Structure-based optimization and derivatization of 2-substituted quinolone-based non-nucleoside HCV NS5B inhibitors with submicromolar cellular replicon potency

Article abstract of DOI:10.1016/j.bmcl.2016.04.042

HCV NS5B polymerase is an attractive and validated target for anti-HCV therapy. Starting from our previously identified 2-aryl quinolones as novel non-nucleoside NS5B polymerase inhibitors, structure-based optimization furnished 2-alkyl-N-benzyl quinolones with improved antiviral potency by employing privileged fragment hybridization strategy. The N-(4-chlorobenzyl)-2-(methoxymethyl)quinolone derivative 5f proved to be the best compound of this series, exhibiting a selective sub-micromolar antiviral effect (EC₅₀ = 0.4 μM, SI = 10.8) in Huh7.5.1 cells carrying a HCV genotype 2a. Considering the undesirable pharmacokinetic property of the highly substituted quinolones, a novel chemotype of 1,6-naphthyridine-4,5-diones were evolved via scaffold hopping, affording brand new structure HCV inhibitors with compound 6h (EC₅₀ (gt2a) = 2.5 μM, SI = 7.2) as a promising hit. Molecular modeling studies suggest that both of 2-alkyl quinolones and 1,6-naphthyridine-4,5-diones function as HCV NS5B thumb pocket II inhibitors.

Full text of DOI:10.1016/j.bmcl.2016.04.042

Bioorganic & Medicinal Chemistry Letters xxx (2016) xxx–xxx
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Structure-based optimization and derivatization of 2-substituted
quinolone-based non-nucleoside HCV NS5B inhibitors with
submicromolar cellular replicon potency
a,
Yu Cheng a,y, Jian Shen b,y, Run-Ze Peng a, Gui-Feng Wang b, Jian-Ping Zuo b, , Ya-Qiu Long
⇑
⇑
a CAS Key Laboratory of Receptor Research, Shanghai Institute of Materia Medica, Chinese Academy of Sciences, Shanghai 201203, China
b State Key Laboratory of Drug Research, Shanghai Institute of Materia Medica, Chinese Academy of Sciences, Shanghai 201203, China
a r t i c l e i n f o
a b s t r a c t
Article history:
HCV NS5B polymerase is an attractive and validated target for anti-HCV therapy. Starting from our
previously identified 2-aryl quinolones as novel non-nucleoside NS5B polymerase inhibitors, structure-
based optimization furnished 2-alkyl-N-benzyl quinolones with improved antiviral potency by employ-
ing privileged fragment hybridization strategy. The N-(4-chlorobenzyl)-2-(methoxymethyl)quinolone
derivative 5f proved to be the best compound of this series, exhibiting a selective sub-micromolar
Received 29 February 2016
Revised 14 April 2016
Accepted 16 April 2016
Available online xxxx
antiviral effect (EC50 = 0.4 lM, SI = 10.8) in Huh7.5.1 cells carrying a HCV genotype 2a. Considering the
undesirable pharmacokinetic property of the highly substituted quinolones, a novel chemotype of
1,6-naphthyridine-4,5-diones were evolved via scaffold hopping, affording brand new structure HCV
inhibitors with compound 6h (EC50 (gt2a) = 2.5 lM, SI = 7.2) as a promising hit. Molecular modeling
studies suggest that both of 2-alkyl quinolones and 1,6-naphthyridine-4,5-diones function as HCV
NS5B thumb pocket II inhibitors.
Keywords:
HCV NS5B polymerase
2-Alkyl quinolone
1,6-Naphthyridine-4,5-dione
Non-nucleoside inhibitor
Direct acting antiviral
Allosteric site
Ó 2016 Elsevier Ltd. All rights reserved.
As a leading cause for liver fibrosis, cirrhosis, and hepatocellular
carcinoma,1 chronic hepatitis C virus (HCV) infection has been
estimated to affect about 160 million people worldwide and causes
approximately 500,000 deaths annually from HCV-related liver
diseases.2,3 For a long span of time, the standard of care (SOC) for
HCV treatment has been based on a combination of pegylated
interaction of these agents with the highly mutational virus leads
to the emergence of drug resistance.
HCV non-structural protein 5B (NS5B), a RNA dependent RNA
polymerase, plays a critical role in HCV genome replication.16
The absence of a counterpart of this enzyme in human body makes
it an attractive therapeutic target. NS5B inhibitors are classified
into nucleoside inhibitors (NIs) that bind to the active site and
non-nucleoside inhibitors (NNIs) that bind to the allosteric sites.
The right-hand-like structure of NS5B polymerase can be morpho-
logically divided into ‘thumb’, ‘palm’, and ‘finger’ domains.17 Until
recently, five allosteric sites have been identified for NS5B poly-
merase and two of them reside in the thumb domain (thumb site
I, TSI and thumb site II, TSII).18 A broad range of TSII inhibitors
emerged in the last few years,10 but none of them has been
approved for clinical use. Therefore, this underexploited area offers
great opportunities for developing new antivirals with distinct
mechanism to treat drug-resistant HCV infection.
Our research group has been working on the drug-like privi-
leged scaffolds to generate novel chemotherapeutic agents, espe-
cially interested in the construction and biological application of
quinolone scaffold.19–22 The quinolones have been reported to be
the NS5B TSII inhibitors (Fig. 1, compounds 1–3),23–25 as confirmed
by the cocrystal structure of ligand 2 bound to NS5B polymerase
(PDB code 3PHE).23 Based on the complex structure, for the first
interferon
a (pegIFN-a) and ribavirin (RBV), which cured approxi-
mately 50% of treated patients after weekly injections for
12 months, but sometimes caused severe adverse reactions with
low sustained viral responses (SVR) against HCV genotype 1
strains.4,5 Direct-acting antiviral agents (DAAs), an interferon-free
treatment selectively targeting essential proteins in HCV life cycle,6
were first validated as an effective approach for HCV therapy by
the market approval of NS3/4A protease inhibitors (telaprevir7
and boceprevir8) in 2011. The past five years have witnessed a vast
number of DAAs including NS4B inhibitors,9 NS5B inhibitor10
(sofosbuvir11 and dasabuvir12) and NS5A inhibitors (ledipasvir13
and ombitasvir14). The multiple combination of theses DAAs have
achieved high SVR in clinical trials.15 However, intense drug
discovery efforts are still urgently needed because consecutive
⇑
Corresponding authors. Tel./fax: +86 21 50806876.
E-mail addresses: jpzuo@simm.ac.cn (J.-P. Zuo), yqlong@simm.ac.cn (Y.-Q. Long).
These authors contributed to this work equally.
y
http://dx.doi.org/10.1016/j.bmcl.2016.04.042
0960-894X/Ó 2016 Elsevier Ltd. All rights reserved.
Please cite this article in press as: Cheng, Y.; et al. Bioorg. Med. Chem. Lett. (2016), http://dx.doi.org/10.1016/j.bmcl.2016.04.042

2
Y. Cheng et al. / Bioorg. Med. Chem. Lett. xxx (2016) xxx–xxx
O
N
O
replicon. Then, further structural optimization was focused on
the improvement of the drug-likeness of the lead compound
through scaffold reshuffling. According to the binding mode of
the 2-alkyl quinolone series with the NS5B polymerase, the
solvent-exposing C-6/C-7 portion was truncated and the C-2
substituent was constrained with the C-3 moiety to deliver a brand
new 1,6-naphthyridine-4,5(1H,6H)-dione scaffold, maintaining
similar pharmacophores to that of quinolone 4. Preliminary SAR
study was conducted to determine an optimal substitution to
achieve the key interactions with the two hydrophobic pockets.
O
N
O
F
O
F
O
N
Cl
N
Cl
N
N
S
O
1
2
Cl
O
HCV EC50(gt1a) = 0.41 μM
EC50(gt1b) = 0.12 μM
HCV IC50(gt1b) = 0.016 μM
O
O
H2N
N
O
N
O
O
F
O
Cl
N
N
N
N
Cl
N
F
F
Cl
3
4
HCV EC50 (gt2a)= 2.02 μM
HCV IC50(gt1b)= 0.069 μM
EC50(gt1b)= 2.2 μM
Figure 1. Representative quinolone-based NS5B polymerase inhibitors.
time we introduced a hydrophobic group at C-2 position on the
quinolone core and identified a new series of 2-aryl-quinolones
to inhibit HCV replication in subgenomic replicon system at
micromolar concentrations, with derivative 4 showing the best bio-
(PDB Code˖3PHE)23
logical profile (EC50 (replicon, gt2a) = 2.02 lM; CC50 = 13.2 l
M).22
The synthesis of 2-alkyl-N-arylquinolone derivatives 5a–f was
described in Scheme 1. Among the quinolone series, compound 1
was prepared as a reference compound according to a reported
procedure.23 The synthetic routes of the target compounds varied
according to the variation on C-7 substituent (5a–c and 5d–f),
which were reported to point to the solvent area and exert remark-
able influence on cellular replicon potency. Starting from commer-
cially available methyl 2-amino-4,5-difluorobenzoate, reductive
amination with p-chlorobenzaldehyde followed by a one-pot
hydrolysis and esterification gave the activated ester 8. Then,
cyclization of 8 with various b-oxo esters formed the 2-alkyl quino-
lone scaffolds 9a–c. Hydrolysis of 9a–c under different conditions
gave quinolone acids with various substitution at C-7, which was
treated with p-cholorobenzyl chloride to furnish 5a–c and 5d–f.
Notably, the C-7 amino substituent in target compounds 5d–f were
introduced through SNAr reaction from the corresponding precur-
sors 10a–b.
Herein we report our further optimization effort to improve the
anti-HCV potency of the 2-arylquinolone series based on the
binding mode of the quinolone NS5B inhibitor, producing
N-benzyl-2-(methoxymethyl)quinolone derivatives with sub-
micromolar anti-HCV potency in genotype 2a replicon and novel
chemotype HCV inhibitors bearing 1,6-naphthyridine-4,5(1H,6H)-
dione scaffold.
Careful examination of the co-crystal structure of quinolone 2
bound to the polymerase TSII site (PDB code: 3PHE)23 and the
structural superimposition between 2 and 4 suggested a prefer-
ence of a bulky aromatic hydrophobic group at N-1 position and
an additional small hydrophobic group at C-2 position to fully
occupy the deep hydrophobic pocket lined by Leu419, Val485,
Ala486, Leu489, Leu497 and Met 423, and gain additional interac-
tion with the surface cleft delineated by Leu419 and Leu497.
Furthermore, we assume that a proper functional group in C-2 sub-
stituent might form a H-bond interaction with an adjacent basic
amino acid residue (Arg501 of gt1b, Lys501 of gt2a) in the
hydrophobic pocket of the protein,26 thus contributing to an
enhanced potency. The resulting 2-alkyl-N-arylquinolones indeed
exhibited an improved anti-HCV potency in genotype (gt) 2a
Scheme 2 describes the preparation of the 1,6-naphthyridine-
4,5(1H,6H)-diones with hydrophobic substituents at the N-1 and
N-6 positions. To the best of our knowledge, this is the first time
to report the design and synthesis of 1,6-naphthyridine-4,5
(1H,6H)-diones and evaluate their biological activities. As depicted
O
O
N
O
O
N
O
O
O
O
F
F
R1
F
F
F
N
F
F
F
F
e,f
d
O
a
b,c
O
O
OMe
NH2
OMe
O
R2
O
R2
Cl
NH
NH
5a-c
9a-c
2
7
8
a
: R = Me
1
2
a
b
Cl
: R =Et, R = Me
Cl
Cl
Cl
b : R2= Et
c : R2= CH2OCH3
1
2
: R =Et, R = Et
c : R1=Me,R2= CH2OCH3
g
,
f
O
N
O
F
O
O
O
F
N
R2
Cl
h
O
N
R3
F
N
R2
Cl
Cl
5d-f
10a-b
2
3
2
d
e
: R =Me, R = Me
a
b
: R = Me
Cl
2
3
2
: R =CH2OCH3, R =Benzyl
: R = CH2OCH3
f : R2=CH2OCH3,R3= Me
Scheme 1. Reagents and conditions: (a) p-chlorobenzyaldehyde (1.25 equiv), NaBH(OAc)3 (2.3 equiv), HOAc, RT, 94%; (b) LiOH, H2O–THF (1:3, v/v), 45 °C, overnight; (c) DCC
(2.0 equiv), HOSu (2.0 equiv), DCM, 95% overall yield for two steps; (d) R2COCH2CO2R1 (2.2 equiv), NaH (2.2 equiv), toluene, 130 °C, 49–74%; (e) NaOH, MeOH–THF–H2O
(3:2:3, v/v), 90 °C, 57–65% for two steps; (f) p-chlorobenzylchloride (3.0 equiv), Cs2CO3 (3.0 equiv), DMF, 60 °C; (g) KOH, THF–H2O (3:1, v/v), 70 °C, 67–78% for two steps; (h)
N-methylpiperazine (10.0 equiv), K2CO3 (3.0 equiv), DMF, 60 °C, 56–74%.
Please cite this article in press as: Cheng, Y.; et al. Bioorg. Med. Chem. Lett. (2016), http://dx.doi.org/10.1016/j.bmcl.2016.04.042

Y. Cheng et al. / Bioorg. Med. Chem. Lett. xxx (2016) xxx–xxx
3
O
O
O
N
O
O
N
O
O
O
O
N
O
NH2
N
O
O
N
R3
NH
e, f, g
d
a
b
c
N
NH
N
O
N
H
R1
R1
N
11
O
R1
12
13
14
6a-v
R2
R2
R2
Scheme 2. Reagents and conditions: (a) 5-(methoxymethylene)-2,2-dimethyl-1,3-dioxane-4,6-dione (1.3 equiv), i-PrOH, 100 °C, 1 h, 65%; (b) diphenyl oxide, 220 °C, 60%; (c)
ArF (10.0 equiv), NaH (4.0 equiv), diphenyl oxide, microwave (150 °C, 150 W), 4 h, 26–80%; (d) concd HCl, 90 °C, 1 h (quantitative); (e) R3Cl or R3Br (3.5 equiv), TBAF
(10.0 equiv), THF, rt, overnight; (f) Zn, HOAc, rt, 1 h, 65–87%; (g) R4COOH (4.0 equiv), HATU (5.0 equiv), DIEA (10.0 equiv), DCM, RT, overnight or R4NCO (1.5 equiv), DCM,
40 °C.
Table 1
R1 and R2 SAR of 2-alkyl quinolones
O
N
O
5
4
1
F
O
6
2
R1
R2
Cl
7
Cl
1, 5a-f
Compd
Structure
HCV-flu(gt2a)
MTT assay
SI*
R1
R2
H
Me
Et
CH2OCH3
Me
CH2OCH3
CH2OCH3
EC50
(l
M)
CC50 (lM)
1
(4-Methyl)morpholinyl
OMe
OMe
4.3
2.4
3.7
11.1
7.2
4.9
0.4
>40
>40
26.8
>40
33.1
>40
4.3
>9.3
5a
5b
5c
5d
5e
5f
>16.7
7.2
>3.6
4.6
8.2
10.8
OMe
(4-Methyl)morpholinyl
(4-Benzyl)morpholinyl
(4-Methyl)morpholinyl
O
O
F
O
422
2.02
13.2
5.5
N
N
Cl
N
F
F
*
SI: selectivity index, SI = CC50/EC50
.
O
N
O
F
O
N
Cl
O
O
R
O
N
O
N
R1
F
O
N
A
N
B
1
Fragment
Scaffold
Hopping
X
N
Cl
Hybridization
N
O
O
R2
F
O
X
R3
N
Cl
N
F
5, 2-alkyl-N-arylquinolones
6
, 1,6-naphthyridine-4,5(1H,6H)-diones
F
4
Figure 2. Structure-based design of 2-alkyl quinolone and 1,6-naphthyridine-4,5(1H,6H)-dione derivatives as potential HCV NS5B polymerase inhibitors.
in Scheme 2, we established a convenient synthesis toward these
novel scaffold compounds. A reversible Michael addition reaction
of 2-methoxypyridin-4-amine with 5-(methoxymethylene)-2,2-
dimethyl-1,3-dioxane-4,6-dione can readily generate 11 which
was transformed into the naphthyridine core 12 through Friedel–
Crafts reaction at high temperature. The N-1 aryl group was
installed via SNAr displacement to give intermediate 13 where
the cross coupling reactions were tried but failed. Removal of
methyl group by acidolysis followed by selective alkylation of
the N-6 atom delivered the target compounds (6a–c) or the
N1-(4-NO2-Ph) precursors, which were then reduced to produce
the amino compounds (6d–m) and further transformed to furnish
amides and ureas (6n–v) via condensation with acids and
isocyanates, respectively.
All the compounds were evaluated in vitro with their antiviral
potency and cytotoxicity in Huh7.5.1 replicon 2a system. The
antiviral activities of 2-substituted quinolone derivatives were
shown in Table 1. Although the reference compound 1 was
reported to show potent antiviral activity against genotype 1 virus
(EC50 (gt1a) = 0.41 lM, EC50 (gt1b) = 0.12 l
M),23 as other NS5B
Please cite this article in press as: Cheng, Y.; et al. Bioorg. Med. Chem. Lett. (2016), http://dx.doi.org/10.1016/j.bmcl.2016.04.042

4
Y. Cheng et al. / Bioorg. Med. Chem. Lett. xxx (2016) xxx–xxx
Table 2
Global SAR exploration of 1,6-naphthyridine-4,5-dione analogs as HCV inhibitors
O
O
R3
6
N
1
N
1'
2'
3'
R1
4'
R2
Compd
Structure
HCV-flu (gt2a)
MTT assay
SIb
R1
R2
R3
Inhibition rate (%), @5
lM
EC50
(
lM)
Cell viability (%), @ 5
lM
CC50 (lM)
6a
6b
6c
6d
6e
6f
20-CF3
30-CF3
NO2
NO2
CN
NH2
NH2
NH2
NH2
NH2
58.6
65.7
—
4.8
—
—
—
—
6.4
2.5
88.2
75.6
—
10.6
—
—
—
—
19.4
17.9
—
2.2
—
—
—
—
3.0
7.2
∗
30-CF3
20-CF3
20-CH3
H
Cl
n.i.a
86.4
84.8
91.6
94.6
79.2
74.3
43.8
52.0
39.8
68.0
78.9
6g
6h
30-Cl
30-CF3
∗
∗
6i
38.7
—
100.0
—
—
Me
Me
Me
6j
51.7
n.i.a
—
—
99.8
96.8
—
—
—
—
30-CF3
NH2
∗
∗
∗
∗
6k
6l
7.7
—
—
100.0
100.0
—
—
—
—
6m
n.i.a
O
O
∗
N
F
6n
6o
H
H
41.7
95.7
—
88.7
46.1
—
N
H
Cl
∗
N
H
4.6
6.4
1.4
F
O
∗
N
N
H
6p
H
52.3
—
68.6
—
Cl
O
∗
O
6q
6r
H
H
42.6
18.8
—
—
88.0
98.4
—
—
N
H
H
H
N
N
∗
O
O
F
∗
20-CH3
20-CH3
55.2
53.1
—
—
72.6
70.4
—
—
N
H
6s
6t
F
H
H
N
N
∗
O
F
O
F
∗
6u
6v
20-CF3
20-CF3
98.2
33.4
1.4
—
38.4
75.4
1.5
—
1.1
F
H
N
H
N
∗
O
F
a
n.i.: no inhibition was observed at the concentration tested.
b
SI: selectivity index, SI = CC50/EC50
.
Please cite this article in press as: Cheng, Y.; et al. Bioorg. Med. Chem. Lett. (2016), http://dx.doi.org/10.1016/j.bmcl.2016.04.042

Y. Cheng et al. / Bioorg. Med. Chem. Lett. xxx (2016) xxx–xxx
5
of the bulky N1-substituent with the hydrophobic pocket, thus
compound 5a–c showed reduced potency with increasing steric
hindrance at C-2 (–Me > –Et > –CH2OCH3). However, When the
privileged fragment of (4-methyl)morpholinyl group was installed
in C-7 position, the introduction of methoxylmethyl (CH3OCH2–)
group at 2-position resulted in a substantial improvement of the
antiviral activity (5f, EC50 = 0.4 lM), by a factor of 11-fold relative
to the reference compound 1. Even compared to our originally
identified 2-arylquinolone inhibitor 4,22 the optimal combination
of the two hydrophobic substituents at position N-1 and C-2 (5f
vs 4) conferred superior anti-HCV potency and selectivity index,
which was consistent well with our initial rational design.
Considering the relatively poor physicochemical property of the
quinolone structures,24 further structural optimization was direc-
ted to evolve drug-like inhibitors through scaffold hopping. Based
on the co-crystal structure of the quinolone 2 with the NS5B pro-
tein, the solvent exposing portion with C-6/C-7 substituents had
no direct contact with the enzyme, though this moiety was bene-
ficial for the replicon potency. On the other hand, the 2-substituent
was involved in the hydrophobic interaction with the protein as
well. Taken together, a new scaffold of 1,6-naphthyridine-4,5-
dione resulted from the B-ring truncation and the cyclization of
the C-2 substituent with C-3 moiety of the quinolone core
(Fig. 2). A global SAR study on N1- and N6-substituent was explored
to determine an optimal substitution pattern for this new scaffold.
First of all, the N6-(4-chlorobenzyl) group was fixed to mimic the
C3-benzyl group of quinolone 1, an exploration of R1 and R2 on
Figure 3. Proposed binding mode of selected 2-alkyl quinolone 5f based on co-
crystal structure of 2 bound to NS5B polymerase thumb pocket II.
thumb II inhibitors functioned,27 it just displayed moderate repli-
cation inhibition against HCV genotype 2a (EC50 = 4.3 lM) in our
assay. Gratifyingly, the incorporation of a small and flexible alkyl
group at C-2 position of the quinolone in the presence of N-1 aryl
substitution indeed exerted an enhancing effect in the cellular
replicon potency (5a, 5f vs 1). As we anticipated, the steric
hindrance of C2-substituent would interfere with the interaction
Figure 4. (A) The modeling of compound 6h bound to NS5B based on the cocrystal structure of 2 with NS5B (PDB: 3PHE). (B) superimposition of compound 6h to the crystal
structure of compound 2. (C) Ligand–enzyme interactions of compound 6h.
Please cite this article in press as: Cheng, Y.; et al. Bioorg. Med. Chem. Lett. (2016), http://dx.doi.org/10.1016/j.bmcl.2016.04.042

6
Y. Cheng et al. / Bioorg. Med. Chem. Lett. xxx (2016) xxx–xxx
N-1 phenyl ring was conducted. As shown in Table 2, most of the
N1-phenyl-N6-(4-chlorobenzyl)-1,6-naphthyridine-4,5-diones (6a–h)
exhibited moderate to high inhibition rate on HCV replication
and low cytotoxicity, except for the 40-cyano substituted analog
(6c). Compared to the amino or nitro group at 40-position, the less
polar nature of cyano group might account for the activity loss. On
the other hand, a hydrophobic substituent at 20-position and/or
30-position of N1-aryl group was beneficial for the antiviral activity
with 30-substitution being superior to the 20-position. The antiviral
activity order as –CF3 > –Cl > –H for 30-substitution (6f, EC50 not
and formed p–p stacking interaction with Trp528, the N-1 aryl
group stretched into another long shallow hydrophobic channel
defined by Leu419, Ala486, Leu489, Pro496 and Leu497. Further-
more, the trifluoromethyl group headed into a small pocket locat-
ing at the bottom as originally expected and the fluorine atom
formed a H-bond interaction with the backbone amide NH of
Leu497 since the distance between (N)H–F was 2.3 Å (Fig. 4A
and C). Overlaying the docking results with the quinolone 2 exhib-
ited an excellent superimposition (Fig. 4B). Compound 6h shared
almost all the pharmacophores with other reported thumb site II
inhibitors. The inconsistence of the relatively reduced cellular
potency of this series might be due to the cell membrane perme-
ability factors of the compounds. Structurally diverse analogs of
the naphthyridinedione were under construction and assay.
In conclusion, we conducted a further structural optimization of
2-substituted quinolones based on the binding mode with NS5B
polymerase. A molecular hybridization of 2-aryl quinolones and
the reported inhibitor 1 lead to a new class of C2-alkyl-N1-aryl qui-
nolones with more than 10-fold improvement in the antiviral
activity. Further drug-like structural optimization on 2-alkyl qui-
nolones led to the identification of a brand new class of naph-
thyridinediones as HCV inhibitors by using scaffold hopping
strategy. Global SAR exploration furnished the most potent com-
pound with low micromolar antiviral activity. The molecular mod-
eling study interpreted the binding modes of the 2-alkyl
quinolones and 1,6-naphthyridine-4,5(1H,6H)-dione analogs as
NS5B thumb pocket II binders.
identified; 6g, EC50 = 6.5
cated a hydrophobic and bulky group at this position might confer
better interaction with the protein by occupying a small
l lM; Table 2) indi-
M; 6h,28 EC50 = 2.5
a
hydrophobic pocket mainly defined by Leu419, Met423, and
Val485.
Although p-chlorobenzyl group was recognized as a privileged
fragment for C-3 position of the quinolone, the N-6 substituent
was investigated as well on the new naphthyridinedione core. Aro-
matic and aliphatic groups with different ring size were examined
(Table 2, 6h–m). Unfortunately, replacement with either an unsub-
stituted benzyl group or aliphatic ring led to a complete loss in
activities (6k–m), probably due to the impaired p–p stacking inter-
action with Trp528. Correspondingly, hydrophobic substituent on
the phenyl ring facilitated the antiviral potency (6h–j) with the
chloro being the best.
Further effort to improve the activity of the 1,6-naphthyridine-
4,5-dione chemotype was switched to the derivatization on the
N1-phenyl ring. Since a sulfone oxygen of compound 2 anchored
the enzyme by forming a key H-bond interaction with backbone
amide hydrogen of Leu497 and the 40-NH2 moiety on N1-phenyl
ring was expected to reside the same region, we designed 40-amide
and 40-urea derivatives (6n–v, Table 2). We assume that the incor-
poration of the carbonyl oxygen as HBA and an additional aromatic
fragment could extend the hydrophobic interaction with the
binding pocket. However, we failed to install an amide or urea
substituent at 40-NH2 in the favorable N1-(30-CF3-40-NH2)phenyl
substituted template, so we tested our idea on 20-sbustituted and
nonsubstituted chemotypes. Disappointingly, the introduction of
the arylcarboxamido or arylcarbonyldiimino moiety didn’t exert
remarkable enhancing effect. Some of the resulting N1-(40-benza-
mido)phenyl-1,6-naphthyridine-4,5-dione derivatives exhibited
Acknowledgments
We are grateful to Ms. Jing Xing at Drug Design and Discovery
Centre (DDDC), Shanghai Institute of Materia Medica for her gener-
ous help in molecular modeling of compound 5f and 6h. This work
was supported by grants from National Science Foundation of
China (81325020, 81361120410, 81321092) and National Program
on Key Basic Research Project of China (2013CB911104).
Supplementary data
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.bmcl.2016.04.
042.
low micromolar replicon potency (6o, EC50 = 4.6
lM; 6u, EC50 =
1.4 M), but with a compromised safety window. As a result,
l
1-(4-amino-3-(trifluoromethyl)phenyl)-6-(4-chlorobenzyl)-1,6-naph-
References and notes
thyridine-4,5(1H,6H)-dione 6h29 showed the best biological profile
(EC50 (replicon, gt2a) = 2.5 lM; CC50 = 17.9 lM) in this series. The
1. Bosetti, C.; Levi, F.; Lucchini, F.; Zatonski, W. A.; Negri, E.; La Vecchia, C. J.
Hepatol. 2007, 46, 827.
2. Lavanchy, D. Clin. Microbiol. Infect. 2011, 17, 107.
3. Lozano, R.; Naghavi, M.; Foreman, K.; Lim, S.; Shibuya, K.; Aboyans, V.;
Abraham, J.; Adair, T.; Aggarwal, R.; Ahn, S. Y., et al. Lancet 2012, 380, 2095.
4. Neukam, K.; Macias, J.; Mira, J. A.; Pineda, J. A. Expert Opin. Pharmacother. 2009,
10, 417.
5. Ghany, M. G.; Strader, D. B.; Thomas, D. L.; Seeff, L. B. Hepatology 2009, 49, 1335.
6. De Clercq, E. Biochem. Pharmacol. 2014, 89, 441.
7. Lin, K.; Perni, R. B.; Kwong, A. D.; Lin, C. Antimicrob. Agents Chemother. 2006, 50,
1813.
8. Venkatraman, S.; Bogen, S. L.; Arasappan, A.; Bennett, F.; Chen, K.; Jao, E.; Liu, Y.
T.; Lovey, R.; Hendrata, S.; Huang, Y. H., et al. J. Med. Chem. 2006, 49, 6074.
9. Cannalire, R.; Barreca, M. L.; Manfroni, G.; Cecchetti, V. J. Med. Chem. 2016, 59,
16.
biological results suggest that the brand new structure 1,6-naph-
thyridine-4,5(1H,6H)-dione could serve as promising new HCV
inhibitors for further development.
In order to throw light on the molecular mechanism of our
designed inhibitors, we performed molecular modeling to propose
the binding mode of the most potent quinolone inhibitor 5f and
the 1,6-naphthyridine-4,5(1H,6H)-dione inhibitor 6h based on
the X-ray co-crystal structure of TSII inhibitor 2 with NS5B poly-
merase (PDB: 3PHE).23,30 For compound 5f, the docking result
described a similar binding mode as that of 2 (Fig. 3). Although
the distance between the two heavy atoms (O (CH2OCH3) of quino-
lone 5f and N in Arg501 of NS5B) was measured as 4.1 Å, we
assumed an H-bond interaction might take place through a water
bridge in physiological environment, to which the significant gain
in the antiviral activity of 5f was attributed.
The docking results of compound 6h (Fig. 4A–C) predicted a
similar binding mode that the two carbonyl groups of the naph-
thyridinedione nucleus anchored the protein through hydrogen
bonding to backbone amide NHs of Tyr477 and Ser476, respec-
tively. The N-6 benzyl group occupied a deep hydrophobic pocket
10. Sofia, M. J.; Chang, W.; Furman, P. A.; Mosley, R. T.; Ross, B. S. J. Med. Chem.
2012, 55, 2481.
11. Gane, E. J.; Stedman, C. A.; Hyland, R. H.; Ding, X.; Svarovskaia, E.; Symonds, W.
T.; Hindes, R. G.; Berrey, M. M. N. Eng. J. Med. 2013, 368, 34.
12. Ivan Gentile, A. R. B.; Guglielmo, Borgia Clin. Trials 2014, 9, 115.
13. Link, J. O.; Taylor, J. G.; Xu, L.; Mitchell, M.; Guo, H.; Liu, H.; Kato, D.; Kirschberg,
T.; Sun, J.; Squires, N., et al. J. Med. Chem. 2014, 57, 2033.
14. DeGoey, D. A.; Randolph, J. T.; Liu, D.; Pratt, J.; Hutchins, C.; Donner, P.; Krueger,
A. C.; Matulenko, M.; Patel, S.; Motter, C. E., et al. J. Med. Chem. 2014, 57, 2047.
15. Poordad, F.; Lawitz, E.; Kowdley, K. V.; Cohen, D. E.; Podsadecki, T.; Siggelkow,
S.; Heckaman, M.; Larsen, L.; Menon, R.; Koev, G., et al. N. Eng. J. Med. 2013, 368,
45.
Please cite this article in press as: Cheng, Y.; et al. Bioorg. Med. Chem. Lett. (2016), http://dx.doi.org/10.1016/j.bmcl.2016.04.042

Y. Cheng et al. / Bioorg. Med. Chem. Lett. xxx (2016) xxx–xxx
7
16. Behrens, S. E.; Tomei, L.; DeFrancesco, R. EMBO J. 1996, 15, 12.
17. Lesburg, C. A.; Cable, M. B.; Ferrari, E.; Hong, Z.; Mannarino, A. F.; Weber, P. C.
Nat. Struct. Biol. 1999, 6, 937.
27. Eltahla, A. A.; Tay, E.; Douglas, M. W.; White, P. A. Antimicrob. Agents Chemother.
2014, 58, 7215.
28. The characterization data of compound 5f: Mp 142–144 °C, 1H NMR (300 MHz,
18. Barreca, M. L.; Iraci, N.; Manfroni, G.; Cecchetti, V. Future Med. Chem. 2011, 3,
1027.
19. Lin, J. P.; Long, Y. Q. Chem. Commun. 2013, 5313.
20. Hu, W.; Lin, J. P.; Song, L. R.; Long, Y. Q. Org. Lett. 2015, 17, 1268.
21. Zhi, Y.; Gao, L. X.; Jin, Y.; Tang, C. L.; Li, J. Y.; Li, J.; Long, Y. Q. Bioorg. Med. Chem.
2014, 22, 3670.
Methanol-d4) d 7.78 (d, J = 13.2 Hz, 1H), 7.51 (d, J = 8.7 Hz, 2H), 7.38 (d,
J = 8.6 Hz, 2H), 7.34 (d, J = 8.6 Hz, 2H), 7.09 (d, J = 8.7 Hz, 2H), 6.76 (d, J = 7.2 Hz,
1H), 5.60 (s, 2H), 5.37 (s, 2H), 4.47 (s, 2H), 3.20 (s, 3H), 3.15–2.77 (m, 4H), 2.76–
2.35 (m, 4H), 2.31 (s, 3H). 13C NMR (151 MHz, DMSO-d6) d 172.6, 167.4, 166.7,
152.5 (d, J = 248 Hz), 147.3, 144.6 (d, J = 10.0 Hz), 138.5, 136.2, 135.4, 133.2,
132.4, 130.6, 129.3, 128.9, 128.3, 120.6 (d, J = 7.0 Hz), 118.4, 111.2, 111.1,
107.4, 69.1, 66.2, 58.5, 56.5, 54.4, 50.2, 49.8, 49.7, 46.0. MS (ESI) m/z 598.2 (M
+H)+.
22. Wang, S. F.; Lin, J. P.; He, P. L.; Zuo, J. P.; Long, Y. Q. Acta Chim. Sinica 2014, 72,
906.
23. Kumar, D. V.; Rai, R.; Brameld, K. A.; Somoza, J. R.; Rajagopalan, R.; Janc, J. W.;
Xia, Y. M.; Ton, T. L.; Shaghafi, M. B.; Hu, H., et al. Bioorg. Med. Chem. Lett. 2011,
21, 82.
24. Kumar, D. V.; Rai, R.; Brameld, K. A.; Riggs, J.; Somoza, J. R.; Rajagopalan, R.;
Janc, J. W.; Xia, Y. M.; Ton, T. L.; Hu, H., et al. Bioorg. Med. Chem. Lett. 2012, 22,
300.
25. Manfroni, G.; Cannalire, R.; Barreca, M. L.; Kaushik-Basu, N.; Leyssen, P.;
Winquist, J.; Iraci, N.; Manvar, D.; Paeshuyse, J.; Guhamazumder, R., et al. J.
Med. Chem. 2014, 57, 1952.
29. The characterization data of copound 6h: Mp 184–186 °C, 1H NMR (400 MHz,
DMSO-d6) d 7.80 (d, J = 7.8 Hz, 1H), 7.61 (d, J = 7.8 Hz, 1H), 7.51 (d, J = 2.3 Hz,
1H), 7.40 (d, J = 8.5 Hz, 2H), 7.38 (dd, J = 8.8, 2.3 Hz, 1H), 7.31 (d, J = 8.5 Hz, 2H),
6.96 (d, J = 8.8 Hz, 1H), 6.12 (s, 2H), 6.03 (d, J = 7.8 Hz, 1H), 5.62 (d, J = 7.8 Hz,
1H), 5.02 (s, 2H). 13C NMR (151 MHz, DMSO-d6) d 176.2, 159.1, 151.6, 147.6,
142.7, 141.0, 136.8, 132.7, 132.6, 130.1, 129.3, 128.9, 128.7, 128.5, 125.9 (q,
J = 5 Hz), 124.8 (q, J = 272 Hz), 118.2, 116.9, 113.4, 110.9 (q, J = 30 Hz), 96.7,
50.6. MS (ESI) m/z 446.3 (M+H)+, 468.3 (M+Na)+.
30. The docking experiments were performed with Maestro 9.2.
26. Biswal, B. K.; Cherney, M. M.; Wang, M.; Chan, L.; Yannopoulos, C. G.; Bilimoria,
D.; Nicolas, O.; Bedard, J.; James, M. N. J. Biol. Chem. 2005, 280, 18202.
Please cite this article in press as: Cheng, Y.; et al. Bioorg. Med. Chem. Lett. (2016), http://dx.doi.org/10.1016/j.bmcl.2016.04.042