Discovery of pyrido[2,3-d]pyrimidine-based inhibitors of HCV NS5A

Article abstract of DOI:10.1016/j.bmcl.2013.04.009

Efforts to improve the genotype 1a potency and pharmacokinetics of earlier naphthyridine-based HCV NS5A inhibitors resulted in the discovery of a novel series of pyrido[2,3-d]pyrimidine compounds, which displayed potent inhibition of HCV genotypes 1a and 1b in the replicon assay. SAR in this system revealed that the introduction of amides bearing an additional 'E' ring provided compounds with improved potency and pharmacokinetics. Introduction of a chiral center on the amide portion resulted in the observation of a stereochemical dependence for replicon potency and provided a site for the attachment of functional groups useful for improving the solubility of the series. Compound 21 was selected for administration in an HCV-infected chimpanzee. Observation of a robust viral load decline provided positive proof of concept for inhibition of HCV replication in vivo for the compound series.

Full text of DOI:10.1016/j.bmcl.2013.04.009

Accepted Manuscript
Discovery of pyrido[2,3-d]pyrimidine-based inhibitors of HCV NS5A
David A. DeGoey, David A. Betebenner, David J. Grampovnik, Dachun Liu,
John K. Pratt, Michael D. Tufano, Wenping He, Preethi Krishnan, Tami J. Pilot-
Matias, Kennan C. Marsh, Akhteruzzaman Molla, Dale J. Kempf, Clarence J.
Maring
PII:
S0960-894X(13)00478-2
DOI:
http://dx.doi.org/10.1016/j.bmcl.2013.04.009
BMCL 20374
Reference:
To appear in:
Bioorganic & Medicinal Chemistry Letters
Received Date:
Revised Date:
Accepted Date:
5 February 2013
29 March 2013
1 April 2013
Please cite this article as: DeGoey, D.A., Betebenner, D.A., Grampovnik, D.J., Liu, D., Pratt, J.K., Tufano, M.D.,
He, W., Krishnan, P., Pilot-Matias, T.J., Marsh, K.C., Molla, A., Kempf, D.J., Maring, C.J., Discovery of pyrido[2,3-
d]pyrimidine-based inhibitors of HCV NS5A, Bioorganic & Medicinal Chemistry Letters (2013), doi: http://
dx.doi.org/10.1016/j.bmcl.2013.04.009
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Discovery of pyrido[2,3-d]pyrimidine-based inhibitors of HCV NS5A.
David A. DeGoey,* David A. Betebenner,‡ David J. Grampovnik, Dachun Liu,
John K. Pratt, Michael D. Tufano,§ Wenping He,# Preethi Krishnan, Tami J. Pilot-
Matias, Kennan C. Marsh, Akhteruzzaman Molla, Dale J. Kempf, Clarence J.
Maring
Global Pharmaceutical Research and Development, AbbVie Inc., North Chicago,
IL 60064, USA
*Corresponding author.
E-mail address: david.degoey@abbvie.com.
‡Present Address: Advocate Condell Medical Center, Libertyville, IL 60048, USA
§Present Address: Department of Chemistry, Northeastern Illinois University,
Chicago, IL 60625, USA
#Present Address: Janssen Research & Development, Raritan, NJ 08869, USA
Efforts to improve the genotype 1a potency and pharmacokinetics of
earlier naphthyridine-based HCV NS5A inhibitors resulted in the discovery of a
novel series of pyrido[2,3-d]pyrimidine compounds, which displayed potent
inhibition of HCV genotypes 1a and 1b in the replicon assay. SAR in this system
revealed that the introduction of amides bearing an additional “E” ring provided
compounds with improved potency and pharmacokinetics. Introduction of a chiral
center on the amide portion resulted in the observation of a stereochemical
dependence for replicon potency and provided a site for the attachment of
functional groups useful for improving the solubility of the series. Compound 21
was selected for administration in an HCV-infected chimpanzee. Observation of a
robust viral load decline provided positive proof of concept for inhibition of HCV
replication in vivo for the compound series.
It’s estimated that more than 170 million people are chronically infected
with the hepatitis C virus (HCV) worldwide. While the infection is often
asymptomatic, chronic infection can lead to cirrhosis, hepatocellular carcinoma or
liver failure. HCV is classified into six genotypes (1–6), with subgenotypes 1a and
1b most commonly found in the United States, Japan, and Western Europe.
Treatment of genotype 1 (GT 1) patients with pegylated interferon alpha (pegIFN-
α) and ribavirin is associated with significant tolerability issues and poor efficacy
(40-50% cure rate). Combination of an NS3-4A protease inhibitor such as
telaprevir or boceprevir with pegIFN-α and ribavirin has provided improved
efficacy and is now the standard of care. However, there still remains a large
unmet medical need for therapies to treat patients who are unable to tolerate or
did not respond to previous therapy with pegIFN-α + ribavirin. The combination of

new direct-acting antivirals (DAAs) holds significant promise for improved cure
rates and shortened treatment duration.1
The genome of HCV encodes several nonstructural (NS) proteins. While
inhibition of the enzymatic activities of NS3/4A protease and NS5B polymerase
has been the subject of intense drug discovery efforts, the NS5A protein has no
known intrinsic enzymatic activity, although it is essential for viral RNA
replication.2 The discovery of compounds that select mutations in NS5A
established it as a potential drug target.3,4 NS5A inhibitors with dimeric structures
have recently become the subject of intense research, due to their potent
antiviral activity and clinical validation from human clinical trials.5-8 Recent
discovery efforts at Abbott to identify HCV inhibitors using the subgenomic
replicon assay yielded potent naphthyridine inhibitors such as 1 that lacked
activity against HCV protease and polymerase (Figure 1).9 Studies to elucidate
the mode of action concluded that the compounds were associated with NS5A,
and analysis of resistant variants selected during in vitro passage revealed a
number of mutations within the NS5A gene. Efforts to improve the GT 1a potency
of 1 resulted in the discovery of compound 2, which displayed potent inhibition of
GT 1a and 1b.10 While 2 displayed improved potency, it lacked sufficient
metabolic stability and oral pharmacokinetics for further advancement as a lead
series. Herein, we describe our efforts to identify additional SAR in this system
resulting in the discovery of amides bearing an additional “E” ring (Figure 1),
which displayed both improved GT 1a potency and pharmacokinetics.
Figure 1. Design of HCV NS5A Inhibitors
The key acid intermediate 6 was prepared starting from methyl 4-chloro-3-
nitrobenzoate according to Scheme 1. Nucleophilic aromatic substitution of
methyl 4-chloro-3-nitrobenzoate with 4-aminothiophenol and subsequent
protection of the amino group provided 3. After reduction of the nitro group, the
pyrido[2,3-d]pyrimidine 5 was formed by reaction with the appropriate pyrido
dimethylamidine in hot acetic acid, as described previously.10 Saponification of
the ester gave the acid 6 which was allowed to react with a variety of amines to

give the protected amides 7. Removal of the Boc group with trifluoroacetic acid
provided the final products 8.
O
O
O
O
O
O
CN
O
O
N
N
N
c
a,b
d
HN
N
H2N
O2N
O2N
S
5
S
S
Cl
N
N
NHBoc
NHBoc
NHBoc
4
3
e
H
H
N
O
N
O
N
O
N
OH
R
R
g
f
HN
N
HN
N
HN
N
S
S
S
N
N
NH2
N
NHBoc
N
NHBoc
8
6
7
Scheme 1. Reagents and conditions: (a) 4-aminothiophenol, K2CO3, 90 °C; (b)
Boc2O, dioxane, 90 °C; (c) Fe powder, NH4Cl, aqueous EtOH, reflux; (d) AcOH,
120 °C, 20 min.; (e) LiOH, THF, H2O; (f) O-benzotriazol-1-yl-N,N,N’,N’-
tetramethyluronium tetrafluoroborate [TBTU], N,N-diisopropylethylamine, DMSO,
RNH2, rt; (g) TFA.
Amide compounds were tested in the GT 1a and GT1b subgenomic
replicon assays9 and the data are shown in Table 1. In an effort to improve the
drug-like properties of polyaromatic lead molecules 1 and 2, sp3 hybridized
carbons were introduced to disrupt the planarity of the system. Introduction of a
chiral center as in α-methyl benzyl amine analogs 10 and 11 resulted in
observation of a distinct stereochemical dependence for their potency. The S-
isomer 10 was several fold more potent than either the R-isomer 11 or the achiral
analog 9. The gem-disubstituted analogs 12 and 13 also displayed weaker
potency than 10. Similar stereochemical dependence was observed for the ethyl
analogs 14 and 15. The introduction of a hydroxyl group on these amide side
chains (16-19) was poorly tolerated. When a quaternary center was introduced,
however, the R-isomer 21 retained high potency, particularly against GT 1b.
Further exploration of the substituents led to the discovery of additional
analogs with potent activity against both subgenotypes, such as the CF3 analog
22, p-fluoro analog 25, and p-bromo analog 37. In general substitution at the 4-
position of the phenyl ring yielded the most potent analogs, such as 25, 28, 34
and 37, compared to the corresponding 2- or 3-substituted analogs. Cyclization
of the R1 or R2 group onto the phenyl E ring resulted in the indane analogs 38
and 39, which also displayed strong stereochemical dependence for potency.
Interestingly, the more active enantiomer 39 (R isomer) had the opposite
stereochemistry compared to the more active acyclic enantiomer 10 (S isomer).

The introduction of heterocyclic E rings, such as pyridine in analogs 40-42
and an isosteric thiophene in analog 43, was generally not well tolerated. The
larger naphthalene analog 44 and a saturated cyclohexane analog 45 also
proved to be less potent.
Table 1.
In vitro activity of amide analogs against HCV GT 1a and GT 1b in the replicon
assay
Replicon EC50 (µM)
no.
9
10
11
12
Stereochem.
R1
H
Me
H
R2
H
H
Me
Me
E
GT 1a
0.150
0.045
0.180
0.284
GT 1b
0.130
0.007
0.100
0.181
-
S
R
-
Ph
Ph
Ph
Ph
Ph
Me
13
-
1.07
0.700
*
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
S
R
R
S
S
R
S
R
R
RS
RS
S
RS
RS
S
RS
S
S
S
S
RS
S
RS
S
Et
H
-CH2OH
H
-CH2CH2OH
H
H
Et
H
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
0.035
0.266
4.57
3.03
1.47
0.033
0.127
0.314
0.237
0.242
0.347
0.060
0.002
0.002
0.020
0.015
0.002
0.030
0.010
0.005
0.040
0.004
0.052
0.020
0.021
0.041
0.028
0.009
0.004
-CH2OH
H
-CH2CH2OH
-CH2OH
3.27
1.20
Me
-CH2OH
CF3
Me
Me
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
0.188
0.013
0.280
0.092
0.010
0.300
0.100
0.022
0.610
0.024
0.089
0.370
0.210
0.066
0.169
0.033
0.016
2-F-Ph
3-F-Ph
4-F-Ph
2-Me-Ph
3-Me-Ph
4-Me-Ph
2-OMe-Ph
3-OMe-Ph
4-OMe-Ph
2-CF3-Ph
3-CF3-Ph
4-CF3-Ph
2-Br-Ph
3-Br-Ph
4-Br-Ph
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
38
S
0.451
0.054

39
R
0.016
0.013
40
41
42
43
44
45
RS
RS
RS
RS
S
Me
Me
Me
Me
Me
Me
H
H
H
H
H
H
2-pyridine
3-pyridine
4-pyridine
0.919
1.424
1.172
0.378
0.340
0.240
0.164
0.377
0.331
0.159
0.030
0.045
2-thiophene
2-naphthalene
cyclohexane
S
Compound 10 was chosen as the starting point for additional SAR studies
on other regions of the molecule. Data from these studies are shown in Table 2.
Replacement of the A/B ring isopropyl group with H, methyl or trifluoromethyl (46,
47 and 48, respectively) resulted in less potent compounds, while the t-butyl
analog 49 retained similar potency to 10. Replacement of the sulfur linkage
between the C and D rings with oxygen (50) resulted in a significant loss of
activity (50-fold for GT1a and 100-fold for GT1b). The sulfur atom’s larger size,
greater polarizability, longer bond length, and lower bond angle (C-S-C)
compared to oxygen may result in significantly different topologies for 10 and 50,
and may account for the potency difference.11 Replacement of the D-ring amino
group with a hydroxyl group (51) resulted in a >10-fold loss of potency. While
methyl carbamate analog 52 retained good potency, both the amide and
sulfonamide analogs (53 and 54, respectively) were less potent.
Table 2.
In vitro activity of amide analogs against HCV GT 1a and GT 1b in the replicon
assay
Replicon EC50 (µM)
no.
46
47
48
49
50
51
52
53
54
R1
H
X
S
S
S
S
O
S
S
S
S
Y
NH2
NH2
NH2
NH2
GT 1a
GT 1b
7.89
6.25
Me
CF3
t-Bu
iPr
iPr
iPr
iPr
iPr
1.20
0.60
0.20
0.25
0.046
2.31
0.003
0.79
NH2
OH
0.69
0.12
NHCO2Me
NHCOMe
NHSO2Me
0.042
5.03
0.349
0.004
0.149
0.051
On the basis of their HCV subgenomic replicon potencies, compounds
were chosen for pharmacokinetic evaluation in male Sprague-Dawley rats (Table
3). Compound 10 demonstrated low clearance and moderate oral bioavailability.

Compound 21 demonstrated low clearance, although the oral bioavailability was
lower than that observed for 10. Compounds 22 and 37, which displayed uniform
1a and 1b potency, demonstrated similar pharmacokinetics to 10. Potent analogs
28 and 45 displayed higher clearance than compound 10. Methyl carbamate
analog 52 demonstrated moderate clearance, although it was not orally
bioavailable. Because of its superior rat pharmacokinetics, compound 10 was
advanced to dog pharmacokinetic studies. The results were similar to rat, with
the observation of low clearance, a good half life and similar oral bioavailability.
Table 3.
Pharmacokinetic parameters of selected compoundsa
IV
PO
no.
10
21
22
28
37
45
52
10
Species
rat
t1/2
1.6
1.1
2.9
0.7
1.9
0.5
0.4
3.2
Cl
t1/2
2.7
2.7
1.8
2.6
2.0
0.8
-
Cmax
1.48
0.32
0.93
0.16
0.33
0.20
0.0
AUC
6.49
0.90
3.89
0.60
3.84
0.25
0.0
F
0.33
0.75
0.24
1.10
0.34
1.29
0.77
0.26
42.4
13.3
18.7
12.2
24.9
5.0
rat
rat
rat
rat
rat
rat
dog
0.0
3.0
0.49
2.35
18.8
a Units: t1/2 (h); Cl (L/h/kg); Cmax (µg/mL); AUC (µg*h/mL); F (%). Doses: 5 mg/kg in rat; 2.5
mg/kg in dog.
Because adaptive mutations in NS5A that increase replication occur
during development of the replicon cell culture assay,2 it remained to be
demonstrated that compounds in this series would exert an antiviral effect in vivo
against the native virus. Two compounds, 10 and 21, were selected for further
characterization for potential administration in an HCV infected chimpanzee proof
of concept study. Compound 21 retained high potency in the presence of plasma
proteins (0.034 µM EC50 in the GT 1b subgenomic replicon assay in the
presence of with 40% human plasma), while compound 10 was considerably
weaker (1.2 µM). In addition, 21 demonstrated higher solubility(19.7 µM for
compound 21 and 5.0 µM for compound 10) in aqueous buffer and was more
suitable for IV administration in a chimpanzee. Both 10 and 21 had short half-
lives in monkey (ca. 1 h), while 21 displayed a good half-life in a chimpanzee (4.1
h). As a result, 21 was chosen for evaluation in an HCV GT1b-infected
chimpanzee. Intravenous administration of compound 21 (2.5 mg/kg, 5 single
doses, q8h) resulted in a significant viral load decline (1.65 log10 over the first 32
hours of dosing) as well as the development of resistance-associated variants in
NS5A (K26R, Q24R + K26R, L28M, Y93C + K26R).9 Mutations at tyrosine 93
have been found to confer resistance to multiple classes of NS5A inhibitors,3,4
and the Y93C mutant demonstrated 900-fold reduced susceptibility to compound
21. These observations provided positive proof of concept that compounds in this
series exert an antiviral effect in vivo.
In conclusion, we discovered new pyrido[2,3-d]pyrimidine-based HCV
NS5A inhibitors such as compound 10 that displayed potent inhibition of GT 1a
and 1b as well as oral bioavailability in rats and dogs. The introduction of a chiral

center resulted in the observation of a stereochemical dependence for HCV
replicon potency and provided a site for the introduction of functional groups
useful for improving the solubility of the series. On the basis of its high potency
against GT 1b replicon in the presence of 40% human plasma and good aqueous
solubility, compound 21 (A-972338) was selected for administration in an HCV
infected chimpanzee study. Observation of a robust viral load decline provided
positive proof of concept for inhibition of HCV replication in vivo for the
compound series.
Acknowledgement
The authors gratefully acknowledge the work of Dr. Robert Lanford for studies
conducted in the HCV-infected chimpanzee.
Disclosure Statement
This research was sponsored by AbbVie. AbbVie contributed to the research,
and interpretation of data, writing, reviewing, and approving the publication. DD,
DG, DL, JP, PK, TP-M, KM, AM, DK and CM are employees of AbbVie, and own
AbbVie stock. DB, MT, and WH own AbbVie stock.
References
1. For a recent review see: Nature Outlook: Hepatitis C. Nature 2011, 474, S1.
2. Tellinghuisen, T. L.; Marcotrigiano, J.; Rice, C. M. Nature 2005, 435, 374.
3. Schmitz, U.; Tan, S.-L. Recent Patents on Anti-Infective Drug Discovery 2008,
3, 77.
4. Lemm, J. A.; O'Boyle, D.; Liu, M.; Nower, P.T.; Colonno, R.; Deshpande, M.S.;
Snyder, L.B.; Martin, S.W.; St Laurent, D.R.; Serrano-Wu, M.H.; Romine, J.L.;
Meanwell, N.A.; Gao, M. J. Virol. 2010, 84, 482.
5. Gao, M.; Nettles, R. E.; Belema, M.; Snyder, L. B.; Nguyen, V. N.; Fridell,R. A.;
Serrano-Wu, M. H.; Langley, D. R.; Sun, J.-H.; O’Boyle II, D. R.; Lemm, J. A.;
Wang, C.; Knipe, J. O.; Chien, C.; Colonno, R. C.; Grasela, D. M.; Meanwell, N.
A.; Hamann, L. G. Nature 2010, 465, 96.
6. Lemm, J. A.; Leet, J. E.; O’Boyle II, D. R.; Romine, J. L.; Huang, X. S.;
Schroeder, D. R.; Alberts, J.; Cantone, J. L.; Sun, J.-H.; Nower, P. T.; Martin, S.
W.; Serrano-Wu, M. H.; Meanwell, N. A.; Snyder, L. B.; Gao, M. Antimicrob.
Agents Chemother. 2011, 55, 3795.
7. Chang, W.; Mosley, R. T.; Bansal, S.; Keilman, M.; Lam, A. M.; Furman, P. A.;
Otto, M. J.; Sofia, M. J. Bioorg. Med. Chem. Lett. 2012, 22, 2938.
8. Shi, J.; Zhou, L.; Amblard, F.; Bobeck, D. R.; Zhang, H.; Liu, P.; Bondada, L.;
McBrayer, T. R.; Tharnish, P. M.; Whitaker, T.; Coats, S. J.; Schinazi, R. F.
Bioorg. Med. Chem. Lett. 2012, 22, 3488.
9. Full details will be provided elsewhere. Krishnan, P. et al. Manuscript in
preparation.
10. Krueger, A. C.; Madigan, D. L.; Beno, D. W.; Betebenner, D. A.; Carrick, R.;
Green, B. E.; Wenping He, W.; Liu, D.; Maring, C. J.; McDaniel, K. F.; Mo, H.;
Molla, A.; Motter, C. E.; Pilot-Matias, T. J.; Tufano, M. D.; Kempf, D. J. Bioorg.
Med. Chem. Lett. 2012, 22, 2212.

11. Oae, S.; Tamagaki, S. In Organic Sulfur Chemistry: Structure and
Mechanism; Oae, S., Ed.; CRC Press, Inc., Boca Raton, 1992; Vol. 1, pp 1-26.
Graphical Abstract