Discovery of a Series of 2′-α-Fluoro,2′-β-bromo-ribonucleosides and Their Phosphoramidate Prodrugs as Potent Pan-Genotypic Inhibitors of Hepatitis C Virus

Article abstract of DOI:10.1021/acs.jmedchem.8b01300

Hepatitis C virus (HCV) nucleoside inhibitors display pan-genotypic activity, a high barrier to the selection of resistant virus, and are some of the most potent direct-acting agents with durable sustained virologic response in humans. Herein, we report, the discovery of β-d-2′-Br,2′-F-uridine phosphoramidate diastereomers 27 and 28, as nontoxic pan-genotypic anti-HCV agents. Extensive profiling of these two phosphorous diastereomers was performed to select one for in-depth preclinical profiling. The 5′-triphosphate formed from these phosphoramidates selectively inhibited HCV NS5B polymerase with no inhibition of human polymerases and cellular mitochondrial RNA polymerase up to 100 μM. Both are nontoxic by a variety of measures and display good stability in human blood and favorable metabolism in human intestinal microsomes and liver microsomes. Ultimately, a preliminary oral pharmacokinetics study in male beagles showed that 28 is superior to 27 and is an attractive candidate for further studies to establish its potential value as a new clinical anti-HCV agent.

Full text of DOI:10.1021/acs.jmedchem.8b01300

Subscriber access provided by Gothenburg University Library
Article
Discovery of a Series of 2’-#-Fluoro,2’-#-Bromo-
Ribonucleosides and their Phosphoramidate Prodrugs
as Potent Pan-Genotypic Inhibitors of Hepatitis C Virus
SEEMA MENGSHETTI (BAGMARE), Longhu Zhou, Ozkan Sari, Coralie De Schutter,
Hongwang Zhang, Jong Hyun Cho, Sijia Tao, Leda Bassit, Kiran Verma, Robert Domaoal,
Maryam Ehteshami, Yong Jiang, Reuben Ovadia, Mahesh Kasthuri, Olivia Ollinger Russell,
Tamara McBrayer, Tony Whitaker, Judy Pattassery, Maria Luz Pascual, Lothar Uher,
Biing Y. Lin, Sam Lee, Franck Amblard, Steven J. Coats, and Raymond Felix Schinazi
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.8b01300 • Publication Date (Web): 17 Jan 2019
Downloaded from http://pubs.acs.org on January 19, 2019
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a service to the research community to expedite the dissemination
of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in
full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully
peer reviewed, but should not be considered the official version of record. They are citable by the
Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore,
the “Just Accepted” Web site may not include all articles that will be published in the journal. After
a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web
site and published as an ASAP article. Note that technical editing may introduce minor changes
to the manuscript text and/or graphics which could affect content, and all legal disclaimers and
ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or
consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W.,
Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 63
Journal of Medicinal Chemistry
1
2
3
4
Schinazi, Raymond; Emory University, Department of Pediatrics
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 63
1
2
3
4
5
6
7
8
9
Discovery of a Series of 2’-a-Fluoro,2’-b-
Bromo-Ribonucleosides and their
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Phosphoramidate Prodrugs as Potent Pan-
Genotypic Inhibitors of Hepatitis C Virus
1
Seema Mengshetti, Longhu Zhou, Ozkan Sari, Coralie De Schutter, Hongwang Zhang,
1
1
1
1
1
1
1
1
1
Jong Hyun Cho, Sijia Tao, Leda C. Bassit, Kiran Verma, Robert A. Domaoal, Maryam
1
1
1
1
1
Ehteshami, Yong Jiang, Reuben Ovadia, Mahesh Kasthuri, Olivia Ollinger Russell,
1
2
2
2
2
Tamara McBrayer, Tony Whitaker, Judy Pattassery, Maria Luz Pascual, Lothar Uher,
2
2
1
1
Biing Y. Lin, Sam Lee, Franck Amblard, Steven J. Coats, and Raymond F. Schinazi
1*
1
Center for AIDS Research, Laboratory of Biochemical Pharmacology, Department of
2
Pediatrics, Emory University School of Medicine, Atlanta, GA 30322, USA; Cocrystal
Pharma, Inc., Tucker, GA 30084, USA
KEYWORDS:
2’-modifications, anti-HCV drug, phosphoramidate prodrug, ribo
nucleoside, NS5B inhibitor
ACS Paragon Plus Environment

Page 3 of 63
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
ABSTRACT: HCV nucleoside inhibitors display pan-genotypic activity, a high barrier to
selection of resistant virus and are some of the most potent direct-acting agents with
durable sustained virologic response in humans. Herein, we report, the discovery of β-D-
2′-Br,2′-F-uridine phosphoramidate diastereomers 27 and 28, as nontoxic pan-genotypic
anti-HCV agents. Extensive profiling of these two phosphorous diastereomers was
performed to select one for in depth preclinical profiling. The 5′-triphosphate formed from
these phosphoramidates selectively inhibited HCV NS5B polymerase with no inhibition of
human polymerases and cellular mitochondrial RNA polymerase up to 100 µM. Both are
non-toxic by a variety of measures and display good stability in human blood and favorable
metabolism in human intestinal microsomes and liver microsomes. Ultimately, a
preliminary oral pharmacokinetics study in male beagles showed that 28 is superior to 27
and is an attractive candidate for further studies to establish its potential value as a new
clinical anti-HCV agent.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
INTRODUCTION
An estimated 180 million people worldwide are infected with hepatitis C virus (HCV)
and about 71 million people have chronic HCV infection. Among those with chronic
infection about 15-30% will develop cirrhosis within 20 years and have a significant
increased risk for developing an end-stage chronic liver disease, such as hepatocellular
1
carcinoma (HCC). About four hundred thousand people die each year as a result of HCV
infection, mostly from cirrhosis and hepatocellular carcinoma. Just a few years ago, the
combination therapy with pegylated interferon (Peg-IFN)-alpha plus ribavirin (RBV) was
2
used to treat chronically infected patients. However, the combination of a long treatment
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 63
1
2
3
4
5
6
course, variable efficacy across genotypes, a poor safely profile and low patient tolerability
has resulted in very limited use of this therapy.
7
8
Advances in understanding HCV viral replication cycle revealed multiple proteins or
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
3
steps in viral life cycle that could be pursued as potential targets for antiviral therapy.
Among them, NS3/4A protease, NS5A and NS5B RNA dependent RNA (RdRp)
polymerase were extensively studied and as a result, direct acting antivirals (DAA)
4
targeting these viral proteins now forms the mainstay of current HCV treatment. Early
NS3/4A protease inhibitors (telaprevir and boceprevir) and NS5A inhibitor (daclatasvir)
exhibited high potency, but generally showed a low barrier to resistance. Moreover their
activity profile was limited to specific genotypes and associated with significantly high rate
5
of side effects. On the other hand, sofosbuvir (SOF), a nucleoside inhibitor of NS5B RdRp,
6
quickly became the backbone of interferon free HCV treatment due to its high potency,
pan-genotypic activity, high barrier to resistance selection and low incidence of side
7
effects. Thus, SOF, in combination with velpatasvir or voxilaprevir (QD) is now part of
8
the most prescribed single tablet regimens for the treatment of HCV genotypes 1-6.
Another pan-genotypic combination regimen comprised of glecaprevir and pibrentasvir
with a 8-week (BID) treatment course has also been approved for adults who have been
previously treated with an NS5A or NS3/4A protease inhibitor and also for new subjects
without cirrhosis and preferably low viral load.Although DAAs have improved the HCV
treatment prospects significantly, the treatment course remains long at anywhere from eight
to twelve weeks. Furthermore, treatment has been dampened due to high cost of currently
available DAA. One company, Gilead, markets the only nucleoside (SOF) and does not yet
permit other competing companies to use SOF as part of their fixed dose combination with
ACS Paragon Plus Environment

Page 5 of 63
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
their DAAs. Therefore, there remains a need for additional safe, pan-genotypic nucleoside
analogs with a high barrier to resistance that could be formulated with other non-Gilead
DAAs. These should be cost effective and when combined with potent pangenotypic DAAs
could lead to a shorter duration of treatment while maintaining very high cure rates.
NS5B RdRp is responsible for replication of HCV RNA and is conserved across all the
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
3
genotypes and, as such, is considered a target of choice for HCV treatment. Nucleoside
inhibitors, such as SOF, are anabolized intracellularly to their 5’-triphosphate form and
compete as alternative substrates of the NS5B polymerase. Once incorporated into the
9
growing viral RNA chain, elongation is blocked and RNA replication is stopped. During
the last decade, various structural modifications have been made to natural nucleosides
6,10,11
with the aim of developing potent and selective anti-HCV nucleoside analogs. Among
those structural modifications, 2’-variations showed higher selectivity rates toward the
HCV polymerase versus human polymerases, thus crossing the first big hurdle for
nucleoside analogs. But so far, SOF (1) is the only nucleoside inhibitor approved by the
FDA despite a large number of nucleoside analogs that reached human clinical trials
(Figure 1). Many of these analogs along with other known preclinical HCV nucleoside
inhibitors possess a 2’-α-OH or isosteric group such as a fluorine atom and many also
11, 12 , 13
contain a 2’-b-C-methyl.
Molecular modeling studies suggest that the 2’-C-methyl
group of the incorporated nucleoside analog might exerts steric clash to the next incoming
14
nucleotide substrate. We theorized that replacement of the methyl group with other groups
of similar size, such as halogens, may exert similar steric effects in the 2’-b-position
resulting in new HCV selective inhibitors.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 63
1
2
3
4
Recently, we reported b-D-2’-deoxy-2’-b-chloro-2’-a-fluoro uridine prodrug, 2, as non-
5
6
15
toxic, potent, pan-genotypic anti-HCV NS5B inhibitor (Figure 1). Pinho et al. reported
7
8
9
that the monophosphate prodrug of β-D-2’-deoxy-2’-dichloro-uridine showed activity
against HCV NS5B (G1b) while in our hands β-D-2’-deoxy-2’-dibromo uridine nucleotide
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
16
prodrug showed micromolar potency. Herein, we describe the synthesis of b-D-2’-deoxy-
2’-b-bromo-2’-a-fluoro nucleosides and their phosphoramidate prodrugs, 3, as potent, pan-
genotypic anti-HCV agents and select the uridine analog as a potential clinical
17
development candidate (Figure 1).
Figure 1. Selected potent HCV antiviral agents and targeted b-D-2’-deoxy-2’-b-bromo-2’-
a-fluoro nucleoside prodrugs.
RESULTS AND DISCUSSION
Chemistry. For the preparation of the 2’-bromo-2’-fluoro nucleoside analogs, we chose
15
to capitalize on our group’s recently developed protocol for the unsymmetrical synthesis
of 2-dihalogenated compounds. Deoxyribonolactone 4 served as the starting material for
ACS Paragon Plus Environment

Page 7 of 63
Journal of Medicinal Chemistry
1
2
3
4
synthesis of purine and pyrimidine nucleosides and was readily obtained by oxidation of
5
6
18
commercially available 2-deoxy-D-ribose (Scheme 1). Protection of 4 with tert-
7
8
9
butyldiphenylsilyl groups provided lactone 5 in 85% yield. Fluorination of lactone 5 with
N-fluorodibenzenesulfonimide (NFSI) in presence of LiHMDS furnished 2-deoxy-2-
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
18
fluoroarabinolactone 6 in 29% yield. Lower reaction yields were obtained in the
fluorination reaction due to a competing beta-elimination of tert-butyldiphenylsilyl
alkoxide from the enolate along with formation of the 2-gem-difluoro (~5%) compound
and a 25% recovery of starting material, 5. Bromination of compound 6 using N-
bromosuccinimide (NBS) in presence of LiHMDS afforded the diastereomeric mixture of
2-bromo-2-fluoro lactones 7 and 8 in approximately 1:1 ratio which were separated by
flash chromatography on silica gel. The stereochemistry of these 2-dihalogenated
compounds were determined at a later stage in the synthesis. Subsequent reduction of 7
with lithium-tert-butoxyaluminium hydride (LiAl(t-OBu)3H) gave lactol 9 as a mixture of
anomers (a/b ratio 1.2/1) in a 90% yield. The lactol 9 was converted to 1-mesylate 10 as
an anomeric mixture by treatment with methanesulfonyl chloride and triethylamine in
quantitative yield.
a
Scheme 1. Synthesis of 2-Br,2-F Lactone and Mesylate 10
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 63
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
a
Reagents and conditions: a) TBDPSCl, imidazole, DMF, rt, 24 h, 85%; b) NFSI,
LiHMDS, THF, -78 °C, 1 h, 29%; c) NBS, LiHMDS, THF, -78 °C, 40 min, b-anomer 29%
and a-anomer 31%; d) Li(t-BuO)3AlH, THF, 2 h, 0 °C to rt, 90%; e) MsCl, Et3N, DCM, 0
°C to rt, 1 h, 99%.
Coupling of 1-mesylate 10 with silylated uracil in presence of TMSOTf afforded the N-
glycosylated product 11 as an inseparable mixture of a/b (ratio 2/1). Removal of the silyl
groups and flash silica gel chromatographic separation gave b-anomer 12 in 21% and a-
anomer in 46% yield. Phosphorylation of nucleoside 12 using phenyl-L-isopropylalaninyl
19
phosphorochloridate 13 gave phosphoramidate prodrug 14 as mixture of Rp/Sp isomers (7:3
ratio, assignment unknown). Following a similar protocol for the coupling and
deprotection, we separated both a/b isomer of N-4-benzoyl protected cytosine
nucleoside 16.
The lower yields and unfavorable anomer ratios we observed in these coupling reactions
deserves some further comments. We explored the sugar protecting groups, solvents and
20,21
Lewis acids quite extensively and found the above conditions, while not fully optimized,
to provide the best yields and anomer ratios. It is worth noting that our initial attempts to
ACS Paragon Plus Environment

Page 9 of 63
Journal of Medicinal Chemistry
1
2
3
4
3
use Mitsunobu conditions using key lactol 9 and N -benzyluracil yielded a mixture of major
5
6
O-glycosylated and minor N-glycosylated a and b products.
7
8
9
16
Deprotection of the N-4-amino group of b- with methanolic ammonia furnished fully
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
deprotected nucleoside 17 in 93% yield. Phosphoramidate prodrug 18 was obtained as a
mixture of Rp/Sp isomers (ratio 6:4, assignment unknown) by reaction of nucleoside 17 with
phenyl-L-isopropylalaninyl phosphorochloridate, 13 in presence of t-BuMgCl.
a
b
Scheme 2. Synthesis of 2’- -F,2’- -Br Pyrimidine Nucleosides and Their
a
Corresponding Prodrugs
a
4
Reagents and conditions: a) i) Uracil or N -benzoylcytosine, BSA, DCE, 60 °C, 30 min;
ii) TMSOTf, DCE, 80 °C, 5 h, 55-57%; b) 1M TBAF in THF, THF, 0 °C to rt, 1 h, U
analog: b-anomer 21%, a-anomer 46%, C analog: b-anomer 24%, a-anomer: 46%; c)
NMI, THF, rt, 4 h, 45%; d) NH3, MeOH, overnight, 93%; e) 13, t-BuMgCl, THF, 0 °C to
rt, 3 h, 21%.
Mitsunobu conditions were found to provide the best yields for coupling of purine bases
to the 2-Br, 2-F sugar. Hence, coupling of lactol 9 with bis-Boc-adenine gave a mixture of
a/b isomers 19 (a/b ratio 3/7) in 39% isolated yield (Scheme 3). Removal of silyl groups
and chromatographic separation of a/b isomers gave b-anomer 20 in 44% yield. Further
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 63
1
2
3
4
5
6
7
8
9
treatment of 20 with BCl3 afforded adenine nucleoside 21 in 69% yield. Reaction of 20 with
phenyl L-isopropylalaninyl phosphorochloridate 13 in presence of NMI followed by
removal of the Boc groups using 50% TFA in DCM afforded prodrug 22 as a mixture of
Rp/Sp isomers (ratio 7:3, assignment unknown). Likewise, the coupling of lactol 9 with bis-
Boc-2-amino-6-benzyloxypurine afforded a mixture of a/b isomers 23 (a/b ratio 3/7) in
40% yield (Scheme 4). Again, removal of silyl groups and chromatographic separation of
a/b isomers gave b isomer 24 in 44% yield. The benzyl and Boc protection where removed
in one step using BCl3 tofurnish guanine nucleoside 25 in 65% yield. Reaction of compound
24 with phenyl L-isopropylalaninyl phosphorochloridate 13 in presence of NMI followed
by Boc group removal using BCl3 afforded prodrug 26 as a mixture of Rp/Sp (ratio ~ 1:1)
isomers in 20% over two steps.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
a
a
b
Scheme 3. Synthesis of 2’- -F,2’- -Br Adenosine Nucleoside 21 and its Prodrug 22
ACS Paragon Plus Environment

Page 11 of 63
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
a
Reagents and conditions: a) bis-N-Boc adenine, DIAD, PPh3, THF, rt, 24 h, 39%; b) 1M
TBAF in THF, THF, 0 °C to rt, 1 h, b-anomer 44%, a-anomer: 25%; c) 1M BCl3, DCM, -
78 °C to rt, 1 h, 69%; d) (i)13, NMI, THF, 0 °C to rt, 3 h; (ii) 50% TFA-H2O, 0 °C to rt,
overnight, 40% over two steps.
a
a
b
Scheme 4. Synthesis of 2’- -F,2’- -Br Guanine Nucleoside 25 and its Prodrug 26
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 63
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
a
Reagents and conditions: a) 6-O-Bn-bis-N-Boc guanine, DIAD, PPh3, THF, 0 °C to rt,
24 h, 40%; b) 1M TBAF in THF, THF, 0 °C to rt, 1 h, b-anomer 60%, a-anomer: 19%; c)
1M BCl3, DCM, -20 C to rt, 2 h, 65% d) (i)13, NMI, THF, 0 °C to rt, 3 h; (ii) 1M BCl3,
DCM, , -78 °C to rt, 3 h, 20% over two steps.
o
Identification of both a and b anomers was determined by 2D-NOE experiments (Figure
2). In all cases, NOEs between H1’ and H4’ in the b-nucleoside and H1’ with H3’/H5’ in
a-nucleoside were observed.
Figure 2. a/b Anomer assignment for nucleoside 12.
ACS Paragon Plus Environment

Page 13 of 63
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Single crystal of 12 were grown from ethanol by slow evaporation of the solvent. The
crystal structure showed a b orientation of both the uracil group and bromine atom and an
a orientation of fluorine atom and thus confirming S-configuration at 2’-position and
correct anomer selection from the glycosylation reaction (Figure 3).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 3. The ORTEP drawing of nucleoside 12 from X-ray crystal analysis.
Antiviral Profile. Parent nucleosides 12, 17, 21 and 25 and their corresponding
phosphoramidate prodrugs 14, 18, 22 and 26 were evaluated for inhibition of HCV
22
genotype 1b RNA replication in Huh-7 cells using a subgenomic HCV replicon system.
Cytotoxicity in Huh-7 cells was determined simultaneously by extraction and amplification
23
of both HCV RNA and cellular ribosomal RNA (rRNA). In addition, cytotoxicity was
determined in primary human peripheral blood mononuclear (PBM) cells, human
24,25
lymphoblastoid cells (CEM), and African Green monkey Vero cells. The results are
summarized in Table 1. The uracil, adenine, and guanine nucleoside analogs 12, 21, and
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 63
1
2
3
4
5
6
7
8
9
25 were devoid of anti-HCV activity up to 10 μM, while the corresponding monophosphate
prodrugs of the uridine and guanine analogs 14 and 26 were active in the submicromolar
range (EC50’s of 0.4 and 0.6 μM, respectively). Only the cytosine nucleoside analog 17
displayed activity (EC50 = 3.9 μM) without the aid of a phosphoramidate prodrug. However,
formation of its monophosphate prodrug, 18 allowed for a 5-fold increase of potency. This
pattern of activity among the parent nucleoside analogs and their corresponding prodrugs
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
15
is quite similar to what was observed with our recently reported 2’-Cl,2-F series and the
10,21,26
2’Me,2’-F series to which SOF belongs. In addition, none of the compounds tested,
except for reference compound 2’-C-Me-cytidine (NM-107), displayed any cytotoxicity
versus our panel of cell lines.
Uridine is the only RNA nucleoside in which the de novo synthesis of its monophosphate
form does not proceed through the parent nucleoside. Indeed, the biosynthesis of uridine
monophosphate in humans occurs via decarboxylation of 6-carboxy uridine
monophosphate (orotidylate) which is catalyzed by orotidylate decarboxylase. This
intricate pathway explains why uridine nucleoside analogs are generally not converted to
their triphosphate forms and therefore need to be administered as monophosphate prodrugs
to display activity. This also gives uridine derived nucleoside monophosphate prodrugs an
inherent safety advantage in fighting liver diseases since it is well established that
phosphoramidate prodrugs are largely absorbed and metabolized to their monophosphate
forms in the liver. The polarity of the mono-, di-, and tri-nucleotide phosphates traps them
in liver tissue until they are ultimately released to systemic circulation as the parent
27
nucleoside analog and eliminated from the body. These properties, unique to uridine
analogs, can effectively reduce systemic exposure to uridine analog phosphates and
ACS Paragon Plus Environment

Page 15 of 63
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
potential unwanted off target effects. With these considerations combined with the superior
potency and lack of cellular toxicity in our panel of cell lines, the uridine phosphoramidate,
14, was chosen for further study.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Table 1. HCV Genotype 1b Replicon Activity and Cytotoxicity of Synthesized
Nucleosides and Their Phosphoramidate Prodrugs
*
Anti-HCV EC50 µM (SD )
*
Cytotoxicity, CC50 µM (SD )
Compound
EC50
> 10
EC90
> 10
Huh-7
> 10
> 10
> 10
> 10
> 10
> 10
> 10
>10
PBM
> 100
> 100
> 100
> 100
> 100
> 100
> 100
> 100
> 100
CEM
> 100
> 100
> 100
> 100
> 100
> 100
> 100
> 100
> 100
Vero
> 100
> 100
> 100
> 100
> 100
> 100
> 100
> 100
> 100
12
14
27
28
17
18
21
22
25
0.5 (0.05)
0.7 (0.1)
0.9 (0.01)
2.4 (0.1)
0.16 (0.002) 0.28 (0.003)
4.0 (1.6)
0.6 (0.1)
> 10
9.5 (0.3)
1.8 (0.2)
> 10
7.1 (0.2)
> 10
> 10
> 10
> 10
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 63
1
2
3
4
5
6
7
8
9
0.6 (0.1)
2.8 (0.2)
0.5 (0.1)
1.8 (0.3)
9.9 (0.5)
0.9 (0.03)
> 10
> 100
> 100
> 100
> 100
26
2’-C-Me-C
> 100
60 (20) 22 (15)
SOF
> 100 > 100
> 100
> 100
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
*
The SD were determined in Excel using the STDEV function (calculated using the "n-
1" method) based on the average of three independent EC50 and EC90 values.
As phosphoramidate 14 is a diastereomeric mixture at the phosphorous center, each
stereoisomer was studied independently to determine which may be the most promising to
select for human clinical development. The Sp and Rp phosphorous diastereomers were
initially separated by column chromatography providing compounds 27 and 28 (Figure 4).
When larger amounts of phosphorous single diastereomers were needed the Sp and Rp
diastereomers were synthesized by the known literature protocol utilizing isopropyl ((R or
28
S)-(perfluorophenoxy)-(phenoxy)phosphoryl)-L-alaninate (Scheme 5)
Use of this
synthetic approach also allowed us to assign the absolute stereochemistry to 27 and 28.
a
Scheme 5. Synthesis of Diastereomers 27 and 28
ACS Paragon Plus Environment

Page 17 of 63
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
a
Reagents and conditions: a) t-BuMgCl, THF, -5 to 4 °C, 16 h, 68%.
The single diastereomers 27 and 28 were first evaluated for inhibition of HCV genotype
1b RNA replication in Huh-7 cells using a subgenomic HCV replicon system. As noted
with SOF, the two diastereomers have different potencies, but again no cytotoxicity was
seen in any of the four cell lines tested (Table 1). It is interesting to note that in the case of
9
SOF, clone A cells were used and an 18-fold difference was seen for the diastereomers
whereas in our case the difference was only 2.7-fold and only 1.7-fold at the EC90.
Next 27, 28 and SOF were tested versus chimeric replicons GT3a, GT4a, GT5a and
polymerase mutation S282T associated with drug resistance (built on a GT1b backbone),
GT1a and GT1b stably transfected replicons systems in Huh-7 cells and finally in a GT2a
infectious assay (Table 2). While both diastereomers were pan genotypic versus all
genotypes tested, 28 was more potent than 27 against all genotypes tested. In this assay a
bigger difference with 27 and 28 was noted versus GT1b with a 3.8-fold difference at the
EC50 and a 7.2-fold at the EC90. Similar to SOF, compound 28 displayed an 8 to 9-fold
decrease in potency versus the S282T mutant while isomer 27 was 38 times less potent. On
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 63
1
2
3
4
a highly positive note, we found that in these assays, phosphoramidate 28 was equipotent
5
6
or more potent versus SOF at both the EC50 and EC90 levels.
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Table 2. Potency of 27, 28 and SOF in Huh7 Cells Against Various HCV Genotypes and
the Mutant S282T Virus.
*
EC50 and EC90 (μM) and (fold increase versus GT1b-WT)
(Chimeric replicons, 1a/1b replicons and 2a infectious assay)
GT1a
GT2a
GT1b-WT
GT1b/3a
GT1b/4a
GT1b/5a
GT1b/S282T
Cmpd
EC50
0.12 ± 0.07 (1.3) 0.16 ± 0.0 (1.7) 0.092 ± 0.0 (1) 0.23 ± 0.01 (2.5) 0.17 ± 0.04 (1.8) 0.21 ± 0.09 (2.3) 3.6 ± 0.12 (38)
27
EC90
EC50
EC90
0.48 ± 0.004 (0.9)0.83 ± 0.01 (1.6) 0.53 ± 0.7 (1)
1.6 ± 0.02 (3.1) 0.87 ± 0.23 (1.6) 0.80 ± 1.5 (1.5) 21 ± 0.3 (40)
0.055 ± 0.0 (2.3) 0.046 ± 0.15 (1.9)0.024 ± 0.0 (1) 0.038 ± 0.0 (1.5) 0.021 ± 0.0 (0.9) 0.030 ± 0.01 (1.3) 0.18 ± 0.01 (8)
0.19 ± 0.017 (2.6)0.077 ± 0.0 (1.0) 0.074 ± 0.002 (1) 0.20 ± 0.001 (2.7) 0.13 ± 0.003 (1.7) 0.083 ± 0.02 (1.1) 1.2 ± 0.1 (16)
28
EC50
EC90
0.061 ± 0.0 (1.8) 0.043 ± 0.01 (1.3)0.034 ± 0.0 (1) 0.064 ± 0.0 (1.8) 0.058 ± 0.0 (1.7) 0.044 ± 0.01 (1.3) 0.30 ± 0.02 (9)
0.32 ±0.06 (2.1) 0.13 ± 0.02 (0.9) 0.15 ± 0.03 (1) 0.37 ± 0.01 (2.5) 0.15 ± 0.02 (1.0) 0.32 ± 0.15 (2.2) 4.4 ± 0.62 (29)
SOF
*
The SD were determined in Excel using the STDEV function (calculated using the "n-
1" method) based on the average of three independent EC50 and EC90 values.
Next, we looked at the functional strength of inhibitors 27 and 28 by evaluating the
corresponding 5’-triphosphate, 12-TP, versus a panel of HCV NS5B polymerases (Table
3). While the three inhibitors tested were all pan genotypic versus GT1-6 the IC50’s were
somewhat surprising in that NM-107-TP was by far the most potent when compared to 12-
TP and 2’-Me,2’-F UTP; which is not the case in cell culture nor in humans (in the case of
NM-107-TP). These observations underscore the inherent obstacles in predicting clinical
efficacy based on in vitro data. Also 12-TP was noticeably less potent versus 2’-Me, 2’-F
ACS Paragon Plus Environment

Page 19 of 63
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
UTP which one would not predict based on the cell culture data presented above. All three
nucleoside 5’-triphosphate analogs had a loss of activity versus S282T with NM-107-TP
having a 570-fold loss in activity versus this mutation. Interestingly, although 12-TP had a
70-fold loss in activity versus the S282T mutation we have been unable to select this
mutation in cell culture after multiple attempts.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Table 3. 12-TP IC (µM) versus Various Genotypes the NS5B Mutant S282T
50
IC µM (SD)
50
GT1b
Cmpd
GT1b
GT1a
GT2a
GT3a
GT4a
GT5a
GT6a
S282T
0.64
(0.018)
0.38
(0.0065)
1.2
(0.17)
1.8
(0.078)
1.8
(0.048)
1.2
(0.061)
0.59
(0.025)
45
(2.3)
12-TP
IC50
1
0.59
1.9
2.8
2.8
1.9
0.92
70
fold
change
2’-
0.18
0.096
(0.0015)
0.29
(0.035)
0.38
(0.013)
0.57
(0.016)
0.43
(0.012)
0.13
(0.0022)
7.3
(0.41)
Me,2’- (0.0012)
F-UTP
IC50
1
0.53
1.6
2.1
3.2
2.4
0.72
41
fold
change
2’-Me- 0.063
0.025 0.19
(0.00018) (0.020)
0.047
(0.0021)
0.07
(0.0013)
0.097
(0.004
6)
0.047
(0.0008
9)
36
(1.7)
UTP
(NM-
107-
TP)
(0.0015)
1
0.40
3.0
0.75
1.1
1.5
0.75
570
IC50
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 63
1
2
3
4
5
fold
change
6
7
8
9
Cellular Pharmacology. To better understand the difference in cell culture potency of
the two diastereomers 27 and 28, and also to explain the discrepancy between the cell
culture data for 28 and SOF versus the enzymology data for their TPs, we undertook a
study on the cellular uptake and egress profile of 27, 28 and SOF in Huh-7 cells, and
determined the Tmax and half-lives of nucleoside 5’-triphosphate metabolites. Uptake
incubations were performed at 10 µM for 2, 4, 8, 12, 24, 36 and 48 h (Figure 5). Compound
28 provided higher levels of NTP versus 27 at every time point studied except 48 h and
also more rapidly produces NTP (Tmax = 12 h and 24 h, respectively), thus explaining the
difference in HCV replicon potency for the two diastereomers. Also, at every time point
(except 48 h) 28 delivers substantially more NTP intracellularly versus SOF, again
explaining the discrepancy observed above between the cell culture and enzymology data.
Next, to understand if the difference on potency and NTP formation for 27 and 28 was
an artifact of the Huh-7 replicon system or if it might behave this way in other cell systems
and ultimately in humans, we repeated the above uptake study but this time with primary
human hepatocytes (Figure 5). The observed NTP levels for all phosphoramidates
derivatives was substantially higher than that observed in Huh-7 cells and more
importantly, 27 and 28 gave virtually identical NTP levels and Tmax’s (8 h). On the other
hand, SOF was slower to deliver its NTP with a Tmax of 12 h, but it ultimately delivered
NTP levels that were higher versus 27 and 28.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment

Page 21 of 63
Journal of Medicinal Chemistry
1
2
3
4
5
Level of UTP analog in Huh-7 cells
6
7
8
9
1,600
1,200
800
400
0
27
28
SOF
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
0
12
24
36
48
Time (h)
Level of UTP analog in primary human
hepatocytes
27
28
SOF
5,000
4,000
3,000
2,000
1,000
0
0
12
24
36
48
Time (h)
Figure 5. Cellular uptake of 27, 28 and SOF in Huh-7 cells and primary human hepatocytes
6
(pmol/10 cells).
The egress profile of 27, 28 and SOF were next studied in Huh-7 cells to determine their
T1/2. Egress incubation was done in triplicate at 10 µM with pre-treatment for 24 h based on
the above uptake studies, then a change to new media followed by harvesting cells at 0, 2,
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 63
1
2
3
4
5
6
7
8
9
4, 8, 12, 24, 36 and 48 h (Figure 6). As one might predict 27 and 28 did not show much
difference in egress profile (T1/2 = 15.4 h and 15.7 h, respectively) but both had shorter half-
lives versus SOF (T1/2 = 28.2 h). We also did a similar egress study in primary human
hepatocytes in triplicate at 10 µM with pre-treatment for 12 h and 27, 28 and SOF show
similar egress profiles with T1/2 of 15.3 h, 15.0 h and 15.3 h, respectively. Armed with this
information, selection of one diastereomer to move forward could not be achieved based
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
9
solely on cell culture potency and enzymology data as was done with SOF.
Level of UTP analogs in Huh-7 cells
1,400
27
28
SOF
1,200
1,000
800
600
400
200
0
0
12
24
36
48
Time (h)
ACS Paragon Plus Environment

Page 23 of 63
Journal of Medicinal Chemistry
1
2
3
4
5
6
Level of UTP analogs in primary human
hepatocytes
7
8
9
27
8,000
6,000
4,000
2,000
0
28
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
SOF
0
12
24
36
48
Time (h)
Figure 6. Cellular egress profiles of prodrug 27, 28 and SOF in Huh-7 cells and primary
6
human hepatocytes (pmol/10 cells).
Cytotoxicity Profile. Nucleoside analogs are known to potentially exhibit mitochondrial
toxicity.29 Mitochondrial dysfunction commonly involve the peripheral nervous system
and brain, the retina, cardiac and other muscles; and the endocrine, renal, gastrointestinal,
hematologic, and hepatic systems which can ultimately result in stroke, pancreatitis, lactic
acidosis with liver failure, hepatic steatosis, and myopathy.30 Therefore, the potential
liability for the phosphoramidates 14, 27, 28 and the parent nucleoside 12 were evaluated
and compared to SOF for their effects on mitochondrial DNA levels. HepG2 cells were
propagated in the presence of nucleotide analogs (up to 50 μM) for 14 days prior to
quantification of mitochondrial COXII DNA (mtDNA) and ribosomal DNA using real-
time PCR. Lamivudine (3TC) and β-D-2′,3′-dideoxycytidine (ddC) (at 10 μM) were used
as negative and positive controls, respectively (Table 6). At the end of the 14-day assay,
neither the parent nucleoside, 12 nor the prodrug diastereomer mixture, 14 showed
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 63
1
2
3
4
5
6
7
8
9
measurable mitochondrial toxicity up to 50 μM in HepG2 cell line (Table 6) whereas ddC,
as anticipated, was highly toxic at 10 µM. Similar suppression of nuclear DNA for 28 and
SOF at 50 μM was noted, with 28 having more impact on mitochondrial DNA levels versus
SOF. Lactic acid levels were also measured in the culture supernatant after 14 days of
incubation with each drug. Increased production of lactic acid (generally above 100% when
normalized to ribosomal DNA control) is a marker for the HepG2 cells being under stress
and associated with mitochondrial toxicity.31 In this study, we did not observe increased
lactic acid production with the parent nucleoside 12 nor its phosphoramidate diastereomer
mixture 14 up to 50 μM. Conversely, at 50 μM an increased lactic acid production was
observed for 28, SOF and the positive control, ddC (at 10 μM) (Table 6). However, no
mitochondrial toxicity was noted with 28 when evaluated at 10 µM.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
HepG2 cells are highly proliferative immortalized cells that derive a significant
proportion of their energy from glycolysis rather than mitochondrial oxidative
phosphorylation. The use of glucose containing media with this cell line can mask the
effects of potential mitochondrial toxicants. This is referred to as the Crabtree effect and
can be evaluated by comparing the cytotoxic effect of drug candidates in glucose versus
galactose supplemented media. In galactose containing media the HepG2 cells are forced
to generate energy by mitochondrial oxidative phosphorylation so toxic effects of a drug
on mitochondrial function is more pronounced. Therefore, 14 was also tested in HepG2
cells for 3 days in glucose and galactose supplemented media up to 100 μM and no toxicity
signal was observed in either media. Extending this assay to 14 days for 27 and 28 showed
no significant difference between using glucose or galactose supplemented assay media
and again both 27 and 28 were not toxic up to 100 μM.
ACS Paragon Plus Environment

Page 25 of 63
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Pluripotent hematopoietic stem cells are found in the primary bone marrow of healthy
adults. These cells proliferate and differentiate into all mature hematopoietic cells. When
cultured in methylcellulose, individual progenitors called colony-forming cells proliferate
and differentiate to form colonies of identifiable progeny. Inhibition of the burst forming
unit-erythroid (BFU-E) lineage will present in-vivo as red blood cell loss and anemia.
Inhibition of colony-forming unit-granulocyte, macrophage (CFU-GM) lineage will
present in-vivo as neutropenia.32
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
While 12, 14 and 27 were devoid of BFU-E and CFU-GM bone marrow toxicity up to
100 μM versus three separate donors, when 28 was tested we found an IC50 = 72 μM for
BFU-E and an IC50 = 76 μM for CFU-GM. However, no marked cytotoxicity was noted at
more physiologic concentrations of 10 µM. In all of these bone marrow toxicity assays we
found the positive control AZT exhibited the expected high toxicity in both erythroid and
myelomonocytic progenitors at the concentration tested, and 3TC exhibited no toxicity
with both progenitors with an IC50’s of > 100 µM.
Table 6. Effects of Compounds 12, 27, 28, and SOF on Mitochondrial (Mt), Nuclear
DNA Levels and Lactic Acid Production in HepG2 Cells (14-day assay)
IC50, µM
MtDNA /
nDNA
MtDNA Content
Lactic
Conc.,
% Inhibition
Cmpd
% of control acid production
µM MtDNA / nDNA
(Range)
72
(% of control)
14 ± 3.7 /
50
< 1 ± 0.004
69 ± 14
12
> 50 / > 50
(63-83)
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 63
1
2
3
4
5
6
7
8
9
39 ± 5.8
100
(99 - 104)
65
140 ± 17
100 ± 6.2
90 ± 14.6
120 ± 1.1
220 ± 6.5
79 ± 17
14
27
50
50
> 50 / > 50
> 50 / > 50
/ 40 ± 3.9
31 ± 2.8 /
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
< 1 ± 0.002
(65 - 65)
86
23 ± 7.9 /
11 ± 6.5
31 ± 5.6 /
39 ± 2.9
58 ± 7.7 /
72 ± 9.0
12.5
25
(79 - 93)
110
46 ± 8.0 /
32 ± 6.0
28
(110 - 110)
150
50
(150 - 150)
96
6.7 ± 5.1 / 3.1 ±
6.4
12.5
(90 - 100)
> 50 /
SOF
45 ± 2.5
130
44 ± 2.8 / 58 ±
3.0
50
10b
10b
180 ± 12.5
83 ± 35
(120 - 150)
3TC
ddC
<10 / <10
84 ± 9.0 /
50 ± 4.5
> 10 / > 10
< 10 / < 10
140
7.3
200 ± 11
Untreated
control
10b
0 / 0
N/A
100 (72 - 140)
100 ± 5.0
Cmpd, compound; Conc., concentration; MtDNA, mitochondrial DNA; nDNA, nuclear
DNA, N/A, not applicable
ACS Paragon Plus Environment

Page 27 of 63
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Compound 27 and 28 were also evaluated for cytotoxicity in HepaRG cells in a 14-day
assay and the CC50 values were 7.2 µM and 35 µM, respectively (Table 7). While there is
a mild toxicity signal in this HepaRG cell line, we found SOF to have a CC50 of 44 µM,
very similar to what we found with 28. The cytotoxicity potential of 27 and 28 was also
evaluated in primary human hepatocytes up to 200 µM and found IC50’s of 140 µM and 46
µM, respectively. We tested in vitro for nephrotoxicity in HK-2 cells in a 3-day assay and
found that 28, 27 and SOF all had CC50’s > 300 µM. We evaluated the potential for cardiac
toxicity by incubating 28, 27 and SOF with human embryonic stem cell-induced ventricular
cardiomyocytes and all three compounds had CC50’s > 100 µM (Table 7).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Next, 27 and 28 were evaluated for potential effects on hERG (human ether-à-go-go-related
gene) potassium channels using CHO cells stably expressing hERG potassium channels at
room temperature utilizing the whole-cell patch clamp technique. Both compounds were
determined to have IC50 values greater than 30 µM while the positive control, amitriptyline
had an IC50 of 3 µM (Table 7). For comparison, SOF has been reported to be free from
hERG channel inhibition,33,34 although symptomatic bradycardia has been reported when
co-administered with amiodarone.35 However, these effects were found to be non-hERG
related.36
Table 7. Cytotoxicity (IC50; µM)of Compounds 27 and 28 in Various Cell Lines
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 63
1
2
3
4
5
6
7
8
9
HepRG
Bone
HepG2
3 day
HepG2
14 day
Bone
14 day
glucose
versus
galactose
7.2
Primary
HK-2
cells
marrow
CFU-
GM
Cardio-
Cmpd
marrow
BFU-E
human
hERG
glucose / glucose /
galactose galactose
myocytes
hepatocytes
3 day
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
27
28
> 100
> 100
> 100
> 100
> 100
72*
> 100
76**
140
46
> 300
> 300
> 100
> 100
> 30
> 30
35
*Donor specific data 15.8, >100, >100 µM
**Donor specific data 26.5, >100, >100 µM
A Mini Ames study was conducted to evaluate 27 and 28 ability to induce reverse
mutations both in the presence and absence of S9 mix at the histidine locus in the genome
of four strains of Salmonella typhimurium (TA98, TA100, TA1535, and TA1537) and at
the tryptophan locus in the genome of Escherichia coli WP2 uvrA (pKM101). For 27 and
28 cytotoxicity or precipitate was not observed under any of the test conditions. Both test
articles did not induce more than 2-fold increase in strain TA98, TA100, and WP2 uvrA
(pKM101) nor 3-fold increase in strain TA1535 and TA1537 in the mean number of
revertant colonies at any dose level relative to the concurrent negative/solvent control,
either in the presence or absence of the S9 mix and no dose response was observed with
any strain.
+
In contrast, for all the bacterial tested strains used in this study, the mean number of his
+
and trp revertant colonies observed for the negative/solvent control was comparable to the
laboratory historical negative control data. All positive controls induced the expected
increase more than three-fold in the mean number of revertant colonies, in the presence
and absence of S9 mix, when compared to the concurrent negative/solvent control. This
ACS Paragon Plus Environment

Page 29 of 63
Journal of Medicinal Chemistry
1
2
3
4
33
study concluded that, like SOF, both 27 and 28 were negative for mutagenicity under the
5
6
conditions of this study.
7
8
9
Selectivity for cellular RNA polymerase. We also tested whether nucleoside 5’-
triphosphate analog of 14 was a substrate for the human mitochondrial RNA polymerase
(POLRMT). Each nucleoside triphosphate was incubated at 100 μM with POLRMT
enzyme and the appropriate DNA/RNA primer/template hybrid and incorporation was
evaluated at 2 h. NTP analog incorporation was normalized to that of natural rNTP
substrates. As shown in Table 8, 12-TP was incorporated 7.4% as compared to natural
UTP. This value was comparable to the active metabolite of SOF (3.1%) and 2’-C-Me-
UTP (9.8%). Similar to 2’-C-methyl-2’-F-UTP, the active metabolite of SOF, 12-TP was
an exceedingly poor substrate for POLRMT.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Table 8. Incorporation of 12-TP into the Mitochondrial RNA Polymerase (POLRMT)
POLRMT %
incorporation
Inhibitor
12-TP
2’-F,2’-C-Me-UTP
2’-C-Me-UTP
UTP
7.4 ± 1.3
3.1 ± 1.4
9.8 ± 4.7
100
The effect of 12-TP with human DNA polymerases was evaluated (Table 9). Inhibition
of cellular DNA polymerases α, β and γ was determined in vitro using commercially
available enzymes and appropriate DNA primers and templates and incubated with
ACS Paragon Plus Environment