Diastereoselective synthesis of aryloxy phosphoramidate prodrugs of 3'-deoxy-2',3'-didehydrothymidine monophosphate

Article abstract of DOI:10.1021/jm100817f

The first diastereoselective synthesis of aryloxy phosphoramidate prodrugs of 3'-deoxy-2',3'-didehydrothymidine monophosphate (d4TMP) is reported. In our approach, (S)-4-isopropylthiazolidine-2-thione 1 was used as a chiral auxiliary to introduce the ster

Full text of DOI:10.1021/jm100817f

J. Med. Chem. 2010, 53, 7675–7681 7675
DOI: 10.1021/jm100817f
Diastereoselective Synthesis of Aryloxy Phosphoramidate Prodrugs
of 30-Deoxy-20,30-didehydrothymidine Monophosphate§
Cristina Arbelo Roman,† Jan Balzarini,‡ and Chris Meier*,†
†Organic Chemistry, Department of Chemistry, Faculty of Science, University of Hamburg, Martin-Luther-King-Platz 6, 20146 Hamburg,
Germany, and ‡Rega Institute for Medical Research, Katholieke Universiteit Leuven, Minderbroedersstraat 10, B-3000 Leuven, Belgium
Received July 2, 2010
The first diastereoselective synthesis of aryloxy phosphoramidate prodrugs of 30-deoxy-20,30-didehy-
drothymidine monophosphate (d4TMP) is reported. In our approach, (S)-4-isopropylthiazolidine-
2-thione 1 was used as a chiral auxiliary to introduce the stereochemistry at the phosphorus atom. In the
last step of the developed reaction sequence, the nucleoside analogue d4T was introduced to a stereo-
chemically pure phosphordiamidate which led to the formation of the almost diastereomerically
pure phosphoramidate prodrugs 8a-d (g95% de). As expected, the individually prepared diaste-
reomers of the phosphoramidate prodrugs showed significant differences in the antiviral activity.
Moreover, the difference was strongly dependent on the aryl substituent attached to the phosphor-
amidate moiety.
Introduction
nucleoside analogues. So far, all these pronucleotides cannot
beprepared in the formof single diastereomers because of lack
of control of the stereochemistry at the phosphorus atom
during synthesis; e.g., the reported synthesis of the phosphor-
amidates afforded the prodrugs as a mixture of two diaste-
reomers in an almost 1:1 ratio.4 Moreover, the chromato-
graphic separation of the diastereomers was often impossible
or a very difficult task to achieve. Liquid chromatography on
chiral stationary phases provided successful results in the
separation of the diastereomers in only a few cases.10,11
However, in those cases in which it was possible to separate
the diastereomers of the phosphoramidates or the cycloSal
compounds, the biological activity was found to be strongly
dependent on the configuration at the phosphorus atom.
For example the SP-diastereomer of the 9-[2-(R)-phospho-
nomethoxypropyl]adenine prodrug (GS-7340) is 10-foldmore
potent against HIV than the RP-diastereomer.12 The biological
activity against other viruses like hepatitis C was also dependent
on the phosphorus configuration. As mentioned above, phos-
phoramidates of 20-C-methylcytidine were synthesized as mix-
tures first and then separated into individual diastereomers
using reversed phase HPLC for antiviral evaluation. The anti-
viral evaluation of these separated diastereomers proved that
the more lipophilic diastereomer was also about 8-fold more
antivirally active than the other diastereomer.13 Additionally,
not only the antiviral activity but also the cytostatic activity
depended on the phosphorus configuration. 1-Naphthyl phos-
phoramidates of [5-(E-bromovinyl)]-20deoxyuridine were syn-
thesized as a mixture, and after separation, the diastereomers
were evaluated in a breast cancer cell line. Again, the SP-
diastereomer was about 10 times more active than the diaste-
reomeric mixture, whereas the RP-diastereomer was slightly less
active than the mixture.11 All these examples show the impor-
tance of the phosphorus configuration on the biological activity.
For this reason the development of a diastereoselective synthesis
of phosphoramidate prodrugs would be of great interest. Here,
we report on a successful chemical approach.
Chiral phosphorus compounds are important target struc-
tures. It has been proven that the configuration at the phos-
phorus atom has an influence on the biological activity while
interacting with biomolecules. Therefore, the stereoselective
synthesis of such compounds is an important aim to achieve.
Stec et al. reported the oxathiaphospholane method, a none-
nzymatic approach that allows the diastereoselective synthesis
of phosphorothioate oligonucleotides.1 Wada et al. developed a
new approach to synthesize diastereomerically pure backbone-
modified DNA analogues, which is based on the oxazaphos-
pholidine method.2 Another important application of chiral
phosphorus compounds is as pronucleotides that act as lipo-
philic nucleotide precursors in antiviral chemotherapy. Pro-
nucleotides are membrane-permeable nucleotide precursors
that allow the intracellular delivery of nucleoside monophos-
phates. The 50-monophosphates are converted into the 50-di-
and 50-triphosphates, which are the ultimate biological active
compounds. Several pronucleotide strategies have been re-
ported inorder toimprovethe biologicalactivityof nucleoside
analogues,3 e.g., the phosphoramidates,4,5 the cycloSal-phos-
phate triesters,6 the mixed-S-acyl-2-thioethyl compounds,7
the HepDirect technique,8 and recently the bis(acyloxybenzyl)
approach for the intracellular delivery of nucleoside diphos-
phates.9 All these pronucleotide approaches use different
means of activation, e.g., by enzymatic reactions or chemical
cleavage. However, the first four compounds are derivatives
of phosphoric acid that contain four different residues at-
tached to the phosphorus center. As a consequence, the phos-
phorus atom is a chiral center that leads to diastereomers of
the pronucleotides due to the additional chiral centers in the
§
€
Dedicated to Professor Jurgen Heck on the occasion of his 60th
birthday.
*To whom correspondence should be addressed. Phone:
þ49-40-42838-4324. Fax: þ49-40-42838-5592. E-mail: chris.meier@
chemie.uni-hamburg.de.
r
2010 American Chemical Society
Published on Web 10/14/2010
pubs.acs.org/jmc

7676 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 21
Roman et al.
Scheme 1. General Synthesis Strategy toward Diastereomerically Pure Phosphoramidate Prodrugs (RP)-8a-d and (SP)-8a-d
Results and Discussion
Chemistry. The diastereoselective synthesis is based on the
chiral auxiliary (S)-4-isopropylthiazolidine-2-thione 1. The use
of the thiazolidine-2-thione 1 was the result of an in-depth
developing process that will be summarized in a further pub-
lication. Besides the stereochemical induction, the auxiliary has
to fulfill two further important criteria: (a) it should lead to
chromatographically separable diastereomers at the phosphor-
diamidate level; (b) it has to be suitable in the selective replace-
ment by the nucleoside (or an analogue) in the last step of the
reaction sequence under mild reaction conditions. Scheme 1
outlines the concept of the synthesis route that was studied
using four phenol derivatives 3a-d and the methyl ester of
L-alanine 5.
Figure 1. Synthesis of aryl-[(S)-4-isopropylthiazolidine-2-thione]-
phosphorochloridate derivatives 4a-d.
Table 1. 31P NMR Chemical Shifts and Diastereomeric Excess of 4a-d
method 1
method 2
δ (31P)
conversion
of 2 (%)b
%
deb
conversion
of 2 (%)b
%
deb
compd
(ppm) a
This approach requires (S)-4-isopropylthiazolidine-2-
thione 1 as a chiral auxiliary, which was synthesized accord-
ing to the procedure of Delaunay et al.14 The chiral auxiliary
1 was reacted in CH2Cl2 at 0 °C with 3 equiv of phosphoryl
chloride and 1.1 equiv of NEt3 to give the phosphorodichlo-
ridate 2. 31P NMR spectroscopy of the crude product showed
only one phosphorus containing compound (δ= 3.70 ppm).
The crude product of phosphorodichloridate 2was pure enough
to be used in the following reaction without further purification.
Next, phosphorodichloridate 2 was reacted with phenol deriva-
tives 3a-c and 1-naphthol 3d to give the chiral phosphoro-
chloridates 4a-d(Figure 1). The diastereomeric ratio of this key
step was dependent on the phenol derivatives 3a-d and on the
reaction conditions (Table 1).
Phosphorochloridates 4a-d were synthesized by two differ-
ent methods and were isolated as an up to 17:1 mixture of two
diastereomers. According to method 1, acetone and DBUa
at -91 °C for 25-45 min were used. In method 2 acetone and
NEt3 at -91 °C for 2 h followed by 1 h at room temperature
were used in order to complete the reaction. Both methods
have an influence on the diastereomeric excess (de) in the
cases in which 4-methylphenol 3a and 4-methoxyphenol 3b
were used. Method 2 gave better diastereomeric excess of
phosphorochloridates 4a,b than method 1. In contrast, the
use of 2-methylphenol 3c and 1-naphthol 3d led to signifi-
cantly lower diastereoselectivity. Most striking, in the case of
4a
4b
4c
4d
-1.58, -1.41
-1.03, -0.88
-2.16, -1.66
-1.39, -1.28
93
83
64
79
81
72
51
36
77
83
66
90
88
89
28
3
a 31P NMR chemical shifts of major diastereomers are in italic font.
b Conversion and de determined by 31P NMR of the crude mixture.
1-naphthol 3d the diastereoselectivity was completely lost
when method 2 was used. The conversion of 2 varied between a
moderate 64% to an excellent 93% independent of the meth-
od. Compounds 4 were found to be too unstable to be purified
or separated into the diastereomers by chromatography.
The synthesis of phosphordiamidates (RP)- and (SP)-6a-d
was carried out with L-alanine methyl ester hydrochloride 5
and the diastereomerically enriched mixture of phosphoro-
chloridates 4a-dobtained according to method 1. This L-amino
acid was used because previously L-alanine had been identified
as being the most effective amino acid to be used in the phos-
phoramidate approach.15
The synthesis required 1 equiv of phosphorochloridate
4a-d, 1 equiv of L-alanine methyl ester hydrochloride 5, and
3 equiv of NEt3 in CH2Cl2. After the addition of NEt3 at 0 °C
the reaction mixture was stirred for 16 h at room temperature
(Figure 2).
Table 2 summarizes the conversion of starting materials 4
and the diastereomeric excess of the phosphordiamidates
6a-d. Again, the results demonstrate that the aromatic resi-
due attached to the phosphorus atom in the first reaction
a Abbreviations: DBU, 1,8-diazabicyclo[5.4.0]undec-7-ene.

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 21 7677
played an important role with regard to the observed diaste-
reoselectivity. In the case of phosphordiamidates 6a,b the
determined diastereomeric excess was identical to that of the
precursors 4a,b. However, 2-methoxyphenol and 1-naphthol
phosphordiamidate derivatives 6c,d were obtained in even
lower diastereoselectivity than phosphorochloridate 4c,d.
Nevertheless, at this stage the phosphordiamidates 6a-d were
found to be sufficiently stable to be purified and separated in
the stereochemically pure diastereomers by column chroma-
tography. The diastereomeric excess of 6a-d was determined
by 31P NMR and found to be g95%. The diastereomers were
first assigned as the “major” and the “minor” diastereomer,
which differ in the configuration at the phosphorus atom.
Interestingly, except compound 6c the “minor” diastereomers
6a,b,d were crystalline compounds, whereas all “major” dia-
stereomers were obtained as oils. X-ray crystal structure
analyses of the minor diastereomers 6a,b,d proved the SP-
configurationat the phosphorus atom(Figure 3).16 Therefore,
the “major” diastereomer must have the RP-configuration.
Phosphordiamidates (RP)-6a-d or (SP)-6a-d were reacted
separately with 30-deoxy-20,30-didehydrothymidine (d4T,
stavudin, Zerit) 7 to give phosphoramidate prodrugs (SP)-8a-d
and (RP)-8a-d, respectively (Figure 4). The reactions were
carried out in a 1:1 mixture of THF/CH3CN, with 1 equiv of
phosphordiamidate derivatives 6a-d, 1.5 equiv of d4T 7, and
3 equiv of tert-butylmagnesium chloride. The Grignard re-
agent was used in analogy to Uchiyama’s method.17
Table 3 shows the isolated yields and the diastereomeric
purity of phosphoramidate prodrugs (SP)-8a-d. Taking the
generally accepted addition-elimination mechanism for
Figure 2. Synthesis of aryl-N-[(S)-alaninyl]-[(S)-4-isopropylthia-
zolidine-2-thione]phosphordiamidate derivatives 6a-d.
Table 2. 31P NMR Chemical Shifts and Diastereomeric Excess of 6a-d
compd
δ (31P) (ppm)a
conversion of 4 (%)b
% deb,c
6a
6b
6c
6d
-0.86, 1.43
-0.54, 1.82
-0.96, 0.66
-0.32, 1.21
78
60
54
62
81
72
30
28
a 31P NMR chemical shifts of the major diastereomers are in italic
font. b Conversion and de determined by 31P NMR of the crude mixture.
c The starting materials are the phosphorochloridate derivatives 4a-d
obtained with method 1.
Figure 4. Synthesis of phosphoramidate prodrugs (SP)- and (RP)-
8a-d.
Figure 3. ORTEP plot of phosphordiamidate derivatives (SP)-6a,b,d.

7678 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 21
Roman et al.
Table 3. Yields and Diastereomeric Excess of Phosphoramidate Prodrugs (SP)- and (RP)-8a-d
compd
δ (31P) (ppm)
yielda (%)
% deb,c
compd
δ (31P) (ppm)
yielda (%)
% deb,d
(SP)-8a
(SP)-8b
(SP)-8c
(SP)-8d
3.25
3.59
3.01
3.27
13
14
7
g95
g95
94
(RP)-8a
(RP)-8b
(RP)-8c
(RP)-8d
2.59
2.96
2.65
2.87
22
38
30
29
g95
g95
g95
g95
16
g95
a Isolated yields. b de determined by 31P NMR after purification. c The starting materials are the phosphordiamidate derivatives (RP)-6a-d with
g95% diastereomeric excess. d The starting materials are the phosphordiamidate derivatives (SP)-6a-d with g95% diastereomeric excess.
Table 4. Antiviral Activity of Phosphoramidates 8a-d
EC50 (μM)a
CEM/0c
CEM/TK-d
HIV-2 (ROD)
C3H/3T3
MSV
CC50 (μM)b
CEM/0
compd
HIV-1 (IIIB)
HIV-2 (ROD)
(SP)-8a
(RP)-8a
(SP)-8b
(RP)-8b
(SP)-8c
(RP)-8c
(SP)-8d
(RP)-8d
d4T 7
0.12 ( 0.049
0.82 ( 0.0
0.15 ( 0.014
1.1 ( 0.34
1.3 ( 0.21
1.3 ( 0.14
0.15 ( 0.0071
0.12 ( 0.049
0.84 ( 0.078
0.14 ( 0.014
0.46 ( 0.30
0.14 ( 0.085
0.86 ( 0.049
1.1 ( 0.44
0.032 ( 0.022
2.1 ( 1.7
0.048 ( 0.0085
1.9 ( 2.1
1.9 ( 1.2
2.9 ( 2.3
0.13 ( 0.0
0.25 ( 0.12
132 ( 4.9
1.8 ( 0.6
1.8 ( 1.2
2.0 ( 0.0
5.3 ( 5.7
9.8 ( 7.0
4.0 ( 2.7
0.53 ( 0.13
0.72 ( 0.11
2.5 ( 1.4
79 ( 6.4
>250
77 ( 7.8
235 ( 21
>250
1.3 ( 0.14
>250
0.16 ( 0.0071
0.45 ( 0.52
0.75 ( 0.49
70 ( 1.4
107 ( 5.7
>250
a Antiviral activity in T-lymphocytes: 50% effective concentration. b Cytostatic activity: 50% cytostatic concentration. c Wild-type CEM/0 cells.
d Thymidine kinase-deficient CEM TK- cells.
substitutions at the phosphorus atom into consideration, the
stereochemistry in (RP)-6 should be inverted to give (SP)-8.
This conclusion is supported by studies with phosphorami-
dates of BVdU.11 The diastereomeric excess was determined
by 31P NMR after column chromatography and found to be
equal to or even greater than 95%. In the case of phosphor-
amidate prodrug (SP)-8c the diastereomeric excess was 94%.
Therefore, the values correspond to g95% de of the starting
material (RP)-6a-d. The diastereomeric excess was not deter-
mined by 31P NMR of the crude mixture because of the low
yields of (SP)-8a-d. For comparison the diastereomeric mix-
ture of 8a was synthesized, according to the published proce-
dure, in moderate yield of 37% in a 1:1.1 ratio.4 This diaste-
reomeric mixture of phosphoramidate 8a appears as only one
spot on the TLC plate and was found to be inseparable by
column chromatography.
Table 3 also summarizes the isolated yields and the dia-
stereomeric purity of phosphoramidate prodrugs (RP)-8a-d.
The diastereomeric excess was again determined by 31P NMR
after column chromatography and corresponded to g95% de
of the starting material (SP)-6a-d. Compared to the reaction
using (RP)-6, it is obvious that the reaction of (SP)-6 led to
twice as high chemical yields for unknown reasons.
Although the reactions gave convincing stereoselectivities,
the synthesis of phosphoramidate prodrugs 8a-d is still
associated with moderate chemical yields. One reason for
the moderate yields is that the starting materials 6a-d and 7
are not completely converted. Even after 5 days the starting
materials still remained in considerable amounts in the reac-
tion mixture besides the formed product. It seems that the
reaction went to an equilibrium.
Therefore, this fact clearly indicates that the chemical yield
of the reaction can be improved. In order to investigate the
influence of the base on the reaction, the Grignard reagent
tert-butylmagnesium chloride was replaced by sodium hy-
dride or DBU. These experiments were done with phosphor-
diamidate derivative (RP)-6a to get phosphoramidate pro-
drug 8a (Figure 4, R = 4-Me-Ph). However, the replacement
of the Grignard reagent by NaH or DBU led to significant
loss of stereoselectivity at the phosphorus atom. Although
the starting material (RP)-6a had a de of g95%, the phos-
phoramidate prodrug 8a was isolated as a 1:0.6 mixture of
the diastereomers without improving the chemical yield (7%
and 20%, respectively). Most striking, the complexation of
sodium hydride with 15-crown-5 led to a complete racemiza-
tion at the phosphorus atom (28% chemical yield).
Antiviral Evaluation. (RP)- and (SP)-phosphoramidates
8a-d were evaluated for their anti-HIV activity in vitro. All
compounds showed activity against HIV-1 and HIV-2 in
wild-type CEM/0 cell cultures within the same concentration
range as the parental d4T (Table 4). It is also interesting to
notice that the SP and RP diastereomers fully keep their anti-
viral potential in thymidine kinase-deficient CEM cell cul-
tures whereas the parental d4T loses its activity by more
than 100-fold. This observation points to the efficiency of the
prodrugs to intracellularly release d4TMP. Thereby, the need
of the presence of TK for drug activation is circumvented.
The anti-HIV activity data of diastereomers 8a,b proved
the importance of the diastereoselective synthesis. Diaste-
reomers (SP)-8a,b were found to be more active than (RP)-8a,b
(65-fold in the case of 8a and 40-fold for 8b against HIV-2 in
CEM/TK-deficient cells). There seems to be a tendency of
higher activity in HIV-2 (ROD) infected wild-type CEM cells
(∼3-fold) of (SP)-8a and (SP)-8b but a lower activity of (RP)-
8a and (RP)-8b against HIV-2 in CEM TK- than in CEM/
0 cell cultures. The reason for this observation is unclear. Since
CEM TK- represents a clone derived from CEM/0, it cannot
be excluded that uptake or processing of certain diastereo-
mers is more or less efficient than in the wild-type cells than
the other diastereomers. However, it remains to be proven
whether the observed differences are significant or not. Earlier
studies in our laboratories showed that the more active deriv-
ative of the separated diastereomers of 3-methyl-cycloSal-
d4TMP (EC50=0.06 μM) was endowed with an antiviral
potency of the same order of magnitude as compounds (SP)-8a
and (SP)-8b.18 Interestingly, the 3-methyl-cycloSal compound

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 21 7679
with the lower antiviral activity lost only 5- to 10-fold of the
antiviral activity compared to the more active diastereomer.
Surprisingly, the diastereomers 8c,d showed almost identical
antiviral values. Since compounds 8 vary only in the attached
phenol, there should be a difference in the activation of these
compounds. Two possible options can play a role: (i) the ortho-
methyl phenyl or the naphthyl substituent interferes differently
with the activating enzymes (e.g., carboxyesterases) and/or the
particular phenyl/naphthyl substituents induce an altered com-
pound conformation that significantly influences the cellular
uptake of the compounds. Differences in the activation pro-
cess of different phosphoramidates of d4T19 and other nucleo-
side analogues20,21 have already been reported, and thus, this
may support the first option. Currently, incubation experi-
ments in CEM-cell extracts are performed to address this
possibility. The diastereomeric preference of SP versus RP for
8a and 8b also seems cell-type-specific, since this effect was not
observed for their inhibitory activity against Moloney sarcoma
virus (MSV) induced transformation of murine C3H/3T3
cells (Table 4).
HPLC system consisting of a VWR-Hitachi L-2130 pump, auto-
sampler, and a VWR-Hitachi UV detector L-2455. The column
used was a Nucleodur C18 Isis, 5 μm (Macherey-Nagel). Elution
was performed using a water/acetonitrile (Sigma-Aldrich, HPLC
grade) eluent: 0-80% CH3CN (0-60 min), 80-0% CH3CN
(60-65 min), 0% CH3CN (65-70 min), flow rate 0.5 mL/min.
UV detection was at 265 nm. The purity of phosphoramidate
prodrugs 8 was checked by means of HPLC and was in all cases
g95%.
Preparation of [(S)-4-Isopropylthiazolidine-2-thione]phosphorodi-
chloridate 2. A solution of (S)-4-isopropylthiazolidine-2-thione 1
(0.80 g, 4.96 mmol) and phosphoryl chloride (1.39 mL, 14.8 mmol)
in CH2Cl2 (12 mL) was cooled to 0 °C. NEt3 (0.76 mL, 5.45 mmol)
in CH2Cl2 (12 mL) was added dropwise. Following the addition,
the reaction mixture was allowed towarmtoroomtemperatureand
stirred for 16 h. The solvent was removed under reduced pressure,
and the residue was suspended in CH2Cl2 (1mL) andEt2O(15mL).
The precipitated triethylammonium chloride was filtered under
nitrogen and washed with Et2O (5 mL). The solvent of the filtrate
was removed under reduced pressure to give the crude product 2 as
an oil. Further purification was not possible because of high
reactivity, but the 31P NMR spectrum showed only one signal at
3.70 ppm.
General Procedure A: Preparation of Aryl-[(S)-4-isopropyl-
thiazolidine-2-thione]phosphorochloridates 4. Method 1. A solu-
tion of [(S)-4-isopropylthiazolidine-2-thione]phosphorodichlo-
ridate (2, 1.0 equiv) and the phenol or 1-naphthol derivative (3,
1.0 equiv) in acetone was cooled to -91 °C. DBU (1.0 equiv) was
added dropwise. After 25-45 min (dependent on 3) at -91 °C
the reaction was quenched with saturated ammonium chloride
solution and extracted with CH2Cl2 three times. The combined
organic layers were dried over magnesium sulfate and concen-
trated under reduced pressure. The resulting residue was puri-
fied by column chromatography on silica gel (petroleum ether
50-70/EtOAc 2:1).
Method 2. A solution of [(S)-4-isopropylthiazolidine-2-thione]-
phosphorodichloridate (2, 1.0 equiv) and the respective phenol
derivative (3, 1.0 equiv) in acetone was cooled to -91 °C. NEt3
(1.0 equiv) was added dropwise. The reaction mixture was stirred
2 h at -91 °C, followed by 1 h at room temperature. The reaction
was quenched with saturated ammonium chloride solution and
extracted with CH2Cl2 three times. The combined organic layers
were dried over magnesium sulfate and concentrated under re-
duced pressure. The resulting residue was purified by column chro-
matography on silica gel (petroleum ether 50-70/EtOAc 2:1).
General Procedure B: Preparation of Aryl-N-[(S)-alaninyl]-
[(S)-4-isopropylthiazolidine-2-thione]phosphordiamidates 6. A sus-
pension of the aryl-[(S)-4-isopropylthiazolidine-2-thione]phos-
phorochloridate derivative (4, 1.0 equiv) and L-alanine methyl
ester hydrochloride 5 (1.0 equiv) in CH2Cl2 was cooled to 0 °C.
NEt3 (3.0 equiv) was added dropwise. Following the addition,
the reaction mixture was allowed to warm to room temperature
and stirred for 16 h. The reaction was quenched with saturated
ammonium chloride solution and extracted with CH2Cl2 three
times. The combined organic layers were dried over magnesium
sulfate and concentrated under reduced pressure. The resulting
residue was purified by column chromatography on silica gel
(petroleum ether 50-70/EtOAc 2:1).
Conclusion
A diastereoselective synthesis of phosphoramidate prodrugs
8a-d of d4TMP is described. The synthetic route uses (S)-
4-isopropylthiazolidine-2-thione 1 as chiral auxiliary, which is
converted in three steps into the key intermediates 6a-d. These
compounds were obtained in up to 81% de. By column chro-
matography the diastereomeric purity was increased to g95%.
X-ray analysis of three different intermediates 6 proved that
the RP diastereomer was obtained preferentially. The phos-
phordiamidate derivatives (RP)-6a-d and (SP)-6a-d were
reacted separately with d4T 7 to give phosphoramidate pro-
drugs (SP)-8a-d and (RP)-8a-d as almost diastereomerically
pure compounds (g95% de). The antiviral tests of diastereo-
mers 8a,b showed significantly different antiviral potencies in
CEM/TK-deficient cells and thus verify the importance of the
phosphorus configuration on the eventual antiviral activity.
The obtained results proved that the diastereoselective synthe-
sis of phosphoramidate prodrugs 8a-d has been achieved
leading to almost diastereomerically pure compounds although
the chemical yields of 8a-d require further improvement.
Further optimization of this step is now in progress.
Experimental Section
All experiments were conducted under scrupulously dry condi-
tions and under nitrogen atmosphere. Solvents CH2Cl2, CH3CN,
and NEt3 were distilled from CaH2 and stored over molecular
sieves. Et2O and THF were distilled from sodium or potassium
benzophenone and stored over molecular sieves. Acetone was
distilled from phosphorus pentoxide and stored over molecular
sieves. DBU was distilled under nitrogen and stored over molecular
sieves. Petroleum ether 50-70, EtOAc, CH2Cl2, and CH3OH
employed in chromatography were distilled before use. For col-
umn chromatography, silica gel 60, 230-400 mesh, was used. Thin
layer chromatography was performed on precoated aluminum
plates 60 F254 with 0.2 mm layer of silica gel containing a fluo-
rescence indicator. NMR spectra were recorded on a 400 or 500
General Procedure C: Preparation of 50-O-(30-Deoxy-20,
30-didehydrothymidinyl)-O-(aryl)-N-[(S)-methoxyalaninyl]phosphor-
amidates 8. A solution of d4T 7 (1.5 equiv) in THF/CH3CN (1:1)
was cooled to 0 °C. tert-Butylmagnesium chloride (3.0 equiv,
1.7 M solution in THF) was added dropwise. Following the addi-
tion, the reaction mixture was allowed to warm to room tempera-
ture and stirred for 30 min. A solution of aryl-N-[(S)-alaninyl]-[(S)-
4-isopropylthiazolidine-2-thione]phosphordiamidate 6 (1 equiv)
in THF/CH3CN (1:1) was added to the nucleoside suspension at
0 °C. Following the addition, the reaction mixture was allowed to
warm to room temperature and stirred for 5 days. The reaction
was quenched with saturated ammonium chloride solution and
MHz spectrometer (AMX 400, 400, or DRX 500). All 1H and 13
C
NMR chemical shifts are quoted in ppm and were calibrated on
solvent signals. 31P NMR chemical shifts are quoted in ppm using
H3PO4 as external reference. High resolution mass spectra were
obtained with a VG Analytical VG/70-250F spectrometer (FAB,
matrix was m-nitrobenzyl alcohol). HR-ESI spectra were obtained
with an Agilent Technologies ESI-TOF 6224 spectrometer. Ana-
lytical HPLC was carried out on a VWR-Hitachi LaChromElite

7680 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 21
Roman et al.
extracted with CH2Cl2 three times. The combined organic layers
were dried over magnesium sulfate and concentrated under
reduced pressure. The resulting residue was purified by column
chromatography on silica gel (CH2Cl2/CH3OH 39:1). The pro-
duct was freeze-dried.
dure C was used with 2-methylphenyl-N-[(S)-alaninyl]-[(S)-4-iso-
propylthiazolidine-2-thione]phosphordiamidate ((SP)-6c, 0.15 g,
0.36 mmol), d4T 7 (0.12 g, 0.54 mmol), tert-butylmagnesium chlo-
ride (0.64 mL, 1.08 mmol), and 4.8 mL of THF/CH3CN. The pro-
duct (RP)-8c (0.05 g, 30%) was obtained as a colorless foam. 1H
NMR [ppm] (400 MHz, CDCl3): δ = 8.52 (brs, 1H), 7.29-7.26
(m, 1H), 7.22 (d, 1H), 7.21-7.16 (m, 1H), 7.16-7.10 (m, 1H),
7.09-7.03 (m, 1H), 7.02-6.98 (m, 1H), 6.29-6.25 (m, 1H), 5.93-
5.87 (m, 1H), 5.02-4.95 (m, 1H), 4.25 (dd, 2H), 4.07-3.95 (m,
1H), 3.71 (s, 3H), 3.70-3.69 (m, 1H), 2.28 (s, 3H), 1.86 (d, 3H),
1.38 (d, 3H). 31P NMR (162 MHz, CDCl3): δ = 2.65.
50-O-(30-Deoxy-20,30-didehydrothymidinyl)-O-(4-methylphenyl)-
N-[(S)-methoxyalaninyl]phosphoramidate (SP)-8a. General proce-
dure C was used with 4-methylphenyl-N-[(S)-alaninyl]-[(S)-4-iso-
propylthiazolidine-2-thione]phosphordiamidate ((RP)-6a, 0.17 g,
0.4 mmol), d4T 7 (0.13 g, 0.6 mmol), tert-butylmagnesium chlo-
ride (0.70 mL, 1.2 mmol), and 5.3 mL of THF/CH3CN. The pro-
duct (SP)-8a (0.02 g, 13%) was obtained as a colorless foam. 1H
NMR [ppm] (400 MHz, CDCl3): δ=8.52 (brs, 1H), 7.32 (brs,
1H), 7.09 (d, 2H), 7.06-7.01 (m, 3H), 6.37-6.33 (m, 1H), 5.87
(d, 1H), 5.03 (brs, 1H), 4.41-4.26 (m, 2H), 4.02-3.91 (m, 1H),
50-O-(30-Deoxy-20,30-didehydrothymidinyl)-O-(1-naphthyl)-N-[(S)-
methoxyalaninyl]phosphoramidate (SP)-8d. General procedure C
was used with 1-naphthyl-N-[(S)-alaninyl]-[(S)-4-isopropylthiazo-
lidine-2-thione]phosphordiamidate ((RP)-6d, 0.14 g, 0.31 mmol),
d4T 7 (0.10 g, 0.46 mmol), tert-butylmagnesium chloride (0.54 mL,
0.92 mmol), and 4.1 mL of THF/CH3CN. The product (SP)-8d
(0.02 g, 19%) was obtained as a colorless foam. 1H NMR [ppm]
(400 MHz, CDCl3): δ = 8.06-8.00 (m, 2H), 7.87-7.83 (m, 1H),
7.69-7.65 (m, 1H), 7.55-7.49 (m, 2H), 7.48-7.44 (m, 1H), 7.39 (t,
1H), 7.29-7.27 (m, 1H), 7.04-7.01 (m, 1H), 6.38-6.34 (m, 1H),
5.90-5.87 (m, 1H), 5.09-5.05 (m, 1H), 4.48-4.34 (m, 2H), 4.09-
3.98 (m, 1H), 3.63 (s, 3H), 3.58 (t, 1H), 1.62 (d, 3H), 1.26 (d, 3H).
31P NMR (162 MHz, CDCl3): δ = 3.27.
3.70 (s, 3H), 3.63 (t, 1H), 2.30 (s,3H), 1.81 (s,3H), 1.31 (d, 3H). 31
P
NMR (162 MHz, CDCl3): δ = 3.25.
50-O-(30-Deoxy-20,30-didehydrothymidinyl)-O-(4-methylphenyl)-
N-[(S)-methoxyalaninyl]phosphoramidate (RP)-8a. General proce-
dure C was used with 4-methylphenyl-N-[(S)-alaninyl]-[(S)-4-iso-
propylthiazolidine-2-thione]phosphordiamidate ((SP)-6a, 0.20 g,
0.4 mmol), d4T 7 (0.16 g, 0.7 mmol), tert-butylmagnesium chlo-
ride (0.87 mL, 1.4 mmol), and 6.5 mL of THF/CH3CN. The pro-
duct (RP)-8a (0.05 g, 22%) was obtained as a colorless foam. 1H
NMR [ppm] (400 MHz, CDCl3): δ = 8.16 (brs, 1H), 7.26-7.25
(m, 1H), 7.04-7.02 (m, 4H), 7.02-6.98 (m, 1H), 6.30-6.26 (m,
1H), 5.92-5.88 (m, 1H), 5.02-4.97 (m,1H), 4.29-4.24 (m, 2H),
4.05-3.92 (m, 1H), 3.71 (s, 3H), 3.58 (t, 1H), 2.31 (s, 3H), 1.86 (d,
3H), 1.36 (d, 3H). 31P NMR (162 MHz, CDCl3): δ = 2.59.
50-O-(30-Deoxy-20,30-didehydrothymidinyl)-O-(4-methoxyphenyl)-
N-[(S)-methoxyalaninyl]phosphoramidate (SP)-8b. General proce-
dure C was used with 4-methoxyphenyl-N-[(S)-alaninyl]-[(S)-4-iso-
propylthiazolidine-2-thione]phosphordiamidate ((RP)-6b, 0.10 g,
0.23 mmol), d4T 7 (0.08 g, 0.35 mmol), tert-butylmagnesium chlo-
ride (0.41 mL, 0.69 mmol), and 3.1 mL of THF/CH3CN. The pro-
50-O-(30-Deoxy-20,30-didehydrothymidinyl)-O-(1-naphthyl)-
N-[(S)-methoxyalaninyl]phosphoramidate (RP)-8d. General pro-
cedure C was used with 1-naphthyl-N-[(S)-alaninyl]-[(S)-4-iso-
propylthiazolidine-2-thione]phosphordiamidate ((SP)-6d, 0.19 g,
0.43 mmol), d4T 7 (0.14 g, 0.64 mmol), tert-butylmagnesium
chloride (0.75 mL, 1.28 mmol), and 5.7 mL of THF/CH3CN. The
product (RP)-8d (0.06 g, 29%) was obtained as a colorless foam.
1H NMR [ppm] (400 MHz, CDCl3): δ = 8.75 (brs, 1H), 8.15-
7.99 (m, 1H), 7.92-7.77 (m, 1H), 7.72-7.60 (m, 1H), 7.59-7.45
(m, 3H), 7.44-7.33 (m, 1H), 7.31-7.19 (m, 1H), 7.07-6.93 (m,
1H), 6.34-6.20 (m, 1H), 5.97-5.83 (m, 1H), 5.09-4.93 (m, 1H),
4.43-4.22 (m, 2H), 4.17-3.97 (m, 1H), 3.96-3.79 (m, 1H), 3.65
(s, 3H), 1.79 (s, 3H), 1.40-1.28 (m, 3H). 31P NMR (162 MHz,
CDCl3): δ = 2.87.
Antiretroviral Evaluation. Human immunodeficiency virus type
1 (HIV-1) was originally obtained from a persistently HIV-infected
H9 cell line, as described previously, and was kindly provided by Dr.
R. C. Gallo (then at the National Institutes of Health, Bethesda,
MD). Virus stocks were prepared from the supernatants of HIV-
1-infected MT-4 cells. HIV-2 (strain ROD) was kindly provided by
Dr. L. Montagnier (then at the Pasteur Institute, Paris, France), and
virus stocks were prepared from the supernatants of HIV-2-infected
MT-4 cells. CEM cells were obtained from the American Type
CultureCollection (Rockville, MD). CEMcells wereinfected with
HIV as previously described.15,18 Briefly, 4 ꢀ 105 CEM cells/mL
were infected with HIV-1(IIIB) or HIV-2(ROD) at ∼100 CCID50
(50% cell culture infective dose) per mL of cell suspension. The
thymidine kinase-deficient CEM cell cultures were also infected
with HIV-2(ROD). Then 100 μL of the infected cell suspensions
was transferred into 96-well microtiter plate wells and mixed with
100 μL of the appropriate dilutions of the test compounds. After
4-5 days, giant cell formation was recorded microscopically in the
HIV-infected cell cultures. The 50% effective concentration is
defined as the compound concentration required to inhibit virus-
induced cytopathicity by 50%. The 50% cytostatic concentration
is defined as the compound concentration required to inhibit
CEM cell proliferation by 50%, as derived by counting the cell
numbers in the presence of different compound concentrations by
use of a Coulter particle counter ZI (Analysis, Gent, Belgium).
For Moloney sarcoma virus (MSV) assays, C3H/3T3 cells were
seeded at 20 000 cells/mL into wells of tissue culture cluster plates
(48 wells/plate). Following a 24 h incubation period, cell cultures
were infected with 80 focus-forming unitsof MSV during 120 min,
whereafter the culture medium was replaced by 1 mL of fresh
medium containing appropriate concentrations of the test com-
pound. After 6 days, transformation of the cells was examined
1
duct (SP)-8b (0.02 g, 14%) was obtained as a colorless foam. H
NMR [ppm] (400 MHz, CDCl3): δ = 8.68 (s, 1H), 7.32 (d, 1H),
7.10-7.05 (m, 2H), 7.04-7.01 (m, 1H), 6.84-6.79 (m, 2H), 6.37-
6.33 (m, 1H), 5.90-5.86 (m, 1H), 5.05-5.00 (m, 1H), 4.40-4.26 (m,
2H), 4.02-3.91 (m, 1H), 3.77 (s, 3H), 3.70 (s, 3H), 3.64 (t, 1H), 1.82
(d, 3H), 1.30 (d, 3H). 31P NMR (162 MHz, CDCl3): δ = 3.59.
50-O-(30-Deoxy-20,30-didehydrothymidinyl)-O-(4-methoxyphenyl)-
N-[(S)-methoxyalaninyl]phosphoramidate (RP)-8b. General proce-
dure C was used with 4-methoxyphenyl-N-[(S)-alaninyl]-[(S)-4-iso-
propylthiazolidine-2-thione]phosphordiamidate ((SP)-6b, 0.11 g,
0.26 mmol), d4T 7 (0.08 g, 0.38 mmol), tert-butylmagnesium chlo-
ride (0.45 mL, 0.77 mmol), and 3.4 mL of THF/CH3CN. The pro-
1
duct (RP)-8b (0.04 g, 38%) was obtained as a colorless foam. H
NMR [ppm] (400 MHz, CDCl3): δ = 8.48 (brs, 1H), 7.25-7.24 (m,
1H), 7.14-7.08 (m, 2H), 7.02-6.98 (m, 1H), 6.85-6.79 (m, 2H),
6.29-6.26 (m, 1H), 5.92-5.88 (m, 1H), 5.02-4.96 (m, 1H), 4.29-
4.23 (m, 2H), 4.03-3.92 (m, 1H), 3.77 (s, 3H), 3.71 (s, 3H), 3.65
(t, 1H), 1.86 (d, 3H), 1.36 (d, 3H). 31P NMR (162 MHz, CDCl3):
δ = 2.96.
50-O-(30-Deoxy-20,30-didehydrothymidinyl)-O-(2-methylphenyl)-
N-[(S)-methoxyalaninyl]phosphoramidate (SP)-8c. General proce-
dure C was used with 2-methylphenyl-N-[(S)-alaninyl]-[(S)-4-iso-
propylthiazolidine-2-thione]phosphordiamidate ((RP)-6c, 0.13 g,
0.31 mmol), d4T 7 (0.10 g, 0.46 mmol), tert-butylmagnesium chlo-
ride (0.54 mL, 0.92 mmol), and 4.1 mL of THF/CH3CN. The
product (SP)-8c (0.01 g, 7%) was obtained as a colorless foam. 1H
NMR [ppm] (400 MHz, CDCl3): δ = 8.36 (brs, 1H), 7.29-7.28
(m, 1H), 7.24-7.21 (m, 1H), 7.20-7.17 (m, 1H), 7.15-7.10 (m,
1H), 7.08-7.05 (m, 1H), 7.04-7.02 (m, 1H), 6.39-6.36 (m, 1H),
5.93-5.90 (m, 1H), 5.07-5.02 (m, 1H), 4.39-4.28 (m, 2H),
4.05-3.95 (m, 1H), 3.72 (s, 3H), 3.60 (t, 1H), 2.26 (s, 3H), 1.74
(d, 3H), 1.34 (d, 3H). 31P NMR (162 MHz, CDCl3): δ = 3.01, 2.65
(dr = 44.7:1, 95% de).
50-O-(30-Deoxy-20,30-didehydrothymidinyl)-O-(2-methylphenyl)-
N-[(S)-methoxyalaninyl]phosphoramidate (RP)-8c. General proce-

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 21 7681
microscopically. The EC50 was defined as the compound concen-
tration required to inhibit MSV-induced cell transformation
by 50%.
(9) Jessen, H. J.; Schulz, T.; Balzarini, J.; Meier, C. Bioreversible
protection of nucleoside diphosphates. Angew. Chem. 2008, 120,
8847–8850; Angew. Chem., Int. Ed. 2008, 47, 8719-8722 .
(10) Allender, C. J.; Brain, K. R.; Ballatore, C.; Cahard, D.; Siddiqui,
A.; McGuigan, C. Separation of individual antiviral nucleotide
prodrugs from synthetic mixtures using cross-reactivity of a molecu-
larly imprinted stationary phase. Anal. Chim. Acta 2001, 435, 107–
113. Mesplet, N.; Saito, Y.; Morin, P.; Agrofoglio, L. A. Liquid chromato-
graphic separation of phosphoramidate diastereomers on a polysaccha-
ride-type chiral stationary phase. J. Chromatogr., A 2003, 983, 115–124.
(11) Congiatu, C.; Brancale, A.; Mason, M. D.; Jiang, W. G.; McGuigan,
C. Novel potential anticancer naphthyl phosphoramidates of
BVdU: separation of diastereoisomers and assignment of the abso-
lute configuration of the phosphorus center. J. Med. Chem. 2006, 49,
452–455.
(12) Chapman, H.; Kernan, M.; Prisbe, E.; Rohloff, J.; Sparacino, M.;
Terhorst, T.; Yu, R. Practical synthesis, separation, and stereo-
chemical assignment of the PMPA pro-drug GS-7340. Nucleosides,
Nucleotides Nucleic Acids 2001, 20, 621–628. Chapman, H.; Kernan,
M.; Rohloff, J.; Sparacino, M.; Terhorst, T. Purification of PMPA
amidate prodrugs by SMB chromatography and X-ray crystallography
of the diastereomerically pure GS-7340. Nucleosides, Nucleotides
Nucleic Acids 2001, 20, 1085–1090.
(13) Gardelli, C.; Attenni, B.; Donghi, M.; Meppen, M.; Pacini, B.;
Harper, S.; Di Marco, A.; Fiore, F.; Giuliano, C.; Pucci, V.; Laufer,
R.; Gennari, N.; Marcucci, I.; Leone, J. F.; Olsen, D. B.; MacCoss,
M.; Rowley, M.; Narjes, F. Phosphoramidate prodrugs of 20-
C-methylcytidine for therapy of hepatitis C virus infection. J. Med.
Chem. 2009, 52, 5394–5407.
(14) Delaunay, D.; Toupet, L.; Le Corre, M. Reactivity of β-amino
alcohols with carbon disulfide. Study on the synthesis of 2-oxazo-
lidinethiones and 2-thiazolidinethiones. J. Org. Chem. 1995, 60,
6604–6607.
Acknowledgment. We thank Isabelle Nevoigt for solving
the X-ray structures and Leen Ingels and Lizette van Berckelaer
for excellent technical assistance. C.M. thanks the University of
Hamburg, Germany, for support. J.B. received a grant from
Katholieke Universiteit Leuven, Belgium (Grant GOA 10/14).
Supporting Information Available: 1H, 31P, and 13C NMR
spectroscopic data; mass spectrometric data; melting points of
phosphordiamidates (SP)-6a-c; analytical HPLC data of phos-
phoramidate prodrugs 8. This material is available free of
charge via the Internet at http://pubs.acs.org.
References
(1) Nawrot, B.; Rebowska, B.; Michalak, O.; Bulkowski, M.; Blaziak,
D.; Guga, P.; Stec, W. J. 1,3,2-Oxathiaphospholane approach to
the synthesis of P-chiral stereodefined analogs of oligonucleotides
and biologically relevant nucleoside polyphosphates. Pure Appl.
Chem. 2008, 80, 1859–1871.
(2) Oka, N.; Yamamoto, M.; Sato, T.; Wada, T. Stereocontrolled
synthesis of oligonucleoside phosphorothioates and PO/PS-chi-
meric oligonucleotides by using oxazaphospholidine. Nucleic Acids
Symp. Ser. 2008, 52, 335–336. Iwamoto, N.; Oka, N.; Wada, T.
Stereocontrolled synthesis of backbone-modified oligonucleotides via
diastereopure H-phosphonate intermediates. Nucleic Acids Symp. Ser.
2008, 52, 333–334.
(15) McGuigan, C.; Tsang, H.-W.; Cahard, D.; Turner, S.; Velazquez,
S.; Salgado, A.; Bidois, L.; Naesens, L.; De Clercq, E.; Balzarini, J.
Phosphoramidate derivatives of d4T as inhibitors of HIV: The
effect of amino acid variation. Antiviral Res. 1997, 35, 195–204.
(16) The crystal structures (SP)-6a,b,d are allocated to the deposition
numbersCCDC774812-774814 atthe Cambridge Crystallographic
Data Centre.
ꢀ
(3) McGuigan, C.; Meier, C.; Perigaud, C.; Wagner, C. R. Pro-nucleotide
development and delivery of biologically active nucleotide analogues.
Mini-Rev. Med. Chem. 2004, 4, 371-419.
(4) McGuigan, C.; Pathirana, R. N.; Balzarini, J.; De Clercq, E.
Intracellular delivery of bioactive AZT nucleotides by aryl phos-
phate derivatives of AZT. J. Med. Chem. 1993, 36, 1048–1052.
(5) Fries, K. M.; Joswig, C.; Borch, R. F. Synthesis and biological
evaluation of 5-fluoro-20-deoxyuridine phosphoramidate ana-
logues. J. Med. Chem. 1995, 38, 2672–2680. Tobias, S. C.; Borch,
R. F. Synthesis and biological studies of novel nucleoside phosphor-
amidate prodrugs. J. Med. Chem. 2001, 44, 4475–4480.
(17) Uchiyama, M.; Aso, Y.; Noyori, R.; Hayakawa, Y. O-Selective
phosphorylation of nucleosides without N-protection. J. Org.
Chem. 1993, 58, 373–379.
(18) Meier, C.; Lorey, M.; De Clercq, E.; J. Balzarini, J. cycloSal-20,30-
dideoxy-20,30-didehydrothymidinemonophosphate(cycloSal-d4TMP):
synthesis and antiviral evaluation of a new d4TMP delivery system.
J. Med. Chem. 1998, 41, 1417–1427.
(6) Meier, C. cycloSal-phosphates as chemical Trojan horses for
intracellular nucleotide and glycosylmonophosphate delivery;
chemistry meets biology. Eur. J. Org. Chem. 2006, 5, 1081–1102.
(19) Siccardi, D.; Gumbleton, M.; Omidi, Y.; McGuigan, C. Stereospe-
cific chemical and enzymatic stability of phosphoramidate triester
prodrugs of d4T in vitro. Eur. J. Pharm. Sci. 2004, 22, 25–31.
(20) Venkatachalam, T. K.; Samuel, P.; Uckun, F. M. Stereochemical
influence on lipase-mediated hydrolysis and biological activity of
stampidine and other stavudine phosphoramidates. Bioorg. Med.
Chem. 2005, 13, 1764–1773.
(21) Venkatachalam, T. K.; Samuel, P.; Qazi, S.; Uckun, F. M. Effect of
change in nucleoside structure on the activation and antiviral
activity of phosphoramidate derivatives. Bioorg. Med. Chem.
2005, 13, 5408–5423. Mesplet, N.; Saito, Y.; Morin, P.; Agrofoglio,
L. A. HPLC and mass spectrometry analysis of the enzymatic hydro-
lysis of anti-HIV pronucleotide diastereomers. Bioorg. Chem. 2003,
31, 237–247.
ꢀ
(7) Giradet, J. L.; Perigaud, C.; Aubertin, A. M.; Gosselin, G.; Kim,
A.; Imbach, J. L. Increase of the anti-HIV activity of d4T in human
T-cell culture by the use of the SATE pronucleotide approach.
Bioorg. Med. Chem. Lett. 1995, 5, 2981–2984. Peyrottes, S.; Egron,
D.; Lefebvre, I.; Gosselin, G.; Imbach, J. L.; Perigaud, C. SATE pronu-
cleotide approaches: an overview. Mini-Rev. Med. Chem. 2004, 4,
395–408.
(8) Erion, M. D.; Reddy, K. R.; Boyer, S. H.; Matelich, M. C.; Gomez-
Galeno, J.; Lemus, R. H.; Ugarkar, B. G.; Colby, T. J.; Schanzer,
J.; van Poelje, P. D. Design, synthesis, and characterization of a
series of cytochrome P450 3A-activated prodrugs (HepDirect
prodrugs) useful for targeting phosph(on)ate-based drugs to the
liver. J. Am. Chem. Soc. 2004, 126, 5154–5163.