Synthesis of diastereomerically pure nucleotide phosphoramidates

Article abstract of DOI:10.1021/jo201492m

Prodrugs of therapeutic nucleoside monophosphates masked as phosphoramidate derivatives have become an increasingly important class of antiviral drugs in pharmaceutical research for delivering nucleotides in vitro and in vivo. Conventionally, phosphoramidate derivatives are prepared as a mixture of two diastereomers. We report a class of stable phosphoramidating reagents containing an amino acid ester and two phenolic groups, one unsubstituted and the other with electron-withdrawing substituents. The reagents can be isolated as single diastereomers and reacted with the 5′-hydroxyl group of nucleosides through selective nucleophilic displacement of the substituted phenol to prepare single diastereomer phosphoramidate products. This method has been used to prepare the HCV clinical candidate PSI-7977 in high yield and high diastereomeric purity.

Full text of DOI:10.1021/jo201492m

ARTICLE
pubs.acs.org/joc
Synthesis of Diastereomerically Pure Nucleotide Phosphoramidates
Bruce S. Ross,* P. Ganapati Reddy, Hai-Ren Zhang, Suguna Rachakonda, and Michael J. Sofia
Pharmasset, Inc., 303A College Road East, Princeton, New Jersey 08540, United States
S Supporting Information
b
ABSTRACT: Prodrugs of therapeutic nucleoside monophos-
phates masked as phosphoramidate derivatives have become an
increasingly important class of antiviral drugs in pharmaceutical
research for delivering nucleotides in vitro and in vivo. Con-
ventionally, phosphoramidate derivatives are prepared as a
mixture of two diastereomers. We report a class of stable
phosphoramidating reagents containing an amino acid ester and two phenolic groups, one unsubstituted and the other with
electron-withdrawing substituents. The reagents can be isolated as single diastereomers and reacted with the 50-hydroxyl group of
nucleosides through selective nucleophilic displacement of the substituted phenol to prepare single diastereomer phosphoramidate
products. This method has been used to prepare the HCV clinical candidate PSI-7977 in high yield and high diastereomeric purity.
’ INTRODUCTION
chemistry for our synthesis, the chloro reagent 2 is produced as
an even mixture of two diastereomers. Logically, it would be
more convergent to develop a synthesis of a single isomer of a
phosphoramidating reagent which then could be used to make
the desired nucleoside phosphoramidate diastereomer by nu-
cleophilic displacement. If any chiral separation were needed, it
could be done on the relatively inexpensive reagent. Unfortu-
nately, reagent 2 was not sufficiently stable. As commonly done
for other phosphoramidating reagents,9 2 was prepared as a
crude mixture just prior to use. It is possible to purify the reagent
on silica gel if care is taken to avoid contact with any moisture or
alcohols, but it would not be practical to purify it by chiral
chromatography. Consequently, a stable phosphoramidate re-
agent that was sufficiently reactive to allow ester bond formation
with a nucleophilic 50-hydroxyl group while maintaining the
desired phenolic phosphate ester intact was required. Upon
completion of our work, Meier10 published a method to prepare
single diastereomers of phosphoramidates using a chiral auxiliary
on the phosphoramidating reagent. This method was limited by
the need to chromatographically purify the reagent, modest
yields, and demonstration only on a nucleoside substrate with
no competing 20- or 30-hydroxyl groups. Here we describe the
preparation of a series of stable, crystalline single diastereomer
phosphoramidating reagents and a general method for the
efficient preparation of diastereomerically pure 50-nucleotide
phosphoramidates.
Antiviral nucleosides must be converted to their respective
50-O-triphosphate nucleotides through three separate kinases in
order to inhibit the viral polymerases. Often, the first kinase for
conversion of the nucleoside to its respective 50-O-monophos-
phate nucleotide is the most discriminating. Nucleotides them-
selves do not have acceptable pharmacokinetic properties for
drug development. Medicinal chemists have long sought to
circumvent the first phosphorylation step by creating prodrugs
of the monophosphate. In the case of phosphoramidates,1 the
monophosphate is masked by substitution with an aromatic ester
and an amine such as an amino acid ester. Phosphoramidate
prodrug strategies have been applied widely in the development
of treatments for hepatitis C,2 hepatitis B,3 and HIV.4 This
prodrug construct provides suitable pharmacokinetic properties
for in vivo absorption and in the case of treatments for HCV and
HBV provides liver targeting characteristics. Once in the liver, the
phosphoramidate is metabolized to the monophosphate nucleo-
tide through a cascade of steps starting with the cleavage of the
amino acid ester by esterases.5 Our second generation nucleoside
inhibitor, PSI-79776 (3, Scheme 1), utilizes this strategy and has
demonstrated clinical proof7 of concept. PSI-7977 is a single Sp
isomer that is obtained by the first selective crystallization of a
nucleotide phosphoramidate from the diastereomeric mixture
and is the more potent diastereomer. However, this separation-
by-crystallization approach is an inefficient method to use in the
last step of a multistep synthesis. Therefore, we undertook an
effort to develop a diastereoselective synthesis of PSI-7977.
Unlike the wealth of information available for stereoselective
synthesis of carbon centers, the stereoselective synthesis of
phosphorus centers8 and especially phosphoramidates is limited.
For a phosphorus(V) electrophilic reagent such as 2, the
phosphorus stereochemistry is already set assuming that the
reagent is stable. Nucleophilic substitution on the phosphorus
center leads to an inversion of configuration. Using conventional
’ RESULTS AND DISCUSSION
In our search for a desirable reactive phosphoramidating
reagent, we were aware that Hayakawa11 et al. tested a series of
phosphorochloridate diesters and phenolic phosphate triesters as
phosphorylation reagents used to make nucleoside phosphate
Received: July 18, 2011
Published: September 14, 2011
r
2011 American Chemical Society
8311
dx.doi.org/10.1021/jo201492m J. Org. Chem. 2011, 76, 8311–8319
|

The Journal of Organic Chemistry
ARTICLE
Scheme 1. Conventional Preparation of Phosphoramidates
Scheme 2. Synthesis of 4-Nitrophenyl Reagent 5 and 6
triesters. Interestingly, they reported that the chloride leaving
group could be replaced with a p-nitrophenyl ester even in the
presence of a competing but albeit less reactive o-chlorophenyl
ester substitution. To promote nucleophilic displacement, a
strong base was needed to deprotonate the 50-hydroxyl group.
tert-Butylmagnesium chloride had been shown to afford 50-O-
nucleoside regioselectivity in the construction of 50-O-
phosphotriesters12 and 50-O-phosphoramidates.13 Therefore,
we decided to test a phosphoramidating reagent containing a
p-nitrophenyl ester and to use tert-butylmagnesium chloride as
the base.
The desired nitrophenyl phosphoramidating reagent could be
readily prepared by reacting commercial p-nitrophenyl dichloro-
phosphate and isopropyl L-alanate in the presence of triethyl-
amine, followed by phenol to give a 1:1 mixture of diastereomers
5 and 6 as a chromatographically homogeneous oil in 83% yield
(Scheme2).
Nucleoside 1 was then reacted with the reagent mixture in the
presence of two equivalents of tert-butylmagnesium chloride in
THF. One equivalent of the Grignard base was consumed by
deprotonation of the relatively acidic hydrogen on the uracil base.
The resulting precipitated salt of the starting nucleoside redis-
solved as the reaction progressed. The reaction proceeded to
approximately 65% completion after 2 days at ambient tempera-
ture despite active reagent remaining. 31P NMR of the crude
reaction revealed high regioselectivity for the desired 50-hydroxyl
group with no detectable remaining monosubstitution on the 30-
hydroxyl group, but 5% of 30,50-bis-substituted byproduct. More
interestingly, 31P NMR of the crude reaction mixture revealed
that there was a 3:1 diastereoselectivity for the Sp isomer 3 as well
as a corresponding 1:3 (Sp/Rp) ratio of the remaining reagent
diastereomers. After chromatographic purification, the recovered
yield was 47% of the diastereomeric mixture of 3 and 4.
Further investigation of the p-nitrophenyl phosphoramidate
reagent preparation led to identifying conditions that allowed for
crystallization of a single diastereomer in 96% diastereomeric
excess in 44% yield of theory for one isomer without chroma-
tography. Single crystal X-ray analysis of the crystalline diaster-
eomer 5 revealed it to have the Sp configuration. The Rp
nitrophenol reagent 6 was readily isolated from the mother
liquor through supercritical fluid chromatography (SFC) using
a chiral stationary phase to afford an oil that slowly solidified upon
standing. Considering the change in stereochemical nomen-
clature priorities, it is the reagent having the Sp configuration
that should lead, after inversion through a nucleophilic displace-
ment reaction with 1, to the desired Sp product 3, PSI-7977. This
is in fact what we observed. The transference of diastereomeric
excess of reagent 5 to the product was complicated by the
diastereoselectivity for Sp product 3 and a slight racemization
of the phosphorus center of the reagent that was dependent on
the level of water contamination in the reaction mixture. For
product 3, trace levels of the Rp diastereomer 4 could easily be
reduced by crystallization. The Rp reagent 6 gave the Rp
phosphoramidate product 4, in a similar yield.
Even with diastereomerically pure reagent 5, the reaction on
the nucleoside 1 still did not go beyond 80% completion
(Scheme 3). A procedure was developed to isolate the product
only by crystallization to give 99.7% pure 3 in 40% yield, which
was a significant improvement over the conventional synthesis6
crystallized yield of 15%. Attempts to push the reaction to
completion by heating or adding more base or reagent were
not productive. A survey of alternative bases failed to find one
superior to tert-butylmagnesium chloride. Isopropylmagnesium
chloride provided similar results, but methylmagnesium bromide
gave a complex mixture. DBU and DMAP led first to reaction on
the 30 hydroxyl and racemization of the reagent phosphorus
center. Potassium tert-butoxide racemized the reagent and failed
to react. Sodium hydride did not give any product. Sodium
hexamethyldisilazane gave a low yield of product as well as the 30-
regiosiomer and 30,50-bis byproducts. Adding cosolvents such as
acetonitrile led to lower yields.
Nucleoside substrates without a competing free hydroxyl
would be expected to give higher yields. For example, protection
of the 30-hydroxyl group on 1 with tert-butyldimethylsilyl did
allow the standard reaction to proceed faster and upon heating to
45 °C allowed the reaction to reach completion, affording the 30-
protected product in 70% isolated yield. From an efficiency point
of view, with the additional steps for selective 30-hydroxyl
protection and deprotection, the overall yield was not improved.
8312
dx.doi.org/10.1021/jo201492m |J. Org. Chem. 2011, 76, 8311–8319

The Journal of Organic Chemistry
ARTICLE
Scheme 3. Synthesis of PSI-7977 (3) using Sp Nitrophenyl Reagent (5)
Table 1. Model Reaction To Compare Relative Reactivity of Reagents
Ar-OH
pKa
reactivity rank
isomer
2,4-dinitrophenol
pentafluorophenol
2-chloro-4-nitrophenol
2-nitrophenol
4.1
5.5
5.4
7.2
7.1
7.8
1
2
3
4
5
6
7, Sp
8, Sp
9, Sp
11, Sp/Rp mix
5, Sp
4-nitrophenol
2,4-dichlorophenol
12, Sp/Rp mix
Clean desilylation in the presence of the phosphoramidate also
proved challenging.
In our desire to increase the rate and extent of reaction, we
considered more reactive reagents. Increasing the electrophilicity
of the phosphorus center could be accomplished by adding
electronegative substituents to the phenolic leaving group. One
relative measure of this electronic effect is the pKa of the phenol
derivatives in water. The pKa of phenol itself is 10.0, whereas
4-nitrophenol has a pKa of 7.1. We prepared a series of analogous
reagents with other acidic phenolic leaving groups in a similar
series of reactions as shown in Scheme 2, except that we started
with commercial phenyl dichlorophosphate, followed by addi-
tion of amino acid ester and then the test phenol. Single
diastereomers were isolated by crystallization for 7, 8, and 9.
The Rp isomer of the reagent from 2-chloro-4-nitrophenol, 10,
could also be isolated through fractional crystallization. The
stereochemistry of 7 and 9 was assigned on the basis of the
ability to produce 3. Reagent 8 stereochemistry was assigned on
the basis of single crystal X-ray analysis. In order to test the
reagents for their relative reactivity, we chose the 20,30-O-
isopropylidene derivative of adenosine (13) as a model substrate
as it did not have the complications of a competing hydroxyl
group or an ionizable hydrogen that could form a salt with
reduced solubility. We did observe a positive general correlation
between greater acidity of the phenolic leaving group and
reactivity (Table 1, Figure 1). The 2-nitrophenyl reagent (11)
was more active than 4-nitrophenyl reagent (4) despite similar
pKa values of the leaving phenol. 2,4-Dinitrophenol was the most
acidic, and its reagent derivative 7 was most reactive with 13.
However, when tested with nucleoside 1, there was less dis-
crimination between the 50- and 30-hydroxyl groups, leading to a
Figure 1. Time course of model reactions to compare relative reactivity
of reagents.
higher proportion of 30,50-bis substitution. The pentafluorophe-
nyl reagent (8) was the optimal reagent identified in this series. A
separate reaction with reagent 8 and the adenosine derivative 13
was performed and resulted in a 96% isolated yield of 14. The
reaction of reagent 8 with 1 could be driven to completion in a
few hours at ambient temperature with good selectivity for the
desired 50-hydroxyl group. Reagent 8 could also be prepared in a
higher yield and crystallized readily as a single Sp diastereomer
with a diastereomeric excess greater than 99.0%. Therefore we
selected reagent 8 for further reaction optimization with 1 to
produce 3.
The main challenges in reaction optimization were solu-
bility of the nucleoside magnesium salt and balancing reaction
8313
dx.doi.org/10.1021/jo201492m |J. Org. Chem. 2011, 76, 8311–8319

The Journal of Organic Chemistry
ARTICLE
Scheme 4. Synthesis of 3 (PSI-7977) Using Pentafluorophenyl Reagent 8
Scheme 5. Synthesis of Diastereomerically Pure Cytidine and Guanosine Phosphoramidate Derivatives
completion versus minimizing the 30,50-bisphosphoramidate
byproduct. The magnesium salt of product 3 is more soluble in
the reaction solvent than the salt of nucleoside 1, leading to a
higher effective concentration of 3 and more exposure of its 30-
hydroxyl group to the reagent. Cooling the reaction to ꢀ5 °C
and slow addition of the reagent limited both the remaining
unreacted starting material 1 and the 30,50-bis byproduct to
5ꢀ8% as monitored by HPLC. The Rp product (4) was present
at less than 0.5%. Neither 30-monophosphoramidate nor the
pentafluorophenyl analogue of 4 was observed when compared
to prepared standards by HPLC. Lack of the latter side product
indicated a very high selectivity for displacing pentafluorophenol
instead of phenol in the reagent.
A multigram scale demonstration synthesis of 3 (Scheme 4)
started with the addition of 2.1 equiv of tert-butylmagnesium
chloride in THF at ꢀ5 °C to 1 followed by the addition of 1.2
equiv of reagent 8 and warming the reaction to 5 °C for 18 h. This
provided 3 as 85% of the total nucleosidic material present in the
reaction mixture by HPLC. Aqueous workup followed by two
crystallizations of the product produced a 68% isolated yield of
the desired 3 as a 99.8% pure crystalline material and with a
diastereomeric excess of 99.7%.
and a greater amount of the 30,50-bis substitution byproduct. The
purine analogue 20,30-O-isopropylidine guanosine, 17, had no
competing hydroxyl groups on the sugar but did contain an
ionizable hydrogen on the base unit. As for 1, an extra equivalent
of t-BuMgCl was added to allow the reaction to reach comple-
tion, producing an isolated yield of 75% of 18.
’ CONCLUSION
In summary, we have developed a novel synthetic method for
the preparation of diastereomerically pure nucleotide phosphor-
amidates and have demonstrated its use on representative
derivatives of all four classes of nucleosides. This method was
applied to the synthesis of the anti-HCV clinical agent PSI-7977,
3. Along with our earlier published work on the synthesis of 20-
fluoro-20-C-methyl nucleosides, this method now allows for the
efficient synthesis of PSI-7977 without the use of any chiral
catalysts or chiral auxiliaries and without any chromatography.
This method can be applied to the preparation of a variety of
stable phosphoramidation reagents and may find great utility in
medicinal chemistry exploration of nucleotide phosphorami-
dates. Optimization and scope of the chiral phosphoramidate
reaction is being explored and will be reported in the future.
This phosphoramidate methodology has the potential to be
broadly applicable for a variety of reagents and substrates. Those
reagents that do not crystallize as a single isomer can be separated
by chiral chromatography. Even when used as a mixture, this
method is still superior to existing phosphoramidate methods in
regioselectivity, yield, and with the convenience of a shelf-stable
reagent. Cytidine and guanosine nucleosides can also be used.
For example, as shown in Scheme 5, reaction with reagent 8 and
the cytidine analogue 15 (PSI-6130)14 led to an isolated yield of
53% of product 16. Although 15 had no ionizable base hydro-
gens, it had poor solubility in THF, which led to a slower reaction
’ EXPERIMENTAL SECTION
General Analytical Methods. Reactions were monitored by thin
layer chromatography with 250 μm silica gel plates and visualized by UV
light or by charring in 5% sulfuric acid in methanol. NMR spectra were
recorded in CDCl3, DMSO-d6, or CD3OD as noted on a 400 MHz
spectrometer with a multinuclear probe. Optical rotations were mea-
sured using a polarimeter at ambient temperature and 589 nm. Melting
points are uncorrected. HPLC purity was determined on a C18 reverse
phase column with a gradient of methanol in 10 mM ammonium acetate
8314
dx.doi.org/10.1021/jo201492m |J. Org. Chem. 2011, 76, 8311–8319

The Journal of Organic Chemistry
ARTICLE
aqueous buffer at pH 4.0. All solvents, reagents, and the starting
nucleosides 13 and 17 were used as received from commercial vendors.
(S)-2-[(4-Nitro-phenoxy)-phenoxy-phosphorylamino] Pro-
pionic Acid Isopropyl Ester Diastereomeric Mixture (5 and 6).
To a stirred solution of 4-nitrophenyl dichlorophosphate (12.8 g, 50
mmol) in dichloromethane (100 mL) was added a solution of phenol
(4.71 g, 50 mmol) and triethylamine (Lot 1, 7.7 mL, 55 mmol) in
dichloromethane (100 mL) at ꢀ78 °C over a period of 20 min. The
mixture was stirred at this temperature for 30 min and then transferred to
another round-bottom flask containing L-alanine isopropyl ester hydro-
chloride (8.38 g, 50 mmol) in dichloromethane (100 mL) at 0 °C. To the
mixture was added a second lot of triethylamine (Lot 2, 14.6 mL, 105
mmol) overa periodof15min. The mixture wasstirredat 0°C for 1 h, and
then the solvent was evaporated. The residue was triturated with ethyl
acetate (150 mL), and the white solid was filtered off. The filtrate was
concentrated under reduced pressure to give pale yellow oil. The crude
compound was chromatographed using 0ꢀ20% ethyl acetate/hexanes
gradient to give product (17 g, 83%) as a mixture of diastereomers 5 and 6
isopropyl ether to provide suitable crystals for single crystal X-ray analysis
which confirmed the Sp sterereochemistry. [α]25D (c 1.00, CHCl3) +1.7;
1
mp 62ꢀ66 °C; 31P NMR (162 MHz, DMSO-d6) δ ꢀ0.31; H NMR
(400 MHz, DMSO-d6) δ 8.30ꢀ8.27 (m, 2H), 7.49(d, J = 8.8 Hz, 2H),
7.41ꢀ7.37(m, 2H), 7.23ꢀ7.19 (m, 3H), 6.66 (dd, J = 13.6, 10.0 Hz, 1H),
4.86ꢀ4.78 (m, 1H), 3.97ꢀ3.86 (m, 1H), 1.19 (d, J = 7.2 Hz, 3H), 1.10
(d, J = 6.4 Hz, 6H); 13C NMR (CDCl3, 100 MHz) δ 172.4 (d, J = 8.4 Hz),
155.5 (d, J = 6.1 Hz), 150.2 (d, J = 6.9 Hz), 144.6, 129.8, 125.6, 125.4, 120.7
(d, J = 5.3 Hz), 120.1 (d, J = 4.5 Hz), 69.5, 50.5, 21.6 (d, J = 7.6 Hz), 20.9
(d, J = 3.8 Hz); HRMS-ESI (m/z): calcd for C18H22N2O7P [M + H]+
409.1165, found 409.1150.
Separation of (S)-2-[(R)-(4-Nitro-phenoxy)-phenoxy-phos-
phorylamino] Propionic Acid Isopropyl Ester (6) from 5 and 6
Mixture. A sample of the mixture of diastereomers (4.8 g) enriched with
the Rp isomer was subjected to SFC using a chiral column (2 ꢁ 15 cm)
and elutedwith 35%isopropanol in carbon dioxideat100 bar. Aninjection
loading of 4 mL of sample at a concentration of 17 mg/mL of methanol
was used. The Rp isomer (6) eluted first. The appropriate fractions of the
multiple runs were combined and concentrated under reduced pressure to
give2.9g of 6 as a lightyellow viscous oil thatsolidified upon standing over
several days, mp 42ꢀ43 °C and 1.9 g of the Sp isomer 5 as a white solid.
Both isomers were >99.9% pure by analytical SFC. Analytical data for 6:
[α]20D (c 1.0, MeOH) ꢀ5.6 ; 31P NMR (162 MHz, DMSO-d6) δ ꢀ0.47;
1H NMR (400 MHz, DMSO-d6) δ 8.30ꢀ8.27 (m, 2H), 7.46ꢀ7.38
(m, 4H), 7.27ꢀ7.20 (m, 3H), 6.68 (dd, J = 13.8, 10.2 Hz, 1H), 4.86ꢀ4.77
(m, 1H), 3.97ꢀ3.86 (m, 1H), 1.20 (d, J = 7.2 Hz, 3H), 1.10(dd, J = 6.2,
2.2 Hz, 6H); 13C NMR (100 MHz, CDCl3) δ 172.8 (d, J = 6.8 Hz), 155.8
(d, J = 6.8 Hz), 150.5 (d, J = 6.8 Hz), 144.8, 130.1, 125.8, 125.7, 121.1
(d, J = 5.3 Hz), 120.3 (d, J = 5.3 Hz), 69.7, 50.8, 21.9, 21.8, 21.0 (d, J =
5.3 Hz); MS (ESI) m/z 407 (M ꢀ 1)+; HRMS-ESI (m/z): calcd for
C18H22N2O7P [M + H]+ 409.1165, found 409.1167.
(S)-2-[(S)-(2,4-Dinitro-phenoxy)-phenoxy-phosphorylamino]
Propionic Acid Isopropyl Ester (7). Phenyl dichlorophosphate
(10.0 g, 47.4 mmol) was dissolved in dry dichloromethane (60 mL), and
the solution was cooled to ꢀ78 °C. A solution of 2,4-dinitrophenol (8.72
g, 47.4 mmol) and triethylamine (7.27 mL, 52.1 mmol) in dichloro-
methane (20 mL) was slowly added at ꢀ78 °C over a period of 30 min.
The reaction mixture was warmed to 0 °C and stirred for 2.5 h at this
temperature before L-alanine isopropyl ester (7.95 g, 47.4 mmol) was
added in one lot. After stirring for 40 min at 0 °C, triethylamine
(13.9 mL, 99.54 mmol) was added and stirred for additional 3 h at
0 °C. The reaction mixture was evaporated under reduced pressure, and
the residue was suspended in TBME (100 mL). The triethylamine
hydrochloride salt was removed by filtration, and the filtrate was
concentrated under reduced pressure to give yellow syrup. 31P NMR
of the crude sample indicated mixture of two diastereomers in the ratio
of 1:1. A mixture of EtOAc/hexanes (1:1, 50 mL) was added, and the
mixture was allowed to stir for 15 h. The white solid thus formed was
filtered off and washed with EtOAc/hexanes (1:1, 20 mL) and dried
under vacuum to give 6.0 g (28%) of 7. The phosphorus stereochemistry
was assigned as Sp on the basis of subsequent reactions to produce 3,
which has a known phosphorus stereochemistry. Based on 1H NMR, the
ratio of Sp/Rp isomers was 15.7:1 (88% de). Mp 104ꢀ106 °C; 31P NMR
(CDCl3, 162 MHz) δ ꢀ1.82; 1H NMR (CDCl3, 400 MHz) δ 8.82ꢀ8.81
(m, 1 H), 8.43ꢀ8.40 (m, 1 H), 7.89ꢀ7.86 (m, 1 H), 7.36ꢀ7.32 (m, 2 H),
7.23ꢀ7.19 (m, 3 H), 4.96 (hepta, J = 6.4 Hz, 1 H), 4.19ꢀ4.08 (m, 2 H), 1.42
(d, J= 6.4 Hz, 3 H), 1.20 (d, J=6.4Hz, 3H);13CNMR(CDCl3, 100MHz)
δ 172.3 (d, J = 8.4 Hz), 149.8 (d, J = 6.8 Hz), 148.4 (d, J = 5.3 Hz), 143.6,
140.9 (d, J = 6.8 Hz), 130.0, 128.8, 125.8, 123.9 (d, J = 2.3 Hz), 121.6, 120.1
(d, J = 5.4 Hz), 69.6, 50.6, 21.6 (d, J = 5.3 Hz), 20.8 (d, J = 4.6 Hz); HRMS-
ESI (m/z): calcd for C18H21N3O9P [M + H]+ 454.1010, found 454.1003.
(S)-2-[-(S)-(2,3,4,5,6-Pentafluoro-phenoxy)-phenoxy-phos-
phorylamino] Propionic Acid Isopropyl Ester (8). To a stirred
solution of L-alanine isopropyl ester hydrochloride (8.0 g 47.7 mmol) in
1
in about 1:1 ratio. 31P NMR (162 MHz, CDCl3) δ ꢀ2.05, ꢀ2.10; H
NMR (400 MHz, CDCl3) δ 8.22 (d, J = 9.2 Hz, 2H), 7.41ꢀ7.33(m, 4H),
7.26ꢀ7.18(m, 3H), 5.05ꢀ4.96(m, 1H), 4.14ꢀ4.05(m, 1H), 3.93ꢀ3.88(m,
1H), 1.38(d, J = 6.8 Hz, 3H), 1.22 (dd, J = 6.2 and 3.0 Hz, 6H); MS (ESI)
m/z 407 (M ꢀ 1)+.
(S)-2-[(S)-(4-Nitro-phenoxy)-phenoxy-phosphorylamino]
Propionic Acid Isopropyl Ester (5). L-Alanine isopropyl ester
hydrochloride (330 g, 1.97 mol) was predried by coevaporation with
toluene (2 ꢁ 400 mL) under reduced pressure and then dried in a
vacuum oven (50 °C, 0.2 mmHg, 17 h). To a stirred solution of
4-nitrophenyl dichlorophosphate (500.0 g, 1.95 mol) in anhydrous
dichloromethane (3.0 L) was added a solution of phenol (183.8 g,
1.95 mol) and triethylamine (300 mL, 2.15 mol) in dichloromethane
(900 mL) at ꢀ60 °C internal temperature over a period of 3 h. The
mixture was stirred at this temperature for additional 30 min and then
allowed to warm to ꢀ5 °C over 2.5 h. The predried amino acid ester was
added between ꢀ5 and 0 °C under an atmosphere of nitrogen over
10 min. The mixture was stirred at 0 °C for 40 min, and additional
triethylamine (571 mL, 4.10 mol) was added over a period of 40 min
at 0 °C. The mixture was stirred between 0 and 10 °C for 3 h, and then the
white solid (triethylamine hydrochloride) was filtered off and rinsed with
dichloromethane (3 ꢁ 300 mL). The filtrate was concentrated under
reduced pressure, and the residue was triturated with TBME (tert-
butylmethyl ether, 4 L). The additional triethylamine hydrochloride salt
was filtered off and washed with TBME (3 ꢁ 150 mL). The filtrate was
concentrated under reduced pressure to give clear light brown color oil.
The residue was coevaporated with hexanes (2 ꢁ 140 mL) to remove any
residual TBME and further dried under vacuum at 40 °C for 2 h. The dry
residue was dissolved in diisopropyl ether (1.1 L) and stirred at 5 °C in an
iceꢀwater bath. Small amount of seeds of the desired Sp isomer 5 was
added to the solution and the mixture was stirred at 5 °C for 22 h to form
moderately thick slurry. This was allowed to stand in a freezer (ꢀ10 °C)
for 44 h The product was collected via filtration and washed with a
precooled mixture of diisopropyl ether and hexanes (1:1, 3 ꢁ 190 mL).
The white solid was dried under vacuum (0.5 mmHg) at ambient
temperature until a constant weight was obtained to give 227.2 g (yield:
29%). The ratio of two diastereomers Sp:Rp was 9.7/1 based on
31P NMR (162 MHz, DMSO-d6, δ ꢀ0.31 (desired Sp isomer 5), ꢀ0.47
(Rp isomer 6). The product was recrystallized from diisopropyl ether
(840 mL) with seeds of Sp isomer 5. A white solid was formed within
2 h, and the flask was allowed to stand in a freezer (ꢀ10 °C) for 16 h. A
white and fine crystalline solid obtained was filtered, washed with precooled
diisopropyl ether (3 ꢁ 50 mL), and dried under vacuum (0.5 mmHg) to
give a white fluffy solid (177.7 g, 22% overall yield or 44% overall yield based
on theoretical yield of the Sp isomer 5) with diastereomeric ratio of 48/1
based on 31P NMR (96% de). A small sample was recrystallized from
8315
dx.doi.org/10.1021/jo201492m |J. Org. Chem. 2011, 76, 8311–8319

The Journal of Organic Chemistry
ARTICLE
anhydrous dichloromethane (50 mL) was added triethylamine (10.0 g,
13.8 mL, 99.0 mmol) at ꢀ70 °C over 15 min dropwise. To this mixture
was added a solution of phenyl dichlorophosphate (10.0 g, 47.3 mmol)
in anhydrous dichloromethane (50 mL) over 1 h. The reaction mixture
was stirred at this temperature for additional 30 min and then allowed to
warm to 0 °C over 2 h and stirred for 1 h. To this mixture was added a
solution of 2,3,4,5,6-pentafluoro phenol (8.72 g, 47.3 mmol) and
triethylamine (5.27 g, 7.26 mL, 52.1 mmol) in dichloromethane
(30 mL) over 20 min. The crude mixture was allowed to stir at 0 °C
for 4 h, and the white solid (triethylamine hydrochloride) was filtered off
and washed with dichloromethane (1 ꢁ 25 mL). The filtrate was
concentrated under reduced pressure, the residue was triturated with
TBME (150 mL), and the triethylamine hydrochloride salt was removed
by filtration. The cake was washed with TBME (2 ꢁ 25 mL), and the
combined filtrate was concentrated under reduced pressure to give 22 g
of crude solid containing an even mixture of diastereomers. The mixture
was triturated with 20% EtOAc in hexanes (100 mL) and collected by
filtration to give 7.4 g (34%) of 8 as a white solid (>98% de as
determined by NMR). A small portion was crystallized from the mixture
of 2-propanol and TBME to give suitable crystals for single crystal X-ray
isomer 10: [α]25 (c 1.00, CHCl3) +26.6; mp 77ꢀ80 °C; 31P NMR
D
(CDCl3, 162 MHz) δ ꢀ2.02; 1H NMR (CDCl3, 400 MHz) δ 8.32ꢀ8.31
(m, 1H), 8.13ꢀ8.10 (m, 1 H), 7.73ꢀ7.71 (m, 1 H), 7.38ꢀ7.34 (m, 2 H),
7.28ꢀ7.19 (m, 3 H), 5.02 (m, 1 H), 4.21ꢀ4.11 (m, 1 H), 4.01ꢀ3.95 (m, 1
H), 1.40 (d, 3 H), 1.25ꢀ1.22 (m, 6 H); 13C NMR (CDCl3, 100 MHz)
δ 172.4 (d, J = 9.1 Hz), 150.1 (d, J = 6.8 Hz), 142.7ꢀ142.5 (m),
140.2ꢀ140.0 (m), 139.3ꢀ139.0 (m), 137.8ꢀ137.4 (m), 136.8ꢀ136.5
(m), 129.8, 125.6, 120.0 (d, J = 4.5 Hz), 69.6, 50.6, 21.6 (d, J = 2.2 Hz),
20.9 (d, J = 4.6 Hz); HRMS-ESI (m/z): calcd for C18H21ClN2O7P
[M + H]+ 443.0769, found 443.0760.
(2S)-Isopropyl 2-(((2-Nitrophenoxy)(phenoxy)phosphoryl)
amino)propanoate (11). To a stirred solution of phenyl dichloro-
phosphate (30 g, 142.2 mmol) in dichloromethane (150 mL) was added
a solution of o-nitrophenol (19.76 g, 142.2 mmol) and triethylamine
(19.8 mL, 142.2 mmol) in dichloromethane (150 mL) over a period
of 1 h at ꢀ70 °C. The mixture slowly warmed to 0 °C over 2 h. To the
mixture was added L-alanine isopropyl ester hydrochloride salt (26.2 g,
156.3 mmol) in one lot followed by a solution of triethylamine (43.7 mL,
313.4 mmol) in dichloromethane (150 mL) over 20 min, and the
reaction mixture was stirred for additional 1 h. The solid was filtered, and
the filtrate was concentrated. The residue was purified by column
chromatography (20% EtOAc/hexanes) on a silica gel to give 11 as a
mixture of diastereomers (14.4 g, 25% yield). 31P NMR (CDCl3, 162
MHz) δ ꢀ1.55 (isomer I), ꢀ1.76 (isomer II); 1H NMR (CDCl3, 400
MHz) δ 7.94ꢀ7.90 (m, 1 H), 7.67ꢀ7.63 (m, 1 H), 7.57ꢀ7.54 (m, 1 H),
7.33ꢀ7.26 (m, 3 H), 7.23ꢀ7.14 (m, 3 H), 5.04ꢀ4.89 (m, 1 H),
4.21ꢀ4.04 (m, 2 H), 1.38 (d, 3 H, isomer I), 1.33 (d, 3 H, isomer II),
1.23ꢀ1.17 (m, 6 H); 13C NMR (CDCl3, 100 MHz) δ 172.6 (s), 172.5
(s), 150.3ꢀ150.2 (m), 143.7ꢀ143.6 (m), 141.4 (s), 134.5ꢀ134.4 (m),
129.8 (d), 125.7ꢀ125.2 (m), 123.2ꢀ123.0 (m), 120.3ꢀ120.2 (m),
69.4ꢀ69.3 (m), 68.3 (s), 50.5ꢀ50.3 (m), 22.9 (s), 21.6ꢀ21.5 (m),
21.0ꢀ20.9 (m); HRMS-ESI (m/z): calcd for C18H22N2O7P [M + H]+
409.1165. Observed: 409.1154.
(2S)-Isopropyl 2-(((2,4-Dichlorophenoxy)(phenoxy)phos
phoryl)amino)propanoate (12). To a stirred solution of phenyl
dichlorophosphate (10.0 g, 47.4 mmol) in dry dichloromethane (60 mL)
was added a solution of 2,4-dichlorophenol (7.73 g, 47.4 mmol) and
triethylamine (7.27 mL, 52.1 mmol) in dichloromethane (20 mL)
at ꢀ78 °C over 30 min. The reaction mixture was allowed to warm
to 0 °C and stirred for additional 2.5 h. To the above mixture was
added L-alanine isopropyl ester hydrochloride (7.95 g, 47.4 mmol)
in one lot followed by triethylamine (13.9 mL, 99.54 mmol). The
mixture was stirred for an additional 3 h at 0 °C, and the solvent
was evaporated under reduced pressure. The residue was chroma-
tographed using ethyl acetate/hexanes gradient to give 12 as color-
less viscous oil (13.6 g, 66% yield, 1:1 mixture of diastereomers).
31P NMR (CDCl3, 162 MHz) δ ꢀ1.52 (isomer I), ꢀ1.54 (isomer
II); 1H NMR (CDCl3, 400 MHz) δ 7.47ꢀ7.44 (m, 1 H), 7.42ꢀ7.39
(m, 1 H), 7.35ꢀ7.30 (m, 2 H), 7.24ꢀ7.15 (m, 3 H), 5.05ꢀ4.94
(m, 1 H), 4.19ꢀ4.08 (m, 1 H), 3.96ꢀ3.89 (m, 1 H), 1.41ꢀ1.35
(m, 1 H), 1.24ꢀ1.19 (m, 6 H); 13C NMR (CDCl3, 100 MHz)
172.5ꢀ172.4 (m), 150.4ꢀ150.2 (m), 145.6ꢀ145.4 (m), 130.4ꢀ
129.6 (m), 127.9 (d), 126.2 (d), 125.2 (s), 122.5ꢀ122.4 (m), 120.3ꢀ
120.1 (m), 69.3 (d), 50.5ꢀ50.4 (m), 21.6 (s), 21.5ꢀ20.9 (m). HRMS-
ESI (m/z): calcd for C18H21Cl2NO5P 1:1 [M + H]+ 432.0529, found
432.0521.
analysis, which confirmed the Sp sterereochemistry. [α]25 (c 1.00,
D
CHCl3) +3.8; mp 130ꢀ134 °C; 31P NMR (CDCl3, 162 MHz) δ ꢀ0.50;
1H NMR (CDCl3, 400 MHz) δ 7.38ꢀ7.34 (m, 2 H), 7.27ꢀ7.24 (m, 2
H), 7.23ꢀ7.19 (m, 1 H), 5.04 (m, J = 6.4 Hz, 1 H), 4.18ꢀ4.09 (m, 1 H),
4.00ꢀ3.95 (m, 1 H), 1.45 (d, J = 7.2 Hz, 3 H), 1.25 (d, J = 6 Hz, 3 H),
1.24 (d, J = 6 Hz, 3 H); 13C NMR (CDCl3, 100 MHz) δ 172.4 (d, J = 9.1
Hz), 150.1 (d, J = 6.8 Hz), 142.7ꢀ142.5 (m), 140.2ꢀ140.0 (m),
139.3ꢀ139.0 (m), 137.8ꢀ137.4 (m), 136.8ꢀ136.5 (m), 129.8, 125.6,
120.0 (d, J = 4.5 Hz), 69.6, 50.6, 21.6 (d, J = 2.2 Hz), 20.9 (d, J = 4.6 Hz);
HRMS-ESI (m/z): calcd for C18H18F5NO5P [M + H]+ 454.0837, found
454.0829.
(S)-2-[(S)-(2-Chloro-4-nitro-phenoxy)-phenoxy-phospho-
rylamino] Propionic Acid Isopropyl Ester (9) and (S)-2-
[(R)-(2-Chloro-4-nitro-phenoxy)-phenoxy-phosphorylamino]
Propionic Acid Isopropyl Ester (10). Phenyl dichlorophosphate
(10.0 g, 47.3 mmol) was dissolved in 50 mL of dry dichloromethane and
cooled to 0 °C. After addition of solid L-alanine isopropyl ester HCl salt
(7.94 g, 47.3 mmol), the reaction mixture was cooled to ꢀ70 °C and then
treatedwithtriethylamine (13.8mL, 94.6 mmol) dissolved in 50 mLofdry
dichloromethane. The resulting mixture was stirred for 30 min at this
temperature and then allowed to warm to 0 °C. A solution of 2-chloro-4-
nitrophenol (8.2 g, 47.3 mmol) and triethylamine (6.6 mL, 47.3 mmol) in
20 mL of dry dichloromethane was added over 5ꢀ10 min and stirred for
additional 2 h. The solution was filtered, and the filtrate was concentrated
under reduced pressure. The resulting residue was suspended in 50 mL of
TBME and stirred for 10 min at room temperature. The white solid was
removed by filtration, and the filtrate was concentrated under reduce
pressure. Column chromatography of the residue using dichloromethane
as an eluent gave the desired product (12.2 g, Sp/Rp ratio of 2:1 as
determined by 31P NMR) as a pale yellow solid. The product was
recrystallized twice from EtOAc/hexane (2:3) to give pure 9 (5.2 g,
25% yield, >98% de). Stereochemistry at phosphorus center was assigned
as Sp on the basis of the ability of the product to make 3. Upon cooling the
mother liquor to ꢀ5 °C, the Rp isomer 10 was crystallized out as a white
solid that was collected by filtration (1.5 g, 7% yield, >98% de). Sp isomer
9: [α]25D (c 1.00, CHCl3) ꢀ16.1; mp 70ꢀ72 °C; 31P NMR (CDCl3, 162
MHz)δꢀ1.97;1H NMR(CDCl3, 400 MHz)δ8.33(m, 1H), 8.13ꢀ8.10
(m, 1 H), 7.73ꢀ7.71 (m, 1 H), 7.36ꢀ7.33 (m, 2 H), 7.25ꢀ7.18 (m, 3 H),
5.00 (m, 1 H), 4.19ꢀ4.10 (m, 1 H), 4.02ꢀ3.97 (m, 1 H), 1.43 (d, 3 H),
1.23ꢀ1.21 (m, 6 H); 13C NMR (CDCl3, 100 MHz) δ 172.3 (d, J = 8.3
Hz), 151.9 (d, J = 5.3 Hz), 150.0 (d, J = 7.6 Hz), 144.4, 129.9, 126.4 (d,
J = 8.4 Hz), 126.1, 125.7, 123.4, 121.5 (d, J = 2.3 Hz), 120.2 (d, J = 4.6 Hz),
69.6, 50.6, 21.6 (d, J = 7.6 Hz), 21.0 (d, J = 4.5 Hz); HRMS-ESI (m/z):
calcd for C18H21ClN2O7P [M + H]+ 443.0769, found 443.0759. Rp
(S)-Isopropyl (((2R,3R,4R,5R)-5-(2,4-Dioxo-3,4-dihydropy-
rimidin-(2H)-yl)-4-fluoro-3-hydroxy-4-methyltetrahydrofur-
an-2-yl)methoxy)(phenoxy)phosphorylamino)propanoate
Diastereomeric Mixture (3 and 4). To a stirred solution of
1-((2R,3R,4R,5R)-3-fluoro-4-hydroxy-5-hydroxymethyl-3-methyl-tetra-
hydro-furan-2-yl)-1H-pyrimidine-2,4-dione (130 mg, 0.5 mmol) in dry
THF (1.5 mL) was added a 1.0 M solution of tert-butylmagnesium
chloride (1.05 mL, 1.05 mmol, 2.1 equiv) at room temperature over a
8316
dx.doi.org/10.1021/jo201492m |J. Org. Chem. 2011, 76, 8311–8319

The Journal of Organic Chemistry
ARTICLE
period of 5 min. After 30 min, a solution of (S)-2-[(4-nitro-phenoxy)-
phenoxy-phosphorylamino] propionic acid isopropyl ester (1:1 mixture
of isomers 5 and 6, 408 mg, 1.0 mmol) in THF (1.5 mL) was added
dropwise over a period of 5 min. The mixture was allowed to stir at room
temperature for 48 h and then quenched with saturated aqueous NH4Cl
(20 mL). The mixture was partitioned between ethyl acetate (50 mL)
and water (20 mL). The combined organic extract was dried over
anhydrous sodium sulfate, filtered, and concentrated under reduced
pressure to give a pale yellow residue. Column chromatography of the
residue using 0ꢀ5% MeOH/dichloromethane gradient gave a white
foamy solid (125 mg, 47% yield, mixture of Sp/Rp diastereomers 3 and 4
in about 3.05:1.0 ratio, respectively). Analytical data of the product was
consistent with the original mixture of diastereomers.6
(S)-Isopropyl 2-((R)-(((2R,3R,4R,5R)-5-(2,4-Dioxo-3,4-dihy-
dropyrimidin-1(2H)-yl)-4-fluoro-3-hydroxy-4-methyltetra-
hydrofuran-2-yl)methoxy)(phenoxy)phosphorylamino)-
propanoate (4). To a stirred solution of 1-((2R,3R,4R,5R)-3-fluoro-
4-hydroxy-5-hydroxymethyl-3-methyl-tetrahydro-furan-2-yl)-1H-pyr-
imidine-2,4-dione (1, 260 mg, 1 mmol) in dry THF (6 mL) was added a
1.7 M solution of tert-butylmagnesium chloride (1.23 mL, 2.1 mmol, 2.1
equiv) at room temperature over a period of 5 min. After 30 min, a
solution of (S)-2-[(R)-(4-nitro-phenoxy)-phenoxy-phosphorylamino]
propionic acid isopropyl ester (6, Rp isomer) in THF (3 mL) was
added dropwise over a period of 3 min. The mixture was allowed to stir at
room temperature for 96 h and then quenched with saturated aqueous
NH4Cl (10 mL). The mixture was partitioned between ethyl acetate
(50 mL) and water (2 ꢁ 20 mL). The combined organic extract was
dried over anhydrous sodium sulfate, filtered, and concentrated under
reduced pressure to give a pale yellow residue (490 mg). The residue was
chromatographed using 0ꢀ5% MeOH/dichloromethane gradient to
give product 4 as a white solid (160 mg, 30% yield). Analytical data of the
product was consistent with standard sample of 4.6
(S)-Isopropyl 2-((S)-(((2R,3R,4R,5R)-5-(2,4-Dioxo-3,4-dihydro-
pyrimidin-1(2H)-yl)-4-fluoro-3-hydroxy-4-methyltetrahydrofur-
an-2-yl)methoxy)(phenoxy)phosphorylamino)propanoate (3)
Using Reagent 5. 1-((2R,3R,4R,5R)-3-Fluoro-4-hydroxy-5-hydroxy-
methyl-3-methyl-tetrahydro-furan-2-yl)-1H-pyrimidine-2,4-dione (1, 10.0 g,
38.5 mmol) was dried under vacuum at 50 °C for 20 h and dissolved in dry
THF (200 mL) in a two neck 1 L round-bottom flask equipped with an
addition funnel, nitrogen inlet, and magnetic stir bar. The flask was immersed
in a tap water bath at ambient temperature, and then was added a 1.7 M
solution of tert-butylmagnesium chloride (47.5 mL, 80.7 mmol, 2.1 equiv)
over a period of 30 min (slightly exothermic). After an additional 30 min, a
solution of (S)-2-[(S)-(4-nitro-phenoxy)-phenoxyphosphorylamino] pro-
pionic acid isopropyl ester (5) in THF (20 mL) was added in over a period of
15 min. The mixture was allowed to stir at room temperature for 60 h. TLC of
the reaction mixture indicated ∼20% of unreacted starting material. The
reaction mixture was quenched with saturated aq ammonium chloride
(5 mL). The solvent was evaporated under reduced pressure, and the residue
was dried under high vacuum to give yellow foam. The residue was suspended
in ethyl acetate (400 mL) and then washed with water (2 ꢁ 50 mL). The
organic layer was washed with 5% aqueous sodium bicarbonate (6 ꢁ 50 mL),
until TLC of the organic layer indicated the absence of nitrophenol. The
organic layer was washed with water (3 ꢁ 50 mL) and brine (50 mL), dried
over anhydrous sodium sulfate, and concentrated to give a light brown solid,
18.2 g. Dichloromethane (45 mL) was added to the solid and heated at reflux
for 3 h, then allowed to cool to ambient temperature, and stirred for 65 h. The
white solid was filtered, washed with 1:1 (v/v) isopropyl ether/dichloro-
methane (20 mL), and dried under high vacuum to give white solid, 9.2 g.
The solid was redissolved in dichloromethane (275 mL), heated to reflux,
and filtered hot through filter paper. The dichloromethane was concentrated
at atmospheric pressure to a volume of 200 mL. The solution was cooled
to ambient temperature and allowed to stand for 17 h. The white needles
obtained were filtered and dried under high vacuum for 17 h to give 3
(8.1 g, 40% yield, 99.7% HPLC purity). Analytical data of the product was
consistent with the standard sample of 3.6
(S)-Isopropyl 2-((S)-(((2R,3R,4R,5R)-5-(2,4-Dioxo-3,4-dihy-
dropyrimidin-1(2H)-yl)-4-fluoro-3-hydroxy-4-methyltetra-
hydrofuran-2-yl)methoxy)(phenoxy)phosphorylamino)
propanoate (3) Using Reagent 8. To a stirred suspension of
1-((2R,3R,4R,5R)-3-fluoro-4-hydroxy-5-(hydroxymethyl)-3-methyl-
tetrahydrofuran-2-yl)pyrimidine-2,4(1H,3H)-dione (1, 5.0 g, 19.1
mmol, dried under vacuum at 50 °C for 20 h) in dry THF (75 mL)
was added a 1.7 M solution of tert-butylmagnesium chloride in THF
(23.7 mL, 40.35 mmol) using an addition funnel over a period of 30 min
at ꢀ5 °C. The white suspension was stirred at this temperature for
30 min and then warmed to ambient temperature (20 °C) at which
temperatureit was stirred for additional 30 min. Thereaction mixturewas
cooled to 5 °C, and then was added a solution of (S)-2-[(S)-(2,3,4,5,
6-pentafluorophenoxy)-phenoxyphosphorylamino] propionic acid iso-
propyl ester (8, 10.45 g, 23.06 mmol) in THF (50 mL) over a period
of 30 min. The mixture was stirred at 5 °C for 18 h, cooled to ꢀ5 °C,
and then quenched with 2 N HCl (25 mL). Toluene (100 mL) was
added to the mixture and warmed to room temperature. After 20 min
the layers were separated. The organic layer was washed with 1 N HCl
(2 ꢁ 10 mL), water (10 mL), 5% aqueous Na2CO3 (4 ꢁ 10 mL), water
(2 ꢁ 10 mL), and brine (10 mL). All of the aqueous layers were
re-extracted with toluene (20 mL) and washed with 5% aqueous
Na2CO3 (2 ꢁ 5 mL), water (10 mL), and brine (5 mL). The combined
organic layer was dried over anhydrous sodium sulfate, filtered, and
evaporated to an approximate volume of 20 mL. Dichloromethane
(20 mL) was added to the solution, and the mixture was stirred at room
temperature for 18 h. The solid was filtered, washed with 1:1 TBME/
dichloromethane mixture (2 ꢁ 10 mL), and dried under high vacuum to
give white solid (7.7 g). HPLC of the solid indicated 98.21% of 3, 0.18%
of unreacted nucleoside 1 and 0.67% of 30,50-bis-phosphoramidate
impurity. The above solid was redissolved in dichloromethane (77 mL,
heated in a pressure vessel at 55 °C) and allowed to stand at room
temperature for 20 h. The crystalline solid was filtered washed with cold
dichloromethane (5 mL, 0 °C) and dried under high vacuum to give
pure product as a white solid (6.9 g, 68% yield) matching a standard
sample of 3.6 HPLC analysis showed a total purity of 99.79% with
impurities composed of 0.07% unknown and 0.14% of the Rp isomer,
resulting in a diastereomeric excess of 99.72%.
(S)-Isopropyl 2-(((S)-(((3aR,4R,6R,6aR)-6-(6-Amino-9H-purin-
9-yl)-2,2-dimethyltetrahydrofuro-[3,4-d][1,3]dioxol-4-yl)
methoxy)(phenoxy)phosphoryl)amino)propanoate (14). To
a stirred solution of ((3aR,4R,6R,6aR)-6-(6-amino-9H-purin-9-yl)-2,2-
dimethyltetrahydrofuro[3,4-d][1,3]dioxol-4-yl)methanol (13, 307 mg,
1 mmol) in dry THF (3 mL) was added a 1.7 M solution of tert-
butylmagnesium chloride in THF (0.706 mL, 1.2 mmol) over a period
of 3 min at room temperature. The white suspension was stirred at
this temperature for 30 min, and then was added a solution of (S)-
2-[(S)-(2,3,4,5,6-pentafluorophenoxy)-phenoxyphosphorylamino] pro-
pionic acid isopropyl ester (8, 544 mg, 1.2 mmol) in THF (3 mL) over a
period of 3 min. The mixture was stirred at this temperature for 18 h. The
reaction mixture was quenched with methanol (1 mL), solvent was
evaporated, and the residue was chromatographed using 0ꢀ10%
methanol/dichloromethane gradient to give pure 14 as a white amor-
phous solid, mp 43ꢀ65 °C (552 mg, 96% yield). There was no other
isomer detectable by 31P or 1H NMR. Stereochemistry was assigned as
Sp on the basis of the known stereochemistry of the regent. [α]20D (c 1.0,
MeOH) ꢀ25.7; 31P NMR (162 MHz, CDCl3) δ 2.81; 1H NMR (400
MHz, CDCl3) δ 8.32 (s, 1H), 7.94 (s, 1H), 7.29ꢀ7.25 (m, 2H),
7.17ꢀ7.10 (m, 3H), 6.10 (d, J = 2.0 Hz, 1H), 5.92 (bs, 2H), 5.38(dd,
J = 6.4, 2.0 Hz, 1H), 5.07 (dd, J = 6.0, 3.2 Hz, 1H), 4.94 (m, J = 6.4 Hz,
1H), 4.51ꢀ4.47 (m, 1H), 4.37ꢀ4.32 (m, 1H), 4.28ꢀ4.22 (m, 1H), 4.02
(t, J = 10.4 Hz, 1H), 3.94ꢀ3.84 (m, 1H), 2.10 (bs, 1H), 1.61 (s, 3H),
8317
dx.doi.org/10.1021/jo201492m |J. Org. Chem. 2011, 76, 8311–8319

The Journal of Organic Chemistry
ARTICLE
1.38 (s, 3H), 1.23 (d, J = 6.8 Hz, 3H), 1.18 (dd, J = 5.6, 1.6 Hz, 6H);
13C NMR (100 MHz, CDCl3): 173.4 (d, J = 7.6 Hz), 156.1, 153.5, 150.8
(d, J = 6.8 Hz), 149.4, 140.0, 129.9, 125.2, 120.4 (d, J = 4.6 Hz), 120.3,
114.8, 91.2, 85.8 (d, J = 8.6 Hz), 84.5, 81.8, 69.5, 66.5(d, J = 5.5 Hz), 50.5,
27.4, 25.6, 21.9 (d, J = 5.3 Hz), 21.2 (d, J = 5.3 Hz). HRMS-ESI (m/z):
calcd for C25H34N6O8P [M + H]+ 577.2170, found 577.2163.
General Experimental for Conversion of 13 to 14 (Model
Study). To a stirred solution of nucleoside (1 mmol) in dry THF
(3 mL) was added a 1.7 M solution of tert-butylmagnesium chloride in
THF (0.706 mL, 1.2 mmol) over a period of 3 min at room temperature.
The white suspension was stirred at this temperature for 30 min,
and then was added a solution of the (S)-2-aryloxy-phenoxyphosphorylamino]
propionic acid isopropyl ester (1.2 mmol) in THF (3 mL) over a
period of 3 min. The progress of the reaction was monitored
by HPLC (column: Luna 3 μm C8; 50 mm ꢁ 4.6 mm; flow rate:
1.0 mL/min; mobile phase: 2ꢀ98% 0.1% TFA in CH3CN/0.1% TFA
in water; UV: 254 nm; retention times: 7.71 min for product and
4.99 min for nucleoside starting material) at 15, 45, 90, and 180 min time
points.
(S)-Isopropyl 2-(((S)-(((2R,3R,4R,5R)-5-(4-Amino-2-oxopyri-
midin-1(2H)-yl)-4-fluoro-3-hydroxy-4-methyltetrahydrofur-
an-2-yl)methoxy)(phenoxy)phosphoryl)amino)propanoate
(16). To a stirred suspension of 4-amino-1-((2R,3R,4R,5R)-3-fluoro-4-
hydroxy-5-(hydroxymethyl)-3-methyltetrahydrofuran-2-yl)pyrimidin-2
(1H)-one (15,14 259 mg, 1 mmol) in dry THF (3 mL) was added a 1.7 M
solution of tert-butylmagnesium chloride in THF (0.71 mL, 1.2 mmol)
over a period of 3 min at room temperature. The white suspension was
stirred at this temperature for 30 min, and then was added a solution
of(S)-2-[(S)-(2,3,4,5,6-pentafluorophenoxy)-phenoxyphosphorylamino]
propionic acid isopropyl ester (8, 544 mg, 1.2 mmol) in THF (3 mL) over
a periodof3 min. The mixture was stirredatthis temperature for18 h. The
reaction mixture was quenched with methanol (1 mL), solvent was
evaporated, and the residue was chromatographed using 0ꢀ15% metha-
nol/dichloromethane gradient to give pure 16 as a white amorphous solid,
mp 67ꢀ95 °C (445 mg, 53% yield). There was no other isomer detectable
by 31P or 1H NMR. Stereochemistry was assigned as Sp on the basis of the
’ ASSOCIATED CONTENT
Supporting Information. 1H and 13C NMR spectra for
S
b
compounds 5ꢀ12, 14, 16, and 18; single crystal X-ray data for
5 and 8 in CIF format; and complete refs 2a, 2b, 6, and 7. This
material is available free of charge via the Internet at http://pubs.
acs.org.
’ AUTHOR INFORMATION
known stereochemistry of the regent. [α]20D (c 1.0, MeOH) +57.8 ; 31
P
NMR (162 MHz, CD3OD) δ 4.57; 1H NMR (400 MHz, CD3OD) δ 7.63
(d, J = 7.2 Hz, 1H), 7.39ꢀ7.36 (m, 2H), 7.28ꢀ7.26 (m, 2H), 7.22ꢀ7.19
(m, 1H), 6.24 (bd, J = 12.4 Hz, 1H), 5.84 (d, J = 7.6 Hz, 1H), 4.95 (m, J =
6.2 Hz, 1H), 4.53 (br d dd, J = 11.2, 5.2 Hz, 1H), 4.38 (ddd, J = 12.0, 5.8,
3.6Hz, 1H), 4.11ꢀ4.08(m, 1H), 3.95ꢀ3.87 (m, 2H), 1.34 (dd, J= 7.2 Hz,
3H), 1.29 (d, J = 22.4 Hz, 3H), 1.20 (d, J = 6.0 Hz, 6H); 13C NMR (100
MHz, CD3OD) δ 173.2 (d, J = 5.3 Hz), 166.3, 156.9, 150.9 (d, J = 6.0 Hz),
140.4, 129.7, 125.1, 120.2, (d, J = 5.3 Hz), 101.3, 99.5, 95.5, 89.8, 79.5, 71.9
(d, J = 17.4 Hz), 69.0, 64.6, 50.5, 20.8, 20.7, 19.5 (d, J = 6.1 Hz), 15.8 (d, J =
25.1 Hz). HRMS-ESI (m/z): calcd for C22H31FN4O8P [M + H]+
529.1865, found 529.1849.
Corresponding Author
*E-mail: bruce.ross@pharmasset.com.
’ ACKNOWLEDGMENT
We thank Dr. Patrick Carroll at the University of Pennsylvania
for providing the single crystal x-ray data of the chiral reagents 5
and 8 and Drs. Vinod Uchil and K. Sreenivasulu Reddy of
Escientia Life Sciences for providing technical assistance.
’ REFERENCES
(S)-Isopropyl 2-(((S)-(((3aR,4R,6R,6aR)-6-(2-Amino-6-oxo-
1H-purin-9(6H)-yl)-2,2-dimethyl-tetrahydrofuro[3,4-d][1,3]-
dioxol-4-yl)methoxy)(phenoxy)phosphoryl)amino)propanoate
(18). To a stirred solution of 2-amino-9-((3aR,4R,6R,6aR)-6-(hydro-
xymethyl)-2,2-dimethyltetrahydrofuro[3,4-d][1,3]dioxol-4-yl)-1H-pur-
in-6(9H)-one (17, 323 mg, 1 mmol) in dry THF (3 mL) was added a
1.7 M solution of tert-butylmagnesium chloride in THF (1.24 mL, 2.1
mmol) over a period of 3 min at room temperature. The white suspension
was stirred at this temperature for 30 min and then was added a solution
of(S)-2-[(S)-(2,3,4,5,6-pentafluorophenoxy)-phenoxyphosphorylamino]
propionic acid isopropyl ester (8, 544 mg, 1.2 mmol) in THF (3 mL) over
a periodof3 min. The mixture was stirredatthis temperature for18 h. The
reaction mixture was quenched with methanol (1 mL), solvent was
evaporated, and the residue was chromatographed using 0ꢀ10% metha-
nol/dichloromethane gradient to give pure 18 as a white amorphous solid,
mp 101ꢀ136 °C (445 mg, 75% yield). There was no other isomer
detectable by 31P or 1H NMR. Stereochemistry was assigned as Sp on the
basis of the known stereochemistry of the regent. [α]20D (c 1.0, MeOH)
(1) Siddiqui, A. Q.; Balatore, C.; McGuigan, C.; Pathirana, R. N.;
Balzarini, J.; Declercq, E. J. Med. Chem. 1999, 42, 393.
(2) Selected examples: (a) Gardelli, C.; et al. J. Med. Chem. 2009,
52, 5394–5407. (b) McGuigan, C.; et al. Bioorg. Med. Chem. Lett. 2010,
20, 4850–4854. (c) Zhou, X.-J.; Pietropaolo, K.; Chen, J.; Khan, S.;
Sullivan-Boylai, J.; Mayers, D. Antimicrob. Agents Chemother. 2011,
55, 76–81. (d) Mehellou, Y.; Balzarini, J.; McGuigan, C. ChemMedChem
2009, 4, 1779–91. (e) McGuigan, C.; Kelleher, M. R.; Perrone, P.;
Mulready, S.; Luoni, G.; Daverio, F.; Rajyaguru, S.; Le Pogam, S.; Najera,
I.; Martin, J. A.; Klumpp, K.; Smith, D. B. Biorg. Med. Chem. Lett. 2009,
19 (15), 4250–4254.
(3) Gudmundsson, K. S.; Wang, Z.; Daluge, S. M.; Johnson, L. C.;
Hazen, R.; Condreay, L. D.; McGuigan, C. Nucleosides, Nucleotides
Nucleic Acids 2004, 23, 1929–1937.
(4) Venkatachalam, T. K.; Qazi, S.; Uckun, F. M. Arzneim. Forsch.
2006, 56, 136–151.
(5) Murakami, E.; Tolstykh, T.; Bao, H.; Niu, C.; Micholochick
Steuer, H. M.; Bao, D.; Chang, W.; Espiritu, C.; Bansal, S.; Lam, A. M.;
Otto, M. J.; Sofia, M. J.; Furman, P. A. J. Biol. Chem. 2010, 285,
34337–34347.
(6) Sofia, M. J.; et al. J. Med. Chem. 2010, 53, 7202–7218.
(7) Nelson, D. R. et al. European Association for the Study of Liver
Disease, Berlin, Germany, March 30ꢀAprit 3, 2011, abstract 1372.
(8) (a) Nakayama, K.; Thompson, W. J. J. Am. Chem. Soc. 1990,
112, 6936–6942. (b) Wozniak, L. A; Okruzsek, A. Chem. Soc. Rev. 2003,
32, 158–169.
1
+3.5; 31P NMR (162 MHz, DMSO-d6) δ 4.67; H NMR (400 MHz,
DMSO-d6) δ 10.73 (bs, 1H), 7.83 (s, 1H), 7.35ꢀ7.31 (m, 2H),
7.18ꢀ7.13 (m, 3H), 6.55 (bs, 1H), 6.03ꢀ5.98 (m, 2H), 5.16 (dd, J =
6.0, 2.0 Hz, 1H), 5.12 (dd, J = 6.4, 2.8 Hz, 1H), 4.79 (m, J = 6.4 Hz, 1H),
4.31ꢀ4.22 (m, 2H), 4.04ꢀ3.99 (m, 1H), 3.80ꢀ3.69 (m, 1H), 1.49 (s,
3H), 1.29 (s, 3H), 1.15 (d, J = 6.8 Hz, 3H), 1.09 (dd, J = 6.0, 1.5 Hz, 6H);
13C NMR (100 MHz, DMSO-d6) δ 173.3 (d, J = 4.6 Hz), 157.4, 154.4,
151.3 (d, J = 6.8 Hz), 151.0, 136.9, 130.3, 125.3, 120.8 (d, J = 4.6 Hz),
117.7, 113.9, 89.2, 85.8 (d, J = 8.4 Hz), 84.5, 81.8, 68.6, 66.4, 50.4, 27.7,
25.9, 22.1 (d, J = 3.8 Hz), 20.3 (d, J = 6.1 Hz). HRMS-ESI (m/z): calcd for
C25H34N6O9P [M + H]+ 593.2127, found 593.2160.
(9) Lehsten, D. M.; Baehr, D. N.; Lobl, T. J.; Vaino, A. R. Org. Process
Res. Dev. 2002, 6, 819–822.
(10) Roman, C. A; Balzarini, J.; Meier, C. J. Med. Chem. 2010, 53,
7675–7681.
8318
dx.doi.org/10.1021/jo201492m |J. Org. Chem. 2011, 76, 8311–8319

The Journal of Organic Chemistry
ARTICLE
(11) Uchiyama, M.; Aso, Y.; Noyori, R.; Hayakawa, Y. J. Org. Chem.
1993, 58, 373–379.
(12) Howes, P. D.; Slater, M. J.; Wareing, K. Nucleosides, Nucleotides
Nucleic Acids 2003, 22 (5ꢀ8), 687–689.
(13) Perrone., P.; Luoni, G. M.; Kelleher, M. R.; Daverio, F.; Angell,
A.; Mulready, S.; Congiatu, C.; Rajyaguru, S.; Martin, J. A.; Leveque, V.;
Le Pogam, S.; Najera, S.; Klumpp, K.; Smith, D. B.; McGuigan, C. J. Med.
Chem. 2007, 50, 1840–1849.
(14) Wang, P.; Chun, B.-K.; Rachakonda, S.; Du, J.; Khan, S.; Shi, J.;
Stec, W.; Cleary, D.; Ross, B. S.; Sofia, M. J. J. Org. Chem. 2009,
74, 6819–6824.
8319
dx.doi.org/10.1021/jo201492m |J. Org. Chem. 2011, 76, 8311–8319