem. 2010, 2, 121.