Stereoselective and Divergent Aza-Adenosine and Aza-Guanosine Syntheses from Xylofuranose, the Key Fragments of a STING Cyclic Dinucleotide Agonist

Article abstract of DOI:10.1021/acs.joc.1c00984

The stereoselective and divergent synthesis of two aza-nucleosides is reported. Starting from xylofuranose 9, aza-adenosine 2 was prepared in 13 steps and 7% overall yield, and aza-guanosine 3 was prepared in 13 steps and 7.8% overall yield. Compared to t

Full text of DOI:10.1021/acs.joc.1c00984

pubs.acs.org/joc
Article
Stereoselective and Divergent Aza-Adenosine and Aza-Guanosine
Syntheses from Xylofuranose, the Key Fragments of a STING Cyclic
Dinucleotide Agonist
Changxia Yuan,* Adrian Ortiz, Zhongmin Xu, Jason Zhu, Michael A. Schmidt, Amanda Rogers,
and Martin Eastgate
Cite This: https://doi.org/10.1021/acs.joc.1c00984
Read Online
sı
*
Metrics & More
Article Recommendations
Supporting Information
ACCESS
ABSTRACT: The stereoselective and divergent synthesis of two aza-nucleosides is reported. Starting from xylofuranose 9, aza-
adenosine 2 was prepared in 13 steps and 7% overall yield, and aza-guanosine 3 was prepared in 13 steps and 7.8% overall yield.
Compared to the original syntheses, some advantages of these new routes are significant yield improvement, overall step-count
reduction, an optimized protecting group strategy, the development of a versatile platform for nitrogenous base incorporation, and
the elimination of hazardous reagents (e.g., benzyl isocyanate, Et3N·HF).
(5) Some of the reagents, such as benzyl isocyanate and Et3N·
HF, were highly hazardous and nonoptimal for manufacturing.
New Route Analysis and Synthesis of Intermediate
21. Embracing the challenges (Scheme 3) and recognizing that
both aza-nucleosides, besides nitrogenous bases at C-1′,
possessed identical atom connections and diastereochemistry,
we envisioned that 7 could serve as a common intermediate to
INTRODUCTION
■
In recent years, biopharmaceutical companies have and will
continue to allocate resources in immune oncology for the
treatment of cancers.1 Among these efforts, one promising
strategy is activating STING (STimulator of INterferon
Genes), fostering an immune response to achieve a durable
anticancer outcome.2 One class of active pharmaceutical
ingredients (APIs) under development in this area consists
of cyclic thiophosphorous dinucleotides, such as 1, featuring
two unnatural 3′-aza-nucleosides cyclized with two stereogenic
thiophosphates on both 3′-N and 5′-O (Scheme 1).3 This
challenging structure required an efficient synthesis to enable a
commercial-scale API supply. In this report, we discuss our
efforts to develop new routes to both key fragments, aza-
nucleosides 2 and 3, with excellent diastereoselectivity and
good yield.
both fragments via a Vorbruggen reaction. Further retro-
̈
synthetic analysis then led to acetonide 8, followed by final
simplification to widely commercially available protected
xylofuranose 9.
The synthesis commenced with selective benzoyl protection
of xylofuranose 9 at C5′-OH, which proved scalable with an
85% in-process yield. Instead of isolating the resulting mono-
Bz protected product, the stream was telescoped into the next
step (Scheme 4). While the subsequent oxidation failed using
IBX, Dess-Martin, and Moffatt conditions, Swern conditions
worked moderately well (i.e., 60% yield at −70 °C) to afford
decagram quantities of ketone 10 for early route exploration.
Further screenings revealed that PIDA-TEMPO in CH2Cl2
delivered 10 in 71% in-process yield over the two telescoped
RESULTS AND DISCUSSION
■
The first-generation syntheses of aza-nucleosides 2 and 3
started from naturally occurring adenosine 4 and guanosine 5,
respectively (Scheme 2). The routes contained multiple
challenges for future development: (1) Both routes were
lengthy and low-yielding: 13 steps/4% overall yield for
adenosine 2 and 15 steps/2% overall yield for guanosine 3.
(2) The current routes had high process mass intensities
(PMI) and long lead times. (3) Some transformations suffered
from low chemo- and/or regioselectivity (details in SI-1). (4)
Both routes contained multiple protection−deprotection steps.
Special Issue: Excellence in Industrial Organic Syn-
thesis 2021
Received: April 27, 2021
https://doi.org/10.1021/acs.joc.1c00984
J. Org. Chem. XXXX, XXX, XXX−XXX
© XXXX American Chemical Society
A

The Journal of Organic Chemistry
pubs.acs.org/joc
Article
Scheme 1. Cyclic Dinucleotide and Key Aza-Nucleoside Fragment
Scheme 2. Summary of First-Generation Synthesis of 2 and 3
Scheme 3. Retrosynthetic Analysis of 2 and 3
Scheme 4. Initial Functionalization of Xylofuranose 9
Scheme 5. Acetonide Deprotection and Cyclization
steps. Imine formation then proceeded smoothly with
benzylamine in TFE, followed by reduction with sodium
borohydride in MeOH at 0 °C to give benzylamine 11 with a
49% isolated yield.4 Poor conversion was observed during
imine formation using other common solvents, with or without
boron or zinc-based Lewis acids. Benzylamine 11 was then
reacted with either phenyl chloroformate or methyl chlor-
oformate to afford 12 and 13, respectively.
To prepare precursor 7 for Vorbru
̈
ggen chemistry,5
deprotection of the acetonide and C1′-OH activation were
required. Brønsted acid-mediated deprotection conditions,
such as TFA−water, AcOH, or PPTS-MeOH worked poorly
on 12, resulting in decomposition (Scheme 5). BCl3 was
B
https://doi.org/10.1021/acs.joc.1c00984
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Scheme 6. Detour of Vorbruggen Chemistry and New Proposal
pubs.acs.org/joc
Article
̈
Scheme 7. Summary of Early Steps and Vorbruggen Chemistry
̈
Scheme 8. Optimized Vorbruggen Chemistry and Deprotection
̈
Scheme 9. Challenges in First-Generation Protection−Deprotection
effective, but it led to OPh adduct 15 rather than desired
hemiacetal 14. Gratifyingly, using TFA-THF-H2O conditions,
methyl carbamate 13 smoothly deprotected to give hemiacetal
14 with a minimal C1′ methoxy adduct. The reaction of 14 in
solution with either benzoyl chloride in the presence of Et3N,
DMAP, levamisole, or lutidine was unsuccessful. Eventually,
the protection worked efficiently using benzoic anhydride in
the presence of both DMAP and Et3N to give 16 in 55% yield
over the two steps. This process was reproducible on the gram
scale to produce high-quality product as a single diastereomer
after workup and crystallization from toluene/hexane.
Vorbruggen chemistry. On the basis of this hypothesis, we
̈
envisioned that releasing the ring strain in the carbamate (i.e.,
new target 18) would prove beneficial.
During the subsequent development, we found dibenzyla-
mino sugar 19 to be stable and robust and thus used this
compound as the handle for future development (Scheme 7).
While direct reductive amination of ketone 10 with dibenzyl-
amine to afford 19 was unsuccessful, 19 was readily prepared
from 11 in 81% yield. After TFA deprotection of 19, acetal 20
was worked up and the concentrated crude product was
subjected to Bz2O, Et3N, and DMAP in warm toluene to
deliver 21 in 90% yield, isolated as a pale-brown crystalline
To our surprise (Scheme 6), subjecting 16 to various
Vorbru
̈
ggen conditions using different nitrogenous bases (i.e.,
solid. The Vorbruggen chemistry between 21 and 6-
̈
6-chloropurine, 6-benzoylaminopurine, adenine, and 6-benzoy-
ladenine) and activating agents (i.e., TMSOTf, SnCl4) led only
to trace product 17, with most of the starting material intact
(details in SI-2). Interestingly, a closer examination of 2-D
NMR data supported that bicycle 16 was very rigid and a 1,3
interaction between C5′ and C1′ likely discouraged the desired
benzoyladenine (HABz) 23, activated by BSA and TMSOTf
in MeCN, worked very well to give adenosine nucleoside 22 in
88% isolated yield. In summary, the synthesis of key
intermediate 21 was accomplished in six chemical trans-
formations, with only three isolations (i.e., 11, 19, and 21), no
chromatography, and a 25% overall yield.
C
https://doi.org/10.1021/acs.joc.1c00984
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
pubs.acs.org/joc
Article
Scheme 10. Second-Generation End-Game Approach to 31
Scheme 11. Adenosine 2 End-Game Summary
Scheme 12. Guanosine 3 Synthesis Route Development
Aza-Adenosine 2 Synthesis. Having obtained a proof-of-
̈
concept for the challenging Vorbruggen chemistry, additional
afforded 25, though with varying conversion.8 Unfortunately,
any scale-up efforts (>100 mg) resulted in complicated
mixtures for this transformation. Furthermore, since subse-
quent TBS protection produced 26 and 27 as an inseparable
mixture and benzoyl protection led to complex mixtures that
required tedious chromatography, C5′ acetyl protection was
deemed unsuitable for long-term development.
On the basis of the previous knowledge, we obtained a proof
of concept for an improved endgame that started with the TBS
protection of alcohol 24 (Scheme 10). The TBS protection
produced 29 on a small scale (<500 mg). Because of stability
challenges associated with 29, the reaction stream was
telescoped into bis-benzoyl protection using catalytic DMAP,
BzCl, and TEA in CH2Cl2, followed by mono-Bz deprotection
with ammonia in EtOAc to afford 31 in 80% yield over three
steps.
development revealed a few key factors. First, 6-benzoylami-
nopurine (23) should be premixed with 21 and BSA to avoid
precipitation. Second, the amount of TMSOTf was critical. At
least 1.0 equiv of TMSOTf was required for consistent
conversion (≥95%)6 and low impurity formation.7 After the
workup, crude 22 was globally hydrolyzed with K2CO3 in the
mixture of THF, MeOH, and water to deliver diol 24 on a
gram scale in 81% yield (Scheme 8).
We next needed to selectively protect the most reactive
primary alcohol in 24, followed by sequential protection of the
secondary alcohol and aniline nitrogen (Scheme 9). To our
surprise, selective monoprotection of diol 24 proved
challenging under numerous conditions (details in SI-3),
resulting in complex mixtures in most cases. During the
screening process, enzymatic conditions using Novozyme 435
D
https://doi.org/10.1021/acs.joc.1c00984
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
pubs.acs.org/joc
Article
Piecing all of the knowledge together, the final optimized
route to aza-nucleoside 2 is shown in Scheme 11. Starting from
21, optimization of the temperature, solvent volume, and
reagent equivalents led to consistent performance of the
synthesis, isolating intermediates 37, 38, 41, and 3 by
crystallization and proceeding in 31% overall yield on gram
scales with no chromatography.
CONCLUSIONS
Vorbruggen reaction on the decagram scale. For subsequent
̈
■
benzoyl deprotection, we found that a 3:1 aqueous
KOH:MeOH ratio was optimal to minimize the reaction
stalling during the formation of 24 and ultimately developed a
two-step telescope to convert 21 to 24, by performing a
crystallization without the need for any chromatography. For
the TBS protection of 24, the optimal conditions involved the
use of EtiPr2N at −10 °C in THF, followed by the telescoped
benzoyl protection/deprotection to afford 31 as an amorphous
solid in 80% yield over the three steps. It is worth mentioning
that, for the TBS protection step, controlling the reaction
temperature to −10 °C reduced the exotherm and changing
the solvent from DCM to THF improved the safety
robustness. Monodesilyation of 31 proceeded without issue
using TFA/water/CH2Cl2. After a few rounds of hydro-
genation screening, we chose to telescope the crude solution of
32 using 50 wt % Pd/C in EtOH to afford 2 in 51% isolated
yield over the two steps. In brief summary, the synthesis of aza-
nucleosides 2 proceeded in six chemical transformations from
common intermediate 21, with only three isolations (i.e., 24,
31, and 2), no chromatography, and a 28% overall yield.
Guanosine Synthesis. With the success of the aza-
adenosine 2 synthesis, we applied our experience to the
synthesis of the aza-guanosine nucleoside 3 (Scheme 12). Built
We have developed a second-generation synthesis of aza-
nucleosides 2 and 3 with significantly improved efficiency. As
summarized in Table 1, the new routes are highly advanta-
Table 1. Route Comparison for Step Count, Overall Yield,
PMI, and Isolation Number
on the prior success, we explored the Vorbruggen reaction with
̈
multiple guanosines, revealing that N2 acetylguanine 34 and
guanine 35 led to sluggish reactivity and/or poor selectivity.
2 old route 2 new-route 3 old route 3 new route
Fortunately, the Vorbruggen reaction using 6-chloropurine 33
̈
total steps
overall yield**
PMI
13
13
15
13
afforded 36 in over 95% conversion and was isolatable by
chromatography on small scale.9 Further exploration of the
next steps led to the decision to telescope crude 36 into the
hydrolysis with TFA to afford 37 in 70% isolated yield over the
two steps with no need for chromatography.10 Because of the
low solubility of monobenzoyl intermediates and 38, initial
attempts to prepare 38 using inorganic bases in various
solvents failed. Fortunately, the addition of hydrazine in hot
IPA led to bis-benzoyl hydrolysis to afford diol 38 in 81% yield.
These three steps were readily conducted on a multigram scale
to prepare more than 10 g of 38. To mirror the experience in
the aza-adenosine route, we initially investigated the enzymatic
conditions to selectively acylate the 5′-primary alcohol in 38
but were unsuccessful. Switching to the TBSOTf conditions
delivered the triTBS-protected product 39 with excellent
conversion. As 39 was unstable to chromatographic
purification, we successfully telescoped 39 into a reaction
4%
2400
9
7%
1000
6
2%
20000
8
7.8%
1300
7
a
b
a
b
isolation number
a
Average number based on the scale-up batches (0.5−1.1 kg),
considering that the crystallizations are not fully optimized.
Estimation based on the current conditions, considering that the
crystallizations are not fully optimized.
b
geous considering that (1) the overall step count was reduced
from 28 (13 + 15) to 20 (6+7+7) steps through the current
divergent synthesis, (2) the overall yield was improved from
less than 4% to 7% for aza adenosine 2 and from 2% to 7.8%
for aza-guanosine 3, (3) the overall PMI was reduced by >90%
for 3 and by 50% for 2,13 and (4) the number of overall
isolations were reduced from 17 (9 (2 first gen) + 8 (3
firstgen)) to 9 (2 (21) + 3 (2 sec gen) + 4 (3 sec gen)).
Further efforts toward the kilogram-scale synthesis of 2 and 3
are currently under development and will be reported in due
course.
with iBuCOCl and Hunig base to simultaneously install the
̈
isobutyl group and deprotect the guanine TBS moiety. This
observation was consistent with the prior finding11 that the
presence of the guanine TBS group assisted the acetylation. In
addition, we found that isobutyryl chloride was superior to
EXPERIMENTAL SECTION
■
General Experimental Information. All reactions were
performed under a nitrogen atmosphere using anhydrous techniques
unless otherwise noted. The xylofuranose is a known compound that
is widely available from vendors, for both laboratory and non-
laboratory use. The compound was received and used as is. 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 a Bruker DRX-400 or
isobutyl anhydride and Hunig’s base was optimal. In the last
̈
step of three-step telescope, we found that the primary TBS
group in 41 was surprisingly robust and using TFA led to the
optimal balance between reaction time and yield. Overall, this
three-step telescope reaction from 38 afforded 41 in 65%
isolated yield after crystallization from iPrOH/H2O. Finally,
guanosine substrate 41 readily debenzylated to 3 in 85% yield
with Pd/C under H2 in EtOH at ambient temperature.12 In
summary, starting from 21, we prepared 3 through a seven-step
E
https://doi.org/10.1021/acs.joc.1c00984
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
pubs.acs.org/joc
Article
DRX-500 instrument and are referenced to residual undeuterated
solvents. The following abbreviations are used to explain multi-
plicities: br, broad; s, singlet; d, doublet; t, triplet; q, quartet; and m,
multiplet. High-resolution mass spectra (HRMS) were recorded on a
Thermo Orbi-trap ESI Discovery instrument.
8.07−7.95 (m, 2H), 7.61−7.49 (m, 1H), 7.46−7.32 (m, 8H), 7.32−
7.19 (m, 2H), 7.18−7.08 (m, 2H), 5.86 (d, J = 3.7 Hz, 1H), 5.18 (br
d, J = 16.3 Hz, 1H), 4.98−4.75 (m, 2H), 4.70−4.60 (m, 2H), 4.0−
4.24 (m, 1H), 4.02−3.80 (m, 1H), 1.69 (br s, 3H), 1.41 (br s, 3H).
13C{1H} NMR (101 MHz, CDCl3) δ 166.3, 155.7, 151.4, 139.0,
((3aR,5R,6aS)-2,2-Dimethyl-6-oxotetrahydrofuro[2,3-d][1,3]-
dioxol-5-yl)methyl Benzoate (10). CH2Cl2 (50 mL) was added to a
N2-flushed 250 mL flask. Xylofuranose 9 (50 g, LR) and pyridine (42
mL, 2.0 equiv) were charged into the flask. Benzoyl chloride (39.1 g,
1.08 equiv) was slowly added to the reactor at −10 °C, in a dry ice−
brine bath, over 1 h and quenched the reaction until the TLC
indicated the reaction to finish in 1 h. Water (100 mL) was added to
the cold solution. Layers were partitioned, and the organic stream was
washed with citric acid (30 mL, 10 wt %, aq), NaHCO3 (30 mL, 10
wt %, aq), and brine (20 mL, 15 wt %, aq). The organic crude was
dried over Na2SO4 and concentrated to oil in a 500 mL flask. Into the
crude was charged CH2Cl2 (250 mL). TEMPO (2.05 g, 0.05 equiv)
and PIDA (8.5 g, 2.0 equiv) were added portionwise at 23 °C and
quenched the reaction until the TLC indicated the reaction to finish
in 4 h. The organic crude was washed with sodium sulfite (100 mL),
and the aqueous phase was back-extracted with CH2Cl2 (50 mL × 2).
The combined organic phase was washed with brine (50 mL), and the
isolated organic layer was dried over Na2SO4. The crude was filtered
and solvent-swapped with MTBE (200 mL) and added to heptane
(200 mL) at 40 °C in 1 h, and then the slurry was further cooled to rt
for 2 h before filtration to give 10 as an off-white solid, after drying to
138.9, 133.3, 130.0, 129.6, 129.0, 128.7, 128.6, 128.1, 127.8, 127.3,
125.9, 121.8, 113.2, 104.1, 104.0, 80.2, 79.7, 74.4, 63.5, 62.7, 58.6,
50.0, 27.0, 26.5. HRMS (ESI) m/z: [M + Na]+ Calcd for
C29H29NO7Na 526.1842; found 526.1840.
((3aR,5S,6R,6aR)-6-(Benzyl(methoxycarbonyl)amino)-2,2-
dimethyltetrahydrofuro[2,3-d][1,3]dioxol-5-yl)methyl Benzoate
(13). In a nitrogen-flushed 40 mL vial, 11 (0.53 g, LR) was dissolved
with CH2Cl2 (8 mL). To the solution, 2,6-lutidine (0.24 mL, 1.5
equiv) and methyl chloroformate (0.13 mL, 1.2 equiv) were added
dropwise at 25 °C. After 30 min, the crude was quenched with
NaHCO3(aq) (5 wt %, 5 mL). The organic layer was isolated and
washed with aqueous citric acid (5 wt %, 5 mL) and aqueous brine
(20 wt %, 5 mL) separately. The organic layer was dried over MgSO4,
filtered, and concentrated. The resulting oil was purified by flash
chromatography with hexane:EtOAc (5:1) to give a colorless oil (0.95
g, 98%). Data for compound 13: 1H NMR (400 MHz, CDCl3) δ 7.98
(d, J = 7.7 Hz, 2H), 7.56 (t, J = 7.4 Hz, 1H), 7.43 (t, J = 7.7 Hz, 2H),
7.39−7.14 (m, 5H), 5.80 (d, J = 3.7 Hz, 1H), 5.18−4.91 (m, 1H),
4.81 (br s, 1H), 4.4.86−4.37 (m, 4H), 4.33−4.13 (m, 1H), 3.84−3.72
(m, 4H), 1.64 (s, 3H), 1.35 (s, 3H). 13C{1H} NMR (101 MHz,
CDCl3) δ 171.4, 157.8, 133.3, 130.1, 130.0, 128.9, 128.6, 127.7, 113.1,
104.1, 74.4, 60.7, 58.3, 53.6, 27.1, 26.4, 21.3, 14.5. HRMS (ESI) m/z:
[M + Na]+ calcd for C24H27NO7Na 464.1680; found 464.1686.
((3aR,4S,6R,6aR)-6-(Benzoyloxy)-3-benzyl-2-oxohexahydrofuro-
[3,4-d]oxazol-4-yl)methyl Benzoate (16). In the N2-flushed 250 mL
flask, 13 (4.0 g) was dissolved in THF (50 mL). To the solution at 50
°C, on a heating mantle, H2O (50 mL) and TFA (50 mL) were
charged in such a sequence dropwise. After 24 h, HPLC confirmed no
more 13, provided that the reaction was quenched with 6 M NaOH
(110 mL, aq) until pH 8 while controlling the temperature to no
more than 40 °C. The crude was agitated for another hour. Then the
solution was extracted with EtOAc (50 mL × 2), and the combined
organic solution was washed with citric acid (23 mL, 5 wt %, aq) and
brine (30 mL), and the organic crude was dried over Na2SO4. The
crude was filtered and solvent swapped with toluene (50 mL). To the
solution was added Et3N (3.8 mL, 3.0 equiv), benzoic anhydride (2.3
g, 1.1 equiv) and DMAP (55 mg, 0.05 equiv) at 35 °C on a heating
mantle. After 14 h, the crude was cooled to rt and washed with citric
acid (25 mL, 5 wt %, aq), NaHCO3 (25 mL, 5 wt % aq), and brine
(30 mL). To the crude oil was slowly added hexane (50 mL) in 1 h at
50 °C. The slurry was cooled to rt and filtered to give an off-white
solid, after drying to give a solid (2.35 g, 55%). Data for compound
1
give 10 (54.5 g, 71%). Data for compound 10: H NMR (400 MHz,
CDCl3) δ 7.93 (d, J = 7.7 Hz, 2H), 7.55 (t, J = 7.2 Hz, 1H), 7.42 (t, J
= 7.7 Hz, 2H), 6.12 (d, J = 4.4 Hz, 1H), 4.70−4.65 (m, 2H), 4.48−
4.39 (m, 2H), 1.49 (s, 3H), 1.41 (s, 3H). 13C{1H} NMR (101 MHz,
CDCl3) δ 207.6, 165.7, 133.3, 129.4, 129.2, 128.4, 114.3, 103.0, 76.1,
63.3, 27.3, 26.9. HRMS (ESI) m/z: [M + Na]+ calcd for C15H17O6Na
315.0839; found 315.0846.
((3aR,5S,6R,6aR)-6-(Benzylamino)-2,2-dimethyltetrahydrofuro-
[2,3-d][1,3]dioxol-5-yl)methyl Benzoate (11). To the N2 flushed 250
mL flask, 10 (30 g), TFE (150 mL), and BnNH2 (13.2 g, 1.2 equiv)
were charged in such a sequence at 10 °C in a cold water batch. After
6 h, HPLC confirmed that no more than 7% of compound 10 was in
the solution. The solution was immediately cooled to 0 °C. In a
separate reactor, MeCN (150 mL) and STAB (65 g, 3.0 equiv) were
added. To the suspension at 23 °C, the imine solution was transferred
in about 10 min. After transfer, the reaction was agitated for an extra
hour. Then the solution was washed with citric acid (150 mL, 5 wt %,
aq) and brine (90 mL). The resulting organic solution was washed
with Na2CO3 (150 L, 15 wt %, aq) and separated while controlling
the pH between 7 and 8. The organic stream was washed with
NaHCO3 (90 mL, 6 wt %, aq) and 20 wt % brine (90 mL). The
solvent of the crude was distilled, and the resulting oil was purified by
flash chromatography with hexane:EtOAc (3:1) to give a yellow oil
(19.3 g, 49%). Data for compound 11: 1H NMR (400 MHz, CDCl3)
δ 7.98 (d, J = 7.3 Hz, 2H), 7.56 (t, J = 7.1 Hz, 1H), 7.46−7.36 (m,
4H), 7.34−7.19 (m, 3H), 5.84 (d, J = 3.8 Hz, 1H), 4.76 (br d, J =
12.2 Hz, 1H), 4.64 (t, J = 4.2 Hz, 1H), 4.45−4.35 (m, 1H), 4.02 (br
d, J = 13.2 Hz, 2H), 3.82 (d, J = 13.2 Hz, 1H), 3.05 (dd, J = 9.9, 4.5
Hz, 1H), 1.59 (s, 3H), 1.40 (s, 3H). 13C{1H} NMR (101 MHz,
CDCl3) δ 166.3, 132.9, 130.0, 129.8, 128.4, 128.3, 128.1, 127.2, 112.1,
104.7, 78.0, 76.9, 63.8, 60.7, 51.9, 26.7, 26.5. HRMS (ESI) m/z: [M +
H]+ calcd for C22H27NO5 384.1805; found 384.1813.
1
16: H NMR (400 MHz, DMSO-d6) δ 7.92 (br d, J = 7.7 Hz, 2H),
7.82 (br d, J = 7.7 Hz, 2H), 7.71−7.59 (m, 2H), 7.54−7.40 (m, 4H),
7.38−7.25 (m, 5H), 6.48 (s, 1H), 5.44 (d, J = 7.6 Hz, 1H), 4.78 (t, J =
6.6 Hz, 1H), 4.66 (d, J = 15.7 Hz, 1H), 4.54 (d, J = 7.6 Hz, 1H),
4.46−4.27 (m, 3H). 13C{1H} NMR (101 MHz, DMSO) δ 165.6,
164.6, 156.5, 136.2, 134.4, 134.0, 129.9, 129.7, 129.5, 129.2, 129.2,
128.3, 128.2, 102.6, 82.5, 81.4, 64.8, 61.2, 46.6. HRMS (ESI) m/z: [M
+ H]+ calcd for C27H24NO7 474.1552; found 474.1547.
((3aR,5S,6R,6aR)-6-(Dibenzylamino)-2,2-dimethyltetrahydro-
furo[2,3-d][1,3]dioxol-5-yl)methyl Benzoate (19). To a nitrogen-
flushed 250 mL flask with 11 (5.1 g), MeCN (50 mL), DIPEA (5.2 g,
3.0 equiv), and BnBr (4.6 g, 2.0 equiv) were added to the reactor. The
reactor was warmed to 70 °C on a heating mantle for 7 h. The
solution was cooled to 20 °C and charged with Et3N (1.3 g, 1.0 equiv)
in 30 min. After 12 h, the solution was added to toluene (200 mL)
and stirred for 30 min. The solution was washed with citric acid (50
mL, 20 wt %, aq) and 20 wt % brine (20 mL). The resulting organic
solution was filtered and concentrated. The resulting oil was purified
by flash chromatography with hexane:EtOAc (5:1) to give 19 as a
yellow oil (5.1 g, 81%). Data for compound 19: 1H NMR (400 MHz,
CDCl3) δ 7.75 (d, J = 7.3 Hz, 2H), 7.44 (t, J = 7.6 Hz, 1H), 7.31−
7.23 (m, 6H), 7.15 (t, J = 7.5 Hz, 4H), 7.10−7.01 (m, 2H), 5.61 (d, J
= 3.7 Hz, 1H), 4.74−4.66 (m, 2H), 4.52 (dt, J = 10.4, 2.2 Hz, 1H),
((3aR,5S,6R,6aR)-6-(Benzyl(phenoxycarbonyl)amino)-2,2-
dimethyltetrahydrofuro[2,3-d][1,3]dioxol-5-yl)methyl Benzoate
(12). In a nitrogen flushed 40 mL vial, 11 (0.84 g) was dissolved
with CH2Cl2 (8 mL). To the solution, 2,6-lutidine (0.20 mL, 1.5
equiv) and phenyl chloroformate (0.33 mL, 1.2 equiv) were added
dropwise at 25 °C. After 30 min, the crude was quenched with
NaHCO3(aq) (5 wt %, 5 mL). The organic layer was isolated and
washed with aqueous citric acid (5 wt %, 5 mL) and aqueous brine
(20 wt %, 5 mL) separately. The organic layer was dried over MgSO4,
filtered, and concentrated. The resulting oil was purified by flash
chromatography with hexane:EtOAc (5:1) to give a colorless oil (1.05
1
g, 95%). Data for compound 12: H NMR (400 MHz, CDCl3) δ
F
https://doi.org/10.1021/acs.joc.1c00984
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
pubs.acs.org/joc
Article
4.24 (dd, J = 12.2, 4.5 Hz, 1H), 4.02 (d, J = 14.1 Hz, 2H), 3.72 (d, J =
14.1 Hz, 2H), 3.06 (dd, J = 10.4, 3.9 Hz, 1H), 1.52 (s, 3H), 1.31 (s,
3H). 13C{1H} NMR (101 MHz, CDCl3) δ 166.4, 139.7, 132.9, 129.9,
129.8, 128.5, 128.4, 128.2, 127.1, 112.7, 104.3, 78.3, 73.9, 63.5, 61.3,
56.1, 26.9, 26.4. HRMS (ESI) m/z: [M + H]+ calcd for C29H31NO5
474.2275; found 474.2262.
min. The solution was aged at 60 °C for 10 h on a heating mantle.
The solution was then cooled to 0 °C in an ice batch, and water (15.0
mL) was introduced slowly to maintain the internal temperature.
Then the suspension was further agitated for 1 h. The solid was
filtered and washed with water (5 mL). The solid was dried and
suspended in MTBE (20 mL) and warmed to 45 °C for 1 h. The
slurry was cooled to 23 °C and filtered. The solid was washed with
MTBE (10 mL) and dried in the oven to afford 24 as a pale-brown
(2R,3R,4R,5S)-5-((Benzoyloxy)methyl)-4-(dibenzylamino)-
tetrahydrofuran-2,3-diyl Dibenzoate (21). To a nitrogen flushed 100
mL flask, 19 (8.0 g), THF (16 mL), and water (8.0 mL) were charged
to stir to a homogeneous solution. TFA (16 mL) was added to the
reactor, and the solution was aged for 8 h at 70 °C on a heating
mantle. The solution was then cooled to 10 °C and diluted with
toluene (50 mL). The layer was separated, and the organic crude was
washed with NaHCO3 (20 mL × 2, 5 wt %, aq). The organic layer
was washed with 20 wt % brine (16 mL), and the solution was used in
the next step without further purifications. To the solution was added
DMAP (0.1 g, 0.05 equiv), Et3N (5.1 g, 3.0 equiv), and Bz2O (7.6 g,
2.0 equiv). The crude was aged at 45 °C on a heating mantle for 16 h.
Into the crude was charged THF (50 mL). The solution was washed
with brine (40 mL, 20 wt %), and the layers were separated. The
organic crude was washed with citric acid + NaCl mixed solution (20
mL, citric acid 5 wt %, NaCl solution 15 wt %, aq) and separated. The
crude was washed again with brine (40 mL, 20 wt %), and the layers
were separated. The solvent of the crude was swapped with 2-MeTHF
(20 mL). At 70 °C, heptane (30 mL) was added dropwise in 30 min.
The slurry was cooled to 25 °C in 2 h. The slurry was filtered, and the
solid was dried in the oven. A brown solid, 21 (9.7 g, 90%), was
isolated. Data for compound 21: 1H NMR (400 MHz, CDCl3) δ 8.20
(d, J = 7.6 Hz, 2H), 7.81 (d, J = 7.7 Hz, 2H), 7.75−7.53 (m, 6H),
7.48−7.36 (m, 7H), 7.28 (t, J = 7.5 Hz, 4H), 7.23−7.10 (m, 4H),
6.50 (s, 1H), 5.92 (d, J = 4.2 Hz, 1H), 4.98 (brd, J = 9.8 Hz, 1H),
4.76 (dd, J = 12.3, 2.0 Hz, 1H), 4.49 (dd, J = 12.3, 3.7 Hz, 1H), 4.22
(d, J = 13.6 Hz, 2H), 4.05 (dd, J = 9.8, 4.1 Hz, 1H), 3.77 (d, J = 13.6
Hz, 2H). 13C{1H} NMR (101 MHz, CDCl3) δ 166.3, 165.7, 164.5,
138.9, 133.9, 133.4, 132.8, 130.0, 129.9, 129.5, 129.3, 129.3, 128.6 (t,
J = 18.3 Hz, 1C), 128.1, 128.5 (t, J = 115.5 Hz, 1C), 98.9, 74.8, 63.4,
58.2, 55.9. HRMS (ESI) m/z: [M + H]+ calcd for C40H36NO7
642.2486; found 642.2487.
((2S,3R,4R,5R)-5-(6-Benzamido-9H-purin-9-yl)-4-(benzoyloxy)-3-
(dibenzylamino)tetrahydrofuran-2-yl)methyl Benzoate (22). Into a
nitrogen-flushed 100 mL flask, MeCN (20 mL) and compound 21
(2.5 g) were charged to give a solution. To the solution were added
HABz (1.86 g, 2 equiv) and BSA (1.19 g, 1.5 equiv). The suspension
was heated to 65 °C for 2 h on a heating mantle, and TMSOTf (0.87
g, 1.0 equiv) was added. The solution was heated to 70 °C for 6 h,
cooled to 25 °C, and quenched with NaHCO3 (10 mL, 5 wt %, aq).
To the mixture, EtOAc (20 mL) was charged, and the organic
solution was separated. The aqueous layer was extracted with EtOAc
(10 mL), and the combined organic layers were washed with
NaHCO3 (10 mL, 5 wt %, aq) and brine (10 mL, 20 wt %, aq). The
organic solution was concentrated (the crude at this point can be used
in the next step) to paste and purified by flash chromatography with
hexane:EtOAc (3:1) to give 22 as a yellow oil (2.6 g, 88%). Data for
compound 22: 1H NMR (400 MHz, DMSO-d6) δ 11.16 (s, 1H), 8.50
(s, 1H), 8.27 (s, 1H), 8.16 (d, J = 8.1 Hz, 2H), 8.03 (d, J = 7.8 Hz,
2H), 7.73 (t, J = 7.0 Hz, 1H), 7.67−7.51 (m, 8H), 7.41−7.31 (m,
6H), 7.30−7.13 (m, 6H), 6.45−6.37 (m, 2H), 5.09 (br d, J = 9.5 Hz,
1H), 4.79 (br d, J = 11.6 Hz, 1H), 4.62 (dd, J = 9.8, 6.1 Hz, 1H), 4.50
(dd, J = 12.4, 3.5 Hz, 1H), 4.16 (br d, J = 14.1 Hz, 2H), 4.03 (d, J =
7.1 Hz, 1H), 3.88 (br d, J = 14.1 Hz, 2H), 3.33 (s, 1H), 1.99 (s, 1H),
1.17 (t, J = 7.2 Hz, 1H). 13C{1H} NMR (101 MHz, DMSO) δ 166.0,
165.9, 165.8, 151.8, 151.7, 151.0, 144.7, 139.9, 134.4, 133.8, 133.7,
132.9, 130.2, 129.6, 129.6, 129.4, 129.0, 128.8, 128.8, 128.7, 127.4,
126.3, 89.2, 76.9, 75.8, 63.2,, 58.7, 55.8. HRMS (ESI) m/z: [M + H]+
calcd for C45H39N6O6 759.2926; found 759.2917.
1
solid (1.05 g, 81%). Data for compound 24: H NMR (400 MHz,
DMSO-d6) δ 8.20 (s, 1H), 8.06 (s, 1H), 7.40 (d, J = 7.5 Hz, 3H),
7.35−7.28 (m, 5H), 7.26−7.18 (m, 2H), 6.02 (d, J = 4.8 Hz, 1H),
5.88 (br s, 1H), 5.41−5.30 (m, 1H), 4.73 (br t, J = 5.6 Hz, 1H), 4.43
(br s, 1H), 4.03 (d, J = 14.1 Hz, 2H), 3.89 (br d, J = 14.1 Hz, 2H),
3.74 (br d, J = 11.6 Hz, 1H), 3.55−3.40 (m, 2H). 13C{1H} NMR
(101 MHz, DMSO) δ 156.6, 152.7, 149.2, 140.6, 140.0, 128.8, 128.6,
127.2, 119.7, 91.0, 82.2, 75.4, 63.0, 59.5, 55.3 .HRMS (ESI) m/z: [M
+ H]+ calcd for C24H27N6O3 447.2139; found 447.2145.
N-(9-((2R,3R,4R,5S)-3-((tert-Butyldimethylsilyl)oxy)-5-(((tert-
butyldimethylsilyl)oxy)methyl)-4-(dibenzylamino)tetrahydrofuran-
2-yl)-9H-purin-6-yl)benzamide (31). To a nitrogen-flushed 50 mL
flask, diol 24 (2.8 g) was suspended in THF (14 mL). To the flask,
EtNiPr2 (2.42 g, 3 equiv) was added at 23 °C. To the suspension,
TBSOTf (3.5 g, 2.5 equiv) was added dropwise at −10 °C in an dry
ice−brine batch. The solution was aged at −10 °C for 1 h and was
added to MeOH (2.5 mL) and aged for 1 h. To the reaction mixture,
water (3 mL) and heptane (20 mL) were charged into the flask.
Crude were extracted and separated. To the aqueous layer was added
heptane (10 mL), and the crude was extracted and separated. The
organic crude was washed with brine (10 mL, 15 wt %, aq) and
separated. The organic was dried over Na2SO4 and filtered. The
organic was concentrated in a 50 mL flask and used in the next step
without further purification. Under nitrogen, CH2Cl2 (14 mL), 2,6-
lutidine (2.69 g, 4.0 equiv), and BzCl (2.65 g, 3.0 equiv) were added.
The solution was agitated at 23 °C for 14 h and charged with MTBE
(28 mL). The organic solution was washed with NaHCO3 (10 mL, 5
wt %, aq) and brine (10 mL, 20 wt %, aq). The resulting organic
mixture was concentrated to paste. To the solution, EtOAc (15 mL)
was added to give a solution. Ammonia in MeOH (1 mL, 7 M) was
added dropwise at 0 °C in an ice bath. The solution was agitated at 0
°C for 7 h. The reaction mixture was washed with citric acid (15 mL,
10 wt %, aq) and brine (10 mL, 20 wt %, aq) and concentrated to
paste (this material can be telescoped to the next step without
isolation) and purified through flash chromatography with hexane:E-
tOAc (3:1) to give 31 as a yellow oil (3.9 g, 80%). Data for
1
compound 31: H NMR (400 MHz, CDCl3) δ 9.53 (br s, 1H), 8.84
(s, 1H), 8.44 (s, 1H), 8.12−8.05 (m, 2H), 7.66−7.46 (m, 4H), 7.39−
7.24 (m, 9H), 6.29 (d, J = 4.3 Hz, 1H), 4.78 (dd, J = 5.9, 4.5 Hz, 1H),
4.60 (brd, J = 5.7 Hz, 1H), 4.12−3.95 (m, 5H), 3.78−3.61 (m, 2H),
0.97 (s, 9H), 0.90 (s, 9H), 0.15 (s, 3H), 0.01 (s, 3H), −0.19 (s, 3H).
13C{1H} NMR (101 MHz, CDCl3) δ 170.0, 165.0, 152.7, 151.5,
149.6, 141.4, 139.6, 133.9, 133.2, 132.7, 130.1, 130.0, 128.8, 128.7,
128.6, 128.4, 128.3, 128.0, 127.2, 122.5, 90.2, 82.4, 78.3, 63.9, 61.2,
60.9, 59.4, 55.3, 26.0, 25.9, 18.5, 17.8, −4.5, −5.0, −5.5, −5.5. HRMS
(ESI) m/z: [M + H]+ calcd for C43H59N6O4Si2 779.4131; found
779.4121.
N-(9-((2R,3R,4R,5S)-4-Amino-3-((tert-butyldimethylsilyl)oxy)-5-
(hydroxymethyl)tetrahydrofuran-2-yl)-9H-purin-6-yl)benzamide
(2). To a nitrogen flushed 50 mL flask, 31 (3.1 g) and CH2Cl2 (30
mL) were charged to give a homogeneous solution at 23 °C. To the
solution, TFA (3 mL) and water (1 mL) was added. The solution was
warmed to 35 °C on a heating mantle for 8 h and then cooled and
diluted with CH2Cl2 (20 mL) at 23 °C. The organic solution was
washed with water (10 mL), NaHCO3 (10 mL, 5 wt %, aq), and brine
(10 mL, 15 wt %). The organic layer was dried over Na2SO4 and
concentrated to a yellow oil. The crude oil was dissolved with EtOH
(20 mL) and transferred to a 50 mL pressure vessel. Pd/C (0.22 g, 5
wt %, 0.5 equiv) was added to the solution, the vessel was refilled with
N2 three times and H2 three times, and the H2 pressure was kept at 30
psi. The reaction was agitated for 18 h at 23 °C. After the hydrogen
was swapped with N2 and air, the crude was filtered through a pad of
(2R,3R,4S,5S)-2-(6-Amino-9H-purin-9-yl)-4-(dibenzylamino)-5-
(hydroxymethyl)tetrahydrofuran-3-ol (24). To a nitrogen-flushed 50
mL flask, 22 (2.2 g), MeOH (9 mL), THF (11 mL), and water (1
mL) were added to give a homogeneous solution. KOH solution (2.9
mL, 1 M, 4.0 equiv) was slowly added to the solution at 25 °C in 10
G
https://doi.org/10.1021/acs.joc.1c00984
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
pubs.acs.org/joc
Article
Celite and the organic solution was concentrated. The paste was
diluted with MTBE (6 mL), and heptane (30 mL) was added slowly
to generate solid precipitate at 0 °C first and then slowly to 25 °C
over 8 h. The suspension was then filtered, and 2 was isolated as an
amorphous solid (0.98 g, 51%). Data for compound 2: 1H NMR (400
MHz, CDCl3) δ 8.77 (s, 1H), 8.18 (s, 1H), 8.02 (d, J = 7.3 Hz, 2H),
7.60−7.54 (m, 1H), 7.53−7.45 (m, 2H), 5.94 (d, J = 5.3 Hz, 1H),
4.85 (t, J = 5.3 Hz, 1H), 4.10−4.06 (m, 1H), 3.99 (dd, J = 12.7, 1.6
Hz, 1H), 3.80−3.69 (m, 2H), 0.82 (s, 9H), −0.10 (s, 3H), −0.24 (s,
3H). 13C{1H} NMR (101 MHz, CDCl3) δ 164.7, 152.3, 150.5, 150.1,
142.6, 133.5, 132.8, 128.8, 127.9, 123.9, 91.2, 87.6, 75.4, 62.7, 53.6,
25.5, 17.8, −5.2. HRMS (ESI) m/z: [M + H]+ calcd for
C23H33N6O4Si 485.2337; found 485.2327.
= 11.5, 1.7 Hz, 1H), 3.43 (dd, J = 11.6, 3.7 Hz, 1H), 3.40 (m, 1H).
13C{1H} NMR (153 MHz, DMSO) δ δ 158.2, 154.9, 151.0, 140.1,
134.8, 128.3, 128.1, 126.8, 116.7, 89.1, 81.3, 75.0, 62.6, 59.2, 54.7.
HRMS (ESI) m/z: [M + H]+ calcd for C24H27N6O4 463.2088; found
463.2070.
N-(9-((2R,3R,4R,5S)-3-((tert-Butyldimethylsilyl)oxy)-4-(dibenzyla-
mino)-5-(hydroxymethyl)tetrahydrofuran-2-yl)-6-oxo-6,9-dihydro-
1H-purin-2-yl)isobutyramide (41). To a nitrogen-backfilled 100 mL
flask, THF (10 mL) and 38 (2.44 g) were charged at 23 °C to give a
suspension. TBSOTf (3.47 g, 2.5 equiv) and Hunig base (3.4 g, 5.0
̈
equiv) were charged at −10 °C and aged for 2 h at 23 °C. MeOH (1
mL) was added dropwise, and the solution was aged for 30 min at 23
°C. MTBE (25 mL) was charged into the solution and extracted with
citric acid (10 mL, 5 wt %, aq) and water (12 mL), and the isolated
organic phase was concentrated under vacuum to 10 mL. To the
crude was added THF (12 mL), and it was concentrated under
vacuum to 5 mL. To the solution, pyridine (7 mL) was charged and
the solution was cooled to 10 °C in a cold water batch. Isobutyryl
chloride (1.69 g, 3.0 equiv) was charged dropwise into the mixture at
10 °C. After 2 h, the solution was warmed to 23 °C and THF (3.0 L)
was charged into the reaction. NH3·H2O (3 mL, 10 wt %, aq) was
charged dropwise into the reaction mixture at 10 °C, and the reaction
was agitated for 6 h until no 39 was observed. The crude was diluted
with MTBE (12 mL) and water (24 mL). Layers were partitioned,
and the aqueous phase was extracted with MTBE (12 mL). The
combined organic phases were washed with critic acid (24 mL × 2, 30
wt %, aq). The organic phase was then washed with water (24 mL, 10
V) and concentrated to 5 mL. To the concentrate, TFA (5 mL) and
water (2.5 mL) were added and agitated for 3 h at 10 °C. DIPEA was
added dropwise to the mixture at 10 °C to adjust the pH of the
mixture to 7−8. Water (24 mL) was added, and the organic phase was
separated. The organic phase was concentrated to 5 mL, and iPrOH
(12 mL) was charged into the mixture. The mixture was concentrated
to 7 mL, and the slurry was filtered. The solid was dried in an oven
under vacuum to give 41 as a light-brown solid (2.22 g, 65%). Data
((2S,3R,4R,5R)-5-(2-Amino-6-oxo-1,6-dihydro-9H-purin-9-yl)-4-
(benzoyloxy)-3-(dibenzylamino)tetrahydrofuran-2-yl)methyl Ben-
zoate (37). To a nitrogen-backfilled 250 mL flask with 21 (7.5 g),
MeCN (60 mL) was added to give a homogeneous solution. To the
solution, HGCl (3.95 g, 2.0 equiv) and BSA (3.56 g, 1.5 equiv) were
added in such an order at 23 °C to generate precipitates. The
suspension was then held to 65 °C for 20 min. TMSOTf (2.6 g, 1.0
equiv) was added, and the resulting homogeneous solution was held
to 70 °C on a heating mantle for 7 h. The solution was cooled to 25
°C and quenched with NaHCO3 (40 mL, 5 wt %, aq). Toluene (60
mL) was added to the crude, and the two layers were separated. The
aqueous layer was back-extracted with toluene (15 mL) and
combined with the first toluene layer. The organic layers were
washed with NaHCO3 (30 mL, 5 wt %, aq) and brine (30 mL, 20 wt
%, aq) and separated. The organic solution of 36 was concentrated
and added to toluene (15 mL). Then the H2O (5 mL) and TFA (40
mL) were placed in the flask. The reaction crude was stirred for at
least 20 h at 40 °C and then cooled to 15 °C before H2O (75 mL)
was added to the mixture at 25 °C. The organic phase and water
phase were separated. The aqueous phase was extracted with toluene
(30 mL) and separated. The two toluene solutions were combined
and washed with water (70 mL), NaHCO3 (50 mL, 5 wt %, aq), and
brine (30 mL, 15 wt %, aq) separately. The final organic crude was
concentrated to 15 mL under vacuum. The concentrated crude was
slowly reversely added to heptane (70 mL) to afford a suspension in 2
h at 25 °C. The solid was filtered and washed with heptane (15 mL)
to give 37 as a light-brown solid (5.49 g, 70%). Data for compound
1
for compound 41: H NMR (600 MHz, DMSO-d6, 25 °C): δ 12.09
(br s, 1H), 11.56 (s, 1H), 8.25 (s, 1H), 7.42 (br d, J = 7.7 Hz, 4H),
7.33 (br t, J = 7.6 Hz, 4H), 7.25 (br t, J = 7.4 Hz, 2H), 6.02 (d, J = 6.8
Hz, 1H), 5.14 (t, J = 5.0 Hz, 1H), 4.85 (br t, J = 7.4 Hz, 1H), 4.53 (m,
1H), 4.33 (d, J = 4.2 Hz, 1H), 3.96 (d, J = 13.6 Hz, 2H), 3.75 (d, J =
13.6 Hz, 2H), 3.66 (m, 1H), 3.50 (dt, J = 11.6, 4.3 Hz, 1H), 3.44 (dd,
J = 7.9, 3.1 Hz, 1H), 2.74 (sept, J = 6.9 Hz, 1H), 1.09 (br d, J = 6.9
Hz, 6H), 0.79 (s, 9H), 0.01 (s, 3H), −0.41 (s, 3H). 13C{1H} NMR
(153 MHz, DMSO) δ 180.1, 154.7, 149.0, 148.2, 139.4, 137.4, 128.5,
128.2, 127.0, 119.8, 88.8, 81.8, 77.2, 63.0, 60.1, 54.1, 34.7, 25.5, 18.8,
18.7, 17.5, −4.8, −5.9. HRMS (ESI) m/z: [M + H]+ calcd for
C34H47N6O5Si 647.3372; found 647.3347.
1
37: H NMR (600 MHz, DMSO-d6, 25 °C): δ 10.84 (v br s, 1H),
8.11 (br d, J = 7.8 Hz, 2H), 7.77 (br d, J = 7.8 Hz, 2H), 7.72 (br t, J =
7.4 Hz, 1H), 7.65 (s, 1H), 7.65 (t, J = 7.4 Hz, 1H), 7.59 (t, J = 7.7 Hz,
2H), 7.47 (t, J = 7.7 Hz, 2H), 7.29 (br d, J = 7.6 Hz, 4H), 7.19 (br t, J
= 7.6 Hz, 4H), 7.13 (br t, J = 7.2 Hz, 2H), 6.16 (overlapped, 2H),
5.00 (m, 1H), 4.73 (dd, J = 12.2, 2.7 Hz, 1H), 4.51 (dd, J = 12.2, 4.8
Hz, 1H), 4.16 (dd, J = 8.3, 5.9 Hz, 1H), 4.05 (d, J = 14.2 Hz, 2H),
3.86 (d, J = 14.2 Hz, 2H). 13C{1H} NMR (153 MHz, DMSO) δ
165.5, 165.0, 156.7, 153.7, 150.6, 139.1, 135.7, 133.9, 133.4, 129.6,
129.2, 129.2, 128.9, 128.9, 128.6, 128.2, 128.2, 127.0, 117.0, 87.2,
76.3, 75.2, 63.8, 59.2, 54.8. HRMS (ESI) m/z: [M + H]+ calcd for
C38H35N6O6 671.2613; found 671.2597.
N-(9-((2R,3R,4R,5S)-4-Amino-3-((tert-butyldimethylsilyl)oxy)-5-
(hydroxymethyl)tetrahydrofuran-2-yl)-6-oxo-6,9-dihydro-1H-
purin-2-yl)isobutyramide (3). To a 50 mL nitrogen-backfilled
pressure bomb, 41 (1.11 g) was dissolved in EtOH (11 mL) at 23
°C. Pd on carbon (0.22 g, 20 wt %) was added to the solution to
afford a suspension at 23 °C, and the reactor was backfilled with H2
three times. The reaction was agitated for 6 h at 23 °C. The reaction
mixture was backfilled with N2 three times and filtered through Celite,
and the cake was washed with EtOH (5.0 mL × 2). The solution was
solvent swapped with MeCN at 10 mL under vacuum at no more than
40 °C on a heating mantle. Water (10 mL) was added dropwise to the
solution, and the slurry was agitated for 30 min. The slurry was
filtered and the cake was washed with MeCN:H2O = 1:3 (2 mL × 2),
followed by MTBE (2.0 mL). The solid was dried in an oven under
vacuum to give 3 as an off-white solid (0.65 g, 81%). Data for
2-Amino-9-((2R,3R,4S,5S)-4-(dibenzylamino)-3-hydroxy-5-
(hydroxymethyl)tetrahydrofuran-2-yl)-1,9-dihydro-6H-purin-6-one
(38). To a nitrogen-backfilled 100 mL flask, IPA (25 mL) and 37 (7.8
g) were added to give a suspension. N2H4 (0.97 g, 2.6 equiv) was
charged into the reactor on a heating mantle at 35 °C. The solution
was heated to 80 °C for 2 h. The reaction mixture was cooled to 25
°C, and water (25 mL) was added dropwise to the flask to give a
suspension at 25 °C. The suspension was filtered and washed with
MeCN/IPA (15 mL, 1/1 = v/v). The solid was dried in an oven
under vacuum to give 38 as a light-brown solid (4.4 g, 81%, 90%
HPLC purity).The solid is ready for the next step. For character-
ization, the solid was then dissolved with NMP (5 mL) and purified
with reverse-HPLC using solvent pair MeCN/H2O (0.05% TFA) to
afford 38 as a white solid with more than 98% purity. Data for
1
compound 3: H NMR (400 MHz, DMSO-d6) δ 7.99 (s, 1H), 5.58
(d, J = 4.5 Hz, 1H), 4.71 (br t, J = 5.0 Hz, 1H), 4.19 (t, J = 5.0 Hz,
1H), 3.50−3.36 (m, 2H), 3.33−3.16 (m, 2H), 3.03 (br s, 1H), 2.50
(quin, J = 6.8 Hz, 1H), 0.83 (d, J = 6.8 Hz, 6H), 0.51 (s, 9H), −0.31
(s, 3H), −0.41 (s, 3H). 13C{1H} NMR (101 MHz, DMSO) δ 180.6,
155.3, 149.2, 148.6, 138.0, 120.5, 87.6, 86.7, 77.4, 61.9, 53.9, 35.2,
26.0, 19.4, 19.3, 18.2, −4.7, −4.7. HRMS (ESI) m/z: [M + H]+ calcd
for C20H35N4O5Si 467.2440; found 467.2433.
1
Compound 38: H NMR (600 MHz, DMSO-d6, 25 °C): δ 7.70 (s,
1H), 7.40 (br d, J = 7.6 Hz, 4H), 7.31 (br t, J = 7.5 Hz, 4H), 7.22 (br
t, J = 7.4 Hz, 2H), 5.82 (d, J = 5.3 Hz, 1H), 4.61 (m, 1H), 4.36 (m,
1H), 4.01 (d, J = 14.1 Hz, 2H), 3.85 (d, J = 14.1 Hz, 2H), 3.66 (dd, J
H
https://doi.org/10.1021/acs.joc.1c00984
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
pubs.acs.org/joc
Article
(c) Vargas, T. R.; Benoit-Lizon, I.; Apetoh, L. Rationale for Stimulator
of Interferon Genes-Targeted Cancer Immunotherapy. Eur. J. Cancer
2017, 75, 86−97.
ASSOCIATED CONTENT
* Supporting Information
The Supporting Information is available free of charge at
https://pubs.acs.org/doi/10.1021/acs.joc.1c00984.
■
sı
(3) Glick, G.; Ghosh, S.; Olhava, E. J.; Roush, W. R.; Jones, R. Cyclic
Dinucleosides for Treating Conditions Associated with STING
Activity Such as Cancer. U.S. Patent US20190031708, January 31,
2019. United States Patent and Trademark Office Web site.
(4) Li, J.; Zhang, Z.; Tang, M.; Zhang, X.; Jin, J. Selective Synthesis
of Isoquinolines by Rhodium(III)- Catalyzed C-H/N-H Functional-
ization with alfa-Substituted Ketones. Org. Lett. 2016, 18, 3898−3901.
First-generation route to access aza-adeonsine 2 and aza-
guanosine 3; Vorbruggen chemistry screening 16 to 17;
̈
selective protection of 25 screening; H and 13C{1H}
NMR spectra for compounds 2, 3, 10, 11, 12, 13, 16, 19,
21, 22, 24, 31, 37, 38, and 41 (PDF)
1
(5) Vorbru
Reaction 1999, 55, 1−629.
(6) The Vorbruggen chemistry worked fine with TMSOTf up to 2.0
̈
ggen, H.; Ruh-Pohlenz, C. Synthesis of Nucleosides. Org.
̈
AUTHOR INFORMATION
Corresponding Author
Changxia Yuan − Chemical Process Development, Bristol
Myers Squibb, New Brunswick, New Jersey 08903, United
States; orcid.org/0000-0001-8442-676X;
Email: Changxia.yuan@bms.com
■
equiv. The safety study shows no hazard for the excess amount of the
activating reagent.
(7) Six isomers with the same molecular weight were generated with
overall 5−8 relative area percent under standard reaction condition,
and the stability of these intermediates limited the structure
elucidation.
(8) Aehle, W. Enzymes in Industry: Production and Applications, 3rd
ed.; Wiley-VCH Verlag, 2007.
Authors
Adrian Ortiz − Chemical Process Development, Bristol Myers
Squibb, New Brunswick, New Jersey 08903, United States;
Present Address: Amgen, 1 Amgen Center Drive,
Thousand Oaks, California 91320, United States
Zhongmin Xu − Chemical Process Development, Bristol Myers
Squibb, New Brunswick, New Jersey 08903, United States
Jason Zhu − Chemical Process Development, Bristol Myers
Squibb, New Brunswick, New Jersey 08903, United States
Michael A. Schmidt − Chemical Process Development, Bristol
Myers Squibb, New Brunswick, New Jersey 08903, United
States; orcid.org/0000-0002-4880-2083
Amanda Rogers − Chemical Process Development, Bristol
Myers Squibb, New Brunswick, New Jersey 08903, United
States
Martin Eastgate − Chemical Process Development, Bristol
Myers Squibb, New Brunswick, New Jersey 08903, United
States; orcid.org/0000-0002-6487-3121
(9) Similar to Vorbruggen chemistry on aza-adenosine, seven
̈
isomers with the same molecular weight were generated with overall
5−8 relative area percent under standard reaction conditions, and the
stability of these intermediates limited the structure elucidation.
(10) Further study showed that 36 was unstable upon storage with
moisture, and the attempt to isolate this intermediate was abandoned.
(11) Jin, S.; Miduturu, C.; McKinney, D.; Silverman, S. Synthesis of
Amine- and Thiol-Modified Nucleoside Phosphoramidites for Site-
Specific Introduction of Biophysical Probes into RNA. J. Org. Chem.
2005, 70, 4284−4299.
(12) The loading of the Pd was relatively low at 20 wt %, and the
reaction kinetics and impurity profiles were comparably much better
than the reduction of 32. Teams are working on the process
optimization and mechanism.
(13) (a) Henderson, R. K.; Kindervater, J.; Manley, J. B. Lessons
Learned through Measuring Green Chemistry Performance. Accessed
27-May-2009 via the ACS GCI Pharmaceutical Roundtable site:
https://acs.confex.com/recording/acs/green09/pdf/free/
4db77adf5df9fff0d3caf5cafe28f496/paper69453_5.pdf. (b) ACS PMI
predictor: https://acsgcipr-predictpmi.shinyapps.io/pmi_calculator/.
(c) Borovika, A.; Albrecht, J.; Li, J.; Wells, A. S.; Briddell, C.; Dillon,
B. R.; Diorazio, L. J.; Gage, J. R.; Gallou, F.; Koenig, S. G.; Kopach,
M. E.; Leahy, D. K.; Martinez, I.; Olbrich, M.; Piper, J. L.;
Roschangar, F.; Sherer, E. S.; Eastgate, M. The PMI Predictor - a
Web App Enabling Green-by-Design Chemical Synthesis. ChemRxiv.
DOI: 10.26434/chemrxiv.7594646.
Complete contact information is available at:
https://pubs.acs.org/10.1021/acs.joc.1c00984
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We thank Richard Fox for insightful comments and editorial
advice on manuscript preparation. We thank Sloan Ayers and
Mike Peddicord for their help with the experiments, including
NMR analysis and HRMS assistant. We thank Guanghui Zhu
for his PACE calculations to generate the PMIs.
REFERENCES
■
(1) (a) Galon, J.; Bruni, D. Approaches to Treat Immune Hot
Altered and Cold Tumours with Combination Immunotherapies. Nat.
Rev. Drug Discovery 2019, 18, 197−218. (b) Adams, J. L.; Smothers,
J.; Srinivasan, R.; Hoos, A. Big Opportunities for Small Molecules in
Immuno-Oncology. Nat. Rev. Drug Discovery 2015, 14, 603−622.
(c) Dempke, W. C. M.; Fenchel, K.; Uciechowski, P.; Dale, S. P.
Second- and Third-Generation Drugs for Immuno-Oncology Treat-
ment-the More the Better? Eur. J. Cancer 2017, 74, 55−72.
(d) Hendriks, L. E.; Besse, B. Windows Open for Cancer
Immunotherapy. Nature 2018, 558, 376−377.
(2) (a) Cross, R. STING Fever is Sweeping through the Cancer
Immunotherapy World. C&EN 2018, 9, 24−26. (b) Ishikawa, H.;
Barber, G. N. STING is an Endoplasmic Reticulum Adaptor that
Facilitates Innate Immune Signaling. Nature 2008, 455, 674−679.
I
https://doi.org/10.1021/acs.joc.1c00984
J. Org. Chem. XXXX, XXX, XXX−XXX