Synthesis of 2′-Methyl-6-methoxyguanosine from the Parent Ribonucleoside Guanosine

Article abstract of DOI:10.1021/acs.joc.8b02194

A short and efficient synthesis of the nucleoside fragment contained in the NS5B nucleoside inhibitor BMS-986094 was achieved in 23% overall yield on a gram scale. The synthesis uses the widely available starting material guanosine via a short sequence ending in a Mukaiyama hydration reaction to establish the key tertiary alcohol moiety and set the C-2′ methyl stereogenic center. This work resulted in a robust and scalable approach to this complex nucleoside.

Full text of DOI:10.1021/acs.joc.8b02194

Subscriber access provided by UNIVERSITY OF TOLEDO LIBRARIES
Article
The synthesis of 2’-Me, 6-methoxyl guanosine
from the parent ribonucleoside Guanosine
Changxia Yuan, Ke Chen, and Martin D. Eastgate
J. Org. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.joc.8b02194 • Publication Date (Web): 25 Sep 2018
Downloaded from http://pubs.acs.org on September 25, 2018
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 16
The Journal of Organic Chemistry
1
2
3
4
5
6
The synthesis of 2’-Me, 6-methoxyl guanosine from the parent ribonucleoside
Guanosine
7
8
9
Changxia Yuan†,*, Ke Chen‡, Martin Eastgate†,*
†Chemical and Synthetic Development, Bristol-Myers Squibb, 1 Squibb Drive, New Brunswick, New
Jersey 08903, United States
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
‡STA Pharmaceuꢀcal, WuXi AppTec Company, 288 Fute Zhong Road, Shanghai, 200131, China
Abstract:
A short and efficient synthesis of the nucleoside fragment contained in the NS5B nucleoside inhibitor
BMS-986094 was achieved in 23% overall yield on gram scale. The synthesis uses the widely available
starting material guanosine, via a short sequence ending in a Mukaiyama hydration reaction to establish
the key tertiary alcohol moiety and set the C-2’ methyl stereogenic center. This work resulted in a robust
and scalable approach to this complex nucleoside.
Introduction:
Hepatitis C virus (HCV) is a blood-borne pathogen which, until recently, chronically infected an
estimated 71-180 million people. Clinical care for patients with HCV-related liver disease has advanced
considerably thanks to an enhanced understanding of the disease, significant improvements in
treatment options and the commercialization of curative regimens1a. Traditional small molecule
therapeutics have led the way in curing HCV, but other options such vaccines also under development1.
Nucleotide analogues were first line treatments for over 40 years2, and early drugs, such as ribavirin (1)
(Figure 1), proved to be essential for treatment of HCV infection,3 despite significant negative side-
effects. More recently, the approval of sofosbuvir (Solvaldi) (2), a nucleoside featuring 2-C-
methylcytidine with a 5’-O-phosphorylamidate diester, marked a milestone in the treatment of HCV with
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 2 of 16
1
2
3
4
curative regimens for geno-type one and geno-type two of HCV diseases being launched (figure 1a)1b, 4
.
Since its regulatory acceptance, multiple pan-genotypic therapies have been approved, moving this once
incurable disease to one which can be treated by a simple course of tablets.
5
6
7
8
9
As part of BMS’ efforts against this disease, we also explored the area of nucleotide based NS5B
inhibitors5, specifically on the development of BMS-986094-01 (3), which showed pan-genotypic
antiviral activity in vitro. Typical of this class, the phosphoramidate 3 is a monophosphorylated prodrug
of the active guanosine nucleotide triphosphate; in this case the nucleoside component featured an
unnatural sugar backbone with a tertiary alcohol at the C-2’ position. We have previously reported on a
highly efficient asymmetric phosphorylation of this nucleoside, to install the phosphoramidate, leading
to the diastereoselective formation of S-(3)6. In that work, we synthesized a single diastereomer of
nucleoside phosphoramidate S-(3) by coupling the nucleoside with the phosphoric acid derivative 4
(where the phosphorous atom does not contain stereochemical information until activated) through a
dynamic kinetic resolution using quinine. This robust and innovative process enabled us to produce API
(active pharmaceutical ingredient) as a crystalline compound in high purity and as a single stereoisomer
(Figure 1b), avoiding an unselective phosphorylation and separation of the resulting diastereoisomers
through extensive chromatography5a.
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 1: a. Representative APIs for the treatment of HCV. b. Final step (API step) in synthesis of S-(3).
Results and Discussion:
While the API coupling chemistry was effective, an improved method to prepare the nucleotide 5 was
required. The chemistry used to prepare nucleotide 5 had several liabilities (scheme 1): (1) the synthesis
begins with conversion of D-fructose 6 to 7 through a Mailard-Amadori reaction in only 19% yield7, and
there were several isolation patents for 78, (2) Chloro nucleoside 9, derived from 89 by Vorbruggen
ACS Paragon Plus Environment

Page 3 of 16
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
reaction, could not be induced to crystallize, which hampered our ability to control the several key
impurities, and (3) It was very challenging to affect 100% conversion of 9 to 5, and even residual traces
of 9 failed to purge during isolation, leading to a daughter impurity which persisted through to the API.
With all of these factors being taken into consideration, it was clear that an alternative synthetic route
was required for the nucleoside 5 to address the above issues. Herein, we detail our work to develop an
improved synthesis of 5 from widely available guanosine (11).
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
Scheme 1: Original synthesis of nucleoside 3.
Early synthetic development:
The goals of this new route were to avoid the problematic 6-chloro impurities, along with the
complexities of the existing chemical processes. Our retrosynthetic analysis is depicted in scheme 2, the
desired molecule only needed the addition of two methyl groups onto the inexpensive, commercial and
natural occurring guanosine 11, with the only differences being the methoxy group on purine and
methyl group at the C-2’ position on the sugar. The challenges posed by this strategy were the efficiency
of guanine methoxylation and the ability to establish the C-2’ methyl group diastereoselectively. We
pursued two approaches for establishing the C-2’ stereocenter on 5. The highest risk approach was to
synthesize ketone 12, where addition carbon nucleophile can add to the carbonyl center to give correct
diastereomer at the C-2’ position, if addition occurs from the side containing the base potentially
through chelation. In the second option, we could utilize olefin 13, and oxidize the tertiary carbon with
the correct stereochemistry.
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 4 of 16
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
Scheme 2: Retrosynthetic analysis for nucleoside 5.
Our exploration commenced from guanosine tetra-acetate 14, in order to understand the C6
methylation leveraging existing known chemistry (scheme 3)10. We attempted the chlorination with
POCl3 at 100 °C, receiving, as expected, consistently low yields (~10%). We expected that the lack of an
additional chloride source led to the observed poor performance. We followed Uznański’s procedural
modification where Et4NCl is added and we could obtain 70% yield of the desired product at 70 °C.
Unfortunately, the reaction always contained mixtures of the desired product 16 and 17 in 2:1 to 4:1
ratio, the product of ethylation by the tetraethylammonium cation. Since the crystallization at this step
was unable to purge the in-process impurities, we carried out the nucleophilic displacement of sodium
methoxide, to afford 18 in 80% yield (based on 14), obtaining high purity (HPLC) after isolation. However,
we found that after potency correction, the yield of 18 dropped significantly because of the presence of
various salts in the solid. We faced the challenge that, while water could readily dissolve these salts, 18
was also extremely water soluble, making an aqueous wash impractical. We hypothesized that by
installing multiple protecting groups, we could decrease the aqueous solubility of the substrate and use
a typical wash to remove salts. However, guanosine 18 gave inconsistent conversion under standard
silylation conditions, even after purification, and resulted in formation of multiple species. Further
attempts off selective 3’ and 5’ protection over a series of common protecting groups (Bn, Bz, Cbz) failed,
leading us to discontinue development of this route.
Scheme 3: Attempt on 6-purine methoxylation optimization and derivatization.
ACS Paragon Plus Environment

Page 5 of 16
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
Alternatively, we tried to test a protection-deprotection approach starting from guanosine 11 (scheme
4). Following Urata’s report11, Guanosine 19 was smoothly prepared by slow addition of tBu2Si(OTf)2 into
a solution of 11 in DMF at 0 °C. With isolated 19 in hand, we had tried different methoxylation methods
resulting in mixed results. Under the previous POCl3 activation condition, we could observe the
formation of 20, after addition of a sodium methoxide solution, but the reaction afforded variable
purities, and low yields. We tried to deprotonate 19 with different bases, in hope of direct methylating
the desired C-6 oxygen, but we observed competitive methylation of N-1, rather than O-6, resulting in a
messy reaction profile containing 21. Lastly, we considered that protecting substrate 19 with a Boc
group on purine could help the purine activation. However, neither 22, 23 nor 24 were produced under
various protecting conditions.
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
Scheme 4: Unsuccessful derivatization of 19 and 20 under different conditions.
New synthetic route:
With these observations in hand, we revised our strategy moving forward. Early results indicated the
phosphoryl chloride activated intermediate 19 in the desired way and was suitable to undergo selective
methoxy displacement. We also observed that a free amino group on guanosine interfered in various
ways during the reaction. From previous studies, we learned that 19 was a stable crystalline compound
which can serve as an intermediate for derivatization. Based on this understanding, we focused on
changing the conditions for activating the purine on intermediate 19 rather than direct protection.
Inspired by the work of Fathi et al, we found that treatment of 19 with 3 equivalent of TFAA in the
presence of 5 volume of pyridine cleanly afforded the pyridinium adduct 26 (scheme 5a)12. This
intermediate was subsequently quenched with LiOMe in methanol to give product 21 after methoxyl
displacement and cleavage of the trifluoro acyl groups. It is worth noting that KOMe or NaOMe provided
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 6 of 16
1
2
3
4
5
6
7
8
9
greater levels of impurities. The ensuing water quench served as both reagent for trifluoroacetamide
hydrolysis and anti-solvent for crystallization of 21. Finally with a method for the direct functionalization
of C-6 in place, we focused on oxidation of alcohol 21 proceeded smoothly with IBD
(diacetoxyiodo)benzene) and TEMPO to give ketone 2713.
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 we developed a robust process on deca-gram scale for the first three-step (scheme 5b). On step 1,
after slow addition of 0.95 equivalent TBS2SiOTf2 into the suspension of guanosine 11 in DMF below 2 °C,
the crude was aged for 1 h at 20 °C and quenched with water to directly crystallize the silyl ether 19 with
excellent yield and purity. Based on the optimized conditions, 21 was also isolated in 90% yield where
we developed the crystallization after workup. In step 3, a robust IBD/TEMPO oxidation in a mixed
solvent of DCM/AcOH delivered the ketone 27 at ambient temperature in 4 h. The product 27 can be
consistently isolated in 78% yield by crystallization with EtOAc and water (V:V ratio 50:1) on 7 gram scale.
Scheme 5: Reaction defining and progress development for silylether ketone 27.
With ketone 27 in hand, we turned our attention to identifying a functionalization strategy (scheme 6).
Well aligned with Piccirilli’s observation14, direct methyl addition to on ketone 27 afforded poor site and
facial selectivity with methyl Grignard, methyl lithium reagent or Me3Al under various conditions. The
Tebbe or Petasis reagents also decomposed 27. An olefination-oxidation strategy was the next option.
Even though our early attempts at the olefination of 27 afforded poor conversion and no more than 50%
yield with butyl lithium and PPh3MeBr, we wondered if salt effects might interfere with the reaction
conversion – given the allylic nature of the anomeric base being created, increasing the probability of
ACS Paragon Plus Environment

Page 7 of 16
The Journal of Organic Chemistry
1
2
3
4
5
6
7
base elimination15. To our delight, after filtration of the solution of Ph3PMeBr and KHMDS in toluene, a
salt-free Wittig reagent was effective in the methyenylation of ketone 27, affording for 100% conversion
and 71% yield of 29 after isolation.
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
Scheme 6: Olefination of ketone 27 to olefin 29.
To complete the synthesis of nucleoside 5, we needed to perform a diastereoselective oxidation on
olefin 29. Oxidation failed when we utilized 29 with conditions such as mCPBA, VO(acac)2/TBHP or
MTO/H2O2, giving partially decomposed mixtures (scheme 7). We assumed that a new substrate, with
the amine on the nuclear base masked, would reduce the impact of oxidation on the heterocyclic moiety
enabling a clean and selective oxidation. To test this assumption, Bis-Boc protection of 29 was
performed to afford 31. Exploring a broad range of oxidation conditions and solvent options, we found
that mCPBA oxidation in the presence of sodium bicarbonate and dichloromethane, yielded epoxidation
product 32 with high diastereoselectivity16. Due to the lack of stability of 32, the crude was subsequently
reduced by LiAlH4 in THF at 0 °C to form 33. Upon treating 33 with TFA, we can obtain desired product 5
after chromatography. Unfortunately, scale-up efforts on this approach met with a significant issue,
primarily due to the stability of 32 and 33. We found epoxide 32 was partially hydrolyzed to form diol 34
during workup, causing various ratio between 33 and 34. Isolation of 33 and 34 by chromatography also
failed due to their instability on silica gel. Triol 33 was also unstable under our scale up conditions
resulting in low and variable yields (~20%) for the three step-telescope reaction. Thus, although this
approach represented a formal proof of concept for an alternate route, the instability of the
intermediates forced us to explore better options.
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 8 of 16
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
Scheme 7: First effort towards end game to achieve the penultimate intermediate 5.
Revisiting substrate 29, we considered a Mukaiyama hydration since it would deliver the desired
oxidation product in a single step without protecting group manipulation17. Moreover, based on our
assumption that the alpha face of the sugar is less hindered (scheme 6), we believed that oxidation
would occur from the desired face. To our delight, under air in the presence of Co(acac)2 and PhSiH3 in
THF, 29 was smoothly converted to the desired tertiary alcohol 35 (Scheme 8). However, the reaction
yield was inconsistent and low-yielding when run on more than 100 mg scale. To further optimize the
condition, we revisited more catalysts, such as Mn(dpm)3, and more solvents, including MeOH, iPrOH, 2-
MeTHF and TFE. We found that TFE delivered highly consistent result, potentially due to the high oxygen
solubility in fluorinated solvents and dielectric constant of TFE15b, tertiary alcohol 35 was consistently
obtained in 75% yield by combination of Co(acac)2 and PhSiH3 under air18. In the subsequent step, TBAF
smoothly deprotected the silyl group to give final product 5 in 75% yield. After further optimization of
the reaction conditions, crude 35 can be carried through the TBAF deprotection after solvent swap from
TFE to THF, without isolation. The isolated yield is 50% for the two-step procedure on gram scale.
Scheme 8: End game to achieve the penultimate intermediate 5.
Conclusion:
ACS Paragon Plus Environment

Page 9 of 16
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
In summary, we have developed a short and efficient synthesis of the nucleoside 5 from widely available
guanosine (11). This route avoided a low-yielding Mailand-Amadori reaction and an expensive,
proprietary isolation technology, multiple impurities and complex reactions, which all complicated the
use of the existing synthesis on scale. This new approach offers robust procedures for preparing the 2’,6
dimethyl guanosine in six steps, with three crystallizations and three telescope-steps in an overall 23%
yield. We believe that this approach may be amenable to SAR activities, enabling a systematic
modification of alkyl groups at C-6 and C-2’ positions facilitating further exploration of this chemotype.
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

The Journal of Organic Chemistry
Page 10 of 16
1
2
3
4
Experimental Section
5
6
7
8
9
General Experimental Information: All reactions were performed under a nitrogen atmosphere using
anhydrous techniques unless otherwise noted. Reagents were used as received from the vendors, unless
otherwise noted. Quoted yields are for isolated material, and have not been corrected for moisture content.
Reactions were monitored by silica TLC or reverse phase HPLC on a Shimadzu system using
CH3CN/H2O/MeOH as the mobile phase (containing either 0.05% TFA, or 0.1% NH4OAc). NMR-
spectra were recorded on Bruker DRX-400 or DRX-500 instruments, and are referenced to residual
undeuterated solvents. The following abbreviates are used to explain multiplicities: br = broad, s =
singlet, d = doublet, t = triplet, q = quartet, m = multiplet. High resolution mass spectra (HRMS) were
recorded on a Thermo Orbi-trap Discovery instrument. Melting points were recorded using a Thomas
Hoover melting point apparatus and are uncorrected. The quantitative analysis of residual palladium
catalyst was performed with a Perkin-Elmer Optima 4300 DV ICP-AES instrument. UPLC (ultra pressure
liquid chromatography) was recorded on Shimazu UPLC work station, with the following condition:
Gradient: Complex; Start % B: 0; 2 Min. 100%; 2.5 Min. 100%; Stop Time: 4 min; Flow Rate: 1 ml/min;
PDA wl1: 220 nm; Solvent Pair: TFA/MeCN/Water; Solvent A: A3 = 0.05% TFA in MeCN:Water (5:95)
S194; Solvent B: B3 = 0.05% TFA in MeCN:water (95:5) S195; column ascentis express C18 2.7 um 2.1
x 50mm; Oven Temperature: 40.
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
2-Amino-9-((4aR,6R,7R,7aS)-2,2-di-tert-butyl-7-hydroxytetrahydro-4H-furo[3,2-d][1,3,2]dioxasilin-6-
yl)-1,9-dihydro-6H-purin-6-one 19:
To a 500 mL round bottom flask was charged with guanosine 11 (20.0 g, 1.0 eq., 70.6 mmol), DMF (100
mL) and 2, 6-lutidine (17.3 mL, 2.1 eq, 148 mmol). The crude was cooled to 0 °C under nitrogen
atmosphere, and di-tert-butylsilylbis(trifluoromethanesulfonate) (30.0 g, 0.96 eq., 67.1 mmol) was added
slowly while maintaining the internal temperature below 5 °C. (Note: The addition was exothermic.)
Upon completion of the addition, the reaction crude was warmed up to 20 °C and stirred until the
reaction finished (monitored by HPLC). After completion of the reaction (1 h), water (100 mL) was
charged over a course of 5 minutes. The resulting white slurry was stirred at 20 °C for additional 30
minutes, filtered, rinsed with 50% DMF/water (2 X 30 mL) then water (30 mL). The wet cake was dried in
vacuum oven (50 °C) till no weight loss was observed (24 h). Compound 19 was obtained as a white
crystalline solid with desired quality (28.4 g, yield 95%). Data for Compound 19: 1H-NMR (400 MHz,
DMSO-D6, 23 °C) δ = 1.01 (s, 9H), 1.07 (s, 9H), 3.93-4.01 (m, 1H), 4.21-4.31 (m, 1H), 4.28-4.43 (m,
2H), 5.69-5.86 (m, 2H), 6.49 (brs, 2H), 7.93 (s, 1H), 10.67 (s, 1H); 13C-NMR (101 MHz, DMSO-D6, 23 °C)
ACS Paragon Plus Environment

Page 11 of 16
The Journal of Organic Chemistry
1
2
3
4
5
δ = 156.6, 153.8, 150.6, 136.5, 116.3, 89.2, 76.3, 73.7, 73.0, 66.9, 27.3, 27.0, 22.2, 20.1; HRMS (ESI-TOF)
m/z: [M+H]+ calcd for C18H29N5O5Si 424.2016; Found 424.2007; M.P. 200-205 °C.
6
7
8
9
(4aR,6R,7R,7aS)-6-(2-Amino-6-methoxy-9H-purin-9-yl)-2,2-di-tert-butyltetrahydro-4H-furo[3,2-
d][1,3,2]dioxasilin-7-ol 21:
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
Compound 19 (5.0 g, 1.0 eq., 11.8 mol) was charged into a 250 mL round bottom flask that had
previously been flushed with nitrogen. To the flask was added pyridine (25 mL) followed by
trifluoroacetic anhydride (5.3 mL, 3.2 eq., 37.8 mmol). After the reaction mixture was stirred at 20 °C for
30 minutes, lithium methoxide (as 10 wt% solution in methanol; 22 mL, 4 eq., 47.2 mmol) and methanol
(50 mL) were added. The resulting reaction mixture was stirred at 20 °C for 6 h. Water (50 mL) was
charged and the reaction resumed for another 4 h. The white slurry was then filtered, washed with 50%
MeOH/water (2 x 15 mL), and dried in vacuum oven (50 °C) till no weight loss was observed (12 h).
Compound 21 was obtained as a white crystalline solid with desired quality (4.7 g, yield 90%). Data for
1
Compound 21: H-NMR (500 MHz, CDCl3, 23 °C) δ = 1.04 (s, 9 H), 1.08 (s, 9 H), 4.03 (s, 3 H), 4.10
(dd, J = 4.9, 9.3 Hz, 1 H), 4.25 (broad, 1 H), 4.47 – 4.53 (m, 3 H), 5.29 (broad, 3 H), 6.06 (s, 1 H), 7.64
13
(s, 1 H); C-NMR (101 MHz, CDCl3, 23 °C) δ = 161.8, 159.2, 152.7, 137.9, 116.4, 90.5, 75.8, 74.8,
73.7, 67.6, 54.0, 27.4, 27.2, 22.8, 20.4; HRMS (ESI-TOF) m/z:[M+H]+ calcd for C19H31N5O5Si
438.2173; Found 438.2163; M.P. 240-245 °C.
(4aR,6R,7aR)-6-(2-Amino-6-methoxy-9H-purin-9-yl)-2,2-di-tert-butyldihydro-4H-furo[3,2-
d][1,3,2]dioxasilin-7(6H)-one 27:
To a 250 mL round bottom flask was charged compound 27 (6.5 g, 1.0 eq. 14.8 mmol), dichloromethane
(130 mL) and acetic acid (0.85 mL, 1.0 eq., 14.8 mmol). The mixture was stirred at 20 °C for 5 minutes to
form a slightly cloudy solution. Iodosobenzene diacetate (5.7 g. 1.2 eq.; 17.8 mmol) was then added,
followed by 2,2,6,6-tetramethylpiperidine N-oxide (0.46 g; 0.2 eq.; 3 mmol). The resulting reaction
mixture was stirred at 20 °C while the reaction progress was monitored by HPLC. After completion of the
reaction (4 h), the crude was concentrated to remove dichloromethane. The crude oil was dissolved in
ethyl acetate (50 mL) to form a homogeneous solution. To this solution was charged with water (1 mL)
over a course of 5 minutes. The resulting white slurry was stirred at 20 °C for additional 30 minutes,
filtered and rinsed with EtOAc (2 x 10 mL). After drying in vacuum oven (50 °C) till no weight loss was
observed (24 h), compound 27 was obtained as a white crystalline solid with desired quality (5.0 g, yield
77%). Data for Compound 27: 1H-NMR (500 MHz, CDCl3, 23 °C) δ = 1.08 (s, 9 H), 1.14 (s, 9 H), 4.08
(s, 3 H), 4.47 – 4.53 (m, 3 H), 4.58 – 4.61 (m, 1 H), 4.90 (broad, 1 H), 6.18 (s, 1 H), 7.07 (s, 1 H), 7.79 (s,
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 12 of 16
1
2
3
13
1 H); C-NMR (125 MHz, CDCl3, 23 °C) δ = 162.3, 156.2, 152.6, 136.5, 117.6, 87.5, 87.2, 78.1, 75.5,
67.6, 54.0, 27.3, 22.8, 22.6 (Note: this intermediate exists as a gem-diol in standard NMR solvents).
HRMS (ESI-TOF) m/z:[2M+H]+ calcd for C38H59N10O10Si2 871.3954; Found 871.3950; M.P. 185-192
°C.
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
9-((4aR,6R,7aS)-2,2-Di-tert-butyl-7-methylenetetrahydro-4H-furo[3,2-d][1,3,2]dioxasilin-6-yl)-6-
methoxy-9H-purin-2-amine
29
and
Boc
protected
9-((4aR,6R,7aS)-2,2-di-tert-butyl-7-
methylenetetrahydro-4H-furo[3,2-d][1,3,2]dioxasilin-6-yl)-6-methoxy-9H-purin-2-amine 32:
To a 25 mL round bottom flask was charged methyltriphenylphosphonium bromide (2.0 g, 2.5 eq., 5.7
mmol) and toluene (10 mL). The flask was flushed with nitrogen and the white slurry was cooled to 0°C.
Potassium hexamethyldisilazide (1.1 g, 2.4 eq., 5.5 mmol) was charged portion wise over a course of 5
minutes. After addition, the resulting crude was stirred at 0 °C for 1 hour, filtered, and rinsed with toluene
(5 mL). The filtrate was transferred to a 50 mL round bottom flask, flushed with nitrogen, and cooled to 0
°C. To this solution was charged with compound 27 (1.0 g, 1.0 eq., 2.3 mmol) as solid. The resulting
suspension was held at 0 °C for 20 minutes before warming up to 20 °C, while the reaction progress was
monitored by HPLC. After completion of the olefination (6 h), di-tert-butydicarbonate (1.3 g, 2.5 eq., 5.7
mmol) was added in one portion, followed by 4-dimethylaminopyridine (10 mg, 2 mol%, 0.05 mmol).
The reaction crude was warmed up to 45°C and held at 45 °C. After completion of the boc-protection (14
h), the crude was concentrated to remove most of toluene to afford a dark brown solid. Purification by
column chromatography (silica gel, eluting with 10 – 25% EtOAc in hexanes) afforded compound 32 as
off-white foam (1.12 g, 71% isolated yield).
Boc protected 9-((4aR,6R,7aS)-2,2-di-tert-butyl-7-methylenetetrahydro-4H-furo[3,2-d][1,3,2]dioxasilin-
1
6-yl)-6-methoxy-9H-purin-2-amine 32: H-NMR (400 MHz, CD3OD, 23 °C) δ = 1.04 (s, 9H), 1.15 (s,
9H), 1.39 (s, 18H), 3.79 (dd, J = 4.0 and 12.0 Hz, 1H), 4.10 (dd, J = 0.8 and 12.0 Hz, 1H), 4.16 (s, 3H),
4.39 (dd, J = 4.0 and 8.0 Hz, 1H), 5.42 (dd, J = 0.8 and 4.0 Hz, 1H), 5.47 (dd, J = 0.8 and 4.0 Hz, 1H),
13
5.68 (ddd, J = 4.0, 8.0 and 12.0 Hz, 1H), 6.67 (dd, J = 0.8 and 4.0 Hz, 1H), 8.49 (s, 1H); C-NMR (125
MHz, CDCl3, 23 °C) δ = 161.5, 152.3, 152.2, 150.7, 145.3, 140.9, 120.3, 111.0, 83.2, 82.9, 77.3, 76.4,
67.6, 54.6, 27.8, 27.4, 27.0, 22.7, 20.1; HRMS (ESI-TOF) m/z: [M+H] calcd for C30H47N5O8Si
634.3272; Found 634.3268; M.P. 150-158 °C.
9-((4aR,6R,7aS)-2,2-di-tert-butyl-7-methylenetetrahydro-4H-furo[3,2-d][1,3,2]dioxasilin-6-yl)-6-
methoxy-9H-purin-2-amine 29: After formation of the olefin confirmed by HPLC, the crude was added
hexanes and stirred for 30 min. The crude was filtered, concentrated to give brown paste. Purification by
ACS Paragon Plus Environment

Page 13 of 16
The Journal of Organic Chemistry
1
2
3
4
5
column chromatography (silica gel, eluting with 10 – 25% EtOAc in hexanes) afforded compound 29 as
1
white solid (2.51 g; 79% isolated yield on 3.20 gram scale). Data for Compound 29: H-NMR (400
6
7
8
9
MHz, CDCl3, 23 °C) δ = 1.02 (s, 9H), 1.12 (s, 9H), 3.77 - 3.71 (m, 1H), 4.09 - 4.01 (m, 4H), 4.41 (dd,
J=9.3, 5.1 Hz, 1H), 4.90 - 4.84 (m, 3H), 5.37 (dd, J=2.9, 1.6 Hz, 1H), 5.56 (t, J=2.3 Hz, 1H), 6.52 (q,
13
J=1.8 Hz, 1H), 7.63 (s, 1H); C-NMR (101 MHz, CDCl3, 23 °C) δ = 160.7, 160.0, 146.1, 138.1, 113.6,
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
109.7, 81.6, 76.3, 76.2, 66.9, 53.2, 27.3, 27.0, 22.2, 19.8; HRMS (ESI-TOF) m/z: [M+H]+ calcd for
C20H32N5O4Si 434.2218; Found 434.2213; M.P. 199-205 °C.
(2R,3S,4R,5R)-2-(2-Amino-6-methoxy-9H-purin-9-yl)-5-(hydroxymethyl)-3-methyltetrahydrofuran-3,4-
diol 5 and (4aR,6R,7S,7aR)-6-(2-amino-6-methoxy-9H-purin-9-yl)-2,2-di-tert-butyl-7-methyltetrahydro-
4H-furo[3,2-d][1,3,2]dioxasilin-7-ol 35:
In a 100 mL flask, compound 29 (1.05 g, 1.0 eq., 2.42 mmol) was dissolved in TFE (16 mL). To the
solution, PhSiH3 (1.25 mL, 4.0 eq., 10.1 mmol,) and Co(acac)2 (1.0 g, 0.16 eq., 3.9 mmol) was added to
give a grey suspension which was stirred vigorously. After 2.5 h, TLC and HPLC showed the reaction
finished. The crude was filtered through celite and concentrated under vacuum. To the crude, THF (20.0
mL) was added to give dark blue solution, TBAF (4.8 mL, 2.0 eq., 4.8 mmol, 1 mol/L) was added to give
a dark brown solution in 1 min. TLC showed the reaction finished in 1.5 h. The crude was concentrated
and purification by column chromatography (silica gel, eluting with 0 – 10% MeOH in CH2Cl2) to give
1
white solid 5 (0.378 g, 50% Yield). Data for Compound 5: H-NMR (400 MHz, MeOH-D4, 23 °C): δ =
8.25 (s, 1H, H8), 5.98 (s, 1H), 4.22 (d, J = 9.1, 1H), 4.05 (s, 3H), 4.02 (ddd, J = 2.1, 5.7 and 8.5, 2H), 3.84
13
(dd, J = 3.0 and 12.1, 1H), 0.94 (s, 3H); C-NMR (101 MHz, CDCl3, 23 °C) δ = 162.9, 162.0, 154.6,
139.5, 115.4, 93.1, 84.3, 80.5, 73.6, 61.3, 54.4, 20.5; HRMS (ESI-TOF) m/z: [M+H]+ calcd for
C12H18N5O5 312.1308; Found 312.1305.
(4aR,6R,7S,7aR)-6-(2-amino-6-methoxy-9H-purin-9-yl)-2,2-di-tert-butyl-7-methyltetrahydro-4H-
furo[3,2-d][1,3,2]dioxasilin-7-ol 35: After formation of the tertiary alcohol 35 confirmed by HPLC, the
crude was filtered through celite and concentrated to give dark blue paste. Purification by column
chromatography (silica gel, eluting with 20 – 35% EtOAc in hexanes) afforded compound 35 as white
1
crystalline solid (101 mg; 75% isolated yield on 0.13 gram scale). Data for Compound 35: H-NMR
(400 MHz, CD3OD, 23 °C): δ = 1.22 (s, 3H), 1.07 (s, 9H), 1.12 (s, 9H), 4.08 (s, 3H), 4.14-4.30 (m, 3H),
4.43 (dd, J= 8.6 and 4.5 Hz, 1H), 5.94 (s, 1H), 7.87 (s,1H); 13C-NMR (101 MHz, CDCl3, 23 °C) δ =
162.9, 162.1, 154.7, 139.2, 115.7, 94.3, 81.5, 79.3, 76.1, 68.7, 54.4, 28.2, 27.8, 23.8, 21.5, 20.2; HRMS
(ESI-TOF) m/z:[M+H]+ calcd for C20H34N5O5Si 452.2329; Found 452.2315, M.P. 210-215 °C.
Associated Content
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 14 of 16
1
2
3
4
Supporting Information:
The Supporting Information is available free of charge on the ACS Publications website at DOI: xxx.
1D NMR (1H, 13C) (PDF), and UPLC (ultra pressure liquid chromatography) data
Author Information
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
Corresponding Author
*E-mail: Changxia.yuan@bms.com, martin.eastgate@bms.com
ORCID
Changxia Yuan: 0000-0001-8442-676X
Martin Eastgate: 0000-0002-6487-3121
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
We thank Neil Strotman for insightful comments and editorial advice on manuscript preparation. We
thank Yande Huang, Sloan Ayers, and Shasha Zhang for their help with the experiments, including NMR
analysis, HRMS assistant and M. P. analysis.
ACS Paragon Plus Environment

Page 15 of 16
The Journal of Organic Chemistry
1
2
3
4
References:
5
6
7
8
9
1 (a) Lee, L. Y.; Tong, C. Y. W.; Wong, T.; Wilkinson, M. New Therapies for Chronic Hepatitis C Infection: A
Systematic Review of Evidence from Clinical Trials. Int. J. Clin. Pract. 2012, 66, 342−355. (b) Shepard, C. W.; Finelli,
L.; Alter, M. Global Epidemiology of Hepatitis C Virus Infection. J. Lancet Infect. Dis. 2005, 5, 558−567. (c) Lavanchy,
D. The Global Burden of Hepatitis C. Liver Int. 2009, 29 (Suppl. 1), 74−81. (d) European Association for the study of
liver. EASL Recommendations on Treatment of hepatitis C 2018. J. Hepatol. 2018, 2, 461-511.
2 Jordheim, L. P.; Durantel, D.; Zoulim, F.; Dumontet, C. Advances in the Development of Nucleoside and
Nucleotide Analogues for Cancer and Viral Diseases. Nat. Rev. Drug Discov. 2013, 12, 447–464.
3 WHO Model List of Essential Medicines, July 2015,
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
http://www.who.int/mediacentre/factsheets/fs164_apr2014/en/.
4 (a) 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.; Keilman, M.; Lam, A. M.; Steuer, H. M.; Niu, C.; Otto, M. J.; Furman, P. A.
Discovery of a β-D-2’-Deoxy-2’- -fluoro-2’-β-C-methyluridine Nucleotide Prodrug (PSI-7977) for the Treatment of
Hepatitis C Virus. J. Med. Chem. 2010, 53, 7202-7218. (b) Singh, H.; Bhatia, H. K.; Grewal, N.; Natt, N. Sofosbuvir: A
Novel Treatment Option for Chronic Hepatitis C infection. J. Pharmacol. Pharmacother. 2014, 5, 278.
5 (a) McGuigan, C.; Madela, K.; Aljarah, M.; Gilles, A.; Brancale, A.; Zonta, N.; Chamberlain, S.; Vernachio, J.;
Hutchins, J.; Hall, A.; Ames, B.; Gorovits, E.; Ganguly, B.; Kolykhalov, A.; Wang, J.; Muhammad, J.; Patti, J. M.;
Henson, G. Design, Synthesis and Evaluation of a Novel Double Pro-drug: INX-08189. A New Clinical Candidate for
Hepatitis C Virus. Bioorganic & Medicinal Chem. Lett. 2010, 20, 4850–4854. (b) Vernachio, J. H.; Bleiman, B.; Bryant,
K. D.; Chamberlain, S.; Hunley, D.; Hutchins, J.; Ames, B.; Gorovits, E.; Ganguly, B.; Hall, A.; Kolykhalov, A.; Liu, Y.;
Muhammad, J.; Raja, N.; Walters, C. R.; Wang, J.; Williams, K.; Patti, J. M.; Henson, G.; Madela, K.; Aljarah, M.;
Gilles, A.; McGuigan, C. INX-08189, a Phosphoramidate Prodrug of 6-O-Methyl-2’-C-Methyl Guanosine, Is a Potent
Inhibitor of Hepatitis C Virus Replication with Excellent Pharmacokinetic and Pharmacodynamic Properties.
Antimicrobial. Agents. Chemotherapy. 2011, 55, 1843–1851.
6 Trans, K.; Beutner, G. L.; Schmidt, M.; Janey, J.; Chen, K.; Rosso, V.; Eastgate, M. D. Development of a
Diastereoselective Phosphorylation of a Complex Nucleoside via Dynamic Kinetic Resolution. J. Org. Chem. 2015,
80, 4994−5003.
7 (a) Hotchkiss, D. J.; Jenkinson, S. F.; Storer, R.; Heinz, T.; Fleet. G. W. J. Amadori Ketoses with Calcium Hydroxide
and the Kiliani Reaction on 1-Deoxy Ketoses: Two Approaches to the Synthesis of Saccharinic Acids. Tetra. Lett.
2006, 47, 315–318. (b) Hotchkiss, D. J.; Soengas, R.; Booth, K. V.; Weymouth-Wilson, A. C.; Eastwick-Field, V. Fleet,
G. W. J. Green Aldose Isomerisation: 2-C-Methyl-1,4-Lactones from the Reaction of Amadori Ketoses with Calcium
Hydroxide Tetra. Lett. 2007, 48, 517–520.
8 Storer, R.; Moussa, A.; Chaudhuri, N.; Waligora, F. Process for the Production of 2’-Branched Nucleosides. 2004,
PCT Int. Appl., 2004052899.
9 Guanosine tetaacetate can be access on multi-kilo scale from the contracted vendor. We cannot evaluate the
availability of this material right now. Guanosine is widely available. More than 100 vendors are selling it and you
can easily access from these vendors.
10 (a) Robins. M. J.; Uznański, B. Nucleic Acid related Compounds. 33. Conversions of Adenosine and Guanosine to
2’, 6’-Dichloro, 2-Amino-6-Chloro, and Derived Purine Nucleosides. Can. J. Chem. 1981, 59, 2601-2607.
11 Mukobata, T.; Ochi, Y.; Ito, Y.; Wada, S-I.; Urata, H. Facile and Efficient Approach for the Synthesis of N2-
Dimethylaminomethylene-2’-O-Methylguanosine. Bioorg. &Med. Chem. Lett. 2010, 23, 129–131.
12 Fathi, R.; Goswami, B.; Kung, P.-P.; Gaffney, B. L.; Jones, R. A. Synthesis of 6-Substituted 2’-Deoxyguanosine
Derivatives Using Trifluoroacetic Anhydride in Pyridine. Tetra. Lett. 1990, 31, 319-322.
13 NMR confirmed 27 formed dimer in methanol and DMSO, the proposed structure was listed in the SI.
14 Li, N.-S.; Lu, J.; Piccirilli. J. A. Synthesis of 2-C-β-Methyl-2-Deoxyguanosine. J. Org. Chem. 2009, 74, 2227–2230.
15 Subramanyam, C. An Unusual Dependency of Counterion during Witting Methylenation of Bis-Heteroaryl
Ketones. Tetra. Lett. 1995, 36, 9249–9252.
16 Undesired diastereomer was not observed in NMR (no less than 20:1 32: undesired diastereomer).
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 16 of 16
1
2
3
4
5
6
7
8
9
17 (a) Hoffmann, R. W. Markovnikov free radical addition reactions, a sleeping beauty kissed to life. Chem. Soc. Rev.
2016, 45, 577–583. (b) Shigehisa, H.; Suwa, Y.; Furiya, N.; Nakaya, Y.; Fukushima, M.; Ichihashi, Y.; Hiroya, K.
Stereocontrolled Synthesis of Trichodermatide A. Angew. Chem. Int. Ed. 2013, 52, 3646–3649.
18 Based on our observation, part of the mass balance can be contributed to decomposition of the sm by over
oxidation and degradation, NMR and LC did not track any significant byproducts.
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