Further studies on hepatitis C virus NS5B RNA-dependent RNA polymerase inhibitors toward improved replicon cell activities: Benzimidazole and structurally related compounds bearing the 2-morpholinophenyl moiety

Article abstract of DOI:10.1016/j.bmcl.2007.03.027

Following the discovery of JTK-109 (1) as a potent inhibitor of hepatitis C virus NS5B RNA-dependent RNA polymerase, [(a) Hirashima, S.; Suzuki, T.; Ishida, T.; Noji, S.; Yata, S.; Ando, I.; Komatsu, M.; Ikeda, S.; Hashimoto, H. J. Med. Chem. 2006, 49, 4721. (b) Hashimoto, H.; Mizutani, K.; Yoshida, A. Int. Patent Appl. WO 01/47883, 2001.] further studies toward the improvement of the cellular potency have been performed. A greater than 40-fold improvement was achieved through replacing the biphenyl moiety with a 2-morpholinophenyl group and the benzimidazole ring with the tetracyclic scaffold to afford compound 7 with an excellent replicon potency (EC₅₀ = 7.6 nM).

Full text of DOI:10.1016/j.bmcl.2007.03.027

Bioorganic & Medicinal Chemistry Letters 17 (2007) 3181–3186
Further studies on hepatitis C virus NS5B RNA-dependent
RNA polymerase inhibitors toward improved replicon cell
activities: Benzimidazole and structurally related
compounds bearing the 2-morpholinophenyl moietyq
Shintaro Hirashima,* Takahiro Oka, Kazutaka Ikegashira, Satoru Noji,
Hiroshi Yamanaka, Yoshinori Hara, Hiroyuki Goto, Ryo Mizojiri, Yasushi Niwa,
Toru Noguchi, Izuru Ando, Satoru Ikeda and Hiromasa Hashimoto*
Central Pharmaceutical Research Institute, Japan Tobacco Inc., 1-1 Murasaki-cho, Takatsuki, Osaka 569-1125, Japan
Received 5 February 2007; revised 6 March 2007; accepted 8 March 2007
Available online 14 March 2007
Abstract—Following the discovery of JTK-109 (1) as a potent inhibitor of hepatitis C virus NS5B RNA-dependent RNA polymer-
ase, [(a) Hirashima, S.; Suzuki, T.; Ishida, T.; Noji, S.; Yata, S.; Ando, I.; Komatsu, M.; Ikeda, S.; Hashimoto, H. J. Med. Chem.
2006, 49, 4721. (b) Hashimoto, H.; Mizutani, K.; Yoshida, A. Int. Patent Appl. WO 01/47883, 2001.] further studies toward the
improvement of the cellular potency have been performed. A greater than 40-fold improvement was achieved through replacing
the biphenyl moiety with a 2-morpholinophenyl group and the benzimidazole ring with the tetracyclic scaffold to afford compound
7 with an excellent replicon potency (EC50 = 7.6 nM).
Ó 2007 Elsevier Ltd. All rights reserved.
Hepatitis C virus (HCV) is a common pathogen that in-
fects an estimated 170 million people worldwide and
that can lead to liver cirrhosis, carcinoma, and hepatic
insufficiency.2 The standard therapy is based on pegylat-
ed interferon-a and ribavirin, a treatment course that is
not necessarily effective for the patients infected with the
most prevalent HCV genotype 1 virus and that is
frequently associated with severe adverse events.3 There-
fore, there is an unmet clinical need for novel anti-HCV
agents, especially against genotype 1 HCV. Among the
encoded HCV non-structural proteins, HCV NS5B
RNA-dependent RNA polymerase (RdRp) is an attrac-
tive target for drug development because it plays a vital
role in HCV viral replication,4 and because correspond-
ing RdRps are not known in mammalian systems. Many
nucleoside and non-nucleoside inhibitors against NS5B
have been described.5
We have recently reported a potent NS5B inhibitor,
JTK-109 (1, Fig. 1), with a low nanomolar biochemical
potency and a submicromolar cellular potency.1a How-
ever, a relatively large ‘shift in potency between bio-
chemical assay and cellular assay’ (EC50/IC50 = 19,
abbreviated as shift in this paper) was observed.
Although cellular activity is generally influenced by fac-
tors such as membrane permeability, metabolism, and
affinity for proteins, the elevated binding of 1 to protein
Cl
A
F
HOOC
N
N
O
B
O
N
C
JTK-109 (1)
Keywords: HCV; NS5B polymerase; Inhibitor.
q
IC50(NS5B:1b) = 0.017 μM
EC50(replicon) = 0.32 μM
EC50/IC50 = 19
Presented in part at the 232nd National Meeting of the American
Chemical Society, San Francisco, CA, September 2006; MEDI-241.
*
Corresponding authors. E-mail addresses: shintaro.hirashima@ims.
jti.co.jp; hiromasa.hashimoto@ims.jti.co.jp
Figure 1. Structure of JTK-109 (1).
0960-894X/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2007.03.027

3182
S. Hirashima et al. / Bioorg. Med. Chem. Lett. 17 (2007) 3181–3186
(>99% in 10% FCS employed in the replicon cell assay)
was considered as the major factor for the large shift.
The hydrophobic biphenyl substructure is known to
promote the increased binding affinity to a wide range
of proteins, especially albumin.6 Indeed, almost all of
the compounds in the previous benzimidazole series
bearing substituted biphenyl groups showed greater
than 20-fold shifts. Thus, replacement of the biphenyl
moiety with a non-biphenyl substructure is clearly neces-
sary to reduce the shift, which may lead to more potent
inhibition in the replicon cells. Another aspect of
increasing the cellular potency is the improvement of
the biochemical potency, which could be achieved by
changing the benzimidazole ring to the other structures.
Herein, we report our results toward highly potent com-
pounds in the replicon cells through the above strategies.
This work led to the identification of compound 7 with a
single digit nanomolar potency (EC50 = 7.6 nM).
In our search for a non-biphenyl substructure, we ini-
tially focused on the A-part of 1 (Table 1).7 Compounds
2a, b were synthesized to deform the shape of the biphe-
nyl functionality. Insertion of the ether linker between
the two phenyl rings (2a) and replacement of the aro-
matic ring with a methoxy group (2b) decreased potency
10-fold against NS5B compared to 1. The biphenyl ether
2a exhibited no improvement in the shift, however, the
shift of 2b was enormously reduced (EC50/IC50 = 3.4).
We therefore speculated that the aromatic property of
the A-part has a deleterious effect on the shift, while a
ring structure directly connected to the B-ring is neces-
sary to maintain the biochemical potency, and hence
we implemented replacement of the phenyl ring with
an aliphatic ring. Substitution with piperidine (2c)
yielded a compound with a modest improvement in
the biochemical potency over 2a or 2b, but still worse
than 1. However, the shift (EC50/IC50 = 5.3) was much
improved. Introduction of either a hydroxyl (2d) or car-
boxylic acid (2e) on the piperidine, or replacement with
piperazines (2f, g), gave compounds with equal bio-
chemical potency, but with larger shifts than 2c. On
the other hand, the morpholine variant (2h) exhibited
Table 1. NS5B enzyme assay IC50 values and replicon cell-based assay
EC50 values for compounds 2 (A-part variation)
F
R1
HOOC
N
N
O
O
N
Compound R1
NS5B
a
Replicon
a
EC50/IC50
IC50 (lM) EC50 (lM)
1
0.017
0.23
0.41
0.15
0.15
0.18
0.13
0.11
0.066
1.0
0.32
3.1
19
13
Cl
Table 2. NS5B enzyme assay IC50 values and replicon cell-based assay
EC50 values for compounds 3 (C-part variation)
2a
2b
2c
2d
2e
2f
O
Cl
O
F
HOOC
N
N
N
O
OMe
1.4
3.4
5.3
9.3
19
R2
0.80
1.4
N
N
N
N
N
N
Compound R2
NS5B
Replicon
EC50/IC50
IC50 (lM) EC50 (lM)
OH
O
2h
0.066
0.31
4.7
N
3.5
COOH
3a
3b
H
Cl
0.13
0.10
1.0
1.2
7.7
12
1.1
8.5
8.6
4.7
>30
5.0
N Me
O
S
O
3c
0.071
0.55
7.7
N
2g
2h
2i
0.95
0.31
N
Ac
O
N
3d
0.090
0.58
6.4
O
O
3e
3f
NMeAc
NHAc
O
0.051
0.070
0.37
0.91
7.3
13
N
O
>30
3g
0.099
1.2
12
N
OMe
Me
2j
0.26
1.3
N
O
3h
3i
3j
Ni-PrAc
CONMe2
CONHi-Pr
0.042
0.068
0.048
0.27
0.44
0.38
6.4
6.5
7.9
a IC50 and EC50 values in all tables are means of at least two inde-
pendent determinations, standard deviation 30%.

S. Hirashima et al. / Bioorg. Med. Chem. Lett. 17 (2007) 3181–3186
3183
ꢀthreefold improved biochemical potency compared to
2c with a similar shift (EC50/IC50 = 4.7). This compound
was as potent as 1 in the cellular assay, although the bio-
chemical potency was ꢀfourfold less than 1. The protein
binding of 2h was evaluated and found to be 90%, sug-
gesting that the decreased shift may be attributed to the
decreased affinity for proteins. Two structurally related
compounds, 2i and 2j, were synthesized to confirm the
appropriateness of the morpholine group. Lactam 2i
caused a negative effect on the biochemical potency
and the shift. The non-ring compound 2j afforded a
fourfold decrease in the biochemical potency compared
to 2h, although the shift was comparable.
and were slightly more potent than 3a (no substituent)
or 3b (non-carbonyl substituent). Compared to the pre-
vious series, the effects of the carbonyl functional groups
were modest. While several compounds (3e, h–j) were as
potent as 2h in the replicon assay, none of them showed
increased cellular potencies.
Although the reduction of the shift was achieved by
replacing the phenyl ring (A-part) with a morpholine
ring, we sacrificed a fourfold loss in the biochemical po-
tency. The initial purpose, improvement of the cellular
potency, remained. Thus, we attempted to enhance the
biochemical potency by changing the benzimidazole ring
to the other structures. In several related inhibitors, sub-
stitution of indole,8 thienopyrrole,9 or tetracyclic scaf-
folds10 for the benzimidazole moiety afforded more
potent inhibition in the biochemical assay. We incorpo-
rated these same structures into compound 2h and syn-
thesized compounds 4–7. All of them exhibited
improved biochemical and cellular potencies compared
to 2h while retaining smaller shifts (Table 3). The rank
order of the biochemical potencies generally correlated
Employing the morpholin-4-yl group as the A-part of
the structure, we turned our attention to the C-part of
the parent compound. Since our previous structure–
activity relationship studies showed that carbonyl/sulfo-
nyl functional groups at the C-part afforded superior
biochemical and the cellular potencies,1a those substitu-
ents (3c–j) were mainly examined in this series (Table 2).
They showed NS5B inhibitory activities similar to 2h
Table 3. NS5B enzyme assay IC50 values and replicon cell-based assay EC50 values for compounds 4–8
Compound
NS5B IC50(lM)
Replicon EC50(lM)
EC50/IC50
O
HOOC
HOOC
N
N
O
4
0.030
0.21
7.0
N
N
O
O
N
N
O
O
5
0.0072
0.052
7.2
O
N
N
HOOC
HOOC
O
6
0.010
0.052
5.2
S
O
N
O
O
N
N
O
7
0.0072
0.0076
1.1
O
N
Cl
O
HOOC
N
O
8
0.019
0.060
3.1
O
N

3184
S. Hirashima et al. / Bioorg. Med. Chem. Lett. 17 (2007) 3181–3186
with the order for the scaffolds reported in the litera-
ture.8c,9a,10 The cellular potency increased in parallel
with the increased biochemical potency. Notably, the
tetracyclic compound 7 showed a significantly improved
cellular potency (EC50 = 7.6 nM) with no shift (EC50/
IC50 = 1.1). Thus, a 42-fold of improvement in the rep-
licon cellular potency was achieved. To verify the effect
of the morpholine group, the biphenyl compound 8 was
prepared. A threefold larger shift compared with 7 was
observed, as was the case in benzimidazole (1 vs 2h).
In contrast to the case of benzimidazole, the morpholine
group did not lose the biochemical potency. The tetracy-
clic ring also seems to be significantly effective in reduc-
ing the shift, judging by the fact that the shifts of 7 and 8
were five- to sixfold smaller than those of 2h and 1,
respectively. All compounds in this study showed low
cytotoxicity (CC50 > 20 lM).
duced with iron to give the aniline 11. Acylation with
4-chlorobutyryl chloride and the subsequent intramolec-
ular alkylation afforded the lactams 12a, c–i. Compounds
12b, j were directly prepared from 9 by Mitsunobu cou-
pling with 17b, j. The benzylalcohols 17 were derived from
the esters 16 in four steps: reduction of the nitro group
with iron; LiAlH4 reduction of the methyl ester; acylation
of the aniline with 4-chlorobutyryl chloride; and cycliza-
tion to form the lactam ring. Compounds 16 were synthe-
sized from 13, 14, or 15 by either amination (16h, j),
alkylation (16b), or Suzuki coupling (16k). Finally,
hydrolysis of the methyl esters of 12 gave 2.
As shown in Scheme 2, 3a–j were synthesized from either
aniline 11h or the phenol 9 via 18. The aniline 11h was
converted to 18b–h in five different ways: (i) for 18b,
replacement of the aniline to the chloride by Sandmeyer
reaction; (ii) for 18c, d, acylation/sulfonylation and sub-
sequent cyclization; (iii) for 18e, g, acylation with the
corresponding acylchlorides and subsequent N-alkyl-
ation with MeI; (iv) for 18f, acetylation; (v) for 18h,
reductive amination with acetone in the presence of
NaBH(OAc)3 and subsequent acetylation. Compound
18a was directly prepared from the phenol 9 by Mitsu-
nobu reaction. Compounds 18i, j were derived from 9
via the carboxylic acid 19, which was synthesized in
three-step sequence: alkylation of the phenol; Pd-medi-
The compounds evaluated in this study were synthesized
as described in Schemes 1–3 . The syntheses of 2a–j were
started from the phenol 9,1b which was converted to 10
in two different ways: (i) for 10a, c–h, alkylation by Mitsu-
nobu reaction with 2-fluoro-5-nitrobenzylalcohol, and
subsequent amination with the corresponding amines or
alkoxylation with 4-chlorophenol; (ii) for 10i, alkylation
with 2-bromo-5-nitrobenzylbromide and subsequent
Pd-mediated amination. The nitro group of 10 was re-
F
F
a, b
R1
MeOOC
MeOOC
N
N
N
N
(for 10a,c-h)
e
OH
O
c, d
(for 10i)
NO2
10a
9
,c-i
i (for 12b,j)
F
R1
R1
HOOC
N
N
R1
O
O
O
f, g
h
O
N
O
N
NH2
11a,c-i
12a-j
2a-j
R1
b (from 13 for 16h,j)
j (from 14 for 16b)
m (from 15 for 16k)
X
R1
OH
COOY
COOMe
e, k, f, l
N
O
NO2
NO2
13: X = F, Y = Me
14: X = OH, Y = H
15: X = Cl, Y = Me
16b,h,j,k
17b,h,j,k
a: R1 = 4-Cl-phenoxy
f: R1 = 4-Me-piperazin-1-yl
g: R1 = 4-Ac-piperazin-1-yl
h: R1 = morpholin-4-yl
b: R1 = OMe
c: R1 = piperidin-1-yl
d: R1 = 4-OH-piperidin-1-yl
i: R1 = 3-oxomorpholin-4-yl
j: R1 = NMe(CH2CH2OMe)
e: R1 = 4-COOH-piperidin-1-yl
k: R1 = 4-Cl-phenyl
Scheme 1. Reagents and conditions: (a) 2-fluoro-5-nitrobenzylalcohol, DIAD, PPh3, THF, rt; (b) amines or 4-chlorophenol, K2CO3, DMSO, 80 °C;
(c) 2-bromo-5-nitrobenzylbromide, K2CO3, DMF, 80 °C; (d) morpholin-3-one, Pd2(dba)3, Xantphos, Cs2CO3, 1,4-dioxane, reflux; (e) Fe, NH4Cl,
THF–EtOH–H2O, reflux; (f) 4-chlorobutyryl chloride, AcONa, AcOH, THF, rt; (g) K2CO3, DMF, 65 °C; (h) 4 N aq. NaOH, THF–MeOH–H2O,
60 °C; (i) 17, DIAD, PPh3, THF, rt; (j) MeI, K2CO3, acetone, reflux; (k) LiAlH4, THF, 0 °C; (l) KOH, EtOH–H2O, 80 °C; (m) 4-Cl-PhB(OH)2,
Pd(PPh3)4, Na2CO3, DME–H2O, reflux.

S. Hirashima et al. / Bioorg. Med. Chem. Lett. 17 (2007) 3181–3186
3185
a (for 18b)
b, c (for 18c,d)
d, e (for 18e,g)
O
O
F
F
MeOOC
MeOOC
N
N
N
N
N
N
O
O
d (for 18f)
f, d (for 18h)
NH2
R2
11h
18a-j
h (for 18a)
g
l (for 18i,j)
O
O
F
F
MeOOC
HOOC
N
N
N
N
N
O
O
i, j, k
N
9
COOH
R2
19
3a-j
a: R2 = H
b: R2 = Cl
f: R2 = NHAc
g: R2 = NMe(COCH2OMe)
-PrAc
i: R2 = CONMe2
j: R2 = CONH
-Pr
c: R2 = 1,1-dioxoisothiazolidin-2-yl h: R2 = N
i
d: R2 = 2-oxopiperidin-1-yl
e: R2 = NMeAc
i
Scheme 2. Reagents and conditions: (a) tert-BuONO, CuCl2, MeCN, 0 °C then rt; (b) Cl(CH2)4COCl or Cl(CH2)3SO2Cl, AcONa, AcOH, THF, rt;
(c) K2CO3, DMF, 60 °C; (d) AcCl or MeOCH2COCl, Et3N, CHCl3, 0 °C then rt; (e) NaH, MeI, DMF, 0 °C then rt; (f) acetone, NaBH(OAc)3, THF,
rt; (g) 4 N aq. NaOH, THF–MeOH–H2O, 60 °C; (h) 2-morpholin-4-ylbenzylalcohol, DIAD, PPh3, THF, rt; (i) 2-bromo-5-tert-butoxy-
carbonylbenzylbromide, K2CO3, 60 °C; (j) morpholine, Pd2(dba)3, BINAP, Cs2CO3, toluene, 100 °C; (k) TFA, CHCl3, rt; (l) amines, PyBopÒ,
Et3N, DMF, rt.
MeOOC
MeOOC
N
OH
OH
OH
OH
N
N
20
21
O
MeOOC
(from 20 and 24h)
(from 21 and 24h)
(from 22 and 24h)
(from 23 and 24h)
(from 23 and 24k)
N
4
5
6
7
8
EtOOC
b, c
S
22
23
R1
R1
N
a
HO
Cl
N
O
O
17h: R1 = morpholin-4-yl
17k: R1 = 4-Cl-phenyl
24h: R1 = morpholin-4-yl
24k: R1 = 4-Cl-phenyl
Scheme 3. Reagents and conditions: (a) SOCl2, Et3N, CHCl3, 0 °C; (b) phenol (20, 21, 22, or 23), K2CO3, DMF, 60 °C; (c) 4 N aq. NaOH, THF–
MeOH–H2O, 60 °C.
ated amination; and deprotection of the tert-butyl
group. Coupling between 19 and the corresponding
amines using PyBopÒ provided 18i, j. The esters 18 were
hydrolyzed to give 3.
(24h, k); coupling with the phenols (20,8c 21,8d 22,9b and
2310); and subsequent hydrolysis of the methyl esters.
In conclusion, we have achieved the improvement of the
replicon cellular potency (42-fold in comparison with 1)
through reducing the shift between the biochemical and
the cellular potencies and improving the biochemical po-
tency. Reduction of the shift was attained by replacing
As shown in Scheme 3, compounds 4–7 and 8 were synthe-
sized from benzylalcohols 17h or 17k, respectively, in
three steps: conversion of the hydroxyl to the chloride

3186
S. Hirashima et al. / Bioorg. Med. Chem. Lett. 17 (2007) 3181–3186
3. Fried, M. W.; Shiffman, M. L.; Reddy, K. R.; Smith, C.;
Marinos, G.; Gonc¸ales, F. L., Jr.; Ha¨ussinger, D.; Diago,
M.; Carosi, G.; Dhumeaux, D.; Craxi, A.; Lin, A.;
Hoffman, J.; Yu, J. N. Engl. J. Med. 2002, 347, 975.
4. Kolykhalov, A. A.; Mihalik, K.; Feinstone, S.; Rice, C. M.
J. Virol. 2000, 74, 2046.
the biphenyl moiety with a 2-morpholinophenyl ring
and the benzimidazole ring with the tetracyclic scaffold.
A further increase in biochemical potency occurred by
substituting the tetracyclic scaffold for the benzimid-
azole ring.
5. Reviews Carroll, S. S.; Olsen, D. B. Infectious Disorders—
Drug Targets 2006, 6, 17; Koch, U.; Narjes, F. Infectious
Disorders—Drug Targets 2006, 6, 31; Beaulieu, P. L. Curr.
Opin. Drug Discov. Dev. 2006, 9, 618.
Acknowledgments
We are grateful to Mr. Mitsumasa Takahashi and Mr.
Eita Nagao for analytical support, Mr. Naoki Ogura
and Ms. Yukiyo Toyonaga for the biological assays,
Mr. Yukihiro Nomura for protein binding study, and fi-
nally Dr. Jun-ichi Haruta for support.
6. Hajduk, P. J.; Bures, M.; Praestgaard, J.; Fesik, S. W.
J. Med. Chem. 2000, 43, 3443.
7. Our assay employed a C-terminally truncated genotype 1b
BK NS5B544 enzyme and Huh-5-2 replicon cells as
previously reported. See Ref. 1a.
8. (a) Ref. 1b; (b) Beaulieu, P. L.; Fazal, G.; Kukolj, G.;
Jolicoeur, E.; Gillard, J.; Poupart, M. A.; Rancourt, J. Int.
Patent Appl. WO 03/010140, 2003; (c) Harper, H.; Pacini,
B.; Avolino, S.; Di Filippo, M.; Migliaccio, G.; Laufer, R.;
De Francesco, R.; Roweley, M.; Narjes, F. J. Med. Chem.
2005, 48, 1314; (d) Oka, T.; Yata, S.; Ikegashira, K.; Noji,
S.; Akaki, T.; Hirashima, S.; Niwa, Y.; Ando, I.; Sato, T.
Int. Patent Appl. WO 05/014543, 2005.
Supplementary data
Supplementary data associated with this article can be
found, in the online version, at doi:10.1016/
j.bmcl.2007.03.027.
9. (a) Ontoria, J. M.; Martin, H. J. I.; Malancona, S.;
Attenni, B.; Stansfield, I.; Conte, I.; Ercolani, C.; Haber-
mann, J.; Ponzi, S.; Di Filippo, M.; Koch, U.; Rowley,
M.; Narjes, F. Bioorg. Med. Chem. Lett. 2006, 16, 4026;
(b) Mizojiri, R.; Hirashima, S.; Oka, T.; Aoki, K.; Noji, S.;
Ando, I.; Sato, T.; Niwa, Y. Int. Patent Appl. WO 05/
049622, 2005.
10. Ikegashira, K.; Oka, T.; Hirashima, S.; Noji, S.; Yama-
naka, H.; Hara, Y.; Adachi, T.; Tsuruha, J.; Doi, S.; Hase,
Y.; Noguchi, T.; Ando, I.; Ogura, N.; Ikeda, S.; Hashim-
oto, H. J. Med. Chem. 2006, 49, 6950.
References and notes
1. (a) Hirashima, S.; Suzuki, T.; Ishida, T.; Noji, S.; Yata, S.;
Ando, I.; Komatsu, M.; Ikeda, S.; Hashimoto, H. J. Med.
Chem. 2006, 49, 4721; (b) Hashimoto, H.; Mizutani, K.;
Yoshida, A. Int. Patent Appl. WO 01/47883, 2001.
2. WHO. Global surveillance and control of hepatitis C. J.
Viral Hepatitis 1999, 6, 35.