Des-A-ring benzothiadiazines: Inhibitors of HCV genotype 1 NS5B RNA-dependent RNA polymerase

Article abstract of DOI:10.1016/j.bmcl.2008.02.064

In our program to discover non-nucleoside, small molecule inhibitors of genotype 1 HCV polymerase, we investigated a series of promising analogs based on a benzothiadiazine screening hit that contains an ABCD ring system. After demonstrating that a methyl

Full text of DOI:10.1016/j.bmcl.2008.02.064

Available online at www.sciencedirect.com
Bioorganic & Medicinal Chemistry Letters 18 (2008) 2735–2738
Des-A-ring benzothiadiazines: Inhibitors of HCV genotype 1
NS5B RNA-dependent RNA polymerase
Pamela L. Donner,* Qinghua Xie, John K. Pratt, Clarence J. Maring,
Warren Kati, Wen Jiang, Yaya Liu, Gennadiy Koev, Sherie Masse,
Debra Montgomery, Akhter Molla and Dale J. Kempf
Antiviral Research, Global Pharmaceutical Research, and Development, Abbott Laboratories, Abbott Park, IL 60064, USA
Received 31 January 2008; revised 19 February 2008; accepted 27 February 2008
Available online 4 March 2008
Abstract—In our program to discover non-nucleoside, small molecule inhibitors of genotype 1 HCV polymerase, we investigated a
series of promising analogs based on a benzothiadiazine screening hit that contains an ABCD ring system. After demonstrating that
a methylsulfonylamino D-ring substituent increased the enzyme potency into the low nanomolar range, we explored a minimum
core required for activity by truncating to a three-ring system. Described herein are the syntheses and structure–activity relationship
of a set of inhibitors lacking the A-ring of an ABCD ring system. We observed that small aromatic rings and alkenyl groups
appended to the 5-position of the B-ring were optimal, resulting in inhibitors with low nanomolar potencies.
Ó 2008 Elsevier Ltd. All rights reserved.
Hepatitis C virus infects over 3% of the world popula-
tion.1 HCV infection is a leading cause of chronic liver
disease and the leading cause of death from liver dis-
ease in the United States.2 Despite the fact that new
infections are decreasing due to better blood screening
and education in the United States and Europe, the
death rate from HCV is expected to continue to rise
through 2015.3 Hepatitis C virus (HCV) is a (+)-strand
RNA virus of the Flaviviridae family and was first iden-
tified in 1989.4 HCV has six major genotype classes,5
with genotypes 1 and 2 being the most prevalent in
the US, Europe and Japan. The goal in treating pa-
tients infected with HCV is eradication of the infection.
The current standard of care for treating HCV is a
combination therapy of pegylated interferon-a (PEG-
IFN-)/ribavirin. Sustained virologic response (SVR)
rates for HCV patients having genotypes 2 or 3 ap-
proach 80%. However, patients infected with genotype
1 HCV do not respond well to this combination ther-
apy, demonstrating SVR rates of <50% even after pro-
longed treatment.
Additionally, nearly 3 out of 4 people treated with PEG-
IFN-/ribavirin experience systemic side effects including
neutropenia, depression, irritability, headaches, nausea
and anemia.6 These factors result in a therapeutic need
for new HCV therapies, particularly those directed at
genotype 1 HCV.
Due to its necessary role in viral replication, NS5B
RNA-dependent RNA polymerase (RdRp) has been
one of the most studied viral protein targets for small
molecule HCV therapy.
O
O
O
O
H
N
O O
H
N
O O
S
S
O
OH N
S
OH N
S
D
C
N
H
N
A
N
B
H
N
O
N
O
2
1
1a IC50
2 nM
7 nM
16 nM
2 nM
8 nM
23 nM
14 nM
3 nM
1b IC50
1a Replicon EC50
1b Replicon EC50
Keywords: HCV polymerase; Benzothiadiazines; HCV genotype 1;
Antivirals.
*
Figure 1. A comparison of enzyme potencies of the 4-ring benzothi-
adiazine system and the truncated 3-ring core.
Corresponding author. Tel.: +1 847 937 3245; fax: +1 847 938
2756; e-mail: Pamela.l.donner@abbott.com
0960-894X/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2008.02.064

2736
P. L. Donner et al. / Bioorg. Med. Chem. Lett. 18 (2008) 2735–2738
Table 1. Biochemical potencies
O O
S
O O
NO2
a
c
S
NO2
H2N
H2N
O O
S
H
N
O
OH N
S
Cl
H2N
O
R
N
H
b
N
O
O O
S
O O
S
H
NH2
N
H2N
H2N
S
Compound
R
1a IC50 (nM) 1b IC50 (nM)
O O
H2N
H2N
5
6
7
2
8
H
Br
274
55
231
150
334
23
3
Phenyl
2-Furyl
2-Thiophene
425
8
Scheme 1. Reagents and conditions: (a) (NH4)2CO3, CuSO4, NH4OH,
120 °C, 4 h, sealed tube; (b) Na2S2O4, 1 N NaOH; (c) CH3SO2Cl,
pyridine, rt, 24 h.
11
36
9
24
6
95
35
OH
N
O
N
S
10
S
a,b
c
O
O
O
N
11
12
13
406
67
483
105
OH
4
d
O
O O
S
O O
480
12622
H
H
N
N
S
OH
N
N
S
O
OH
N
S
O
acetyl
O
O
Br
e
N
H
N
H
14
1
Acetyl
See Figure 1
280
2
799
7
O
N
O
5
6
Scheme 2. Reagents and conditions: (a) NaH, benzyl bromide, DMF,
rt, 20 h; (b) HCO2NH4, PdOH/C, MeOH/THF, 70 °C 2–8 h; (c)
mino D-ring substituent that increased enzyme potency
into the low nanomolar range.9
ðMeSÞ CþMeSOꢀ4 , pyridine, dioxane, 90 °C 1 h; (d) 13, dioxane, 90 °C
3
1 h; (e) N-bromohydantoin, THF, rt.
We then investigated the possibility of truncating this
four-ring system to the minimum core required for activ-
ity. Our models and published X-ray crystal structures11
indicated that the A-ring of benzothiadiazine com-
pounds fit snugly into a small hydrophobic surface of
the binding site. When the A-ring was eliminated, the
potency fell off considerably. Substitutions from the
We have investigated N-1-alkyl-4-hydroxyquinolon-3-yl
benzothiadiazines as inhibitors of HCV NS5B polymer-
ase.7 This class of compounds binds near the active site
of the enzyme in the palm domain.8 Optimization of this
series led to an analog, 1, containing a methylsulfonyla-
O O
S
O O
S
H
H
N
N
OH N
S
OH N
S
O
O
O
Ar
O
Br
a or b
N
N
H
H
N
O
N
O
6
2,7-8
a
a or b
O O
S
H
O O
S
H
N
N
O
OH N
S
OH N
S
O
O
O
O
Y
N
N
H
H
N
O
N
O
14
9-13
Scheme 3. Reagents and conditions: (a) cat. Pd(PPh3)2Cl2, appropriate
tributyltin reagent, THF, 75 °C, 20 h; (b) cat. Pd(PPh3)2Cl2, appro-
priate boronic acid reagent, 4:1.5:1 DME/H2O/EtOH, potassium
carbonate, 110 °C, 25 min, microwave.
Figure 2. A/B ring potion of benzothiadiazine 1 modeled into the
binding site of the enzyme based on reported binding orientation of
analog.11

P. L. Donner et al. / Bioorg. Med. Chem. Lett. 18 (2008) 2735–2738
2737
B-ring into the space previously occupied by the A-ring
produced analogs that maintained nanomolar enzyme
inhibitory potency in both isolated enzyme and cell cul-
ture systems (HCV subgenomic replicon). Figure 1
shows one of the most potent des-A-ring analogs 2 with
an IC50 value of less than 10 nM against genotype 1a.
shown in Scheme 2. 4-Benzyloxy-pyridin-2-ol was alkyl-
ated with 1-bromo-3-methyl butane, followed by depro-
tection to give compound 4. Construction of the
benzothiadiazine ring system was achieved by convert-
ing 4 to the dithioketene acetal12 followed by reaction
with 3 to provide 5. Bromination of 5 produced 6, which
could be coupled with a variety of boronic acids and tin
reagents using Suzuki or Stille conditions as shown in
Scheme 3 to provide analogs 2, 7–14.
Chemistry used to prepare the D-ring is shown in
Scheme 1. 2-Chloro-5-nitrobenzenesulfonamide was
converted to the amino benzenesulfonamide. Reduction
of the nitro moiety followed by a chemoselective sulf-
onylation provided benzothiadiazine ring precursor 3.
The B-ring portion of the molecule was constructed, as
Table 1 shows the activities of these inhibitors in the
HCV polymerase enzymatic assays. The unsubstituted
analog, where R = H, is 100-fold less active compared
Table 2. Replicon potencies in 40% human serum
Compound
EC50 1b replicon 5% FCS (nM)
EC50 1b replicon 40% HS (nM)
Fold Incr.
O O
S
H
N
OH
N
N
S
O O
N
5
163
911
5.6
H
O
O O
S
H
N
OH
N
N
S
O O
N
9
H
8.8
62.5
7.1
O
O O
S
H
N
OH
N
N
S
O O
N
10
5.6
38.2
6.8
H
O
O O
S
H
N
OH
N
N
S
O O
O
N
2
2.5
59.9
24
H
O
O O
S
H
N
OH
N
S
O O
N
H
1
2
736
368
N
N
O
IC50 and EC50 values are means of at least two independent determinations, standard deviation 10%. HCV 1b replicon EC50 values are mea-
surements of RNA production by real-time RT-PCR. HCV 1a replicon EC50 values are measurements using a SEAP reporter gene construct.7,10

2738
P. L. Donner et al. / Bioorg. Med. Chem. Lett. 18 (2008) 2735–2738
to the original four-ring system, 1. Presumably, this is
due to the loss of a hydrophobic contact in the binding
site of the enzyme that the A-ring provides in the 4-ring
system. Figure 2 shows this model of the A/B ring region
of the benzothiadiazine, 1, demonstating a tight hydro-
phobic fit with the surface of the enzyme in this portion
of the molecule. As hydrophobicity in that region is
added back to the molecule, potency increases. Analogs
with vinyl, methylvinyl, thiophene, and furyl substitu-
ents all have IC50 values less than 25 nM against geno-
type 1a. The remaining examples, with larger
substituents, show poorer activity. The size of the
hydrophobic binding site appears to be limited, and is
demonstrated by the fact that the thiophene and furyl
5-membered ring aromatics are in the 10 nM range,
but the analog with a phenyl substituent is 40-fold less
active. Although all of these inhibitors are slightly less
active against the genotype 1b enzyme, the trend is sim-
ilar (see Table 2).
of the antiviral potency of the original 4-ring system, in
both the biochemical and replicon assays, as our models
predicted. This new class of inhibitors can be readily
synthesized from a commercially available 4-benzyl-
oxy-pyridin-2-ol using known chemistry via a dithioke-
tene acetal intermediate followed by coupling with tin
or boronic acid reagents to provide the desired products.
Importantly, these inhibitors are less protein with rela-
tively high potencies in the HCV replicon in the presence
of human serum, and therefore address one of the key
liabilities associated with the benzothiadiazine class with
a 4-ring system of HCV polymerase inhibitors.
Acknowledgment
The authors thank Kent Stewart for modeling work.
References and notes
We also determined the potency in the replicon assays
for these inhibitors. One of the major drawbacks of
the tetracyclic benzothiadiazine system is a high degree
of protein binding. This can be illustrated by comparing
the fold difference in activity in 0% versus 40% human
serum. In Table 2, the difference between the replicon
potency in the presence of 5% FCS with no human ser-
um and the replicon potency in the presence of 40% hu-
man serum is shown for a representative set of inhibitors
in this class. We observed that the truncated des-A-ring
series exhibited a lower protein-binding effect. Thus, the
potencies of these Des-A analogs are typically attenu-
ated by 6- to 24-fold compared to 60- to 360-fold for
the benzothiadiazine four-ring system. Notably, analogs
2, 9, and 10 are 10- to 20-fold more active in 1b replicon
than the tetracyclic analog, 1, in the presence of human
serum.
1. WHO, Hepatitis C. Fact Sheet No. 164. Revised October
2000.
2. Kim, W. R. Hepatology 2002, 36, 30.
3. Pol, S.; Mallet, V. O. Expert Opin. Biol. Ther. 2006, 6, 923.
4. Choo, Q. L.; Kuo, G.; Weiner, A. J.; Overby, L. R.;
Bradley, D. W.; Houghton, M. Science 1989, 244, 359.
5. Simmonds, P. J. Hepatol. 1999, 31, 54.
6. Walker, M. P.; Yao, N.; Hong, Z. Expert Opin. Investig.
Drugs 2003, 12, 1269.
7. Pratt, J. K. Patent Application, 2004; WO2004/041818 A1.
8. Koch, U.; Narjes, F. Curr. Top. Med. Chem. 2007, 7, 1302.
9. Rockway, T. W.; Zhang, R.; Liu, D.; Betebenner, D. A.;
McDaniel, K. F.; Pratt, J. K.; Beno, D.; Montogomery,
D.; Jiang, W. W.; Massie, S.; Kati, W. M.; Middleton, T.;
Molla, A.; Maring, C. J.; Kempf, D. J. Bioorg. Med.
Chem. Lett. 2006, 16, 3833.
10. Pratt, J. K.; Donner, P.; McDaniel, K. F.; Maring, C. J.;
Kati, W. M.; Mo, H.; Middleton, T.; Liu, Y.; Ng, T.; Xie,
Q.; Zhang, R.; Montgomery, D.; Molla, A.; Kempf, D. J.;
Kohlbrenner, W. Bioorg. Med. Chem. Lett. 2005, 15, 1577.
11. Tedesco, R. et al. J. Med. Chem. 2006, 49, 971.
12. Krueger, A.; Madigan, D.; Green, B.; Hutchinson, D.,
et al. Bioorg. Med. Chem. Lett. 2007, 17, 2289.
In our investigation of a set of HCV polymerase inhib-
itors with a benzothiadiazine core, we have discovered
that a three-ring system with small substituents attached
to the 5-position of the B-ring can, indeed, retain much