Synthesis and biological characterization of B-ring amino analogues of potent benzothiadiazine hepatitis C virus polymerase inhibitors

Article abstract of DOI:10.1021/jm801485z

Benzothiadiazine inhibitors of the HCV NS5B RNA-dependent RNA polymerase are an important class of non-nucleoside inhibitors that have received considerable attention in the search for novel HCV therapeutics. Research in our laboratories has identified a

Full text of DOI:10.1021/jm801485z

3174
J. Med. Chem. 2009, 52, 3174–3183
Synthesis and Biological Characterization of B-Ring Amino Analogues of Potent
Benzothiadiazine Hepatitis C Virus Polymerase Inhibitors
John T. Randolph,* Charles A. Flentge, Peggy P. Huang, Douglas K. Hutchinson, Larry L. Klein,† Hock B. Lim,
Rubina Mondal, Thomas Reisch, Debra A. Montgomery, Wen W. Jiang,‡ Sherie V. Masse, Lisa E. Hernandez,
Rodger F. Henry, Yaya Liu, Gennadiy Koev, Warren M. Kati, Kent D. Stewart, David W. A. Beno, Akhteruzzaman Molla, and
Dale J. Kempf
Global Pharmaceutical Research and DeVelopment, Abbott, 200 Abbott Park Road, Abbott Park, Illinois 60064-6217
ReceiVed NoVember 25, 2008
Benzothiadiazine inhibitors of the HCV NS5B RNA-dependent RNA polymerase are an important class of
non-nucleoside inhibitors that have received considerable attention in the search for novel HCV therapeutics.
Research in our laboratories has identified a novel series of tetracyclic benzothiadiazine inhibitors of HCV
polymerase bearing a benzylamino substituent on the B-ring. Compounds in this series exhibit low-nanomolar
activities in both genotypes 1a and 1b polymerase inhibition assays and subgenomic replicon assays.
Optimization of pharmacokinetic properties in rat led to compound 30, which has good oral bioavailability
(F ) 56%) and a favorable tissue distribution drug profile, with high liver to plasma ratios. Compound 30
is a potent inhibitor in replicon assays, with EC50 values of 10 and 6 nM against genotypes 1a and 1b,
respectively.
Introduction
for the majority of infections in the U.S. (75% genotype 1a or
1b), Europe, Asia, and elsewhere across the globe. Other
concerns include the length of therapy (48 weeks to treat
genotype 1 infection) and a number of moderate to severe side
effects experienced by a majority of patients receiving therapy.
Thus, there exists a great need for new agents to combat HCV
infection, in particular those effective for treating patients
infected with genotype 1 virus.
HCV infection is a major concern for healthcare throughout
the world. The most recent estimate of the current size of the
problem is chronic HCV infection in some 2-3% of the world
population (over 170 million infected people).1 Because these
individuals typically remain asymptomatic for 10-20 years after
the initial infection, a large percentage of those currently infected
will likely advance to end-stage liver disease in the coming 1-2
decades.2 This expected increase in the number of HCV infected
individuals experiencing liver complications will result in a
dramatic increase in the number of cases of cirrhosis and
hepatocellular carcinoma throughout the world. In the U.S.
alone, complications due to HCV infection are estimated to
cause over 8000 deaths per year,3 a number that is expected to
increase in the coming decade because of increased mortality
within the population of currently infected individuals.4
The grim statistics of the global HCV infection problem are
in no way relieved by the current line of agents available for
treatment and/or prevention. A clinically proven vaccine for
HCV is not currently available to help control the spread of
HCV infection.5 Furthermore, the task of treating HCV infection
is complicated by the fact that there are six major genotypes,
with numerous subtypes for each genotype, making it difficult
to discover a vaccine and effective therapeutics to cover all of
these variant forms of the virus. The present standard of care is
combination therapy using pegylated interferon and ribavirin.6
While this drug combination is effective for treating patients
infected with genotype 2 or 3 virus (>80% sustained virologic
response, or SVR), it is far less effective for treating other
genotypes, including genotype 1 (<50% SVR) which accounts
Research to discover novel HCV therapeutics has focused in
large part on the identification of compounds that inhibit the
NS5B RNA-dependent RNA polymerase.7,8 A number of small
molecule inhibitors have been identified that bind to several
sites of the enzyme. These research efforts have been aided by
X-ray crystallography and molecular modeling studies of
enzyme-inhibitor complexes that have revealed details of
binding of these molecules to polymerase. One class of non-
nucleoside inhibitors that has received considerable attention
are benzothiadiazine analogues, which bind to a unique site at
the base of the so-called “palm domain”, in closer proximity to
the enzyme catalytic site than other known non-nucleoside
inhibitors.9,10 These compounds inhibit an initial step in viral
RNA synthesis and do not compete with the incorporation of
NTP.
We recently described a series of thiadiazine compounds that
exhibit low nanomolar potency as inhibitors of the NS5B
polymerase of genotype 1 HCV.11 The most active compounds
in this series bear a geminal dialkyl-substitution pattern on the
B-ring (Figure 1). Isoamyl analogue 1 was found to be inferior
to neohexyl analogue 2 both in in vitro activity (compound 2 is
1.5-fold more active than 1 against genotypes 1a and 1b
polymerase) and in pharmacokinetic properties. The advantage
of the neohexyl analogue 2 in vivo is the result of improved
metabolic stability, as it was found that tertiary alcohol 3 was
the major metabolite of 1 in in vitro metabolism studies using
hepatocytes. Thus, the neohexyl substitution pattern in 2
effectively blocks the primary site for oxidative metabolism in
1, resulting in improved exposures on oral administration in
* To whom correspondence should be addressed. Address: Antiviral
Research, Department R4CQ, Building AP52-N, 200 Abbott Park Road,
Abbott Park, IL 60064. Telephone: 847-937-7182. Fax: 847-938-2756.
E-mail: john.randolph@abbott.com.
†
Current Address: Institute for Tuberculosis Research, College of
Pharmacy, University of Illinois at Chicago, 833 S. Wood Street, Chicago,
IL 60612.
‡
Midwestern University, College of Pharmacy, 555 31st Street, Downers
Grove, IL 60515.
10.1021/jm801485z CCC: $40.75
2009 American Chemical Society
Published on Web 04/29/2009

B-Ring Amino Analogues
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 10 3175
Bicyclic intermediate 9 was reacted with tris(methylthio)-
methyl methylsulfate16 to give dithioketeneacetal intermediate
10, properly functionalized for conversion to the desired
thiadiazine tetracycles. This was accomplished by reaction with
o-aminobenzenesulfonamide 11 or 12 to provide thiadiazine
analogues substituted in the 7-position with either a BOC-
protected amino group (13) or a methanesulfonamide group (14),
respectively. The N-O bond of the methoxyamine was cleaved
by reduction using molybdenumhexacarbonyl to provide B-ring
amine compounds 15 and 16.17 Substitution of the amino group
was accomplished using standard conditions of acylation or
reductive alkylation. Compounds derived from BOC-protected
intermediate 15 were further converted to the desired 7-meth-
anesulfonylbenzothiadiazine analogues by BOC removal with
acid hydrolysis, followed by reaction with methanesulfonyl
chloride. In this way, a series of inhibitors represented by
structure 17 having a diverse set of functionality on the amino
group (R1 ) acyl, alkyl, sulfonyl, carbamoyl, etc.) was prepared.
Figure 1. B-Ring dialkylthiadiazine HCV polymerase inhibitors.
Asymmetric synthesis of chiral AB ring portions of these
compounds is shown in Scheme 2. Reaction of allyl meth-
oxyamine intermediate 6 with D-glyceraldehyde acetonide gave
a mixture of two stereoisomeric oxazolidine analogues 18 and
19, which were readily separated by chromatography on silica
gel. Crystallization gave each of these compounds in high
configurational purity as determined by 1H NMR analysis. Chiral
intermediates 18 and 19 were converted to chiral isoamyl-
substituted AB ring fragments 20 and 21 using the methods
previously described for the synthesis of 9. Further elaboration
of each of these enantiomeric methoxyamines to give tetracyclic
benzothiadiazine derivatives was accomplished using the meth-
ods shown in Scheme 1. The resulting chiral inhibitors were
obtained in high optical purity as determined by chiral HPLC
analysis.18
Figure 2. Molecular model showing 1 bound to HCV polymerase,
which revealed that the B-ring methyl group projected toward a polar
region of the enzyme occupied by residues Arg 386 and Ser 367.
rat (3.5-fold improvement in AUC for 2 over 1 following a 5
mg/kg oral dose).
We used molecular modeling to generate a model of inhibitor
1 occupying the thiadiazine binding site of HCV polymerase
(Figure 2). The planar ABCD ring structure was oriented in
like manner to that reported for crystallographic studies of
enzyme-inhibitor complexes of other thiadiazine inhibitors.12
This model places the isoamyl side chain on the B-ring into a
hydrophobic pocket defined by residues Leu 384, Pro 197, and
Tyr 415. This orientation revealed that the methyl side chain
on the B-ring was projected toward a polar region of the enzyme
occupied by residues Arg 386 and Ser 367. There have been
no reports, to date, of inhibitors that occupy this region of the
enzyme. We envisioned a strategy for exploring this region of
the enzyme by replacing the B-ring methyl group with an amino
function that would allow for facile late-stage substitution to
rapidly prepare analogues for testing. It was our hope that such
modifications might result in inhibitors with improved properties,
such as better in vitro potency, improved pharmacokinetics, and/
or improved physicochemical properties.
Neohexyl B-ring amino analogues were synthesized as shown
in Scheme 3. Alkylation of 4 using an organotitanium reagent
generated by the addition of neohexyl Grignard to titanium(IV)
isopropoxide at -78 °C afforded tertiary alcohol intermediate
22.19 Compound 22 was converted to thiadiazine 23 by the two-
step sequence of reactions involving initial formation of the
dithioketeneacetal intermediate using tris(methylthio)methyl
methylsulfate,16 followed by reaction with aminosulfonamide
12. Acetamide 24 was prepared from 23 by a Ritter reaction in
acetonitrile in essentially quantitative yield.20 Acid hydrolysis
removed the acetyl group to give primary amine 25, which was
further converted to secondary amine analogues 26 by reductive
alkylation with a variety of aldehydes. This method provided a
series of compounds for testing having a variety of functionality
extending from the B-ring amino group.
Chemistry
Chiral resolution was used to separate racemic amine 25 into
the individual enantiomers as shown in Scheme 4. Acylation
of 25 with chloroformate 27, prepared from (R)-4-phenylox-
azolidin-2-one by reaction with diphosgene, gave a 1:1 mixture
of diastereomers that was separated by column chromatography
on silica gel.21 This method allowed the desired diastereomer
28, having the S-configuration at the tertiary carbon of the
B-ring, to be isolated in excellent optical purity (>98% de).22
The oxazolidinone ring was displaced using lithium trimethyl-
silylethoxide to give the desired amine protected as the
trimethylsilylethylcarbamate, which was removed with TFA to
give chiral amine 29, isolated as the TFA salt.23 Reductive
alkylation of 29 using 2,6-dimethylbenzaldehyde afforded 30.
Synthesis of an amino derivative of isoamyl analogue 1 is
shown in Scheme 1. 2-Hydroxy-1,4-naphthoquinone (4) was
converted to E-oxime analogue 5 by reaction with meth-
oxyamine in methanol, followed by allylation using an allylin-
dium organometallic reagent generated by the addition of allyl
bromide to a suspension of indium metal in DMF.13 The
resulting allyl intermediate 6 was reacted with isobutyraldehyde
to give oxazolidine intermediate 7.14 The allyl chain of 7 was
extended by olefin metathesis using 2-methyl-2-butene and
Hoveyda-Grubbs second generation catalyst15 to give prenyl
analogue 8. Hydrogenation of 8 using Adam’s catalyst provided
an intermediate with the desired isoamyl chain, which was
further elaborated by acid hydrolysis to give methoxyamine 9.

3176 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 10
Randolph et al.
Scheme 1. Synthesis of Isoamyl-Substituted Analogues of B-Ring Amino Tetracyclic Thiadiazine Inhibitors of HCV Polymerase
Scheme 2. Synthesis of Chiral Isoamyl AB Ring Fragments
Chiral HPLC analysis of 30 confirmed that this compound
contained only trace amounts of the R-enantiomer (<0.5%).
types 1a (H77) and 1b (BK) HCV NS5B was determined by
measuring the amount of H-UTP incorporated into RNA by
3
scintillation counting. IC50 values for inhibitors were calculated
using a standard IC50 equation.
Biology
Biochemical activity of compounds was determined by
measuring their ability to inhibit the formation of RNA products
in a standard polymerase inhibition assay. Inhibition of geno-
Cell-culture activity of compounds was determined using
subgenomic replicons transfected into Huh-7 cells. The ability
of compounds to inhibit replication of genotype 1a (H77)

B-Ring Amino Analogues
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 10 3177
Scheme 3. Synthesis of Neohexyl-Subtituted Analogues of B-Ring Amino Thiadiazine Inhibitors of HCV Polymerase
Scheme 4. Chiral Resolution of Neohexyl B-Ring Aminothiadiazine Inhibitors
and 1b (Con1) replicons expressing the firefly luciferase gene
was determined by measuring the level of luciferase in the
cell lysate using a luminometer. Replicon inhibition activity,
reported as EC50, was measured both in the absence of human
serum (HS) and in the presence of 40% HS in order to
determine the extent to which inhibitor-protein interactions
affect compound potency.
both intravenous and oral administration of a 5 mg/kg dose.
Compounds were formulated using DMSO/PEG400 (1:9,v/v).
Results and Discussion
A series of isoamyl-substituted B-ring amino analogues were
prepared and evaluated for their ability to inhibit RNA synthesis
catalyzed by genotype 1a NS5B.24 Compounds with IC50 < 30
nM were further evaluated for their ability to inhibit the 1a
replicon assay. A summary of SAR for this series is shown in
Pharmacokinetic properties of compounds were determined
using Sprague-Dawley rats (n ) 3 for each experiment) via

3178 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 10
Randolph et al.
Table 2. Comparison of Activities against Genotype 1a and
Table 1. Activity Data for Analogues of Isoamyl-Substituted B-Ring
Aminothiadiazine Inhibitors of HCV Polymerase
Pharmacokinetic Properties in Rat for Chiral Isoamyl B-Ring Amino
Analogues
IC50 (nM)/a
c
d
compd
R
EC50 (nM)b
CLp
t1/2
AUCe
F, %f
31
32
33
34
OMe
H
Bn
6/12
4/37
10/16
7/56
2.7
0.6
0.3
1.7
0.2
0.35
0
0.6
0
34.1
0
13.3
0
2.12
1.16
4.3
Ms
a 1a enzyme activity. b 1a replicon activity. c Clearance (L/(h ·kg))
following 5 mg/kg iv dose in rat. d Half-life in hours following 5 mg/kg iv
dose in rat. e AUC0-∞ following 5 mg/kg oral dose (µg·h/mL). f Percent
bioavailable following 5 mg/kg oral dose in rat.
in replicon, whereas benzylamine 33 lost less than 2-fold activity
in replicon (EC50 ) 16 nM) relative to activity in the enzyme
inhibition assay. Methanesulfonamide 34 gave a similar profile
to unsubstituted amine 32, with potent inhibition in the
biochemical assay (IC50 ) 7 nM) but substantially less activity
in replicon (EC50 ) 56 nM).
Of the compounds tested, only methoxyamine analogue 31
and benzylamine analogue 33 were found to be bioavailable
following oral administration in rat. Compound 33 gave the best
overall pharmacokinetic profile, with better overall exposure, a
longer half-life (1.7 h), and a more moderate rate of clearance
following iv dosing. Primary amine 32 and methanesulfonamide
34 gave no measurable plasma concentrations following oral
dosing.
In an effort to improve the plasma concentration of these
compounds in rat following oral administration, the isoamyl side
chain on the B-ring was replaced with a neohexyl group. This
strategy was based on a discovery made during our investigation
of B-ring dialkyl tetracyclic thiadiazine inhibitors,11 where it
was determined that the neohexyl side chain blocked a site for
oxidative metabolism in the isoamyl side chain (see Figure 1
and the accompanying discussion). Given the success of
benzylamino analogues in the isoamyl series (i.e., compound
33), our attention in the neohexyl series focused on arylmethy-
lamino derivatives. Thus, a series of compounds was synthesized
in parallel from neohexylamine intermediate 25 by reductive
alkylation with a variety of carboxaldehydes to give arylmethyl
derivatives 26 (see Scheme 3).
The SAR for compounds in the neohexyl series of inhibitors
is summarized with the examples shown in Table 3. A
comparison of benzamide 26a and benzylamine 26b confirms
that amide analogues are substantially less active in both the
neohexyl and isoamyl series of inhibitors. In general, it was
found that a wide range of functionality for the arylmethyl group
provided potent inhibition of 1a polymerase. All of the
compounds in this series were found to have IC50 values against
1a polymerase of less than 50 nM. A particularly attractive
property of these compounds was the excellent activity in the
1a replicon assay, which tended to be in good agreement with
activity in the biochemical assay. Replicon EC50 values tended
to be no greater than 2- to 3-fold higher than the enzyme IC50
values, with many compounds giving equal or greater inhibition
a 1a enzyme activity. b 1a replicon activity; ND ) not determined.
Table 1. Primary amine 16 is a potent inhibitor in both the
enzyme (IC50 ) 12 nM) and replicon (EC50 ) 29 nM) assays.
Methoxyamine derivative 14 is equally active to 16 in the
biochemical assay, and 2 times more active in replicon (EC50
) 16 nM). Compound 17a, the N-acetyl derivative of 14, was
several times less active than 14. In general, amide analogues
were found to be somewhat less potent inhibitors of polymerase,
as demonstrated by benzamide 17b (IC50 ) 108 nM). Carba-
mates are more active than amides in the biochemical assay
(see methylcarbamate 17c, IC50 ) 25 nM). However, 17c is
significantly less active as an inhibitor of the replicon assay
(EC50 ) 210 nM). The same is true for urea analogues, such as
17d, which has good enzyme activity (IC50 ) 27 nM) but
relatively poor replicon activity (EC50 ) 576 nM). Sulfonamide
analogues, such as methanesulfonamide 17e, were somewhat
more potent inhibitors of replicon (EC50 ) 69 nM for 17e) than
acylated derivatives such as 17c and 17d.
Alkylation of the B-ring amine resulted in compounds with
improved activity in the replicon assay relative to other
substituted analogues investigated. It was found that a wide
range of alkyl groups provided potent inhibition of polymerase
function. Of the secondary amine analogues investigated, benzyl
analogues such as 17f and 17g gave the best overall perfor-
mance, since they were found to be potent inhibitors in both
the enzyme and replicon assays.
Single enantiomers of a select group of the more potent
compounds from this series were prepared to determine both
in vitro activity and pharmacokinetic properties in rat.18 The
data are summarized in Table 2. The S-enantiomer meth-
oxyamine analogue 31 (IC50 ) 6 nM) was found to be 10 times
more potent as an inhibitor of genotype 1a polymerase than
the R-enantiomer (IC50 ) 64 nM), confirming our expectation
that the S-isomer of these compounds would be the more active
component.22 Amine derivatives 32 and 33 were both found to
be potent inhibitors of 1a polymerase, with the primary amine
32 (IC50 ) 4 nM) showing somewhat better potency than
benzylamine derivative 33 (IC50 ) 10 nM). However, unsub-
stituted amine 32 showed a nearly 10-fold decline in activity

B-Ring Amino Analogues
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 10 3179
Table 3. Activity Data for Analogues of Neohexyl-Substituted B-Ring
Aminothiadiazine Inhibitors of HCV Polymerase
Table 5. Comparison of in Vitro and in Vivo Properties of Compounds
30 and 33
compd
R
Ar
33
30
H
Me
Ph
2,6-di-Me-Ph
polymerase IC50 (nM)a
replicon EC50 (nM)b
compd
1a
1b
1a
1b
1b (HS)c
33
30
10 ( 4
16 ( 7
16 ( 8
19 ( 7
68 ( 31
10 ( 6
15 ( 1
6 ( 0
273
270
d
e
f
t1/2
CLp
Cmax
AUCg
0.6 ( 0.4
F, %h
33 1.7 ( 0.2 1.16 ( 0.09 0.08 ( 0.03
30 3.1 ( 0.6 0.70 ( 0.19 0.62 ( 0.28 4.21 ( 1.30 55.9 ( 17.2
13.3 ( 8.5
1 hi
6 hi
12 hi
33
30
3.59 ( 0.75 (12)
5.86 ( 0.41 (12)
2.35 ( 0.22 (6)
4.36 ( 1.09 (15)
0.88 ( 0.16 (3)
2.11 ( 0.24 (11)
a Inhibition of isolated enzyme activity. b Inhibition of subgenomic
replication in Huh-7 cells. c Inhibition measured in the presence of 40%
human serum. d Half-life in hours following 5 mg/kg iv dose in rat.
e Clearance (L/(h·kg)) following 5 mg/kg iv dose in rat. f Cmax following 5
mg/kg oral dose in rat (µg/mL). g AUC0-∞ following 5 mg/kg oral dose in
rat (µg·h/mL). h Percent bioavailable following 5 mg/kg oral dose in rat.
i Compound concentration in liver (µg/g) at time indicated after dosing;
liver/plasma ratio in parentheses.
a 1a enzyme activity. b 1a replicon activity; ND ) not determined.
of the superior in vivo performance of this compound, dimeth-
ylbenzyl analogue 26l was selected for chiral synthesis to
evaluate the more active S-enantiomer.
Table 4. Pharmacokinetic Properties in Rat for Neohexyl-Substituted
B-Ring Aminothiadiazines
a
b
c
compd
t1/2
CLp
Cmax
AUCd
F, %e
Activity data against genotypes 1a and 1b polymerase in both
enzyme inhibition and replicon assays for chiral neohexyl
analogue 30 are compared with data for chiral isoamyl analogue
33 in Table 5. While both compounds had similar activity as
inhibitors of genotypes 1a and 1b polymerase, neohexyl
analogue 30 was found to be significantly more active than
isoamyl analogue 33 in 1a and 1b replicon assays. However,
when replicon activity was measured in the presence of 40%
human serum, both compounds gave a similar EC50 value. This
large serum-attenuating effect on replicon activity (17-fold and
47-fold for 33 and 30, respectively) was typical for inhibitors
in this series.
Pharmacokinetic studies in rat revealed that the in vivo
performance of neohexyl analogue 30 was significantly im-
proved relative to isoamyl analogue 33. A comparison of iv
and po data is shown in Table 5 and Figure 3. Following a 5
mg/kg iv dose, the plasma half-life increased to 3.1 h for 30,
versus 1.7 h for 33. Furthermore, 30 was cleared from the
plasma more slowly (0.7 L ·h/kg) than 33 (1.16 L ·h/kg). Plasma
exposure of 30 following a 5 mg/kg oral dose was greatly
enhanced relative to 33, with a 7-fold increase in both maximum
concentration (Cmax) and overall exposure (AUC) for 30. The
oral bioavailability for 30 was 55.9%, a 4-fold increase relative
to 33.
26c
26j
26l
26q
26r
26u
3.2
2.2
5.0
1.0
0.5
2.2
1.5
1.2
0.78
1.7
5.9
1.4
0.40
0.08
0.46
0.15
0.0
1.6
0.41
3.5
1.34
0.0
1.02
48.0
8.9
53.7
33.8
0.0
0.18
28.1
a Half-life in hours following 5 mg/kg iv dose in rat. b Clearance (L/
c
(h·kg)) following 5 mg/kg iv dose in rat. Cmax following 5 mg/kg oral
dose in rat (µg/mL). d AUC0-∞ following 5 mg/kg oral dose in rat
(µg·h/mL). e Percent bioavailable following 5 mg/kg oral dose in rat.
in replicon. Benzylamine analogues having a wide range of one
or more substituents all resulted in similar activity in both
biochemical and replicon assays. Potent assay inhibition was
maintained regardless of the position of the substituent or the
nature of the substituting group (electron withdrawing or
donating, charged groups, etc.). Larger substituents, such as the
benzyl ether group in 26h, did result in a significant reduction
in replicon activity. A variety of heteroaryl analogues, as well
as bicyclic aryl and heteroaryl analogues, all gave similar
activities in both assays.
Pharmacokinetic properties in rat for a subset of compounds
from the neohexyl-substituted series of inhibitors are shown in
Table 4. Dimethylbenzyl analogue 26l had properties that were
significantly better than other compounds tested, including a
longer half-life, lower rate of clearance, and greater plasma
exposure and bioavailability following oral dosing. On the basis
Because hepatitis infection affects the liver, it is generally
considered to be a desirable property that HCV drugs have organ

3180 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 10
Randolph et al.
spectrometer. Chemical shifts are reported in ppm (δ) and refer-
enced to an internal standard of tetramethylsilane (δ 0.00 ppm).
1H-1H couplings are assumed to be first-order, and peak multiplici-
ties are reported in the usual manner. MS analysis was conducted
using a Finnigan SSQ7000 (ESI) mass spectrometer.
4-(3,3-Dimethylbutyl)-3,4-dihydroxynaphthalen-1(4H)-one (22). To
a stirred mixture of Mg turnings (2.9 g, 119 mmol) in anhydrous
THF (150 mL) at 0 °C was added neohexyl chloride (14.2 g, 118
mmol), followed by dropwise addition of 1,2-dibromoethane (2.5
mL, 29 mmol). The resulting mixture was stirred at room temper-
ature under N2 for 3 days, during which time a nearly homogeneous
solution was obtained. The mixture was cooled to -78 °C, and
titanium(IV) isopropoxide (33.9 mL, 115 mmol) was added
dropwise. The resulting opaque yellow solution was stirred at -78
°C for 1 h, and then a solution of 2-hydroxy-1,4-naphthoquinone
(5.0 g, 28.7 mmol) in anhydrous THF (80 mL), precooled to 0 °C,
was added dropwise via canula over 20 min. Following the addition,
the resulting mixture was allowed to slowly warm to room
temperature and was stirred under N2 overnight. The mixture was
poured into 1 N aqueous HCl (500 mL) and extracted with ethyl
acetate (2 × 300 mL). The combined organic layers were dried
over Na2SO4, filtered, and concentrated in vacuo. The crude product
was subjected to column chromatography on Florisil, eluting with
a gradient of 1-5% MeOH in CH2Cl2. The still impure product
was purified by column chromatography on silica gel using 1:1
ethyl acetate/hexanes as eluent to give 22 as an orange-yellow solid
Figure 3. Plasma concentrations following administration of a 5 mg/
kg iv or oral dose of 30 or 33 in rat.
distribution properties that favor exposure in the liver. In order
to determine the extent to which these inhibitors are present in
the target organ, we measured the concentration of 30 and 33
in liver following a 5 mg/kg oral dose in rat. Both compounds
gave good liver concentrations at 1, 6, and 12 h after dosing,
with neohexyl analogue 30 giving significantly enhanced levels
relative to isoamyl analogue 33 at each time point. In the case
of neohexyl analogue 30, a liver/plasma ratio of >10 was
maintained through 12 h after dosing. Furthermore, the liver
concentration at 12 h (3.25 µM) for 30 exceeds the measured
1b replicon serum-adjusted EC50 (0.27 µM) by 12-fold, sug-
gesting this compound may be suitable for convenient dosing
regimens.
1
(2.06 g, 28%): H NMR (DMSO-d6) δ 7.82 (d, J ) 6.6 Hz, 1H),
7.61-7.69 (m, 2H), 7.57 (t, J ) 7.0 Hz, 1H), 7.35-7.45 (m, 1H),
5.47 (s, 1H), 4.08 (s, 1H), 1.60-2.07 (m, 2H), 0.74-0.98 (m,
1H), 0.68 (s, 9H), 0.47 (m, 1H).
N-[3-[4-(3,3-Dimethylbutyl)-1,4-dihydroxy-3-oxo-3,4-dihy-
dronaphthalen-2-yl]-1,1-dioxido-4H-1,2,4-benzothiadiazin-7-
yl}methanesulfonamide (23). To a solution of 22 (5.75 g, 22.1
mmol) in anhydrous 1,4-dioxane (100 mL) was added tris(meth-
ylthio)methyl methylsulfate16 (29.0 g, 110 mmol) and pyridine (8.7
mL, 110 mmol). The resulting mixture was stirred at 65 °C for 2 h
and then concentrated in vacuo. The crude product was purified
by column chromatography on silica gel using 1:1 ethyl acetate/
hexanes to give a yellow oil (7.28 g, 90%). The oil was dissolved
in anhydrous 1,4-dioxane (100 mL), and to the solution was added
12 (5.78 g, 21.8 mmol). The resulting mixture was stirred at 80 °C
for 16 h and then cooled to room temperature to give 23 as a
crystalline solid that was collected by filtration and dried (8.5 g,
77%): 1H NMR (DMSO-d6) δ 13.33 (br s, 1H), 10.16 (s, 1H), 8.03
(d, J ) 7.7 Hz, 1H), 7.65-7.75 (m, 2H), 7.47-7.62 (m, 4H), 3.57
(s, 1H), 3.06 (s, 3H), 1.89-2.04 (m, 1H), 1.74-1.89 (m, 1H),
0.87-1.04 (m, 1H), 0.62-0.80 (m, 10H); MS (ESI) m/z 534 (M +
H)+.
N-(1-(3,3-Dimethylbutyl)- 4-hydroxy-3-{7-[(methylsulfonyl)-
amino]-1,2-dioxido-4H-1,2,4-benzothiadiazin-3-yl}-2-oxo-1,2-di-
hydronaphthalen-1-yl}acetamide (24). To a vigorously stirred
solution of 23 (5.7 g, 10.7 mmol) in CH3CN (180 mL) at 0 °C was
added concentrated H2SO4 (60 mL) dropwise over 20 min. The
resulting mixture was allowed to warm to room temperature, stirred
for 4 h, and then poured onto crushed ice (∼400 mL). The mixture
was allowed to warm to room temperature, and the product was
collected by filtration to give 24 as an off-white solid (6.1 g, 99%):
1H NMR (DMSO-d6) δ 13.65 (br s, 1H), 10.26 (s, 1H), 9.15 (s,
1H), 8.12 (d, J ) 7.0 Hz, 1H), 7.69-7.79 (m, 2H), 7.47-7.65 (m,
4H), 3.06 (s, 3H), 2.50 (s, 3H), 1.74-2.06 (m, 2H), 0.84-1.00
(m, 1H), 0.66 (s, 9H), 0.39-0.58 (m, 1H); MS (ESI) m/z 575 (M
+ H)+.
Conclusions
Our continued investigations into novel inhibitors of HCV
NS5B polymerase of the benzothiadiazine class have provided
B-ring amino analogues with promising characteristics as
potential drug candidates for the treatment of HCV infection.
As predicted on the basis of molecular modeling studies of
B-ring dialkylthiadiazine inhibitors, substitution of this site was
well tolerated, with groups ranging in size from amine to
substituted benzylamine giving low-nanomolar inhibitors of
genotypes 1a and 1b polymerase. Benzylamine analogues were
found to be potent inhibitors of replication in subgenomic
replicon assays such that the cell-culture activity for these
compounds was within a few-fold of enzyme inhibition results.
Neohexyl analogue 30 had good pharmacokinetic properties in
rat following oral dosing, with an oral bioavailability of 56%.
Given results previously described, in which the neohexyl side
chain was found to block a primary site for oxidative metabolism
in isoamyl analogues of B-ring dialkylthiadiazine HCV poly-
merase inhibitors, it is likely that the improvement in PK
performance for 30 over isoamyl analogue 33 is the result of
the neohexyl side chain affording a similar advantage for B-ring
aminothiadiazine inhibitors. The tissue distribution properties
of both 30 and 33 make them attractive as potential HCV
therapeutics for patients infected with genotype 1 virus.
N-{3-[4-Amino-1-hydroxy-4-(3,3-dimethylbutyl)-3-oxo-3,4-di-
hydronaphthalen-2-yl]-1,1-dioxido-4H-1,2,4-benzothiadiazin-7-
yl}methanesulfonamide Hydrochloride (25). To a solution of 24
(6.1 g, 10.6 mmol) in 1,4-dioxane (60 mL) was added 4 N aqueous
HCl (30 mL), and the resulting mixture was heated at 80 °C for 5
days. The cooled mixture was concentrated in vacuo to ∼20 mL
and was extracted with 3:1 CH2Cl2/2-PrOH (3 × 50 mL). The
combined organic layers were dried over Na2SO4, filtered, and
concentrated in vacuo to give an oil that was suspended in ethyl
Experimental Section
General Methods. Reagents and solvents, including anhydrous
solvents, were obtained from commercial sources and used as
supplied. Column chromatography was carried out on silica gel.
Chiral chromatography was carried out using a Chiralpak AD-H
1
column (0.46 cm × 25 cm), Daicel Chemical Ind., Ltd. H NMR
spectra were measured using a Bruker AMX 300 (300 MHz) NMR

B-Ring Amino Analogues
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 10 3181
acetate (∼100 mL) to give a solid that was filtered and dried to
and was obtained as a light-yellow solid (1.35 g): 1H NMR (DMSO-
d6) δ 13.60 (br s, 1H), 10.20 (s, 1H), 8.77 (s, 1H), 8.09 (d, J ) 7.0
Hz, 1H), 7.46-7.74 (m, 6H), 7.32-7.41 (m, 2H), 7.18-7.31 (m,
3H), 5.30 (dd, J ) 8.3, 3.1 Hz, 1H), 4.78 (t, J ) 8.5 Hz, 1H), 4.17
(dd, J ) 8.6, 3.1 Hz, 1H), 3.07 (s, 3H), 1.90-2.01 (m, 2H),
0.98-1.13 (m, 1H), 0.81-0.95 (m, 1H), 0.71 (s, 9H); MS (ESI)
m/z 722 (M + H)+.
1
afford 25 as a light-yellow solid (6.0 g, 99%): H NMR (DMSO-
d6) δ 14.44 (br s, 1H), 9.95 (s, 1H), 8.62 (br s, 2H), 8.13 (d, J )
7.7 Hz, 1H), 7.59-7.74 (m, 2H), 7.48-7.59 (m, 2H), 7.42-7.48
(m, 1H), 7.33-7.39 (m, 1H), 3.01 (s, 3H), 1.94-2.12 (m, 1H),
1.76-1.94 (m, 1H), 1.01-1.18 (m, 1H), 0.70-0.80 (m, 1H), 0.69
(s, 9H); MS (ESI) m/z 533 (M + H)+.
N-(1-(3,3-Dimethylbutyl)-4-hydroxy-3-{7-[(methylsulfonyl)-
amino]-1,1-dioxido-4H-1,2,4-benzothiadiazin-3-yl}-2-oxo-1,2-di-
hydronaphthalen-1-yl)benzamide (26a). To a vigorously stirred
suspension of 23 (20 mg, 38 µmol) in PhCN (0.2 mL) at 0 °C was
added concentrated H2SO4 (0.1 mL) dropwise. The resulting mixture
was allowed to warm to room temperature and stirred for 3 h. The
mixture was added to ice (∼2 mL) and was allowed to warm to
room temperature. The resulting mixture was extracted with Et2O
(3 × 2 mL), and the combined organic layers were concentrated to
give a solid. The crude product was purified by crystallization from
N-{3-[4-(S)-Amino-4-(3,3-dimethylbutyl)-1-hydroxy-3-oxo-
3,4-dihydronaphthalen-2-yl]-1,1-dioxido-4H-1,2,4-benzothiadi-
azin-7-yl}methanesulfonamide Trifluoroacetic Acid Salt (29). To
a solution of 2-(trimethylsilyl)ethanol (0.71 mL, 8.6 mmol) in
anhydrous THF (5 mL) under N2 at 0 °C was added a 1.6 M solution
of n-butyllithium in hexane (5.4 mL, 8.6 mmol), dropwise over 10
min. Following the addition, the reaction mixture was stirred for
an additional 30 min at 0 °C before a solution of 28 (0.62 g, 0.86
mmol) in anhydrous THF (5 mL) was added dropwise. Following
the addition, the mixture was allowed to warm to room temperature
and was stirred at room temperature for 16 h. The mixture was
poured into ethyl acetate (50 mL) and washed with 0.1 N aqueous
HCl (50 mL) and H2O (50 mL). The organic layer was dried over
Na2SO4 and concentrated in vacuo. The crude product was purified
by column chromatography on silica gel using 1:4 ethyl acetate/
hexanes as the eluent to give the trimethylsilylethylcarbamate of
29 as a yellow solid (0.40 g, 69%). This solid was dissolved in
CH2Cl2 (4 mL), and the solution was cooled to 0 °C before
trifluoroacetic acid (2 mL) was added dropwise. The reaction
mixture was stirred at 0 °C for 1 h and was then allowed to warm
to room temperature and was stirred for 1 h. The solution was
concentrated in vacuo to afford 29 as a yellow solid (0.38 g,
1
Et2O/hexanes to give 26a as a colorless solid (22 mg, 92%): H
NMR (DMSO-d6) δ 13.67 (br s, 1H), 10.26 (s, 1H), 9.54 (s, 1H),
8.16 (d, J ) 7.7 Hz, 1H), 7.40-7.94 (m, 11H), 3.08 (s, 3H),
2.02-2.28 (m, 2H), 0.91-1.07 (m, 1H), 0.68 (s, 9H), 0.42-0.55
(m, 1H); MS (ESI) m/z 637 (M + H)+.
General Method for Reductive Alkylation of 25 To Give
26b-v.25 N-{3-[4-(Benzylamino)-4-(3,3-dimethylbutyl)-1-hydroxy-
3-oxo-3,4-dihydronaphthalen-2-yl]-1,1-dioxido-4H-1,2,4-ben-
zothiadiazin-7-yl}methanesulfonamide (26b). To a suspension of
25 (0.12 g, 0.21 mmol) in dichloroethane (2 mL) was added N,N-
diisopropylethylamine (73 µL, 0.42 mmol). The mixture was stirred
at room temperature until it was homogeneous (approximately 5
min), and then benzaldehyde (38 µL, 3.2 mmol) and MgSO4 (25
mg) were added. The reaction mixture was heated at 50 °C for
18 h. The mixture was cooled to room temperature and filtered,
and the filtrate was treated with sodium triacetoxyborohydride (72
mg, 0.34 mmol) and AcOH (36 µL, 0.63 mmol). The mixture was
stirred at room temperature for 18 h, diluted with CH2Cl2 (2 mL),
washed with H2O (2 × 2 mL), and dried over Na2SO4. The drying
agent was filtered off, and the filtrate was concentrated in vacuo to
give a crude product that was purified by column chromatography
on silica gel, eluting with 5% MeOH in CH2Cl2 to give 26b as a
light-yellow solid (84 mg, 60%): 1H NMR (DMSO-d6) δ 13.99 (s,
1H), 9.98 (s, 1H), 9.61 (s, 1H), 8.18-8.24 (m, 1H), 7.94 (d, J )
8.1 Hz, 1H), 7.68-7.77 (m, 1H), 7.62 (t, J ) 7.5 Hz, 1H),
7.45-7.56 (m, 2H), 7.28-7.43 (m, 6H), 3.43 (d, J ) 12.9 Hz,
2H), 3.02 (s, 3H), 2.20-2.36 (m, 1H), 1.90-2.05 (m, 1H),
0.98-1.14 (m, 1H), 0.67 (s, 9H), 0.39-0.55 (m, 1H); MS (ESI)
m/z 623 (M + H)+.
2-Oxo-4-(R)-phenyloxazolidin-3-carbonyl Chloride (27). To
a solution of (R)-(-)-4-phenyloxazolidin-2-one (5.0 g, 30.6 mmol)
in anhydrous THF (200 mL) at 0 °C under N2 was slowly added a
1.6 M solution of n-butyllithium in hexane (19.0 mL, 30.4 mmol).
The mixture was stirred at 0 °C for 30 min and then cooled to -78
°C. Trichloromethyl chloroformate (7.8 mL, 45.9 mmol) was added,
and the resulting mixture was stirred at -78 °C for 30 min and
then allowed to warm to room temperature. The mixture was
concentrated in vacuo, and the resulting residue was taken up in
CH2Cl2 (200 mL) and filtered. The filtrate was concentrated in vacuo
to afford 27 as an off-white solid (4.5 g, 65%): 1H NMR (DMSO-
d6) δ 7.29-7.45 (m, 5H), 4.89-4.99 (m, 1H), 4.68 (t, J ) 8.5 Hz,
1H), 4.00 (dd, J ) 8.5, 6.6 Hz, 1H).
1
quantitative): H NMR (DMSO-d6) δ 14.45 (s, 1H), 9.95 (s, 1H),
8.60 (br s, 3H), 8.13 (d, J ) 7.4 Hz, 1H), 7.28-7.72 (m, 6H), 3.01
(s, 3H), 1.95-2.10 (m, 1H), 1.78-1.94 (m, 1H), 0.99-1.17 (m,
1H), 0.65-0.92 (m, 10H); MS (ESI) m/z 533 (M + H)+; [R]D -78°
(c 0.5, MeOH).
N-{3-[4-(S)-[(2,6-Dimethylbenzyl)amino]-4-(3,3-dimethylbu-
tyl)-1-hydroxy-3-oxo-3,4-dihydronaphthalen-2-yl]-1,1-dioxido-
4H-1,2,4-benzothiadiazin-7-yl}methanesulfonamide (30). To a
suspension of 29 (0.38 g, 0.59 mmol) in anhydrous dichloroethane
(6 mL) was added N,N-diisopropylethylamine (0.22 mL, 1.23
mmol). The mixture was stirred at room temperature until it was
homogeneous before 2,6-dimethylbenzaldehyde (0.17 g, 1.23 mmol)
and MgSO4 (0.15 g) were added. The reaction mixture was heated
at 50 °C for 18 h, cooled to room temperature, and filtered. To the
filtrate was added sodium triacetoxyborohydride (0.21 g, 0.99
mmol) and glacial acetic acid (0.11 mL, 1.86 mmol), and the
resulting mixture was stirred at room temperature for 18 h. The
reaction mixture was diluted with CH2Cl2 (25 mL), washed with
H2O (2 × 10 mL), dried over Na2SO4, and concentrated in vacuo.
The crude product was purified by column chromatography on silica
gel using 3% MeOH in CH2Cl2 as the eluent to give 30 as a light-
yellow solid (0.22 g, 58%): 1H NMR (DMSO-d6) δ 13.96 (s, 1H),
9.98 (s, 1H), 9.25 (br s, 1H), 8.22 (dd, J ) 7.7, 1.5 Hz, 1H), 8.03
(d, J ) 7.7 Hz, 1H), 7.70-7.79 (m, 1H), 7.64 (t, J ) 7.4 Hz, 1H),
7.45-7.55 (m, 2H), 7.32-7.42 (m, 1H), 7.14-7.22 (m, 1H),
7.04-7.10 (d, J ) 7.7 Hz, 2H), 3.64-3.79 (m, 1H), 3.38-3.43
(m, 1H), 3.02 (s, 3H), 2.31-2.43 (m, 1H), 2.27 (s, 6H), 1.99-2.12
(m, 1H), 0.94-1.12 (m, 1H), 0.68 (s, 9H), 0.32-0.48 (m, 1H);
MS (ESI) m/z 651 (M + H)+; [R]D -8° (c 0.4, MeOH).
Assay Conditions for Determining HCV NS5B Polymerase
Inhibition. Two-fold serial dilutions of the inhibitors were incubated
with 20 mM Tris-Cl (pH 7.4), 2 mM MnCl2, 1 mM dithiothreitol,
1 mM ethylenediaminetetraacetic acid (EDTA), 60-125 µM GTP,
and 20-50 nM NS5B [HCV genotype 1b (BK, Genbank accession
number AF054247) or HCV genotype 1a (H77, Genbank accession
number AF011751)] for 15 min at room temperature. The reaction
was initiated by the addition of 20 µM CTP, 20 µM ATP, 1 µM
3H-UTP (10 mCi/umol), 5 nM template RNA, and 0.1 U/µL RNase
inhibitor (RNasin, Promega) and allowed to proceed for 2 to 4 h at
room temperature. Reaction volume was 50 µL. The reaction was
terminated by the addition of 1 volume of 4 mM spermine in 10
mM Tris-Cl (pH 8.0), 1 mM EDTA. After incubation for at least
2-Oxo-4-(R)-phenyloxazolidine-3-carboxylic Acid [1-(S)-(3,3-
Dimethylbutyl)-4-hydroxy-3-(7-methanesulfonylamino-1,1-dioxido-
4H-1,2,4-benzothiadiazin-3-yl)-2-oxo-1,2-dihydronaphthalen-1-
yl]amide (28). To a solution of 25 (3.0 g, 5.3 mmol) in anhydrous
CH2Cl2 (50 mL) was added triethylamine (2.2 mL, 15.8 mmol)
followed by 27 (1.25 g, 5.5 mmol). The mixture was stirred at room
temperature for 3 h and then diluted with CH2Cl2 (100 mL) and
washed with 0.1 N aqueous HCl (3 × 50 mL). The organic layer
was dried over Na2SO4, filtered, and concentrated in vacuo. The
crude product was purified by column chromatography on silica
gel using a gradient of 1-7% MeOH in CH2Cl2 as eluent.
Compound 28 eluted as the least polar diastereomeric component

3182 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 10
Randolph et al.
(6) Chandler, G.; Sulkowski, M. S.; Jenckes, M. W.; Torbenson, M. S.;
Herlong, H. F.; Bass, E. B.; Gebo, K. A. Treatment of chronic hepatitis
C: a systematic review. Hepatology 2002, 36, S135–S144.
(7) Beaulieu, P. L. Non-nucleoside inhibitors of the HCV NS5B poly-
merase: progress in the discovery and development of novel agents
for the treatment of HCV infection. Curr. Opin. InVest. Drugs 2007,
8, 614–634.
(8) Koch, U.; Narjes, F. Recent progress in the development of inhibitors
of hepatitis C virus RNA-dependent RNA polymerase. Curr. Top. Med.
Chem. 2007, 7, 1302–1329.
(9) Tedesco, R.; Shaw, A. N.; Bambal, R.; Chai, D.; Concha, N. O.; Darcy,
M. G.; Dhanak, D.; Fitch, D. M.; Gates, A.; Gerhardt, W. G.;
Halegoua, D. L.; Han, C.; Hofmann, G. A.; Johnston, V. K.; Kaura,
A. C.; Liu, N.; Keenan, R. M.; Lin-Goerke, J.; Sarisky, R. T.; Wiggall,
K. J.; Zimmerman, M. N.; Duffy, K. J. 3-(1,1-Dioxo-2H-(1,2,4)-
benzothiadiazin-3-yl)-4-hydroxy-2(1H)-quinolines, potent inhibitors of
hepatitis C virus RNA-dependent RNA polymerase. J. Med. Chem.
2006, 49, 971.
(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. Inhibitors
of HCV NS5B polymerase: synthesis and structure-activity relation-
ships of N-1-heteroalkyl-4-hydroxyquinolon-3-yl-benzothiadiazines.
Bioorg. Med. Chem. Lett. 2005, 15, 1577.
(11) (a) Bosse, T. D.; Larson, D. P.; Wagner, R.; Hutchinson, D. K.;
Rockway, T. W.; Kati, W. M.; Liu, Y.; Masse, S.; Middleton, T.;
Mo, H.; Montgomery, D.; Jiang, W.; Koev, G.; Kempf, D. J.; Molla,
A. Synthesis and SAR of novel 1,1-dialkyl-2(1H)-naphthalenones as
potent HCV polymerase inhibitors. Bioorg. Med. Chem. Lett. 2008,
18, 568–570. (b) Wagner,R.; Larson, D. P.; Beno, D. W. A.; Bosse,
T. D.; Darbyshire, J. F.; Gao, Y.; Gates, B. D.; He, W.; Henry, R. F.;
Hernandez, L. E.; Hutchinson, D. K.; Jiang, W. W.; Kati, W. M.;
Klein, L. L.; Koev, G.; Kohlbrenner, W.; Krueger, A. C.; Liu, J.; Liu,
Y.; Long, M. A.; Maring, C. J.; Masse, S. V.; Middleton, T.;
Montgomery, D. A.; Pratt, J. K.; Stuart, P.; Molla, A.; Kempf, D. J.
Inhibitors of hepatitis C virus polymerase: synthesis and biological
characterization of unsymmetrical dialkyl-hydroxynaphthalenoyl-ben-
zothiadiazines J. Med. Chem., in press.
(12) Inhibitor 1 was modeled in the benzothiadiazine binding site of HCV
polymerase (PDB entry 1C2P: Lesburg, C. A.; Cable, M. B.; Ferrari,
E.; Hong, Z.; Mennarino, A. F.; Weber, P. C. Crystal structure of the
RNA-dependent RNA polymerase from hepatitis C virus reveals a
fully encircled active site. Nat. Struct. Mol. Biol. 1999, 6, 937–943)
with docking of 1 based on the reported binding orientation of a
quinolinone analog (see ref 9).
(13) The Z-oxime analogue was unreactive toward allylation using allylin-
dium. For another example of allylation of methyloximes using
allylindium, see the following: Bernardi, L.; Cere, V.; Femoni, C.;
Pollicino, S.; Ricci, A. Organometallic reactions in aqueous media:
indium-promoted additions to 2-pyridyl and glyoxylic acid oxime
ethers. J. Org. Chem. 2003, 68, 3348–3351.
(14) Compound 6 did not undergo clean olefin metathesis to extend the
allyl chain. This problem was solved by converting 6 to oxazolidine
intermediate 7, which reacted smoothly under the metathesis
conditions.
(15) Garber, S. B.; Kingsbury, J. S.; Gray, B. L.; Hoveyda, A. H. Efficient
and recyclable monomeric and dendritic Ru-based metathesis catalysts.
J. Am. Chem. Soc. 2000, 122, 8168–8179.
(16) Barbero, M.; Cadamuro, S.; Degani, I.; Fochi, R.; Gatti, A.; Regondi,
V. Synthetic application of tris(methylthio)methyl salts. An efficient
route to trithioorthocarboxylic esters from strongly activated aromatic
and heteroaromatic systems. Synthesis 1988, 22–25.
(17) Yamazaki, N.; Atobe, M.; Kibayashi, C. Nucleophilic addition of
methyllithim to chiral oxime ethers: asymmetric preparation of
1-(aryl)ethylamines and application to a synthesis of calcimimetics
(+)-NPS R-568 and its thio analogue. Tetrahedron Lett. 2001, 42,
5029–5032.
(18) A detailed description of methods used for synthesis and analysis of
chiral isoamyl analogues 31-35 is available in the Supporting
Information.
(19) For information on the formation and reactivity of titanium Ate
complexes, see the following: Reetz, M. T. Organotitanium Chemistry.
In Organometallics in Synthesis. A Manual, 2nd ed.; Schlosser, M.,
Ed.; John Wiley & Sons Ltd.: West Sussex, England, 2002; p 832.
(20) Krimen, L. I.; Cota, D. J. The Ritter Reaction. In Organic Reactions;
Dauben, W. G., Ed.; Robert E. Krieger Publishing Co.; New York,
1969; Vol. 17, pp 213-325.
(21) For an initial report on the use of chiral oxazolidonones for the
resolution of racemic amines, see the following: Pirkle, W. H.;
Simmons, K. A. Improved chiral derivatizing agents for the chro-
matographic resolution of racemic primary amines. J. Org. Chem.
1983, 48, 2520–2527.
15 min at room temperature, the precipitated RNA was captured
by filtering through a GF/B filter (Millipore) in a 96-well format.
The filter plate was washed three times with 200 µL each of 2 mM
spermine, 10 mM Tris-Cl (pH 8.0), 1 mM EDTA, and 2 times with
EtOH. After air-drying, 30 µL of Microscint 20 scintillation cocktail
(Packard) was added to each well, and the retained cpm values
were determined by scintillation counting. IC50 values were
calculated by a two-variable nonlinear regression equation using
an uninhibited control and a fully inhibited control sample to
determine the minimum and maximum for the curve.
Assay for Determining Inhibition of Replication of Subge-
nomic HCV Replicons. The inhibitory effects of compounds on
HCV replicon replication were determined by measuring activity
of the luciferase reporter gene. Stable replicon cell lines based on
genotype 1a (H77) and genotype 1b (Con1) sequences were used.
Briefly, replicon-containing cells were seeded into 96-well plates
at a density of 5000 cells per well in 100 µL of DMEM containing
5% FBS. The following day compounds were diluted in DMSO to
generate a 200× stock in a series of eight half-log dilutions. The
dilution series was then further diluted 100-fold in the medium
containing 5% FBS. Medium with the inhibitor was added to the
overnight cell culture plates already containing 100 µL of DMEM
with 5% FBS. In assays measuring inhibitory activity in the
presence of human plasma, the medium from the overnight cell
culture plates was replaced with DMEM containing 40% human
plasma and 5% FBS. The cells were incubated for 3 days in the
tissue culture incubators. For the luciferase assay, 30 µL of Passive
Lysis buffer (Promega) was added to each well, and then the plates
were incubated for 15 min with rocking to lyse the cells. Luciferin
solution (100 µL, Promega) was added to each well, and luciferase
activity was measured with a Victor II luminometer (Perkin-Elmer).
The percent inhibition of HCV RNA replication was calculated for
each compound concentration, and the EC50 value was calculated
using GraphPad Prism 4 software.
Methods for Pharmacokinetic Evaluation of Compounds in
Rat. Compounds were formulated in a DMSO/PEG400 (1:9, v/v)
for single intravenous or oral dosing of 5 mg/mL in overnight fasted
Sprague-Dawley rats (n ) 3). Heparinized plasma samples were
withdrawn at 0.1 (iv only), 0.25, 0.5, 1, 2, 4, 6, 9, 12, and 24 h
after dosing. Liver and plasma samples from separate animals were
collected at 1, 6, and 12 h after oral dose. Plasma and liver drug
concentrations were determined by a liquid chromatography-mass
spectrometry (LC-MS) assay. Mean plasma concentration data
were submitted to multiexponential curve fitting using WinNonlin.
The area under the mean concentration-time curve from 0 to t
hours (time of the last measurable concentration) after dosing
(AUC0-t) was calculated using the linear trapezoidal rule for the
concentration-time profile. The residual area was extrapolated to
infinity, determined as the final measured mean concentration (Ct)
divided by the terminal elimination rate constant (ꢀ), and was added
to AUC0-t to produce the total area under the curve (AUC0-∞).
Supporting Information Available: Experimental details for
the synthesis of isoamyl analogues 14, 16, and 17a-g, chiral
isoamyl analogues 31-35, and neohexyl analogues 26b-v; HPLC
data for all compounds presented in Tables 1-5, including chiral
chromatography conditions and results for compounds 30 and 33.
This material is available free of charge via the Internet at http://
pubs.acs.org.
References
(1) World Health Organization, HCV Factsheet. http://www.who.int/
mediacentre/factsheets/fs164/en.
(2) McHutchinson, J. G. Understanding hepatitis C. Am. J. Manage. Care
2004, 10, S21–S29.
(3) Shepard, C. W.; Finelli, L.; Alter, M. J. Global epidemiology of
hepatitis C virus infection. Lancet Infect. Dis. 2005, 5, 558–567.
(4) Wong, J. B.; McQuillan, G. M.; McHutchinson, J. G.; Poynard, T.
Estimating future hepatitis C morbidity, mortality, and costs in the
United States. Am. J. Public Health 2000, 90, 1562–1569.
(5) Houghton, M.; Abrigani, S. Prospects for a vaccine against the hepatitis
C virus. Nature 2005, 436, 961–966.

B-Ring Amino Analogues
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 10 3183
(22) Molecular modeling studies revealed that the isomer having the S-
configuration would properly position the inhibitor such that the isoamyl
side chain would project into the hydrophobic pocket (see Figure 2). Given
the importance of this group to the activity of the B-ring dialkyl series of
inhibitors (see ref 11), it was anticipated that the S-isomer would be the
more active isomer in the B-ring amino series.
(23) Optical rotation measurement confirmed that 29 was the (-)-
enantiomer {[R]D-72° (c 0.5, MeOH)}, which we expected to be the
desired S-stereoisomer based on polarimetry studies of chiral isoamyl
analogues. Details are available in the Supporting Information.
(24) Our initial investigations of B-ring amino thiadiazine inhibitors of
HCV polymerase focused on a series of compounds bearing an
isoamyl side chain. This initial series of analogues was made prior
to the discovery that a neohexyl side chain on the B-ring offered
advantages over an isoamyl B-ring side chain as the result of
improved metabolic stability.
(25) Experimental details for the synthesis of 26b-v can be found in the
Supporting Information.
JM801485Z