Studies of asymmetric induction in the synthesis of dinucleoside phosphorothioates from 2-oxo-1,3,2-dithiaphospholane nucleoside derivatives

Article abstract of DOI:10.1080/10426500902947898

The nucleoside derivatives, containing at 3'-O or 5'-O-position 2-oxo-1,3,2-dithiaphospholane function were found to react in the presence of DBU with a hydroxyl group of appropriately protected nucleosides to yield, after deprotection, dinucleoside phosphorothioates in moderate yields. An asymmetric induction connected with the formation of new chiral phosphorothioate center has been examined.

Full text of DOI:10.1080/10426500902947898

This article was downloaded by: [Tufts University]
On: 10 October 2014, At: 07:50
Publisher: Taylor & Francis
Informa Ltd Registered in England and Wales Registered Number: 1072954
Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH,
UK
Phosphorus, Sulfur, and Silicon
and the Related Elements
Publication details, including instructions for
authors and subscription information:
http://www.tandfonline.com/loi/gpss20
Studies of Asymmetric
Induction in the
Synthesis of Dinucleoside
Phosphorothioates from 2-
Oxo-1,3,2-dithiaphospholane
Nucleoside Derivatives
Magdalena Olesiak a & Andrzej Okruszek a b
a Department of Bioorganic Chemistry, Centre of
Molecular and Macromolecular Studies , Polish
Academy of Sciences , Łódź, Poland
b Institute of Technical Biochemistry, Faculty
of Biotechnology and Food Sciences , Technical
University of Łódź , Łódź, Poland
Published online: 06 Jun 2009.
To cite this article: Magdalena Olesiak & Andrzej Okruszek (2009) Studies of
Asymmetric Induction in the Synthesis of Dinucleoside Phosphorothioates from 2-
Oxo-1,3,2-dithiaphospholane Nucleoside Derivatives, Phosphorus, Sulfur, and Silicon
and the Related Elements, 184:6, 1548-1560, DOI: 10.1080/10426500902947898
To link to this article: http://dx.doi.org/10.1080/10426500902947898
PLEASE SCROLL DOWN FOR ARTICLE
Taylor & Francis makes every effort to ensure the accuracy of all the
information (the “Content”) contained in the publications on our platform.
However, Taylor & Francis, our agents, and our licensors make no
representations or warranties whatsoever as to the accuracy, completeness,

or suitability for any purpose of the Content. Any opinions and views
expressed in this publication are the opinions and views of the authors, and
are not the views of or endorsed by Taylor & Francis. The accuracy of the
Content should not be relied upon and should be independently verified with
primary sources of information. Taylor and Francis shall not be liable for any
losses, actions, claims, proceedings, demands, costs, expenses, damages,
and other liabilities whatsoever or howsoever caused arising directly or
indirectly in connection with, in relation to or arising out of the use of the
Content.
This article may be used for research, teaching, and private study purposes.
Any substantial or systematic reproduction, redistribution, reselling, loan,
sub-licensing, systematic supply, or distribution in any form to anyone is
expressly forbidden. Terms & Conditions of access and use can be found at
http://www.tandfonline.com/page/terms-and-conditions

Phosphorus, Sulfur, and Silicon, 184:1548–1560, 2009
Copyright © Taylor & Francis Group, LLC
ISSN: 1042-6507 print / 1563-5325 online
DOI: 10.1080/10426500902947898
Studies of Asymmetric Induction in the Synthesis
of Dinucleoside Phosphorothioates from
2-Oxo-1,3,2-dithiaphospholane Nucleoside Derivatives
Magdalena Olesiak1 and Andrzej Okruszek1,2
1Department of Bioorganic Chemistry, Centre of Molecular and
Macromolecular Studies, Polish Academy of Sciences, Lꢀ o´dz´, Poland
2Institute of Technical Biochemistry, Faculty of Biotechnology and Food
Sciences, Technical University of Lꢀ o´dz´, Lꢀ o´dz´, Poland
The nucleoside derivatives, containing at 3ꢀ-O or 5ꢀ-O-position 2-oxo-1,3,2-
dithiaphospholane function were found to react in the presence of DBU with a
hydroxyl group of appropriately protected nucleosides to yield, after deprotection,
dinucleoside phosphorothioates in moderate yields. An asymmetric induction con-
nected with the formation of new chiral phosphorothioate center has been examined.
Keywords Asymmetric induction; DBU assistance; dinucleoside phosphorothioates;
dithiaphospholane ring cleavage
INTRODUCTION
Chemically modified oligonucleotides have recently found wide appli-
cation in biochemistry and molecular biology as indispensible tools for
studying interactions of nucleic acids with other biomolecules1,2 and as
potential therapeutics in antisense/antigene or ribozyme strategies.3–5
Among oligonucleotide analogues most widely used for such stud-
ies are those modified within the internucleotide phosphate group,
and in particular, phosphorothioate oligonucleotides [oligo(nucleoside
phosphorothioate)s] in which one of the nonbridging oxygen atoms is
substituted by sulfur.6 Phosphorothioate oligonucleotides are the most
promising candidates as antisense therapeutics against several viral
and cancer diseases, as indicated by their use in numerous clinical
Received 18 January 2008; accepted 1 March 2008.
Dedicated to Professor Marian Mikolꢀajczyk, CBMiM PAN in Lꢀ o´dz´, Poland, on the
occasion of his 70th birthday.
Address correspondence to A. Okruszek, Institute of Technical Biochemistry, Faculty
of Biotechnology and Food Sciences, Technical University of Lodz, Stefanowskiego 4/10,
90-924 Lodz, Poland. E-mail: andrzej.okruszek@p.lodz.pl
1548

Synthesis of Dinucleoside Phosphorothioates
1549
B 1
B 1
O
O
O
O
x
-
x
S
O
O
O
O
S
P
P
B2
B2
-
O
O
O
O
x
x
O
O
SP
RP
FIGURE 1 Schematic representation of oligo(deoxyribonucleoside phospho-
rothioate)s (X=H) and oligo(ribonucleoside phosphorothioate)s (X=OH) with
opposite configuration (SP or RP) of internucleotide bond.
trials7 and by FDA approval of phosphorothioate 22-mer (Vitravene)
for treatment of CMV retinitis in AIDS patients (August 1998). Un-
fortunately, no other oligonucleotides have been approved as antisense
drugs, in spite of numerous clinical trials.
The first automated synthesis of oligo(deoxyribonucleoside phospho-
rothioate)s was accomplished by Stec et al.8 using a modified phos-
phoramidite approach with a stepwise sulfurization of intermediate
phosphite linkage. An alternative approach was proposed by Froehler,9
who modified the H-phosphonate method by introducing final one-step
sulfurization with a solution of elemental sulfur. The modifications of
the phosphotriester method found only limited application for the syn-
thesis of oligo(deoxyribonucleoside phosphorothioate)s.10 In contrast
to DNA analogues, the synthesis of oligo(ribonucleoside phosphoroth-
ioate)s was described only in a few cases, usually by the H-phosphonate
method.11,12
An intrinsic property of oligo(nucleoside phosphorothioate)s is the
formation of a new center of chirality at each internucleotide phos-
phorus, leading to RP and SP diastereomers (see Figure 1). With an
increasing length of oligonucleotide, the number of diastereomers (m)
grows exponentially (m = 2n for n phosphorothioate bonds).
As an important consequence of aforementioned polydiastere-
omerism of oligo(nucleoside phosphorothioate)s, stereodifferentiated
interactions could occur between particular diastereomers and other
chiral biomolecules such as DNA, RNA, proteins, carbohydrates, or
lipids. This could in turn lead to stereodifferentiated uptake, cellu-
lar trafficking, stability in biological media, pharmacokinetics, and

1550
M. Olesiak and A. Okruszek
toxicity. As a result, particular diastereomers may have different ther-
apeutic properties when applied in antisense, antigene, or ribozyme
strategies.13
The synthesis of oligo(deoxyribonucleoside phosphorothioate)s by
the most commonly used solid support phosphoramidite/sulfurization
approach leads to a more or less random mixture of all possible diastere-
omers. The detailed studies have shown, however, that every individual
coupling step occurs with some stereoselectivity, and the formation of
phosphorothioate center with RP configuration is preferred (usually
52–62% of RP).14
The enzymatic template-directed processes, utilizing DNA or
RNA polymerases of different origin and [SP]-nucleoside 5ꢀ-O-α-
thiotriphosphates as substrates, have been employed for the stereo-
controlled synthesis of several phosphorothioate analogues of DNA15,16
or RNA.17 An important limitation of this methodology is that it can
only be applied for the synthesis of [All-RP] diastereomers of phospho-
rothioate DNA or RNA fragments.
Fully stereocontrolled chemical synthesis of oligo(deoxyri-
bonucleoside phosphorothioate)s was introduced by Stec et al.,
based on the stereospecific reaction of appropriately protected and
previously separated into diastereomers nucleoside-3ꢀ-O-(2-thio-1,3,2-
oxathiaphospholane)s with 5ꢀ -oxygen of nucleosides or growing
oligonucleotide chain.18–21 This methodology, usually referred to
as the oxathiaphospholane approach, allows us to accomplish a
solid-state automated synthesis of phosphorothioate analogues of
oligodeoxyribonucleotides with a predetermined sense of chirality at
each internucleotide phosphorus. The oxathiaphospholane methodol-
ogy found some application for the stereocontrolled synthesis of short
22
oligo(ribonucleoside phosphorothioate)s containing either 3ꢀ,5ꢀ− or
2ꢀ,5ꢀ-internucleotide linkages.23
Several attempts of stereocontrolled synthesis of oligo(nucleoside
phosphorothioate)s were undertaken by other authors, however
with very limited success. By application of the phosphotriester
approach, with separated into diastereomers nucleoside 3ꢀ-O-(S-alkyl-
´
O-p-nitrophenyl)-phosphorothioates as precursors, Lesnikowski and
Jaworska were able to synthesize diastereomerically pure phospho-
rothioate analogues of trithymidine24 and triuridine.25 However,
the method could not be employed for solid-support synthesis. In
alternative studies, Just and coworkers used as substrates nucleoside
3ꢀ-O-cyclic phosphoramidites based on α-D-xylo-furanose26 and L- or
D-tryptophan27 as chiral auxiliaries, whereas Iyer et al. employed nu-
cleoside 3ꢀ-O-bicyclic oxazaphospholidines derived from enantiomers of

Synthesis of Dinucleoside Phosphorothioates
1551
prolinol.28 However, in all cases, only dinucleoside phosphorothioates
could be synthesized in a stereocontrolled way. An improved method
was reported by Wilk et al., who were able to prepare selected
stereoregular oligo(nucleoside phosphorothioate)s of longer nucleotide
sequence by using nucleoside 3ꢀ-O-(N-acyl-1,3,2-oxazaphospholane)s
as starting material.29
RESULTS AND DISCUSSION
All aforementioned methods involved the use of nucleoside deriva-
tives with cyclic P-chiral precursors of phosphorothioate moiety, of-
ten containing additional chiral auxiliary, which had to be separated
into individual P-diastereomers prior to coupling reaction. In this ar-
ticle, we wish to report the results of our studies on the application
of nucleoside derivatives of type 1, containing P-prochiral 2-oxo-1,3,2-
dithiaphospholane function as a precursor of dinucleotide analogues
2, containing internucleotide phosphorothioate linkage (Scheme 1). All
reactions were performed in solution, which is typical to introductory
studies on the dinucleotide stage. Compounds 1 are close analogues
of nucleoside-3ꢀ-O-(2-thio-1,3,2-dithiaphospholane)s, which were devel-
oped by us as substrates for the solid-state synthesis of oligo(nucleoside
phosphorodithioate)s (dithiaphospholane method).30
O
O
S
S
R2O
DBU
S
P
+
R2OH
P
R1O
-
-
R1O
S
1
2
SCHEME 1 The proposed synthesis of phosphorothioate diesters by modified
dithiaphospholane approach. R1 and R2 can be appropriately protected nucle-
oside derivatives.
It was anticipated that in the reaction shown in Scheme 1, with R1O
and R2O being nucleoside residues, a stereoselective formation of in-
ternucleotide phosphorothioate linkage can occur due to the presence
of two nucleoside sugar residues as an “internal” chiral auxiliaries. The
potential of ribose moiety to induce the preferential formation of dinu-
cleoside H-phosphonates of RP-configuration (up to 85% of RP) from P-
prochiral appropriately protected nucleoside-3ꢀ-O-H-phosphonates was
demonstrated by Almer et al.12
The aforementioned hypothesis has been verified using model sub-
strate 3 bearing 2-oxo-1,3,2-dithiaphospholane residue in 5ꢀ-O-position

1552
M. Olesiak and A. Okruszek
of N6-benzoyl-3ꢀ-O-acetyldeoxyadenosine. Compound 3 was synthe-
sized as depicted in Scheme 2 in the reaction of appropriately protected
nucleoside with 2-N, N-diisopropylamino-1,3,2-dithiaphospholane in
the presence of 1H-tetrazole.
O
S
S
S
S
S
S
AdeBz
AdeBz
AdeBz
P
O
O
P
O
O
O
HO
P
NiPr2
t-BuOOH
Benzene
Tetrazole/CH2Cl2
OAc
OAc
OAc
4
3
SCHEME 2 The synthesis of N6-benzoyl-3ꢀ-O-acetyldeoxyadenosine-5ꢀ-O-(2-
oxo-1,3,2-dithia-phospholane) 3.
The PIII intermediate 4 was isolated by column chromatography in
91% yield (31P NMR: δ 150.62 ppm, C6D6) and was further oxidized with
tert-butyl hydroperoxide (5–6 M solution in decane) to give 3, which
was isolated in 87% yield as white solid by precipitation from n-hexane
(31P NMR: δ 80.45 ppm, C6D6). The presence of minor unidentified
contamination was also detected (3–10%; δ 20.9 ppm), which however
could not be removed due to decomposition of 3 during chromatography
on silicagel column. Samples of 3 prepared as above are relatively stable
when stored at –20◦C under argon (31P NMR assay).
The dithiaphospholane 3 was reacted with equimolar amounts
of nucleosides containing free 3ꢀ-OH group in anhydrous ace-
tonitrile solution, in the presence of equimolar amount of 1,8-
diazabicyclo[5.4.0]undec-7-ene (DBU). This compound was previously
selected as optimal activator of alcohols in their reactions with nu-
cleoside oxathia- and dithiaphospholanes.17–22,29,30 The reactions were
performed on 0.1 mmol scale, at room temperature, under 31P NMR
control (Scheme 3).
B
Bz
HO
O
DMTO
O
O
P
Bz
S
S
_
1. CH3COOH
AdeBz
_
O
DMTO
S
O
O
O
O
S
O
O
2. NH4OH
AdeBz
DBU/CH3CN
P
P
+
Ade
O
O
O
O
OH
OAc
3
OH
OAc
SCHEME 3 The synthesis of dinucleoside phosphorothioates from dithiaphos-
pholane 3.

Synthesis of Dinucleoside Phosphorothioates
1553
After 30 min, full disappearance of the signal of substrate 3 was
observed with simultaneous formation of signals corresponding to RP
and SP diastereomers of dinucleoside phosphorothioates (55–57 ppm
range). After evaporation of solvent, the deprotection procedure was
applied. The treatment with 80% aq acetic acid (2 h, r.t.) removed 5ꢀ-
O-dimethoxyl group (DMT), whereas incubation in 30% aq ammonia
(12 h, 55◦C) led to hydrolytic removal of all acyl groups. Dinucleoside
phosphorothioates (a mixture of RP and SP diastereomers) were iden-
tified as major components of the reaction mixture by spectroscopic
(31P NMR chemical shifts) and chromatographic (RP HPLC retention
times) comparison with original samples prepared independently by
phosphoramidite method.14 Thus, the signal of RP isomer appears for
all dinucleoside phosphorothioates at higher frequencies in 31P NMR
(D2O solution), and this isomer shows higher chromatographic mo-
bility under conditions of reversed phase HPLC. The reaction of dithi-
aphospholane 3 with 3ꢀ -OH nucleosides gave desired dinucleoside phos-
phorothioates in ca. 50% yield as determined by integration of NMR
spectra and HPLC profiles, and by measurements of UV absorption
of dinucleotides isolated by preparative HPLC (selected cases). These
measurements also allowed us to determine their RP/SP diastereomeric
ratio for each dinucleotide product. The average values of RP/SP ratios
taken from both NMR and HPLC measurements are listed in Table I.
The analysis of the data collected in Table I reveals that in all ex-
amined cases, a preferential formation of RP isomer was observed for
deprotected dinucleotides; however its excess over SP isomer was not
TABLE I Stereochemical Analysis of Dinucleoside Phosphorothioates
Formed in Reactions of Dithiaphospholane 3 with 3ꢀ-OH Nucleosides
Shown in Scheme 3
RP HPLCd retention
time (min) RP/SP
B
Z
RP/SP (%)a
δ
31P NMR RP/SP (ppm)
Ade
Cyt
Gua
Thy
N6-Bz
N4-Bz
N2-iBu
H
65/35
62/38
55/45
65/35
56.80/56.91b 56.16/55.30c
56.82/57.22b 56.25/55.58c
56.96/57.06b 56.00/55.00c
56.91/57.21b 56.14/55.45c
19.57/21.65
17.44/18.46
17.70/19.12
20.17/21.40
aAverage values for 31P NMR and HPLC measurements.
bCH3CN/CD3CN solution; crude reaction mixture before removal of protecting
groups.
cD2O solution; fully deprotected dinucleotides.
dLinear gradient of acetonitrile in 0,1 M triethylammonium bicarbonate (1%/min);
fully deprotected dinucleoside phosphorothioates.

1554
M. Olesiak and A. Okruszek
extremely prevailing (from 55/45% to 62/38%). The RP/SP ratios were
also measured from integrated 31P NMR spectra of crude reaction mix-
tures before removal of protecting groups (CH3CN/CD3CN solution).
The values agreed well with those obtained for deprotected products,
however in this case the NMR signal resulting from predominant RP
isomer was found at lower resonance frequency (see Table I). The ob-
served “reversal” of relative position of RP/SP isomers in 31P NMR
spectra of protected dinucleoside phosphorothioates is probably caused
by the presence of 5ꢀ-O-dimethoxytrityl group.
In order to increase the asymmetric induction, further ex-
periments were performed with substrates bearing 2-oxo-1,3,2-
dithiaphospholane function in 3ꢀ-O-position of nucleoside. The 5ꢀ-
O-dimethoxytritylthymidine-3ꢀ-O-(2-oxo-1,3,2-dithiaphospholane) (5),
which was chosen as model substrate, was synthesized by phosphityla-
tion of 5ꢀ-O-protected nucleoside as shown in Scheme 4.
Thy
Thy
DMTO
DMTO
O
O
P
S
S
P
NiPr2
Thy
O
DMTO
t-BuOOH
Benzene
O
S
O
O
S
.
.
Tetrazole/CH2Cl2
P
S
S
OH
6
5
SCHEME 4 The synthesis of 5ꢀ-O-dimethoxytritylthymidine-3ꢀ-O-(2-oxo-
1,3,2-dithiaphospholane) 5.
The intermediate PIII dithiaphospholane 6 (isolated quantitatively
by column chromatography, 31P NMR: δ 150.66 ppm, CD3CN) was fur-
ther oxidized with tert-butyl hydroperoxide to give product 5, which
was isolated by precipitation from n-hexane as white solid (31P NMR: δ
79.47 ppm, C6D6; yield 84%).
The reactions of dihiaphospholane 5 with equimolar amounts of nu-
cleosides containing free 5ꢀ-OH group were performed again in anhy-
drous acetonitrile solution in the presence of equimolar amount of DBU,
exactly as described above for reactions of compound 3 (0.1 mmol scale,
30 min at room temperature, 31P NMR control, see Scheme 5).
After 30 min, full disappearance of the signal of substrate 5 was ob-
served with the formation of RP and SP diastereomers of dinucleoside
phosphorothioates (55–57 ppm range). After removal of all protect-
ing groups (procedure as described above), dinucleoside phosphoroth-
ioates (a mixture of RP and SP diastereomers) were identified as major

Synthesis of Dinucleoside Phosphorothioates
1555
Thy
Thy
HO
O
DMTO
O
Thy
DMTO
O
Bz
_
_
HO
O
S
O
O
1. CH3COOH
2. NH4OH
S
O
O
DBU/CH3CN
O
S
O
S
P
+
P
Bz
B
P
O
O
O
O
OAc
OH
OAc
5
SCHEME 5 The synthesis of dinucleoside phosphorothioates from dithiaphos-
pholane 5.
components of the reaction mixture by spectroscopic (31P NMR chemi-
cal shifts) and chromatographic (RP HPLC retention times) comparison
with original samples prepared independently by the phosphoramidite
method.14 As expected, the signal of RP isomer appeared for all dinucle-
oside phosphorothioates at higher frequencies in 31P NMR (D2O solu-
tion), and this isomer showed higher chromatographic mobility under
conditions of RP HPLC. Despite of using strictly anhydrous conditions,
only ca. 50% yield of desired dinucleoside phosphorothioates was ob-
served (31P NMR spectra and HPLC analysis). These measurements
also allowed us to determine their RP/SP diastereomeric ratio for each
dinucleotide product. The average values of RP/SP ratios taken from
both NMR and HPLC measurements are listed in Table II.
An inspection of Table II shows that the shift of dithiaphospholane
ring from 5ꢀ-O-position (in 3) to 3ꢀ-O-position (in 5) reversed the outcome
of asymmetric reaction. Now, a preferential formation of SP isomer was
observed (72% in the case of deoxyguanosine derivative). However, for
TABLE II Stereochemical Analysis of Dinucleoside
Phosphorothioates Formed in Reactions of Ditiaphospholane 5 with
5ꢀ-OH Nucleosides
RP HPLCc retention
time (min) RP/SP
B
Z
RP/SP (%)a
δ
31P NMRb RP/SP (ppm)
Ade
Cyt
Gua
Thy
N6-Bz
N4-Bz
N2-iBu
H
38.5/61.5
41/59
28/72
56.99/57.28
59.62/59.71
57.29/57.33
56.55/56.86
19.03/19.82
17.51/18.34
17,28/18,19
19.61/20.18
44/56
aAverage values for 31P NMR and HPLC measurements.
bD2O solution; fully deprotected dinucleotides.
cLinear gradient of acetonitrile in 0,1 M triethylammonium bicarbonate (1%/min);
fully deprotected dinucleoside phosphorothioates.

1556
M. Olesiak and A. Okruszek
other nucleoside substrates, the asymmetric induction was lower (56–
61.5% of SP isomer).
In the search for improvement of asymmetric induction, 5ꢀ-O-
monomethoxytrityluridine-3ꢀ-O-(2-oxo-1,3,2-dithiaphospholane) 7 was
synthesized, bearing in 2ꢀ-O-position tert-butyldimethylsilyl sub-
stituent. It was hoped that the presence of a bulky 2ꢀ-O-group would in-
crease the chances for stereoselective formation of internucleotide phos-
phorothioate linkage. Thus, commercially available 2ꢀ-O-substituted
uridine was phosphitylated in 3ꢀ-O-position in the manner described
for thymidine derivative (Scheme 6).
Ura
Ura
MMTO
MMTO
O
O
S
S
P
NiPr2
Ura
O
MMTO
OSiMe2tBu
t- BuOOH
OSiMe2tBu
O
O
.
O
.
Tetrazole
P
P
S
S
OSiMe2tBu
S
S
OH
8
7
SCHEME 6 The synthesis of 5ꢀ-O-monomethoxytrityl-2ꢀ-O-tert-butyldi-
methylsilyluridine-3ꢀ-O-(2-oxo-1,3,2-dithiaphospholane) (7).
The intermediate trivalent phosphorus dithiaphospholane 8 was
isolated chromatographically in 93% yield (31P NMR: δ 158.85 ppm,
CD3CN) and oxidized with tert-butyl hydroperoxide to give desired
product 7, which was isolated by precipitation from n-hexane as white
solid (31P NMR: δ 82.38 ppm, CD3CN; yield 79.4%).
The efficiency of dithiaphospholane 7 as stereoselective precursor of
phosphorothioate linkage was tested by its reaction with 5ꢀ-OH deriva-
tives of guanosine and deoxyguanosine (Scheme 7).
Ura
Ura
MMTO
O
MMTO
O
Ura
HO
O
iBu
Gua
X
HO
O
OSiMe2tBu
OSiMe2tBu
-
O
O
S
DBU/CH3CN
1. CH3COOH
O
O
OH
-
O
+
S
P
P
iBu
2. NH4OH
3. HF x Et3N
Gua
O
O
P
S
S
OAc
Gua
O
O
O
X
7
OAc
X = H or OAc
Y
OH
Y = H or OH
SCHEME 7 The synthesis of dinucleoside phosphorothioates from dithiaphos-
pholane 7.
Due to lower reactivity of sterically hindered substrate 7, the re-
actions were carried out for 2 h, with 30% molar excess of protected
guanine nucleosides. For removal of dinucleotide protecting groups, in

Synthesis of Dinucleoside Phosphorothioates
1557
TABLE III Stereochemical Analysis of Dinucleoside Phosphoro-
thioates Formed in Reactions of Ditiaphospholane 7 with Guanine
Nucleosides (Scheme 7)
RP HPLCc retention
time (min) RP/SP
X
Y
RP/SP (%)a
δ
31P NMRb RP/SP (ppm)
H
OAc
H
OH
35/65
46.5/53.5
58.28/58.43
58.06/58.22
13.11/13.55
12.41/12.86
aAverage values for 31P NMR and HPLC measurements.
bCH3CN/CD3CN solution; crude reaction mixture before removal of protecting
groups.
cLinear gradient of acetonitrile in 0,1 M triethylammonium bicarbonate (1%/min);
fully deprotected dinucleotides.
addition to standard acid and ammonia treatment, the product was
reacted with triethylammonium hydrogen fluoride (cleavage of silyl
ether). The results are listed in Table III.
Unfortunately, even the increase of steric hindrance in dithiaphos-
pholane substrate did not enhance the extent of asymmetric induction,
which led to the preponderant formation (65%) of SP isomer of result-
ing phosphorothioate for deoxyguanosine derivative. In the case of pro-
tected guanosine derivative, almost equal proportion of isomers was
observed (53.5% SP).
In conclusion, appropriately protected nucleosides containing 2-oxo-
1,3,2-dithiaphospholane function were found to react with nucleosides
containing free hydroxyl group in anhydrous acetonitrile, in the pres-
ence of equimolar amount of DBU, with the formation (after deprotec-
tion) of dinucleoside phosphorothioates. The products were formed in
moderate yields (ca. 50%), which resulted mainly from the difficulty
in obtaining completely anhydrous reaction conditions. Due to the fact
that initial dithiaphospholane substrates were prochiral, and in the
reaction a new phosphorothioate chiral center was formed, there was
hope for the asymmetric induction and the formation of products highly
enriched in one of the P-isomers. The induction could be caused by
“internal” chiral auxiliaries present in nucleoside sugar moieties. Un-
fortunately, the experiments performed with substrates containing 2-
oxo-1,3,2-dithiaphospholane function either in 5ꢀ-O- or in 3ꢀ-O-position
of deoksyribonucleoside, or even in 3ꢀ-O-position of 2ꢀ-O-bulky substi-
tuted ribonucleoside, did not generate the formation of dinucleoside
phosphorothiotes highly enriched in one of P-diastereomers. The high-
est observed enrichment was 72/28 SP/RP ratio (Table II).

1558
M. Olesiak and A. Okruszek
EXPERIMENTAL
TLC was performed on Kieselgel 60F254 (Merck) with UV (254 nm) de-
tection. Flash column chromatography was run on silica gel 230–400
mesh (Merck). Reverse phase high performance liquid chromatography
(RP HPLC) was made with Gilson chromatograph (model 306) using
analytical column 4.6 × 250 mm packed with ODS Hypersil, 5µ (All-
tech). The columns were eluted with 0.1 M aqueous triethylammonium
bicarbonate (TEAB), pH 7.5, supplemented by linear gradient of ace-
tonitrile from 0 to 20% in 20 min. 31P NMR spectra were recorded
on a Bruker AC 200 spectrometer (200.113 MHz for 1H) and refer-
enced to 85% H3PO4 used as an external standard. Mass spectra were
recorded on Finnigan MAT 95. Dinucleoside phosphorothioate synthe-
ses were performed by modified phosphoramidite method (sulfurisation
with S-TETRA) on an automatic ABI 394 DNA Synthesizer on 1 µ mol
scale.
The analytical grade solvents were purchased from Aldrich and
freshly distilled from CaH2. Analytical grade phosphorus trichlo-
ride, diisopropylamine, 1,2-ethanedithiol, triethylamine, tetrazole, and
DBU were purchased from Aldrich. Nucleosides were purchased from
Pharma Waldhof.
N,N-Diisopropylaminophosphorodichloridite
Into a solution of freshly distilled phosphorus trichloride (67 g, 0.49
mol) in anhydrous benzene/hexane (1:1, 500 mL), a solution of dry
diisopropylamine (99 g, 0.98 mole) in benzene (100 mL) was added
dropwise, with mechanical stirring and cooling at 4◦C. The solution
was then stirred for 1 h at r.t., filtered, and the precipitate was washed
with benzene/hexane (1:1, 250 mL). The filtrate was evaporated, and
the residue was distilled at 0.4 mmHg to yield 78.5 g (79.3%) of colorless
liquid product boiling at 53–5◦C, 31P NMR: δ 169.74 ppm (CDCl3). Lit.31
bp 56–9◦C/0.5 mmHg, 31P NMR: δ 169.0 ppm (C6D6).
2-N,N-Disopropylamino-1,3,2-dithiaphospholane
Into a solution of N, N-diisopropylaminophosphorodichloridite (78.5 g,
0.39 mol) in anhydrous benzene/hexane (1:1, 900 mL), a solution of 1,2-
ethanedithiol (36.5 g, 0.39 mole) and triethylamine (78.5, 0.78 mole)
in 200 mL of benzene was added dropwise, with mechanical stirring
and cooling at 4◦C. After 1 h stirring at r.t., the amine hydrochloride
was filtered off and washed with benzene/hexane (1:1, 400 mL). The
combined filtrates were evaporated, and the residue was distilled at

Synthesis of Dinucleoside Phosphorothioates
1559
0.1 mmHg yielding 79.5 g (91.7%) of colorless liquid boiling at 115◦C,
31P NMR: δ 94.61 ppm (CDCl3). HR MS m/z 223.0596 Da (calcd m/z
223.0618 Da).
5ꢀ-O-Dimethoxytritylthymidine-3ꢀ-O-(2-oxo-1,3,2-
dithiaphospholane) 5
The mixture of 5ꢀ-O-dimethoxytritylthymidine (1.0 g, 1.84 mmol) and
tetrazole (126 mg, 1.84 mmol) was dried on a vacuum line for 12 h.
The mixture was dissolved under argon in 20 mL of anhydrous methy-
lene chloride, and 410 mg (1.84 mol) of 2-N, N-disopropylamino-1,3,2-
dithiaphospholane was added with stirring at r.t. Stirring was contin-
ued overnight, and after evaporation, the trivalent phosphorus product
6 was isolated by flash column chromatography using chloroform with a
gradient of methanol (0–3%) as eluent (white powder, 1.21 g, 31P NMR:
δ 150.66 ppm, CD3CN). The product 6 was dissolved in benzene (10 mL)
and 360 µl of 5–6 M decane solution of tert-butyl hydroperoxide was
added dropwise at r.t. After 1 h the resulting solution was added drop-
wise at 0◦C to the vigorously stirred n-hexane (50 mL). The final prod-
uct 5 was obtained by centrifugation of precipitate and its drying in
vacuo (white powder, 1.04 g, 84%, 31P NMR: δ 79.47 ppm, C6D6). The
compound 5 was stored under argon at –20◦C.
Basically the same procedure was applied for the synthesis of 2-oxo-
1,3,2-dithiaphospholane compounds 3 and 7.
The Synthesis of Dinucleoside Phosphorothioates from
Dithiaphospholane 5: General Procedure
The mixture of 5ꢀ-O-dimethoxytritylthymidine-3ꢀ-O-(2-oxo-1,3,2-
dithiaphospholane) 5 (70 mg, 0.1 mmol) and 3ꢀ-O-acetyl-N-protected-2ꢀ
-deoxyribonucleoside (0.1 mmol) was dried on a vacuum line for 12 h.
Into this mixture a solution of DBU (15 mg, 0.1 mmol) in anhydrous
acetonitrile (1.5 mL) was added with stirring. After 30 min at r.t., the
mixture was analyzed by 31P NMR and evaporated. The residue was
dissolved in 80% aq acetic acid (1 mL). After 2 h at r.t., the solution
was evaporated. The residue was dissolved in 1 mL of acetonitrile and
2 mL of 30% aq ammonia was added. The flask was closely capped and
kept at 55◦C for 12 h. The solution was evaporated and the residue
was analyzed by RP HPLC and 31P NMR.
Basically the same procedure was applied for reactions of compounds
3 and 7 with appropriate nucleoside derivatives.

1560
M. Olesiak and A. Okruszek
REFERENCES
[1] S. Verma and F. Eckstein, Annu. Rev. Biochem., 67, 99–134 (1998).
[2] J. Micklefield, Curr. Med. Chem., 8, 1157–1179 (2001).
[3] E. Uhlmann and A. Peyman, Chem. Rev., 90, 543–584 (1990).
[4] S. Agrawal and R. P. Iyer, Curr. Opin. Biotechnol., 6, 12–19 (1995).
[5] S. Agrawal and E. R. Kandimalla, Mol. Med. Today, 6, 72–81 (2000).
[6] F. Eckstein, Antisense & Nucl. Acid Drug Dev., 10, 117–121 (2000).
[7] E. Wickstrom (Ed.), Clinical Trials of Genetic Therapy with Antisense DNA and
DNA Vectors (Marcel Dekker, Inc., New York, 1998).
[8] W. J. Stec, G. Zon, W. Egan, and B. Stec, J. Am. Chem. Soc., 106, 6077–6079 (1984).
[9] B.C. Froehler, Tetrahedron Lett., 27, 5575–5578 (1986).
[10] I. Barber, J.-L. Imbach, and B. Rayner, Antisense Res. Dev., 5, 39–47 (1995).
[11] S. Agrawal and J.-Y. Tang, Tetrahedron Lett., 31, 7541–7544 (1990).
[12] A. Almer, J. Stawin´ski, and R. Stro¨mberg, Nucl. Acids Res., 24, 3811–3820 (1996).
[13] W. J. Stec, C. S. Cierniewski, A. Okruszek, A. Kobylan´ska, Z. Pawlꢀowska, M.
Koziolꢀkiewicz, E. Pluskota, A. Maciaszek, B. Rebowska, and M. Stasiak, Antisense
ꢀ
& Nucl. Acid Drug Dev., 7, 567–573 (1997).
[14] A. Wilk and W. J. Stec, Nucl. Acids Res., 23, 530–534 (1995).
[15] J. G. Hacia, B. J. Wold, and P. Dervan, Biochemistry, 33, 5367–5369 (1994).
[16] J. Tang, A. Roskey, Y. Li, and S. Agrawal, Nucleosides Nucleotides, 14, 985–990
(1995).
[17] S. Agrawal and J.-Y. Tang, Tetrahedron Lett., 31, 7541–7544 (1990).
[18] W. J. Stec, A. Grajkowski, M. Koziolꢀkiewicz, and B. Uznan´ski, Nucl. Acids Res., 21,
5883–5888 (1991).
[19] W. J. Stec, A. Grajkowski, A. Kobylan´ska, B. Karwowski, M. Koziolꢀkiewicz, K.
Misiura, A. Okruszek, A. Wilk, P. Guga, and M. Boczkowska, J. Am. Chem. Soc.,
117, 12019–12029 (1995).
[20] W. J. Stec, B. Karwowski, M. Boczkowska, P. Guga, M. Koziolꢀkiewicz, M. Sochacki,
M. W. Wieczorek, and J. Blꢀaszczyk, J. Am. Chem. Soc., 120, 7156–7167 (1998).
[21] P. Guga, A. Okruszek, and W. J. Stec, Recent advances in stereocontrolled synthesis
of p-chiral analogues of biophosphates. In Topics in Current Chemistry, 220, J.-P.
Majoral, Ed. (Springer-Verlag, Berlin, Heidelberg, 2002), pp. 169–200.
[22] A. Sierzchalꢀa, A. Okruszek, and W. J. Stec, J. Org. Chem., 61, 6713–6716 (1996).
[23] X.-B.Yang, A. Sierzchalꢀa, K. Misiura, W. Niewiarowski, M. Sochacki, W. J. Stec, and
M. W. Wieczorek, J. Org. Chem., 63, 7097–7100 (1998).
[24] Z. J. Les´nikowski and M. M. Jaworska, Tetrahedron Lett., 30, 3821–3824 (1989).
[25] Z. J. Les´nikowski, Nucleosides Nucleotides, 11, 1621–1638 (1992).
[26] Y. Jin and G. Just, J. Org. Chem., 63, 3647–3654 (1998).
[27] J.-C. Wang and G. Just, J. Org. Chem., 64, 8090–8097 (1999).
[28] R.P. Iyer, M.-J. Guo, D. Yu, and S. Agrawal, Tetrahedron Lett., 39, 2491–2494 (1998).
[29] A. Wilk, A. Grajkowski, L. R. Phillips, and S. L. Beaucage, J. Am. Chem. Soc., 122,
2149–2156 (2000).
[30] A. Okruszek, A. Sierzchalꢀa, K. L. Fearon, and W. J. Stec, J. Org. Chem., 60, 6998–
7005 (1995).
[31] M. E. Piotto, J. N. Granger, Y. Cho, N. Farschtschi, and D. G. Gorenstein, Tetrahe-
dron, 47, 2449–2461 (1991).