3-Heterocyclyl quinolone inhibitors of the HCV NS5B polymerase

Article abstract of DOI:10.1016/j.bmcl.2011.11.013

The discovery and optimization of a novel class of quinolone small-molecules that inhibit NS5B polymerase, a key enzyme of the HCV viral life-cycle, is described. Our research led to the replacement of a hydrolytically labile ester functionality with bio-

Full text of DOI:10.1016/j.bmcl.2011.11.013

Bioorganic & Medicinal Chemistry Letters 22 (2012) 300–304
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
3-Heterocyclyl quinolone inhibitors of the HCV NS5B polymerase
Dange V. Kumar a, Roopa Rai b, , Ken A. Brameld c, Jennifer Riggs d, John R. Somoza e, Ravi Rajagopalan f,
⇑
James W. Janc g, Yu M. Xia h, Tony L. Ton c, Huiyong Hu i, Isabelle Lehoux j, Joseph D. Ho k,
Wendy B. Young i, Barry Hart l, Michael J. Green m
a Myrexis Inc., 305 Chipeta Way, Salt Lake City, UT 84108, USA
b Alios BioPharma, 260 E. Grand Ave., South San Francisco, CA 94080, USA
c Principia Biopharma, 400 E. Jamie Ct., South San Francisco, CA 94080, USA
d Celgene Corp., 4550 Towne Centre Ct., San Diego, CA 92121, USA
e Gilead Sciences Inc., 333 Lakeside Drive, Foster City, CA 94404, USA
f InterMune Inc., 3280 Bayshore Blvd., Brisbane, CA 94005, USA
g Theravance, Inc., South San Francisco, CA 94080, USA
h Crown Bioscience, No.6 Beijing West Road, Taicang City, Jiangsu Province 215400, PR China
i Genentech Inc., 1 DNA Way, South San Francisco, CA 94080, USA
j Allosteros Therapeutics, 1230 Bordeaux Drive, Sunnyvale, CA 94089, USA
k Eli Lilly and Company, 10300 Campus Point Drive, Suite 200, San Diego, CA 9212, USA
l Innovation Pathways, 20 Palo Alto Ave., Palo Alto, CA 94031, USA
m Virobay, Inc., 1490 O’Brien Drive, Menlo Park, CA 94025, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
The discovery and optimization of a novel class of quinolone small-molecules that inhibit NS5B polymer-
ase, a key enzyme of the HCV viral life-cycle, is described. Our research led to the replacement of a hydro-
lytically labile ester functionality with bio-isosteric heterocycles. An X-ray crystal structure of a key
analog bound to NS5B facilitated the optimization of this series of compounds to afford increased activity
against the target enzyme and in the cell-based replicon assay system.
Received 13 September 2011
Revised 2 November 2011
Accepted 3 November 2011
Available online 9 November 2011
Ó 2011 Elsevier Ltd. All rights reserved.
Keywords:
HCV
NS5B polymerase inhibitors
3-Heterocyclyl quinolone analogs
NNI-2 binders
Hepatitis C virus (HCV) infection is a major public health prob-
lem with an estimated 170 million patients worldwide.1 Compro-
mised liver function caused by chronic HCV infection can
progress to cirrhosis and hepatocellular carcinoma and is a leading
cause for liver transplants.2 Until recently, HCV therapy consisted
of a combination of peginterferon alfa-2a/b and ribavirin, a regi-
men that has severe limitations including non-response, relapse,
poor tolerability and long duration of treatment. Two new small
molecule inhibitors of NS3 protease, telaprevir and boceprevir,
have completed human clinical trials and been approved for treat-
ment of HCV.3 Developing novel small molecule inhibitors target-
ing additional aspects of the HCV life cycle is an active area of
research. Both nucleoside and non-nucleoside inhibitors targeting
NS5B polymerase are currently in Phase II clinical trials. It is antic-
ipated that NS5B polymerase inhibitors will be used together with
NS3 protease inhibitors as the field moves ultimately to combina-
tion, all oral therapy for HCV.
Our research led to the discovery of a quinolone-based series of
lead compounds which inhibit HCV NS5B polymerase.4a The opti-
mization of this new structural class of inhibitors was aided by
an X-ray crystallographic structure of the quinolone 1 bound to
the NNI-2 allosteric binding site of this enzyme. We now describe
efforts to optimize this lead, with the initial goal of replacing the
ester functionality with
a
hydrolytically stable bio-isostere5
(Fig. 1) to improve activity and metabolic stability. Additionally,
we describe substitution on the phenyl rings and at the C7 position
of the quinolone to improve the pharmaceutical properties of the
series.6
Analysis of the X-ray structure of 1 bound to NS5b reveals that
the carbonyl of the ester acts as a hydrogen bond acceptor for the
backbone NH of Ser 476 in the NNI-2 binding site. An appropriate
iso-steric heterocycle should position an H-bond acceptor hetero-
atom in a similar position. The X-ray structure also reveals that
the aromatic ring of the benzyl ester fits into a hydrophobic pocket
⇑
Corresponding author.
E-mail address: roopa.rai@hotmail.com (R. Rai).
0960-894X/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2011.11.013

D. V. Kumar et al. / Bioorg. Med. Chem. Lett. 22 (2012) 300–304
301
Ser
476
Ser
476
Tyr
477
Tyr
477
B
O
N
A
O
O
5
D
4
6
R1
R1
F
n
E
O
1'
3
2
7
N
N1
Cl
X
8
N
Me
Y
Hydrophobic
interaction
1
Hydrophobic
interaction
SO2Me
Hydrophobic
interaction
Hydrophobic
interaction
Figure 1. Isosteric replacement of ester with heterocyclic groups.
and is two atoms away from the carbonyl carbon (C10). We there-
fore initially prepared the heterocycles2, 3, and 4 (Table 1) that pos-
sess an H-bond acceptor that corresponds to the position of the
ester carbonyl, and a phenyl ring two (n = 0) atoms away. The syn-
theses of these heterocycles are described in Schemes 1–3.7 Oxadi-
azoles 2 and 3 are both inactive against NS5B polymerase,8 whereas
the oxazoline 4 displayed some activity but was still much less po-
tent than its ester analog. A reasonable explanation for the observed
SAR is apparent from the crystal structure of 1, where there is a 90°
angle between the plane of the ester and the phenyl ring, allowing
the ring to be deeply buried in a hydrophobic pocket. Direct substi-
tution from the heteroaromatic rings of 2 and 3 does not allow the
phenyl to reach deeply into this out-of-plane hydrophobic pocket.
The sp3 linker from the oxazoline ring of 4, on the other hand, al-
lows the required out-of-plane geometry for the phenyl group to
better access the pocket (Fig. 2). Based on this hypothesis, we pre-
pared compounds 5–10 (Schemes 1–3) with a methylene group in-
serted between the heterocycle and the phenyl group to allow the
phenyl group improved access to this hydrophobic pocket.
Our binding hypothesis was supported by the fact that the
1,2,4-oxadiazole 5 was almost equipotent to the ester analog 1.
Compound 10, with the phenyl group two atoms away from the
Table 1
SAR of C-3 heterocyclic quinolones
O
R1
R2
R3
N
8
9
9
#
IC50
(l
M)
EC50
(lM)
GI50 (lM)
Cl
R1
R2
R3
O
N
Ph
2
3
OCH3
OCH3
OCH3
>10
>10
—
—
—
—
N
N
O
p-ClPh
OCH3
OCH3
N
N
N
4
OCH3
4.3
8.91
—
Ph
Ph
O
O
5
6
7
8
OCH3
OCH3
OCH3
OCH3
OCH3
OCH3
OCH3
H
0.041
0.73
1.6
10.3
—
—
—
—
—
N
N
Ph
O
N
N
N
N
p-ClPh
Ph
—
O
1.8
—
O
Ph
N
N
N
N
9
OCH3
OCH3
OCH3
OCH3
>10
>10
—
—
—
—
N
O
Ph
Ph
10
N
O
N
N
N
N
N
N
11
12
F
F
0.023
0.115
0.84
4.4
7.3
9.8
N
N
Ph
O

302
D. V. Kumar et al. / Bioorg. Med. Chem. Lett. 22 (2012) 300–304
O
O
N
O
Y
N
O
N
O
Y
O
N
N
Y
R1
R2
R1
R2
R1
R2
HN
OH
Cl
O
a
X
NH2
b, c
X
Z
Z
Z
7: R1 = R2 = OCH3; X = Cl; Y = H;
N
NH2
Z = Cl
12:
HO
CH3
NH2
HO
1
2
R = F; R = N-CH3-
piperazine; X = H; Y = H; Z = Cl;
e
n
Ph
X
1
2
18:
R = F; R = N-CH3-
d
piperazine;
N
O
O
CH3
X = H; Y = F; Z = CF3
R1
R2
1
2
O
N
Y
19:
n
R = F; R = N-CH3-
piperazine; X = F; Y = F; Z = CF3
N
R1
R2
O
N
Y
N
X
Z
Z
1 = R2 = OCH3; X = Cl; Y = H; Z = Cl; n = 0
6: R1 = R2 = OCH3; X = H; Y = H; Z = Cl; n = 1
3:
R
4: R1 = R2 = OCH3; Y = H; Z = Cl
Scheme 1. Reagents and conditions: (a) Polystyrene bound-PPh3 (PS-PPh3), Cl3CCN, THF, 100 °C, 10 min; (b) DIEA, 150 °C or microwave, 15 min; (c) PS-PPh3, I2, TEA, DCM,
0 °C to rt; (d) Et3N; (e) K2CO3.
CHN2
c
O
N
O
O
N
N
R1
R2
R1
R2
O
N
Cl
8: R1 = OCH3; R2 = H
Cl
a
N
N
O
N
N
O
N
N
NH
R1
R2
N
R1
R2
b
N
N
Cl
Cl
9: R1 = R2 = OCH3
Scheme 2. Reagents and conditions: (a) TMSN3, n-Bu2SnO, toluene, 110 °C, 4 h; (b) Benzyl bromide, K2CO3, DMF, 60 °CC, 4 h; (c) BF3ꢀOEt2, CH2Cl2.
heterocyclic ring, was devoid of NS5b polymerase inhibitory activ-
ity. The isomeric 1,2,4-oxadiazole 6 was also active but 20-fold less
potent than 5, indicating that the oxygen is less capable of forming
the required H-bond than the nitrogen in 5. The 1,3,4-oxadiazole 7
and the oxazole 8 were even less potent while the tetrazole 9 was
inactive.
Despite being a potent inhibitor of the NS5B enzyme, the oxadi-
azole 5 was poorly active in the replicon assay.9 In parallel, com-
pounds were evaluated for their cytotoxic effects in the replicon
cell line.9 To improve the replicon activity, we turned our attention
to improving the biopharmaceutical properties of this class of com-
pounds. The X-ray structure of ester 1 indicates that substituents at
the C7 position point into bulk solvent and positioning hydrophilic
substituents at this site led to good replicon potency in the ester
series.4 We applied the same strategy to the oxadiazole heterocy-
cles 5 and 7. The 6-fluoro-7-(N-methylpiperizine)-analogs of these
compounds, 11 and 12, were synthesized according to Schemes 1
and 3. Compounds 11 and 12 did indeed have improved replicon
potency.4b In particular, the submicromolar replicon potency of
11 encouraged us to continue optimizing this series.
The two phenyl rings were modified with electron withdrawing
substitutents to optimize potency and block potential phenyl ring
oxidative metabolism.7 The results of this work are shown in
Tables 2 and 3. In the 1,2,4-oxadiazole series, replacing the R4 phe-
nyl ring with a 4-F-phenyl (13, Table 2) results in no loss of activity
against NS5B, while replacing it with a 2,4-difluoro phenyl ring
(14) reduces both NS5B and replicon potency. Replacing the
p-chlorine on the N-benzyl group of 11 (Table 2) with 4-trifluoro-
methyl (15) or with 2-fluoro-4-trifluoromethyl (16) maintained
the potency against NS5B and further improved replicon potency.
Compound 17, which has both phenyl rings protected from meta-
bolic oxidation, was highly potent in both assays. These substitu-
tions also improved replicon potency in the 1,3,4-oxadiazole
series, compounds 18 and 19 in Table 3.10
In order to confirm our design paradigm and to guide further
optimization, the structure of 16 bound to NS5B was solved (RCSB

D. V. Kumar et al. / Bioorg. Med. Chem. Lett. 22 (2012) 300–304
303
X
O
N
O
N
N
Y
H3CO
H3CO
OH
O
N
n
O
N
N
Y
O
N
R1
R2
H3CO
H3CO
n
NH2
a
b
HO
Y
X
Z
2
: X = H; Y = H; Z = Cl; n = 0
5: X = H; Y = H; Z = Cl; n = 1
10: X = H; Y = H; Z = Cl; n = 2
Z
Z
R1 = R2 = OCH3
R1 = R2 = F
O W
O
NH
O
N
N
Y
HO
OH
O
N
N
Y
N
F
W
X
H3C
N
F
NH2
N
b
c, a
N
N
X
H3C
N
H3C
Z
Z
11
: X = H; Y = H; Z = Cl; W = H
: X = F; Y = H; Z = Cl; W = H
: X = F; Y = H; Z = Cl; W = F
: X = H; Y = H; Z = CF3; W = H
: X = H; Y = F; Z = CF3; W = H
: X = F; Y = F; Z = CF3; W = H
13
14
15
16
17
Scheme 3. Reagents and conditions: (a) NH2OH, DIEA, aq EtOH; (b)HBTU, DIPEA, 200 °C, microwave; 3 min (c) acetonitile, reflux.
CH3
O
O
N
O
the C-terminus. The structure was refined to 2.15 Å resolution
(R = 0.191, FreeR = 0.243, RMS bonds = 0.03 Å, RMS angles = 0.69°).
Generally, each of the four molecules in the asymmetric unit
showed clear electron density for the inhibitor, although the den-
sity for the N-methyl piperazine, which points into solution was
less well defined. The crystal structure revealed a binding mode
consistent with that observed for compound 1. The ligand occupies
the NNI-2 binding site located in the thumb region (Fig. 3a), mak-
ing several hydrogen bonds and occupying two hydrophobic pock-
ets. The quinolone ring sits on a shelf formed by Ile 482. As
expected, a hydrogen bond exists between the quinolone carbonyl
oxygen atom and the backbone amide NH of Tyr 477. In addition,
the N2 nitrogen atom of the oxadiazole is nicely positioned to re-
ceive a hydrogen bond from the backbone amide NH of Ser 476.
The benzyl group off the oxadiazole occupies a hydrophobic pocket
formed by the collapse of Leu 419, Arg 422, Met 423, and Trp 528. A
second shallow hydrophobic pocket is occupied by the CF3-phenyl
group which contacts Leu 419, Val 485, Ala 486, Leu 489, Leu 497,
and Met 423. Finally, the piperazine ring is solvent exposed, consis-
tent with the suitability of this location for modulating the overall
physico-chemical properties of the molecule.
N
H
N
N
N
2
4
5
Figure 2. Design optimization for best fit of phenyl ring in hydrophobic pocket.
Table 2
Optimization of the 3-(1,2,4-oxadiazole)-quinolone series
O
N
O
N
F
R3
IC50
M)
EC50
M)
GI50
M)
#
(l
(
l
(l
N
N
N
R4
R3
13 4-FPh
14 2,4-DiFPh
15 Ph
16 Ph
17 4-FPh
R4
4-ClPh
4-ClPh
4-CF3Ph
0.019
0.075
0.024
0.014
0.015
1.43
3.68
0.36
0.25
0.23
6.4
4.7
15
7.0
6.0
2-F, 4-CF3Ph
2-F, 4-CF3Ph
An overlay of the X-ray structures for compounds 1 and
16 ( Fig. 3b) confirms the design hypothesis used to drive the
replacement of the benzyl ester with a benzyl oxadiazole. The
methylene linker of the benzyl group allows the phenyl ring to
adopt the preferred out-of-plane geometry and the two rings are
nearly superimposable, despite the longer overall linker length of
the oxadiazole when compared to the ester. The CF3 group at the
phenyl R4 position overlays well with the methylsulfone of 1,
although the benzyl rings are twisted approximately 45° relative
to each other.
The origin of this rotational difference is not readily apparent.
The ortho-fluoro substituent on the N-benzyl ring of 16 does not
make any specific interactions with the proximal residues of the
binding pocket. Thus the twist is likely a consequence of either
an intramolecular interaction between the ortho-fluoro and meth-
ylene of the benzyl group or the influence of the methyl sulfone on
the phenyl ring of 1. In the absence of additional structural data, it
Table 3
Optimization of the 3-(1,3,4-oxadiazole)-quinolone series
N
O
N
F
R3
O
IC50
M)
EC50
GI50
(lM)
#
(
l
(
lM)
N
N
N
R4
R3
R4
18 Ph
19 p-FPh
2-F,4-CF3Ph
2-F,4-CF3Ph
0.038
0.045
0.61
0.73
4.6
5.7
ID code: RCSB068620; PDB ID code: 3UDL). The construct of NS5B
had the N-terminal 21 residues removed and included a His tag at

304
D. V. Kumar et al. / Bioorg. Med. Chem. Lett. 22 (2012) 300–304
compounds are an improvement on a previous series of quinolones
through the replacement of an ester with either a 1,2,4- or a 1,3,4-
oxadiazole. The most potent compounds in this series, 16 and 17,
have replicon EC50s below 250 nM. The continued optimization
of this series will be described in forthcoming papers.
Acknowledgement
The authors thank Dr. Robert Booth (Virobay Inc.) for support of
the research that went into this paper.
References and notes
1. Simmonds, P. J. Gen. Virol. 2004, 85, 3173.
2. Younossi, Z.; Kallman, J.; Kincaid, J. Hepatology 2007, 45, 806.
3. (a) Poordad, F.; McCone, J.; Bacon, B. R.; Bruno, S.; Manns, M. P.; Sulkowski, M.
S.; Jacobson, I. M.; Reddy, K. R.; Goodman, Z. D.; Boparai, N.; DiNubile, M. J.;
Sniukiene, V.; Brass, C. A.; Albrecht, J. K.; Bronowicki, J. P. N. SPRINT-2
Investigators Engl. J. Med. 2011, 364, 1195; (b) Jacobson, I. M.; McHutchison, J.
G.; Dusheiko, G.; Di Bisceglie, A. M.; Reddy, K. R.; Bzowej, N. H.; Marcellin, P.;
Muir, A. J.; Ferenci, P.; Flisiak, R.; George, J.; Rizzetto, M.; Shouval, D.; Sola, R.;
Terg, R. A.; Yoshida, E. M.; Adda, N.; Bengtsson, L.; Sankoh, A. J.; Kieffer, T. L.;
George, S.; Kauffman, R. S.; Zeuzem, S. N. ADVANCE Study Team Engl. J. Med.
2011, 364, 2405.
4. (a) 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.; Lehoux, I.; To, N.; Young, W. B.;
Green, M. J. Bioorg Med Chem Lett 2011, 21, 82; (b) While protein binding and
solubility of the lead ester series was not evaluated, the improvement in
replicon potency with the addition of polar moieties at the C7 position is
attributed to reduced protein binding properties.
5. (a) Saunders, J.; Angus, M.; MacLeod, K. M.; Showell, G. A.; Snow, R. J.; Street, L.
J.; Baker, R. J. Chem. Soc., Chem. Commun. 1988, 1618; (b) Kuduk, S. D.; Chang, R.
K.; Ng, C.; Murphy, K. L.; Ransom, R. W.; Tang, C.; Prueksaritanont, T.;
Freidinger, R. M.; Pettibone, D. J.; Bock, M. G. Bioorg. Med. Chem. Lett. 2005, 15,
3925.
6. Smith, D. A.; Waterbeemd, H.; Walker, D. K. In Pharmacokinetics and Metabolism
in Drug Design; Mannhold, R., Kubinyi, H., Folkers, G., Eds.; Wiliey-VCH:
Weinheim, Germany, 2006. second Ed., Vol. 31.
7. Kumar, D. V.; Rai, R.; Young, W. B.; Hu, H, Riggs, J. R.; Ton, T. L.; Green, M. J.;
Hart, B. P.; Brameld, K. A.; Dener, J. M. US 2007/0287699 and WO 2007/130499.
8. The NS5B polymerase IC50 enzyme activity assay determination: As described
in: McKercher, G.; Beaulieu, P. L.; Lamarre, D.; LaPlante, S.; Lefebvre, S.; Pellerin,
C.; Thauvette, L.; Kukolj, G. Nuc. Acids Res. 2004, 32, 422. Scintillation
proximity assays (SPA) were performed at ambient temperature (22 °C) in
96-well plates using 50 nM of enzyme (a C-terminal 21 residue deletion
mutant of NS5b from genotype 1b BK isolate). Reaction components included
20 mM Tris–HCl pH 7.5, 5 mM MgCl2, 1 mM EDTA, 2 mM DTT, 5% DMSO, 0.01%
Figure 3. (a) X-ray structure of 16 bound to NNI-2 site (PDB ID code: 3UDL). (b)
Overlay of compounds 16 (yellow) and 1 (white), showing a high degree of overlap
between the two ligands and very similar protein conformations. Small differences
in Leu 419 and Ile 482 are highlighted.
BSA and 25 mM KCl. 80
0.5 Ci of [3H]UTP, 1
M UTP, 250 nM 50-biotinylated oligo(rU12) and 0.8
ml poly(rA) to assay wells containing enzyme preincubated with inhibitor.
Reactions were terminated after 120 min by the addition of 20 l of stop
solution containing 150 g/ml tRNA and 10 mg/ml streptavidin-coated SPA
beads (Amersham) in 20 mM Tris–HCl pH 7.5, 25 mM KCl, 0.5 M EDTA and
0.025% sodium azide. After incubation for 30 min, 65 l of 5 M cesium chloride
l
l reactions were initiated by adding a mixture of
l
l
l
g/
l
l
l
is difficult to assess the validity of either of these hypotheses.
However, this observation suggests that the interactions made by
the aromatic portion of the benzyl substituent in this region of
the binding pocket are weak, non-specific, and easily influenced
by subtle chemical changes.
Scheme 1 describes the conversion of a C3 carboxylic acid to
various ester isosteric heterocycles. Activation of the acid as the
acid chloride affords an intermediate that reacts with a hydroxami-
dine, 1,2-amino alcohol or hydrazide to afford a 1,2,4-oxadiazole
(3, 6), 1,3-oxazoline (4), and a 1,3,4 oxadiazole respectively (7, 8,
12, 19). Scheme 2 describes the conversion of the C3 quinolone
nitrile to the C3 oxazole and tetrazole derivatives. Reaction of the
nitrile with TMSN3 affords the intermediate tetrazole that is benzy-
lated to afford compound 9.11 Oxazole 8 is obtained by reacting the
nitrile with 1-diazo-3-phenylpropan-2-one.
were added to the wells and further incubated for 1 h before top-counting in a
Microbeta Trilux plate reader.
9. The cell-based Replicon EC50 determination: HCV replicon-containing cells
(Huh7/Clone A, genotype 1b) were maintained in growth medium (DMEM
medium, Invitrogen), supplemented with 10% Fetal Bovine Serum, non
essential amino acids and 1 mg/mL G418) (Blight K. J., Kolykhalov A. A. and
Rice C. M. (2000) Science 290, 1972–1974). For the HCV replicon assay, Huh7/
Clone A cells were trypsinized from culture flasks, seeded in 1 ml of Clone A
growth medium without G418 at 40,000 cells per well in 24-well plates and
incubated at 37 °C in a humidified CO2 (5%) incubator overnight. Following
overnight incubation, test compound was serially diluted in DMSO and added
to the test system such that the final concentration of DMSO was 0.5% in each
well. For IC50 determinations, compounds were tested at 7 concentrations in
triplicates. Plates were incubated at 37 °C for 48 h. After incubation, cells were
harvested, transferred to 96-well plates, and subjected to total RNA extraction
using the RNA Isolation Kit (RNeasy 96, Qiagen). TaqMan quantitative PCR (RT-
qPCR) was used to quantify the amount of HCV replicon RNA in each sample.
The samples without compound treatment served as a control and the HCV
replicon RNA level from untreated cells was defined as 100%. Compound
inhibitory activity was determined as the ratio of the normalized HCV RNA
amount in treated samples relative to the untreated control. Compound EC50’s
were calculated using a standard 4 parameter curve fit model. The cytotoxicity
of compounds, the GI50,was assayed in parallel in which qPCR of the cellular
host protein GAPDH was used as an indicator of viable cells.
Scheme 3 describes the conversion of quinolone C3 nitriles to
the intermediate hydroxyamidine through reaction with hydroxyl
amine. The resulting hydroxyamidine intermediates are converted
to a variety of 1,2,4-oxadiazole target compounds by reacting with
appropriate carboxylic acids.
10. In general, the 7-position piperazine substituted compounds had low stability
in human liver microsomes with 20–50% remaining after a 1 h incubation.
In conclusion, we have described a new series of allosteric-site
(NNI-2) inhibitors of the HCV NS5B polymerase enzyme. These