Concise syntheses and HCV NS5B polymerase inhibition of (2′R)-3 and (2′S)-2′-ethynyluridine-10 and related nucleosides

Article abstract of DOI:10.1016/j.bmcl.2017.06.064

(2′R)-Ethynyl uridine 3, and its (2′S)-diastereomer 10, are synthesised in a divergent fashion from the inexpensive parent nucleoside. Both nucleoside analogues are obtained from a total of 5 simple synthetic steps and 3 trivial column chromatography purifications. To evaluate their effectiveness against HCV NS5B polymerase, the nucleosides were converted to their respective 5′-O-triphosphates. Subsequently, this lead to the discovery of the 2′-β-ethynyl 18 and -propynyl 20 nucleotides having significantly improved potency over Sofosbuvir triphosphate 24.

Full text of DOI:10.1016/j.bmcl.2017.06.064

Accepted Manuscript
Concise syntheses and HCV NS5B polymerase inhibition of (2’ R)-3 and (2’
S)-2’- ethynyluridine-10 and related nucleosides
Frank Bennett, Alexei V. Buevich, Hsueh-Cheng Huang, Vinay
Girijavallabhan, Angela D. Kerekes, Yuhua Huang, Asra Malikzay, Elizabeth
Smith, Eric Ferrari, Mary Senior, Rebecca Osterman, Lingyan Wang, Jun Wang,
Haiyan Pu, Quang T. Truong, Paul Tawa, Stephane L. Bogen, Ian W. Davies,
Ann E. Weber
PII:
S0960-894X(17)30666-2
DOI:
http://dx.doi.org/10.1016/j.bmcl.2017.06.064
BMCL 25098
Reference:
Bioorganic & Medicinal Chemistry Letters
To appear in:
Received Date:
Revised Date:
Accepted Date:
8 May 2017
22 June 2017
23 June 2017
Please cite this article as: Bennett, F., Buevich, A.V., Huang, H-C., Girijavallabhan, V., Kerekes, A.D., Huang, Y.,
Malikzay, A., Smith, E., Ferrari, E., Senior, M., Osterman, R., Wang, L., Wang, J., Pu, H., Truong, Q.T., Tawa, P.,
Bogen, S.L., Davies, I.W., Weber, A.E., Concise syntheses and HCV NS5B polymerase inhibition of (2’ R)-3 and
(2’ S)-2’- ethynyluridine-10 and related nucleosides, Bioorganic & Medicinal Chemistry Letters (2017), doi: http://
dx.doi.org/10.1016/j.bmcl.2017.06.064
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Bioorganic & Medicinal Chemistry Letters
Concise syntheses and HCV NS5B polymerase inhibition of (2’R)-3 and (2’S)-2’-
ethynyluridine-10 and related nucleosides
Frank Bennetta, Alexei V. Buevichb*, Hsueh-Cheng Huangc* Vinay Girijavallabhana, Angela D. Kerekesa,
Yuhua Huanga, Asra Malikzayc, Elizabeth Smithc, Eric Ferraric, Mary Seniorb, Rebecca Osterman,b
Lingyan Wanga, Jun Wanga, Haiyan Pua, Quang T. Truonga, Paul Tawac, Stephane L. Bogena, Ian W.
Daviesa, and Ann E. Webera
a Merck & Co., Inc., MRL., Department of Medicinal Chemistry, 2015 Galloping Hill Rd, Kenilworth, NJ 07033,USA
b Merck & Co., Inc., MRL., Department of Structure Elucidation, 2015 Galloping Hill Rd, Kenilworth, NJ 07033,USA
cMerck & Co., Inc., MRL., Department of Viirology, 2015 Galloping Hill Rd, Kenilworth, NJ 07033,USA
ARTICLE INFO
ABSTRACT
Article history:
Received
Revised
Accepted
Available online
(2’R)-Ethynyl uridine 3, and its (2’S)-diastereomer 10, are synthesised in a divergent fashion
from the inexpensive parent nucleoside. Both nucleoside analogues are obtained from a total of
5 simple synthetic steps and 3 trivial column chromatography purifications. To evaluate their
effectiveness against HCV NS5B polymerase, the nucleosides were converted to their respective
5’-O-triphosphates. Subsequently, this lead to the discovery of the 2’--ethynyl 18 and -
propynyl 20 nucleotides having significantly improved potency over Sofosbuvir triphosphate 24.
2009 Elsevier Ltd. All rights reserved.
Keywords:
Nucleoside
Hepatitis
Cancer
 Corresponding authors. Tel.: +0-908-740-4602; fax: +0-908-740-6136; e-mail: frank.bennett@merck.com. Tel.: +0-908-740-3990: fax: +0-
908-740-4042; e-mail: alexei.buevich@merck.com. Tel.: +0-908-740-4929; e-mail: hseu-cheng.huang@merck.com.

For more than half a century nucleoside analogues have
found widespread therapeutic application.1 During this period,
base and sugar modified derivatives of the natural products
remain the foundation for the counteraction of both DNA and
RNA virus infections as well as in the treatment of a variety of
cancers. In most cases, these analogues mimic their physiological
counterparts, being metabolised into the corresponding 5’-O-
triphosphates, subsequently incorporated into the growing DNA
or RNA chain and, as a result, inhibit viral replication or cancer
cell division. In his seminal work, in 1959, Prusoff described the
synthesis and biological activities of 5-iodo-2’-deoxyuridine (a
thymidine analogue; IDU, 1, Figure 1), ultimately to become the
first marketed antiviral agent against Herpes Simplex Virus
(HSV).2 From these ground breaking activities, nucleoside
analogues have been designed to combat a variety of viral
diseases3 such as hepatitis B (HBV),4 human immunodeficiency
virus (HIV)5 and other herpes viruses such as human
cytomegalovirus (CMV).3 Most recently, following more than a
decade of investigation of -D-2’--methylribose derivatives by
a variety of groups,6 Sofosbuvir, a monophosphate prodrug that
incorporates -D-2’-deoxy-2’--fluoro-2’--C-methyluridine (2,
Figure 1) was approved by the FDA in late 2013 for the
treatment of hepatitis C (HCV).7
Scheme 1. Reagents and conditions: a) 1,3-dichloro-1,1,3,3-tetrakis(2-
methylethyl)-disiloxane, pyridine, room temperature, 100%; b) TEMPO,
tBuOCl, KBr, NaHCO3, ice bath to room temperature, 83%; c)
Trimethylsilylacetylene, n-BuLi, THF, Et3N, -70°C; d) Silica gel column
chromatography, compound 9, 69%, followed by compound 8, 6%.
Figure 1. Examples of base and sugar modified nucleosides used in the
treatment of DNA and RNA viruses..
The intermediates 8 and 9 offer a variety of opportunities for
further modification. The simplest, of course, is complete
desilylation, achieved with fluoride. to the titled compounds 3
and 10. For example, on exposure to tetra-n-butylammonium
fluoride, the (2’R)-tri-silyl diastereomer 8, provides the ribose
Towards a similar goal, we wished to effectively replace the
methyl group in nucleosides such as 2 with an ethynyl
functionality (3 and 4, Figure 2) and subsequently investigate
their antiviral properties.
modified nucleoside 3 in 64% yield, following purification by
silica gel column chromatography (Scheme 2). Using the same
protocol, the trisilyl protected nucleoside 9, provided the titled
triol 10.
The nucleoside 3 had been previously prepared in 5 steps from
1,3,5-tri-O-benzoyl--ribofuranose)12 Although the overall yield
in the previous route is higher, the present methodology employs
an alternative using inexpensive reagents in a robust fashion.
Figure 2. Ethynyl analogues 3 and 4 of nucleoside 2..
It is commonly known the syntheses of base and ribose
modified nucleosides is a challenging problem, often resulting in
severe restrictions to potential applications and alternative
strategies have been and are currently being explored.8 Our
syntheses began with the readily available and inexpensive parent
nucleoside 5 (Scheme 1). On exposure to 1,3-dichloro-1,1,3,3-
tetrakis(2-methylethyl)-disiloxane in anhydous pyridine, the
natural product generated the 3’, 5’- diprotected bis-silyl ether 6.9
Although the intermediate 6 is commercially available, it was
convenient, at least on larger scales, to prepare and use this
material in crude form. Subsequent TEMPO-mediated oxidation
provided the ketone 7, again, used in the next step without
purification. On treatment with an excess of freshly prepared
lithium trimethylsilylacetylide at -78oC, in THF, the ketone 7 was
converted to the (2’R)-tertiary-alcohol 8, along with the expected
(2’S)-epimer 9, identified by detailed NMR experiments.10
Fortunately, both diastereomers can easily be separated by silica
gel column chromatography, providing gram quantities of each
component for further manipulation.11
Scheme 2. Reagents and conditions: a) Et4NF:H2O, THF, room temperature
followed by silica gel column chromatography, ( 64% for 3 from 8; 57% for
10 from 9).
In addition, the present route provides intermediates such as 8
and 9 that are more amenable for further modification (Scheme
3). For example, selective deprotection of the trimethylsilyl
group in compound 8 was achieved with methanolic potassium
carbonate, providing the terminal acetylene 11.

Scheme 4. Reagents and conditions: a) Propynylmagnesium bromide, THF,
ice bath, 54%; b) DAST, toluene, -20C, 35%; c) Et4NF:H2O, THF, room
temperature, 63%.
.
.
Subsequent exposure to DAST proceeds with inversion,
delivering the fluoride 16. Removal of the silicon protecting
groups provides the nucleoside 17, a close derivative of 3.
As a potential measure of their effectiveness as antiviral
agents, the aforementioned nucleosides were converted to the
corresponding 5’-O-triphosphates 18-2413 and subsequent
inhibitory activity against the wild-type NS5B polymerase
evaluated (Table 1).14
IC50
(M)
NTP


18
19
20
21
22
23
24
0.09
~54
0.16
3.6
9.9
0.43a
1.1a,15
Scheme 3. Reagents and conditions: a) K2CO3, MeOH, room temperature,
40%; b) Boc2O, DMAP, room temperature, 67%; c) n-BuLi, HMPA MeI, -
70°C to room temperature, 53%; d) 20% TFA in CH2Cl2, room temperature,
83%; e) TBAF, room temperature, 71%.
aPrepared and tested as dimethylhexylammonium salts.
Table 1 Activities of the nucleoside triphosphates against NS5B polymerase.
The exposed hydroxyl and base imide are masked as their tert-
butyl carbonate and carbamates 12, prior to alkylation with
iodomethane in the presence of n-butyllithium and HMPA to the
protected nucleoside 13. Finally, removal of the protecting
groups provides the C2’-3-carbon uridine analogue 14.
As expected, the ribo-analogue 18 was considerably more
potent than its xylo-counterpart 19, and approximately 4-fold
more active than the C2’--Me analogue 23. Introduction of a
methyl group at the terminii of the acetylene was tolerated. The
propynyl derivative 20 was slightly less potent than 18, while in
the fluoride series the 3-carbon C2’ analogue 22 was 2-3 times
less active than the ethynyl 21. Importantly, the triphosphate
derived from Sofosbuvir 24 is only 3-fold more potent than than
it’s 2’-acetylene analogue 21, while being an order of magnitude
less effective than the corresponding hydroxy analogue 18.
As mentioned above, an advantage of the current
methodology, at least from a medicinal chemistry perspective, is
the availability of the 2’-(S)-intermediates, such as 9, for further
modification. In a separate approach, treatment of the ketone 7
with propynylmagnesium chloride provided the alcohol 15, as the
exclusive product (Scheme 4).
The anti-HCV activities, evaluated in the 1b genotype replicon
assay and cytotoxicity of the most promising nucleosides were
evaluated and are shown in Table 2.16
EC50
(M)
CC50
(M)
Nuc.


3
14
4
>100
>100
N.T.a
>100
>100
>100
>100
>100
17
aNot Tested.

intermediate (6; 39.9g; 100%). LCMS: MH+, 487.48. This material was used
in the next step (below) without purification.
Table 2. Anti-HCV and Cytotoxicity of Nucleosides.
Sodium bicarbonate (20.67g; 246mmol) was dissolved in water (80ml) and
added to a stirred solution of the crude alcohol (6; 39.9g; 82mmol), TEMPO
(1.92g; 12.3mmol) and potassium bromide (1.46g; 12.3mmol) in
dichloromethane (120ml). The resulting mixture was cooled in an ice bath
and stirred vigorously before the dropwise addition of sodium hypochlorite
(189ml of a 5.75% aq. solution; 176mmol). The reaction was stirred until
LCMS indicated that the reaction was complete (approx. 1h.) then partitioned
between 10% aq. sodium thiosulfate and dichloromethane. The organic phase
was separated, washed with an additional portion of sodium thiosulfate, brine,
dried (MgSO4), and the volatiles removed under reduced pressure to give the
crude ketone (7; 33.0g; 83%) that co-exists with its hydrate form, consistent
with a previous observation in a similar system..11 LCMS: MH+, 485.50 and
(M+H2O+H)+, 503.27. This material was used in the next step (below)
without purification.
Of the nucleosides analyzed (3, 14 and 17) none displayed any
appreciable activity up to a concentration of 100M. To exert
anti-HCV activity, the nucleosides must first be metabolized by
kinases to the 5’-O-triphosphates. Based on experience with
other C2’-uridine nucleosides (including 2, the key component of
Sofosbuvir), the rate-limiting process in this step-wise
transformation, is the initial phosphorylation, and hence the lack
of potency displayed in Table 2 is not surprising.17 Importantly,
none of the nucleosides tested demonstrated cytotoxicity.
In summary, a series of nucleosides bearing an ethynyl or
propynyl substituent at the C2’--position and a hydroxyl or
fluoro as the -substituent were prepared. The synthetic route
provides rapid access to both C2’-epimeric intermediates 8 and 9
and hence allowing flexibility in the design of novel nucleosides.
To evaluate their potential as anti-HCV agents, the corresponding
5’-O-triphosphates were prepared and their inhibition of wild
type HCV NS5B polymerase was evaluated and compared to 24,
the active triphosphate of sofosbuvir. While the fluoro derivative
21 was approximately 3 times less potent, the -substituted
ribose analogues 18 and 20 were 7- and 10-fold more potent than
sofosbuvir triphosphate respectively. Although the nucleosides
did not display anti-viral activity, monophosphate prodrugs, such
as the one demonstrated in Sofosbuvir, have been developed to
by-pass the inefficient initial kinase step. Application of this and
other studies related to the nucleosides described in this paper
will be presented elsewhere.
n-Butyllithium (77ml of a 1.6M solution in hexane; 124mmol) was added
dropwise to a stirred solution of trimethylsilylacetylene (12.16g; 124mmol) in
anhydrous THF (125ml) at -78oC, under an atmosphere of nitrogen. When the
addition was complete, the solution was stirred for a further 30min. and the
crude ketone (7; 20.00g; 41.3mmol) in THF (40ml) was added. After stirring
for a further 3h., the reaction was quenched by the addition of sat. aq.
ammonium chloride followed by EtOAc. The organic layer was separated,
dried (MgSO4) and the volatiles removed under reduced pressure. The
residue was purified by silica gel column chromatography using 0 to 20%
EtOAc in hexanes as eluent to give (2‘-S)-intermediate (9; 16.60g; 69.0%).:
1H NMR (CDCl3):  8.18 (br. s, 1H), 7.87 (d, J=8.2Hz, 1H), 6.05 (s, 1H),
5.70 (dd, J=8.2 and 2.3Hz, 1H), 4.16 (dd, J=13.4 and 1.1Hz, 1H), 4.11 (d,
J=9.3Hz, 1H), 4.01 (dd, J=13.4 and 2.5Hz, 1H), 3.96 (ddd, J=9.3, 2.5 and
1.1Hz, 1H), 2.85 (s, 1H), 0.93-1.13 (m, 28H), 0.20 (s, 9H). 13C NMR (CDCl3)
-0.29 (3C), 12.68, 12.91, 13.07, 13.48, 16.67, 16.83, 17.00, 17.02, 17.22,
17.36, 17.39, 17.53, 59.86, 72.95, 76.60, 81.41, 88.91, 94.73, 100.92, 101.67,
139.75, 151.00, 162.70. LCMS: MH+, 583.36. Followed by the (2‘-R)-isomer
(8; 1.46g; 6.1%): 1H NMR (CDCl3): 8.15 (br. s, 1H), 7.60 (d, J=8.1Hz, 1H),
6.08 (s, 1H), 5.70 (dd, J=8.1 and 2.3Hz, 1H), 4.35 (d, J=8.9Hz, 1H), 4.21 (d,
J=12.5Hz, 1H), (H6, 1H), (H4, 1H), (s, 2OH, 1H), 0.98-1.18 (m, 28H), 0.16
(s, 9H). 13C NMR (CDCl3): -0.65 (3C), 12.40, 12.47, 12.94, 13.81, 16.94,
16.97, 17.07, 17.17, 17.19, 17.20, 17.21, 17.34, 59.56, 74.64, 76.15, 81.26,
90.50, 93.16, 101.39, 101.84, 139.53, 149.73, 162.53. LCMS: MH+, 583.36.
10 Key NOE interactions (red arrows) of compound 8.
References and Notes:
1 Jordheim, L. P.; Durantel, D.; Zoulim, F.; Doumontet, C;. Nature Rev.
Drug Discov, 2013, 12, 447.
2 Prusoff, W. H.; Biochim. Biophys. Acta, 1959, 32, 295.
3 De Clerq, E.; Med. Res. Rev., 2013, 33, 1215.
4 (a) Ohishi, W.; Chayama, K.; Hepatol. Res, 2012, 42, 219. (b) Fung, J.,
Lai, C-L.; Seto, W-K.; Yuen, M-F.; J. Antimicrob. Chemother., 2011, 66,
2715. (c) Kim, S. S.; Cheong, J. Y.; Cho, S. W.; Gut Liver, 2011, 5, 278.
11 For a similar synthetic approach see: Wang, G, Beigelman, L.;
WO2013/096680
5 Kukhanova, W. H.; Mol. Biol. 2012, 46, 768.
6 Sofia, M. J.; Adv. Pharmacol. 2013, 57, 39.
12(a) Harry-O’kuru,; R. E.; Kryjak, E. A.; Wolfe, M. S..; Nucleosides
Nucleotides 1997, 16, 1457. (b) Harry-O’kuru, R. E.; Smith, J. M.; Wolfe, M.
S.; J. Org. Chem., 1997, 62, 1754. 11 (a) Harry-O’kuru,; R. E.; Kryjak, E. A.;
Wolfe, M. S..; Nucleosides Nucleotides 1997, 16, 1457. (b) Harry-O’kuru, R.
E.; Smith, J. M.; Wolfe, M. S.; J. Org. Chem., 1997, 62, 1754.
7 (a) Sovaldi (sofosbuvir): FDA press release dated December 6, 2013.
http://www.fda.gov/newsevents/newsroom/pressannouncements/ucm377888.
htm. (b) Sofia, M. J.; Bao, D.; Chang, W.; Du, J.; Nagarathnam, D.;
Rachakonda, S.; Reddy, P. G.; Ross, B. S.; Wang, P.; Zhang, H-R.; Bansal,
S.; Espiritu, C.; Keilmam, M.; Lam, A. M.; Micolochick Steuer, H. M.; Niu,
C.; Otto, M. J.; Furman, P. A.; J. Med. Chem., 2010, 53, 7202.
13 General Preparation of Nucleoside 5‘-Triphosphates The preparation 5'-
triphosphates.(18-22) were carried under contractual agreement with TriLink
Biotechnologies, San Diego, CA.and analysed as the triethylammonium salts
using general synthetic methods: (a) Ludwig, J.; Acta Biochim. Biophys.
Acad. Sci. Hung.1981, 16, 131. (b) Mishra, N. C.; Broom, A. D.; J. Chem.
Soc., Chem. Commun. 1991, 18, 1276. Triphosphates 23 and 24 were
prepared in a similar manner and examined as the dimethylhexylammonium
salts.
8 Mikhailopulo, I. A; Miroshnikov, A. I.; Mendeleev Commun. 2011, 21, 57.
9 Robust preparation of intermediates 8 and 9,. Uridine (5; 20.00g; 82mmol)
was azeotroped with pyridine (approx. 50ml) and then dissolved in anhydrous
pyridine (180ml). 1,3-Dichloro-1,1,3,3-tetraisopropyldisiloxane (30ml; 82ml)
was added dropwise to this solution, while cooled in an ice bath, under an
atmosphere of nitrogen. The resulting reaction mixture was allowed to warm
to room temperature overnight and patitioned between EtOAc and 10% aq.
HCl. The organic phase was separated, washed with a further portion of 10%
aq. HCl, sat. aq. sodium bicarbonate, brine, dried (MgSO4). The volatiles
were removed under reduced pressure to provide the 1‘,3‘-diprotected uridine
14 Inhibition of HCV NS5B Polymerase by Nucleoside Triphosphate
Analogs. This assay is a modified version of the assay described in
International Publication No. WO2002/057287. Briefly, 50 µL reactions
containing 20 mM HEPES (pH 7.3); 7.5 mM DTT; 20 units/ml RNasIN; 1
µM each of ATP, GTP, UTP and CTP; 20 µCi/mL [33P]-CTP; 10 mM MgCl;

60 mM NaCl; 100 µg/ml BSA; 0.021 µM DCoH heteropolymer RNA
template; and 5 nM NS5B (1b-BKΔ55) enzyme are incubated at room
temperature for 1 hour. The assay is then terminated by the addition of 500
mM EDTA (50 µL). The reaction mixture is transferred to a Millipore DE81
filter plate and the incorporation of labeled CTP is determined using Packard
TopCount. Compound IC50 values can then be calculated from experiments
with 10 serial 3-fold dilutions of the inhibitor in duplicate. The intrinsic
potency (Ki) of an NTP inhibitor is derived from its NS5B IC50 using the
Cheng-Prusoff equation for a competitive inhibitor, as described in Cheng et
al., Biochem Pharmacol 22:3099-3108 (1973): Ki = IC50 / (1+[S]/Km), where
[S] = 1 µM, and Km is the concentration of cognate NTP yielding half-
maximal enzyme activity in the assay absent exogenous inhibitors.
15 Murakami, E.; Niu, C.; Bao, H.; Micolochick Steuer, H. M.; Whitaker, T.;
Nachman, T.; Sofia, M. A.; Wang, P.; Otto, M. J.; Furman, P. A.; J.
Antimicrob. Chemother., 2007, 52, 458.
.
16 Cell-Based anti-HCV Activity and Cytotoxicity. Replicon cells (1b-Con1)
are seeded at 5000 cells/well in 96-well plates one day prior to inhibitor
treatment. Various concentrations of an inhibitor in DMSO are added to the
replicon cells, with the final concentration of DMSO at 0.5% and fetal bovine
serum at 10% in the assay media. Cells are harvested three days post dosing.
The replicon RNA level is determined using real-time RT-PCR (Taqman
assay) with GAPDH RNA as endogenous control. EC50 values are calculated
from experiments with 10 serial twofold dilutions of the inhibitor in triplicate.
To measure cytotoxicity in replicon cells of an inhibitor, an MTS assay is
performed according to the manufacturer's protocol for CellTiter 96 Aqueous
One Solution Cell Proliferation Assay (Promega, Cat # G3580) three days
post dosing on cells treated identically as in replicon activity assays. CC50 is
the concentration of inhibitor that yields 50% inhibition compared to vehicle-
treated cells. Effect of an inhibitor on cellular DNA synthesis is determined
by a scintillation proximity assay. Replicon cells are seeded in 96-well
Cytostar-T Scintillating Microplates (PerkinElmer, Cat # RPNQ0163) one
day prior to inhibitor treatment. Various concentrations of an inhibitor in
triplicate are added with [methyl-14C]-thymidine (PerkinElmer, Cat #
NEC568050UC, final concentration 0.5 Ci/mL media) and incubated for
three days. DNA CC50 is the inhibitor concentration that yields 50%
inhibition of labeled thymidine incorporation as measured by Packard
TopCount compared to vehicle-treated cells.
17 Furman,; P. A.; Muramami, E.; Niu, C..; Lam, A. M.; Espiritu, C.; Bansal,
S.; Bao, H.; Tolstykh, T.; Sreuer, H. M.; Keilman, M.; Zennou, V.; Bourne,
N.; Veselenak, R. L.; Chang, W.; Ross, B. S.; Du, J.; Otto, M. J.; Sofia, M. J.;
Antiviral Res. 2011, 91, 120