2-(3-Thienyl)-5,6-dihydroxypyrimidine-4-carboxylic acids as inhibitors of HCV NS5B RdRp

Article abstract of DOI:10.1016/j.bmcl.2009.06.106

A series of 2-(3-thienyl)-5,6-dihydroxypyrimidine-4-carboxylic acid inhibitors of the hepatitis C virus (HCV) NS5B polymerase enzyme are reported. Sulfonyl urea substituted analogs in this series proved to be the most potent active site non-nucleoside inh

Full text of DOI:10.1016/j.bmcl.2009.06.106

Bioorganic & Medicinal Chemistry Letters 19 (2009) 6245–6249
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
2-(3-Thienyl)-5,6-dihydroxypyrimidine-4-carboxylic acids as inhibitors
of HCV NS5B RdRp
*
Barbara Pacini , Salvatore Avolio, Caterina Ercolani, Uwe Koch, Giovanni Migliaccio, Frank Narjes,
Laura Pacini, Licia Tomei, Steven Harper
IRBM, Merck Research Laboratories Rome, Via Pontina km 30,600, 00040 Pomezia, Rome, Italy
a r t i c l e i n f o
a b s t r a c t
Article history:
A series of 2-(3-thienyl)-5,6-dihydroxypyrimidine-4-carboxylic acid inhibitors of the hepatitis C virus
(HCV) NS5B polymerase enzyme are reported. Sulfonyl urea substituted analogs in this series proved
to be the most potent active site non-nucleoside inhibitors of NS5B reported to date. These compounds
had low nanomolar enzyme inhibition across HCV genotypes 1–3 and showed single digit micromolar
inhibition in the HCV replicon assay. This improved cell-based activity allowed the binding mode of these
compounds to be probed by selection of resistant mutations against compound 21. The results generated
are in broad agreement with the previously proposed binding model for this compound class.
Ó 2009 Published by Elsevier Ltd.
Received 2 April 2009
Revised 10 June 2009
Accepted 10 June 2009
Available online 17 July 2009
Around 2% of the world’s population is infected with hepatitis C
virus (HCV) and at risk of developing serious liver diseases such as
cirrhosis and hepatocellular carcinoma.1 HCV is a leading indica-
tion for liver transplantation and is thought to lead to around
10,000 deaths annually in the United States alone.2 The current
standard of care for HCV, which is based on a combination of inter-
pyrophosphate leaving group on the incoming nucleoside
triphosphate.
The HCV polymerase enzyme is seen as an attractive target for
drug intervention because of its fundamental role in viral replica-
tion. Clinical proof-of-concept for antivirals operating through an
allosteric mechanism of action has now been demonstrated for
several classes of NS5B inhibitors.8 However, allosteric compounds
typically have genotype sensitivities that can result in drastically
reduced effectiveness against HCV from genotypes 2 or 3.9 To date,
clinical reductions in HCV viral load from allosteric compounds has
been limited to patients infected with genotype 1 HCV.10 In con-
trast, the highly conserved active site of NS5B favors the develop-
ment of inhibitors that are active across genotypes. Active site
nucleoside inhibitors such as R7128 have shown comparably
strong clinical effects in patients infected with all three of the main
HCV genotypes (genotypes 1–3).11 Despite the attraction of the ac-
tive site with a view to identifying pan-genotype inhibitors, very
few series of non-nucleoside inhibitors have been reported. 5,6-
Dihydroxypyrimidine-4-carboxylic acids (e.g., 1) that evolved from
a diketoacid lead12 are the most promising active site non-
nucleosides reported to date. Our earlier work in this area (Fig. 1)
demonstrated that replacement of the C2 phenyl group in 1 with
five-membered ring heterocycles (such as 2-thiophene, 2)
improved biochemical potency by an order of magnitude.13 Opti-
mization of 2 led to the 30-substituted thiophene 3 that shows good
NS5B enzyme inhibition (IC50 158 nM) but has poor activity in the
feron-a and ribavarin, can cure HCV infections, but is often inade-
quate. In addition to serious side effects such as fatigue, flu-like
symptoms, and hemolytic anemia3 interferon-based therapy also
has a low success rate (<50%) in the main viral genotype.4 The need
for improved therapies is pressing because although the incidence
of new HCV infections is declining, mortality is expected to in-
crease into the middle the next decade.5
The HCV genome is highly variable in sequence with six main
genotypes that are encoded by (+)-stranded RNA’s containing
around 9.6 kilobases.6 Translation of the viral genome in hepa-
tocytes and post-translational processing events liberate six
non-structural (NS) proteins that comprise the viral replication
machinery. The NS5B enzyme is one of these proteins, and has
been identified as an RNA dependent RNA polymerase. NS5B is
essential for replication since it both effects the synthesis of a
(ꢀ)-stranded HCV RNA template and regenerates the (+)-stranded
genomic RNA.7 The NS5B active site is located in the palm domain
of the enzyme where two conserved aspartic acid residues chelate
a pair of Mg2+ ions that catalyze the nucleotidyl transfer reaction.
The Mg2+ ions have the dual role of activating the 30-OH of the
elongating RNA chain for nucleophilic attack, and stabilizing the
cell-based HCV replicon assay (EC50 10.9 l
M).14 In a continuation
of our pursuit for active-site inhibitors with improved potency
we describe here our work in a related series of 3-thiophene
substituted 5,6-dihydroxypyrimidine-4-carboxylic acids. In addi-
* Corresponding author. Tel.: +39 0691093318.
E-mail address: barbara_pacini@merck.com (B. Pacini).
0960-894X/$ - see front matter Ó 2009 Published by Elsevier Ltd.
doi:10.1016/j.bmcl.2009.06.106

6246
B. Pacini et al. / Bioorg. Med. Chem. Lett. 19 (2009) 6245–6249
OH
N
OH
OH
N
OH
N
OH
N
OH
OH
OH
OMe
OH
OH
N
N
OH
OH
OH
OH
N
N
N
S
N
S
b,c
a
S
S
6b
O
O
5'
O
O
O
NH
NH2
NH
3'
4'
11
O
O
NH
e,c
Cl
d,c
1
2
3
Cl
12
OH
N
OH
N
OH
OH
OH
Figure 1. Previously reported 2-aryl-5,6-dihydroxypyrimidine-4-carboxylic acid
inhibitors of the HCV NS5B polymerase.
N
N
OH
S
S
O
O
NH
NH
tion to more than 10-fold improved biochemical potency, this ser-
ies provided for the first time active-site non-nucleoside inhibitors
that are sufficiently active in the replicon system to allow the study
of resistance to this compound class.
O
O
NH
NH
R
O
S
O
R
13 R=H
14 R=Cl
15-24
2-(3-Thienyl)-5,6-dihydroxypyrimidine-4-carboxylic
acids
were prepared following literature precedent as outlined in
Scheme 1.14 Thus amidoximes 5a–d (available from the corre-
sponding carbonitriles 4a–d) were condensed with dimethyl acet-
ylene dicarboxylate and the resulting adducts were cyclized under
thermal conditions. Routine hydrolysis of the pyrimidine methyl
esters 6a–d furnished the target inhibitors 7–10.
Scheme 2. Reagents and conditions: (a) TFA/CH2Cl2, 25 °C, 99%; (b) 2-Cl–C6H4–
CH2CO2H, NEt3, CDI, DMF, 25 °C; (c) LiOHꢁH2O, THF/H2O, 50 °C, 40–95%; (d) RC6H4-
CH2NCO, Py, 25 °C; (e) ArSO2NHCO2Et, NEt3, toluene, reflux or ArSO2NCO, Py, 25 °C.
Based on previous work14 structure–activity investigations at
the 20- and 40-position of 3-thiophene 7 were of interest. However,
the free amine from N-Cbz deprotection of 6d proved to be unsta-
ble, and compound 10 was the only 20-substituted 3-thiophene
evaluated. In contrast, removal of the N-Boc protecting group from
6b (Scheme 2) afforded the free 40-aminothiophene 11 that reacted
smoothly with electrophiles. Thus treatment of 11 with 2-chloro-
benzoic acid and N,N0-carbonyldiimidazole followed by saponifica-
tion gave the amide 12. Ureas 13–14 were prepared in the same
fashion through reaction of 11 with the appropriate benzyl isocya-
nates. Sulfonyl ureas 15–24 were obtained from commercial sulfo-
nyl isocyanates or with ethyl sulfonyl carbamates (that were
prepared by treatment of the corresponding aryl sulfonamides
with ethyl chloroformate).
NS5B active site. The greater improvement for the 2-thiophene
compound is in line with the somewhat stronger electron with-
drawing effect for this isomer.15 Structure–activity relationships
at the 40-position of the thiophene ring of 7 proved to be somewhat
changed with respect to those at the 30-position of 2-thiophene 2.
An example was the 2-chlorobenzyl amide 12 which brought only
a modest improvement in potency over 7 (rather than the order of
magnitude gain seen in the isomeric 2-thiophene series). The ben-
zyl carbamoyl substituted 9, as well as its 20-isomer 10, each im-
proved potency by around 20-fold with respect to the
unsubstituted analog 7. In contrast, no gain in potency came from
the simple alkyl carbamate in 8. Benzyl ureas, which had given the
most active inhibitors in the 2-thiophene series (i.e., 3) were exten-
sively explored in the 3-thiophene area. Although gains in potency
were achieved with respect to the simple benzyl urea 13 (IC50
873 nM) they were consistently modest; the 2-chlorobenzyl urea
14 again was optimal (IC50 258 nM).
Table 1 summarises our initial exploration of 3-thiophene
substituted 5,6-dihydroxypyrimidine-4-carboxylic acids. In line
with results for the isomeric 2-thiophene analog 2, compound 7
More notable improvements in potency came from introduction
of a sulfonyl urea group at the 40-position of thiophene 7 (Table 2).
The phenyl sulfonyl urea 15 was around 100-fold more active than
7, and was also the first compound in this regioisomeric thiophene
(IC50 6.1
l
M) had improved potency with respect to our original
phenyl substituted lead 1 (IC50 30
l
M).14 This is thought to stem
from acidification of the pyrimidine 5-OH group when five-mem-
bered ring heterocycles are present at the pyrimidone C-2 position,
an effect believed to favorably influence Mg2+ chelation in the
series to show cell-based activity (EC50 32 12 lM). Efforts to
OH
OH
N
OH
Y
Y
N
N
Y
OMe
N
NH2
a
b,c
S
S
S
O
X
X
X
6a-d
4a X, Y=H
5a-d
4b X=NHBoc, Y=H
4c X=NHCbz, Y=H
4d X=H, Y=NHCbz
d
OH
OH
N
Y
2'
7
8
9
X, Y=H
OH
N
X=NHBoc, Y=H
X=NHCbz, Y=H
S
4'
O
X
10 X=H, Y=NHCbz
Scheme 1. Reagents and conditions: (a) NH2OHꢁHCl, TEA, MeOH, 50 °C, 95%; (b) DMAD, CHCl3, 75 °C, 85%; (c) p-xylene, 160 °C, 40%; (d) LiOHꢁH2O, THF/H2O, 50 °C, 80%.

B. Pacini et al. / Bioorg. Med. Chem. Lett. 19 (2009) 6245–6249
6247
Table 1
Table 2
Enzyme inhibition for 3-thiophene substituted 5,6-dihydroxypyrimidine-4-carboxylic
acid; initial SAR at the 20- and 40-positions
Intrinsic potency and cell-based assay results for sulfonyl urea compounds 15–24
OH
OH
OH
OH
N
OH
N
N
S
OH
O
R
N
NH
O
O
O
NH
S
O
a
Compd
R
IC50 (nM)
X
a
b
c
1
30,000 707b
2600 2163
6151 1803
Compd
X
IC50 (nM)
EC50 (1b,
lM)
EC50 (2a,
l
M)
15
16
17
18
19
20
21
22
23
24
Ph
62 13
78 24
32 12
2.6 1.1
16d
2-Cl–Ph
2-CH3–Ph
2-Br–Ph
3-Br–Ph
4-Cl–Ph
1-Naphthyl
2-Naphthyl
2,3-Cl–Ph
Me
6.6c
S
47
1
2
7
74 12
3.4 1.9
38
85
36
48
12
9
2
5
2
1
37d
26d
3.6 0.7e
32d
2.4 0.3
—
3.7c
4.9c
S
S
426 72
O
8
8280b
a
IC50 measured against
DC55 NS5B enzyme from genotype 1b HCV.
Replicon assay run over 96 h in HUH-7 cells harboring genotype 1b HCV.
Full length genotype 2a infectivity assay.
N
H
b
c
O
d
e
n = 1.
n = 18.
S
O
9
326
5
N
H
O
the NS5B enzyme and show low micromolar inhibition in the
cell-based assay.16 The importance of the position of the substitu-
ents on the phenyl ring was again apparent from the 2-naphthyl
compound 22 that had much weaker cell based activity than its
isomer 21.17 The improvement of inhibition observed for com-
pounds with a sulfonyl urea side chain was thought likely to be re-
lated to the acidity of this group. This was confirmed by
preparation of the methyl sulfonyl urea 24 that was 14-fold more
active than the unsubstituted thiophene 7. Potency gains of this
magnitude are not achieved when alkyl groups are attached to a
thiophene through non-acidic linkers; the tert-butylcarbamate 8
is one such example.
O
NH
10
343 11
O
S
Cl
S
S
O
12
13
2610b
N
H
In line with our expectation that inhibitors targeting the highly
conserved active site should block HCV replication for all viral
genotypes, compound 21 displayed potent inhibition against
NS5B isolates from all three of the major HCV genotypes. Thus
O
873 61
N
H
N
H
21 showed potent activity against the
DC21 NS5B enzymes from
genotype 1b (IC50 5.6 nM), 2a (IC50 2.1 nM), 2b (IC50 5.8 nM) and
3a (IC50 3.9 nM). Blockade of HCV in HUH-7 cells stably transfected
with HCV RNA from the 2b genotype was also observed (EC50
S
O
Cl
14
258 98
N
H
N
H
9.6 lM) and cell-based activity was confirmed (for 21, 16 and
23) in a full-length genotype 2a infectivity assay (Table 2). Follow-
ing intravenous administration of compound 21 (3 mg/kg) to rat,
an excellent plasma half life (t1/2 = 15 h) was measured; this re-
flects the expected low volume of distribution (0.1 L/kg) and turn-
over (0.2 mL/min/kg) for this acidic compound that is highly bound
to rat serum proteins (>99%). Oral exposure (3 mg/kg) was
3.3 lM h and though oral bioavailability was low (%Fp.o. = 2), high
levels of systemic exposure could be achieved through intramuscu-
lar administration of the compound (%Fi.m. = 89).
Under a modification to the replicon assay protocol, cell survival
in response to antibiotic neomycin sulfate can be tied to HCV rep-
lication by incorporation of a neomycin phosphotransferase gene
into the HCV RNA.18 Efficient suppression of HCV replication, as ex-
pected by inhibition of NS5B, sensitizes cells to antibiotic and links
cell survival to the development of escape mutations that restore
replication competency. These characteristics have been exploited
for the selection of cell colonies resistant to compound 21 from the
current series. Sequencing of the HCV genomes in the surviving
a
IC50 measured against
n = 1; Unless otherwise stated results in this and subsequent tables are
D
C55 NS5B enzyme from genotype 1b HCV.
b
expressed as mean standard error (or half range for n = 2) for 2–18 independent
determinations.
further improve the sulfonyl urea compounds focused on changes
to the phenyl ring in the thiophene side-chain of 15. The introduc-
tion of substituents at the ortho-position gave analogs 16–18 that,
despite similar biochemical potency to 15, had improved cell-
based activities. In particular, the halogenated compounds 16
and 18 had efficacy in the replicon assay that was improved by
around an order of magnitude. In contrast, compounds with a hal-
ogen substituent at the 3- or 4-positions of the phenyl ring (19–20)
did not bring any improvement over 15. 2,3-Disubstituted aro-
matic rings such as the 1-naphthyl derivative 21 or the 2,3-di-
chloro compound 23 were the most active inhibitors in this
study. Both these compounds are low nanomolar inhibitors of

6248
B. Pacini et al. / Bioorg. Med. Chem. Lett. 19 (2009) 6245–6249
colonies revealed two specific mutations in the NS5B region, P156S
and G152E, each of which was demonstrated by reverse genetics
sufficient to confer resistance to 21. Thus when compound 21
was tested in HUH-7 cells transfected with sub-genomic HCV
RNA harboring either mutation G152E or P156S, it showed at least
a six-fold loss in activity. Compound 21, as well as the related sul-
fonyl urea 18, the urea 3 and the simple pyrimidone 25 were also
tested against an isolated HCV NS5B enzyme containing either the
P156S or G152E mutations. The results in Table 3 illustrate that
sulfonyl ureas 21 and 18 as well as 3 all showed significant impair-
ment of activity against the G152E mutant with IC50 values be-
tween five and eightfold reduced with respect to wild-type. In
contrast, compound 25 (which has an unsubstituted thiophene
ring) showed no decrease in response suggesting a role for the thi-
ophene side-chain in conferring resistance. The loss of cell-based
activity against P156S that had been seen in a cell-based setting
was not recapitulated when this mutation was incorporated into
In large part the resistance data observed are consistent with
the binding model that we previously proposed for these active site
inhibitors. In the C6/N1 amide tautomer (that has been found to
have low ground state energy, and which preliminary SAR sug-
gested may be the active species) the conformation of the sulfonyl-
urea side chain would be dictated by an intramolecular hydrogen
bond between the 3-NH group on the thiophene ring and the N3
atom of the pyrimidone. Modeling of 18 into the pyrophosphate
binding region of the NS5B active site in this planar orientation
(Fig. 2) highlights contacts between the sulfonylurea side chain
and both the P156 and particularly the G152 residues.
In light of the above considerations it became clear that com-
pound 26 (Fig. 3) was an attractive target since it constrains both
the side-chain orientation and pyrimidone tautomerism in our pro-
posed binding model. Unlike the N-methylpyrimidone 27, the ring
constrained analog 26 ought to have a conformation in which the
thiophene and pyrimidone rings are approximately coplanar. The
change of side chain orientation for N-methyl compounds such
as 27 is thought to explain their reduced inhibition; compound
27 (IC50 600 nM) is an 8-fold weaker inhibitor than 16. Compound
27 was prepared as outlined in Scheme 3. Regioselective deproto-
nation of the thiophene 4b (LDA, ꢀ78 °C) followed by alkylation
with TBS protected 1,2-iodoethanol furnished 29, albeit in modest
yield. This compound was elaborated as previously described to
the pyrimidine 31. Protection of the pyrimidine 5-OH as a pivaloate
ester followed by TBS deprotection of the primary alcohol set up
the cyclization step that was smoothly effected (50% yield) under
Mitsunobu-conditions. Compound 33 was elaborated to the target
a genotype 1b
DC55 NS5B enzyme. Apparently the interaction be-
tween 21 and NS5B in the context of a cell-based replication com-
plex is not fully recaptured in the in-vitro assay.19
Table 3
IC50 values for compounds 21, 18, 3 and 25 in
DC55-con1 enzyme either wild type or
single point NS5B mutated
Compd Structure
Wild-type IC50
M)
G152E IC50
M)
P156S IC50
M)
(
l
(
l
(l
OH
N
OH
OH
N
S
O
21
18
3
0.04
0.3
0.04
NH
O
NH
S
O
O
OH
N
OH
N
OH
S
O
0.06
0.4
0.07
NH
O
NH
S
Br
O
Figure 2. Binding model of 18 superimposed to the pyrophosphate in the active site
together with a primer. The thumb, palm and fingers domains are colored in green,
red and blue, respectively. Carbon atoms are shown in green (for 18 carbon atoms
are shown in green and orange), oxygen in red, nitrogen in blue, sulfur in yellow,
phosphorous in violet and bromine in pale green. The two catalytic aspartates and
O
OH
N
Arg 158 are shown. P156 and G152 are located in the
K2 fingertips loop, in close
OH
OH
contact with the specificity portion of 18 but not with its metal chelating portion.
N
S
0.18
0.8
0.17
O
O
N
OH
OH
O
N
NH
O
N
Cl
OH
OH
OH
OH
O
N
N
N
H
N
O
S
S
S
O
O
NH
NH
NH
O
N
O
O
O
NH
S
NH
S
NH
S
Cl
OH
OH
Cl
O
Cl
O
O
N
25°
4.6
3.2
5.6
O
O
O
S
O
28
27
Figure 3.
26
a
Assay run in the presence of MnCl2.

B. Pacini et al. / Bioorg. Med. Chem. Lett. 19 (2009) 6245–6249
6249
TBSO
OH
TBSO
S
TBSO
S
OH
OH
OMe
N
N
N
N
NH2
a
b
c,d
N
S
e
S
O
NH
Boc
4b
NH
Boc
29
NH
Boc
30
NH
Boc
31
TBSO
OH Piv
O
O
N
Piv
O
N
N
h,i,l
f,g
OMe
OMe
26
N
S
S
O
O
NH
Boc
NH
Boc
32
33
Scheme 3. Reagents and conditions: (a) LDA, ICH2CH2OTBS, ꢀ78 °C, 24%; (b) NH2OH, TEA, EtOH, 0.3 M, 80 °C, 30%; (c) DMAD, CHCl3, 90 °C; (d) p-xylene, reflux; (e) Piv-Cl, Py,
25 °C, 32%; (f) THF:Py:HF.Py, ꢀ78 °C; (g) DEAD, PPh3, 25 °C, 50%; (h) CH2Cl2:TFA; (i) 2-Cl–C6H4–SO2NCO, Py, 25 °C; (l) LiOHꢁH2O, THF/H2O, 50 °C, 16%.
3. Foster, G.; Mathurin, P. Antivir. Ther. 2008, 13, 3.
4. Di Bisceglie, A. M.; McHutchinson, J.; Rice, C. M. Hepatology 2002, 35, 224.
5. Garber, K.; Arbor, A. Nat. Biotechnol. 2007, 25, 1379.
inhibitor by removal of the pivaloate and N-Boc protecting groups
followed by installation of the side chain using 2-chloro-
phenylsulfonyl isocyanate; the synthesis was completed by methyl
ester hydrolysis to give 26.
Compound 26 was tested against the genotype 1b NS5B enzyme
which it proved to have extremely weak potency (IC50 10 lM). This
result was somewhat contrary to our expectation based on the
structure activity relationships and modeling hypotheses gener-
ated across several series of inhibitor. A simple explanation may
be that there is insufficient space to accommodate the newly
formed six-membered ring that has been installed into the inhibi-
tor in 26. However, this result may alternatively point to the
importance of an alternative binding orientation, in which a 180°
rotation occurs about the thiophene–pyrimidine bond (such as is
constrained in compound 28).
6. Lahm, A.; Yagnik, A.; Tramontano, A.; Koch, U. Curr. Drug Targets 2002, 3, 281.
7. Appel, N.; Schaller, T.; Penin, F.; Bartenschlager, R. J. Biol. Chem. 2006, 281, 9833.
8. Beaulieu, P. L. Expert Opin. Ther. Patents 2009, 19, 145.
9. De Francesco, R.; Paonessa, G.; Olsen, D. B.; Rowley, M.; Crescenzi, B.;
Habermann, J.; Narjes, F.; Laufer, R. Hep DART 2007. Lahaina, Hawaii.
10. NCT00623649, USA 2008. Available from: http://clinicaltrials.gov/ct2/show/
study/NCT00623649
11. Rodriguez-Torres, M., Lalezari, J.; Gane, E. J.; DeJesus, E.; Nelson, D. R.; Everson,
G.; Jacobson, I.; Reddy, K. R.; McHutchison, J.; Beard, A.; Walker, S.; Symonds,
W.; Berrey, M. M. The 59th Annual Meeting of the American Association for the
Study of Liver Diseases (AASLD), San Francisco, California, USA, 2008.
12. (a) Summa, V.; Petrocchi, A.; Matassa, V. G.; Taliani, M.; Laufer, R.; De
Francesco, R.; Altamura, S.; Pace, P. J. Med. Chem. 2004, 47, 5336; (b) Summa,
V.; Petrocchi, A.; Pace, P.; Matassa, V. G.; De Francesco, R.; Altamura, S.; Tomei,
L.; Koch, U.; Neuner, P. J. Med. Chem. 2004, 47, 14.
13. Stansfield, I.; Avolio, S.; Colarusso, S.; Gennari, N.; Narjes, F.; Pacini, B.; Ponzi,
S.; Harper, S. Bioorg. Med. Chem. Lett. 2004, 14, 5085.
In summary, we have described a series of active site inhibitors
of NS5B that achieved around a 15-fold improvement in biochem-
ical potency over our previous work. Strong enzyme inhibition (c.
14. Koch, U.; Attenni, B.; Malancona, S.; Colarusso, S.; Conte, I.; Di Filippo, M.;
Harper, S.; Pacini, B.; Giomini, C.; Thomas, S.; Incitti, I.; Tomei, L.; De
Francesco, R.; Altamura, S.; Matassa, V. G.; Narjes, F. J. Med. Chem. 2006, 49,
1693.
15. The stronger electron withdrawing effect for the 2-thiophene isomer are
apparent from pKa values for 2-thiophenecarboxylic acid and 3-
thiophenecarboxylic acid which are, respectively. 4.1 and 3.5; see ACD Labs
pKa Database Version 11 for details.
5 nM) was achieved against
DC21 NS5B from HCV genotypes 1–3
and cell-based efficacy was improved to a level that for the first
time allowed resistance to this compound class to be assessed.
Mutation at either of residues P156 and G152 in NS5B was found
sufficient to confer resistance to sulfonylurea 21, and compounds
3 and 18 showed cross-resistance. The binding model previously
proposed for 5,6-dihydroxypyrimidine-4-carboxylic acids was sup-
ported by resistance data, but the weak inhibitor 26 (which was
designed based upon this model) highlights that further studies
are required to fully understand the interaction of this compound
class at the active site.
16. Koch, U.; Narjes, F. Curr. Top. Med. Chem. 2007, 7, 1302.
17. Due to the typically high plasma protein binding (>99%) and low Caco-2
permeability of 5,6-dihydroxypyrimidine-4-carboxylic acids,
a
direct
correlation between physicochemical properties and cell-based activity was
not feasible for the sulfonyl urea compounds. All compounds were judged
soluble and are chemically stable in aqueous media over the 4-day duration of
the HCV replicon assay.
18. Tomei, L.; Altamura, S.; Paonessa, G.; De Francesco, R.; Migliaccio, G. Antivir.
Chem. Chemother. 2005, 16, 225.
19. For details of
a
similar case in which an in-vitro NS5B enzyme assay
more detailed
failed to recapitulate replicon resistance, and for
a
discussion of this topic, see: Tomei, L.; Altamura, S.; Bartholomew, L.;
Bisbocci, M.; Bailey, C.; Bosserman, M.; Celluci, A.; Forte, E.; Incitti, I.;
Orsatti, L.; Koch, U.; De Francesco, R.; Olsen, D. B.; Carroll, S. S.;
Migliaccio, G. J. Virol. 2004, 78, 938.
References and notes
1. McHutchinson, J. G. Am. J. Manag. Care 2004, 10, S21.
2. Houghton, M.; Abrigani, S. Nature 2005, 436, 961.