A new approach to stereospecific synthesis of P-chiral phosphorothioates. Preparation of diastereomeric dithymidyl-(3'-5') phosphorothioates.

Article abstract of DOI:10.1039/b311912b

A new method for stereospecific synthesis of P-chiral phosphorothioates based on intramolecular nucleophile catalysis was developed.

Full text of DOI:10.1039/b311912b

A new approach to stereospecific synthesis of P-chiral phosphorothioates.
Preparation of diastereomeric dithymidyl-(3A-5A) phosphorothioates†
Helena Almer‡,a Tomas Szaboa and Jacek Stawinski*ab
a Department of Organic Chemistry, Arrhenius Laboratory, Stockholm University, S-106 91 Stockholm,
Sweden
b Institute of Bioorganic Chemistry, Polish Academy of Sciences, Noskowskiego 12/14, 61-704 Poznan,
Poland
Received (in Cambridge, UK) 26th September 2003, Accepted 26th November 2003
First published as an Advance Article on the web 19th December 2003
A
new method for stereospecific synthesis of P-chiral
the catalytic group should secure efficient formation of an
internucleotide bond and, due to the intramolecular character of this
phenomenon, also prevent erosion of configuration at the phospho-
rus center during the condensation step. Thirdly, the method would
share other advantages of the underlying phosphotriester chemistry,
e.g. possibility of using various condensing agents, lower sensitiv-
ity to the presence of adventitious water and compatibility with
both solution and solid-phase chemistry.
A mechanistic proposition for the whole transformation that
utilizes 1-oxido-2-picolyl functionality as an intramolecular cata-
lytic group, is outlined in Scheme 1. Activation of a phosphoro-
thioate diester by a condensing agent (CA) will produce a reactive
intermediate, which in two stereospecific steps involving (i) the
intramolecular elimination A of a leaving group and (ii) the
nucleophilic substitution B by a hydroxylic component, collapses
to the product. Since both steps A and B are expected to proceed
with inversion of configuration, the product of the reaction, a
phosphorothioate triester, should be formed with an overall
retention of configuration at the phosphorus center.
While the rate enhancement due to intramolecular catalysis
exerted by a 1-oxido-2-picolyl group attached to a phosphorus
centre finds literature precedents,8,9 the use of a catalytic phosphate
protecting group as means of securing stereochemical integrity of
the phosphorus center during SN2(P) reactions has until now not
been investigated.
Our first objective was thus to find out if the underlying principle
of the method, i.e. that the intramolecular catalysis provided by the
1-oxido-2-picolyl group would result in improved reaction kinetics
while preventing epimerization at the phosphorus center, is sound.
As a model system for our studies we chose a reaction of
diastereomerically pure 5A-O-tert-butyldiphenylsilyl-thymidin-3A-
yl 1-oxido-4-methoxy-2-picolyl phosphorothioates 4a and 4b with
3A-O-tert-butyldiphenylsilylthymidine in the presence of a con-
densing agent (Scheme 2).
phosphorothioates based on intramolecular nucleophile cataly-
sis was developed.
Antisense oligodeoxynucleotides (AOs) are short, single-stranded
DNA fragments that may inhibit the translation of mRNAs into
proteins by forming complexes with a complementary mRNA
target.1 Among various kinds of oligonucleotide analogues that
have been investigated for the purpose of antisense and antigene
therapies,2 it is oligodeoxynucleoside phosphorothioates (PS-
oligos) that, due to their ability to induce the RNase H promoted
degradation of target mRNA and the enhanced resistance to
nuclease degradation in vivo, became the main focus of medical
research.3
A perennial problem of oligonucleoside phosphorothioates
synthesis is chirality of the phosphorus center that manifests itself
in different chemical, biochemical, and biological properties of the
diastereomeric species.3 Despite the efforts of many research
groups during the past years, the solution to the problem of
stereocontrolled synthesis of PS-oligos has eluded a fully accept-
able solution. Early attempts to control stereochemistry at the
phosphorus center by using diastereomerically pure nucleoside
phosphoramidites were not successful4 due to the profound
tendency of tervalent phosphorus compounds to epimerize under
mild acidic conditions. To remedy this problem, a handful of
stereoselective methods for the formation of P-chiral inter-
nucleotide linkages were developed. All of them made use of a
special type of phosphoramidites in which a phosphorus atom is
part of relatively rigid cyclic systems and which usually bear chiral,
bulky auxiliaries attached to the phosphorus center.5
As to a stereospecific formation of P-chiral internucleoside
phosphorothioate linkages, there are only two methods available
for this purpose: the oxathiaphospholane method based on P( )
V
derivatives developed by Stec et al.6 and the N-acyl-oxazaphospho-
lidine approach based on tervalent P(III) compounds elaborated by
Beaucage et al.7 Both methods are highly stereospecific and afford
products of diastereomeric purity > 99%. The possible drawbacks
of these methods, however, are that both require very strong bases
to effect the elongation of an oligonucleotidic chain, and the
preparation of the requisite starting materials is usually long and
rather complicated. The inherent high reactivity of both classes of
compounds make also their separation into individual diaster-
eomers a delicate task and put severe limitations on possible
recycling of the monomers.
To secure simple, efficient, and general access to the key
intermediate, 4-methoxy-1-oxido-2-picolyl phosphorothioate di-
esters of type 4, we developed a dedicated reagent, 4-methoxy-
1-oxido-2-picolyl-(9-fluorenemethyl) phosphorothioate 2, ena-
We argued that a viable approach to a stereospecific synthesis of
oligonucleoside phosphorothioates could be a condensing agent
promoted reaction of nucleosides with phosphorothioate diesters
bearing a nucleophile catalytic group. There are several features
which make this approach appealing. Firstly, by using phosphoro-
thioate diesters we would benefit from an easy to handle and
configurationally stable starting material, that also should be easy
to prepare. Secondly, a powerful nucleophile catalysis provided by
† Electronic supplementary information (ESI) available: synthesis and
characterization of compounds 2 through 6. See http://www.rsc.org/
suppdata/cc/b3/b311912b/
‡ Deceased.
Scheme 1
290
C h e m . C o m m u n . , 2 0 0 4 , 2 9 0 – 2 9 1
T h i s j o u r n a l i s © T h e R o y a l S o c i e t y o f C h e m i s t r y 2 0 0 4

bling
transfer
of
the
phosphorothioate
and
31P NMR experiments allowed us to survey a wide range of
reaction conditions for the formation of triesters 5a and 5b. The
condensation of separate diastereomers 4a or 4b in dichloro-
methane containing pyridine (3 equiv.) with 3A-O-tert-butyldime-
thylsilylthymidine (1.5 equiv.) promoted by a mild condensing
agent NEP-Cl12 (3 equiv.) was fast (ca. 15 min), quantitative and
completely stereospecific ( > 99%).11 Thus, 4a (dP = 57.78 ppm)
afforded exclusively 5a (dP = 67.90 ppm), while 4b (dP = 57.94
ppm) gave exclusively diastereomeric product 5b (dP = 67.24
ppm). No sulfur activation was observed during the course of the
condensation. The diastereomers 5a and 5b were subjected
separately to total deprotection and absolute configurations at the
phosphorus centre in dinucleoside phosphorothioates 6a and 6b
were determined as RP and SP, respectively, by enzymatic
digestion.13,14
In conclusion, we have shown that an intramolecular catalytic
group can secure stereochemical integrity of the phosphorus center
during SN2(P) reactions. On this basis we developed a new method
for stereospecific synthesis of dinucleoside phosphorothioate
diesters, that can be extended to the synthesis of other chiral
phosphorothioates. The thiophosphorylating reagent 2 is stable,
readily accessible, and the method offers a new avenue in the
synthesis of P-chiral thiophosphates.
4-methoxy-1-oxido-2-picolyl moieties to a hydroxylic component
in one step. It sports a 9-fluorenemethyl group as a lipophilic handle
to facilitate chromatographic separation of the P-chiral phosphoro-
thioate triesters 3, and its removal can be effected via b-elimination
without affecting the stereochemical integrity of the phosphorus
center of the produced phosphorothioate diesters 4. The reagent is
a stable, white solid, that can be prepared by condensation of
9-fluorenemethyl phosphonate 110 with 4-methoxy-2-pyridineme-
thanol 1-oxide,8,9 followed by in situ sulfurisation with elemental
sulfur (total yield 63%).
Preparation of separate diastereomers of the starting material 4,
nucleoside phosphorothioate diesters, required for stereospecific
synthesis, commences with thiophosphorylation of 5A-O-tert-
butyldiphenylsilylthymidine with reagent 2 in the presence of
2-chloro-5,5-dimethyl-2-oxo-1,3,2-dioxaphosphinane (NEP-Cl).
The reaction was uneventful and produced almost quantitatively
(31P NMR) nucleoside phosphorothioate triesters 3 as a 6 : 4
mixture of RP and SP diastereomers. Separation of the diaster-
eomeric mixture by silica gel column chromatography furnished
after one run 3a (faster moving diastereomer, ca. 50%), 3b (slower
moving diastereomer, ca. 20%), and a mixture of 3 (ca. 25%). Due
to the stability of phosphorothioate triesters 3, the mixed fractions
can be subjected to re-chromatography or stored for later
separation.
Financial support from the Swedish Research Council is
gratefully acknowledged.
To obtain nucleoside phosphorothioate diesters 4, separate
diastereomers of phosphorothioate triesters 3 were subjected to
deprotection with tert-butylamine. The removal of 9-fluoreneme-
thyl group from 3a and 3b was rapid and clean, and afforded after
silica gel chromatography, diastereomerically pure 4a (ca. 70%)
and 4b (ca. 80%), respectively.
Notes and references
1 J. S. Cohen, Ed.; Oligodeoxynucleotides – Antisense Inhibitors of Gene
Expression; The Macmillan Press Ltd., London, 1989.
2 E. Uhlmann and A. Peyman, Chem. Rev., 1990, 90, 543–584.
3 (a) F. Eckstein, Antisense Nucleic Acid Drug Dev., 2000, 10, 117–121;
(b) L. R. Grillone and R. Lanz, Drugs Today, 2001, 37, 245–255; (c) S.
Akhtar and S. Agrawal, Trends Pharm. Sci., 1997, 18, 12–18; (d) S. L.
Beaucage and R. P. Iyer, Tetrahedron, 1992, 48, 2223–2311; (e) D. Yu,
E. R. Kandimalla, A. Roskey, Q. Zhao, L. Chen, J. Chen and S.
Agrawal, Bioorg. Med. Chem. Lett., 2000, 8, 275–284.
4 W. J. Stec and G. Zon, Tetrahedron Lett., 1984, 25, 5279–5282.
5 (a) R. P. Iyer, M. J. Guo, D. Yu and S. Agrawal, Tetrahedron Lett.,
1998, 2491–2494; (b) M. J. Guo, D. Yu, R. P. Iyer and S. Agrawal,
Bioorg. Med. Chem. Lett., 1998, 8, 2539–2544; (c) Y. Jin, G. Biancotto
and G. Just, Tetrahedron Lett., 1996, 37, 973–976; E. Marsault and G.
Just, Tetrahedron Lett., 1996, 37, 977–980; Y. X. Lu and G. Just,
Tetrahedron, 2001, 57, 1677–1687; (d) J. C. Wang and G. Just,
Tetrahedron Lett., 1997, 38, 3797–3800; (e) Y. X. Lu and G. Just,
Angew. Chem., Int. Ed., 2000, 39, 4521–4524; (f) Y. X. Lu and G. Just,
Tetrahedron, 2000, 56, 4355–4365; (g) J.-C. Wang, G. Just, A. P.
Guzaev and M. Manoharan, J. Org. Chem., 1999, 64, 2595–2598; (h) N.
Oka, T. Wada and K. Saigo, J. Am. Chem. Soc., 2002, 124,
4962–4963.
6 W. J. Stec, A. Grajkowski, M. Koziolkiewicz and B. Uznanski, Nucleic
Acids Res., 1991, 19, 5883–5888.
7 A. Wilk, A. Grajkowski, L. R. Phillips and S. L. Beaucage, J. Am. Chem.
Soc., 2000, 122, 2149–2156.
8 V. A. Efimov, A. A. Buryakova, I. Y. Dubey, N. N. Polushin, O. G.
Chakhmakhcheva and Y. A. Ovchinnikov, Nucleic Acids Res., 1986, 14,
6525–6540.
9 T. Szabó, A. Kers and J. Stawinski, Nucleic Acids Res., 1995, 23,
893–900.
10 Z. W. Yang, Z. S. Xu, N. Z. Shen and Z. Q. Fang, Nucleosides
Nucleotides, 1995, 14, 167–173.
11 The formation of the other diastereomer, if any, was below the detection
level of 31P NMR spectroscopy.
12 R. Strömberg and J. Stawinski, Nucleic Acids Symp. Ser., 1987, 18,
185–188.
Scheme 2 Reagents and conditions: (i) 1. 4-methoxy-2-pyridinemethanol
1-oxide + NEP-Cl/py, 2. H2O, 3. S8; (ii) 5A-TBDMS-T + NEP-Cl/py; (iii) t-
BuNH2/py; (iv) 3A-TBDMS-T + NEP-Cl/py; (v) 1. TEA–PhSH, 2. F2.
Abbreviations: Thy, thymin-1-yl; TBDMS, t-butyldimethylsilyl; NEP-Cl,
2-chloro-5,5-dimethyl-2-oxo-1,3,2-dioxaphosphinane.
13 P. M. J. Burgers and F. Eckstein, Proc. Natl. Acad. Sci. USA, 1978, 75,
4798–4800.
14 M. J. Damha and K. K. Ogilvie, Protocols for Oligonucleotides and
Analogs. Synthesis and Properties, in Methods in Molecular Biology; S.
Agrawal, Ed.; Humana Press, Inc.: Totowa, 1993; pp. 81–114.
C h e m . C o m m u n . , 2 0 0 4 , 2 9 0 – 2 9 1
291