Stereochemistry of internucleotide bond formation by the H-phosphonate method. Part 6: Optimization of the reaction conditions towards highest stereoselectivity

Article abstract of DOI:10.1016/j.tetasy.2008.11.002

The condensations of ribonucleoside 3′-H-phosphonates with simple alcohols and nucleosides are known to be stereoselective reactions that favour formation of D_P(S_P)-diastereomers of the produced H-phosphonate diesters without engagin

Full text of DOI:10.1016/j.tetasy.2008.11.002

Tetrahedron: Asymmetry 19 (2008) 2508–2518
Contents lists available at ScienceDirect
Tetrahedron: Asymmetry
journal homepage: www.elsevier.com/locate/tetasy
Stereochemistry of internucleotide bond formation by the H-phosphonate
method. Part 6: Optimization of the reaction conditions towards highest
stereoselectivity
b
a
Michal Sobkowski a, , Jacek Stawinski , Adam Kraszewski
*
a Institute of Bioorganic Chemistry, Polish Academy of Sciences, Noskowskiego 12/14, 61-704 Poznan, Poland
b Department of Organic Chemistry, Arrhenius Laboratory, Stockholm University, S-106 91 Stockholm, Sweden
a r t i c l e i n f o
a b s t r a c t
The condensations of ribonucleoside 30-H-phosphonates with simple alcohols and nucleosides are known
to be stereoselective reactions that favour formation of DP(SP)-diastereomers of the produced H-phospho-
nate diesters without engaging any chiral auxiliaries. We have investigated various reaction conditions in
order to attain the highest stereoselectivity for the condensation. With an optimal choice of solvents,
reagent concentrations, temperature and the condensing agent used, the diastereomeric excess of the
DP(SP)-isomers of dinucleoside H-phosphonates was enhanced from the initial 50–70% to ca. 85%.
Ó 2008 Elsevier Ltd. All rights reserved.
Article history:
Received 8 October 2008
Accepted 7 November 2008
Available online 29 November 2008
1. Introduction
ularly attractive goal. Wada et al. showed that the oxazaphosphol-
idine approach developed for oligodeoxynucleotides could also be
P-Modified oligonucleotide analogues (e.g., phosphorothioates)
with a defined configuration at the chiral phosphorous centre are
valuable tools in the investigations of interactions between nucleic
acids and other biomolecules. This has stirred the interest of sev-
eral research groups to design methods for the stereocontrolled
formation of P-chiral internucleotide linkages.1–5 To achieve the
desired stereoselectivity, in most of the known methods, the phos-
phorous atom is incorporated into a relatively rigid five- or six-
membered ring system. The necessity of the preparation of unique
synthons dedicated to a particular synthetic scheme seems to be
the major inconvenience of such approaches. Also, due to synthetic
limitations, most of the methods for the stereocontrolled synthesis
of nucleotide analogues are focused on only one type of a phos-
phate moiety modification, usually phosphorothioates and, to a
lesser extent, methylphosphonates or boranophosphates. Only re-
cently, Wada et al. reported preliminary data on a stereocontrolled
synthesis of oligodeoxynucleotide H-phosphonates by an oxaza-
phospholidine approach6 that, by taking advantage of the flexibil-
ity of the H-phosphonate chemistry, can provide various
nucleotide analogues from one common precursor.
applied to oligoribonucleotides,9 but the oxazaphospholidine
derivatives optimized for the highest stereoselectivity in the deoxy
series required significant changes to be effective in the ribo series.
It may be anticipated that also in other approaches to stereocon-
trolled synthesis of oligonucleotide analogues, switching from
DNA to RNA chemistry might not be straightforward or even
possible.
In this context, the stereoselective oligoribonucleotide synthe-
sis via the H-phosphonate approach might appear as an attractive
alternative, as the condensation of ribonucleoside H-phosphonates
with nucleosides (or alcohols) produces DP-diastereomers of the
ribonucleoside H-phosphonate diesters in de (diastereomeric excess)
of 50–70% (Fig. 1).14–17 In this approach, commercially available H-
phosphonate synthons can be used, which make the method partic-
ularly attractive since nucleoside H-phosphonates are known to be
superior substrates for oligonucleotide synthesis in the ribo
series.15,18,19
For the compounds presented herein, the DP-descriptor refers to a structure in
which the P–H bond is directed to the right in the Fischer projection, while for the LP
one, to the left. The full DP/LP notation is described in Refs. 10–13
Another important issue concerns the efficiency and stereo-
chemistry of P-chiral oligonucleotide synthesis in the deoxy versus
ribo series. At present, most of the published methods deal with P-
chiral oligodeoxynucleotide analogues, but the growing interest in
RNAi and its application for therapeutic purposes7,8 have made the
stereocontrolled synthesis of P-chiral oligoribonucleotides a partic-
nucleoside
O
nucleoside
O
O
P
H
H
P
O
OR
OR
* Corresponding author. Tel.: +48 61 8528 503; fax: +48 61 8520 532.
E-mail address: msob@ibch.poznan.pl (M. Sobkowski).
DP
LP
.
0957-4166/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetasy.2008.11.002

M. Sobkowski et al. / Tetrahedron: Asymmetry 19 (2008) 2508–2518
2509
DMTrO
O
DMTrO
DMTrO
B
B
B
O
O
O
PvCl
ROH
O
P
O
P
O
P
OTBDMSi
OTBDMSi
OTBDMSi
H
O
O
H
H
O
O
O
DP(SP)
LP(SP)
Pv
R
1
2
3
3a
equilibrium
, R = nucleosid-5'-yl
DMTr = 4,4'-dimethoxytrityl
TBDMSi = tert-butyldimethylsilyl
B = protected nucleobase
Pv = pivaloyl
mixture of the diastereomers
3b, R = Et
major (~70%): DP(RP)
minor (~30%): LP(SP)
major (75 - 90%): DP(SP)
minor (10 - 25%): LP(RP)
ROH = protected nucleoside
or alcohol
(de ~40%)
(de 50 - 80%)
Figure 1. Stereoselective formation of the DP(SP)-diastereomer of H-phosphonate diesters of type 3.
In the paper by Almer et al.,15 several helpful hints for tuning
the stereoselectivity of the H-phosphonate internucleotide bond
formation were given. Recently, we have resumed the evaluation
of this method in a more systematic way to gain a better under-
standing of the underlying mechanism,16,17,20–24 and to achieve
further improvement of stereoselectivity.
significantly lower stereoselectivity than the other nucleoside H-
phosphonates.15–17
Good stereoselectivity, together with quantitative yields of the
reactions, pointed to 2,6-lutidine as the amine of choice.23 How-
ever, the excellent stereoselectivity observed in the presence of
TEA (de 75% for B = Ura, and 61% for B = CytBz, Fig. 2) prompted
us to carry out similar condensations using 20,30-O-dibenzoyl uri-
dine (1.2 equiv) as a hydroxylic component instead of ethanol
(Fig. 1, R = uridin-50-yl). As expected, under such reaction condi-
tions H-phosphonate diester 3a (B = Ura) was formed with high
stereoselectivity (de 84–88%); however, the yields of H-phospho-
nate diester did not exceed 71%. For the other substrates of type
1 (B = AdeBz, CytBz and Guaibu) the stereoselectivities in the pres-
ence of TEA were also very high (de 84%, 74% and 83%, respec-
tively), but the yields were again low (ca. 65%). For this reason,
we left the investigations of the use of tertiary amines in the con-
densation of H-phosphonates as a subject of separate study.24
2. Results
2.1. Effects of amines on the stereoselectivity
The influence of tertiary amines and pyridine derivatives on the
stereoselectivity of the model reaction (Fig. 1, B = Ura, R = Et) was
discussed in detail in the previous parts of this series.23,24 The most
meaningful results for uridine derivatives are shown in Figure 2
(left panel) along with the results of analogous reactions for cyti-
dine H-phosphonate 1 (B = CytBz), which is known to react with a
Figure 2. Diastereomeric excess (de) of the DP(SP)-diastereomers of the H-phosphonate diesters 3b (Fig. 1, B = Ura and B = CytBz) formed in the presence of several amines.
The reaction conditions: 0.05 mmol of 1 (B = Ura) + EtOH (3 equiv) + amine (3 equiv) + PvCl (1.5 equiv) in DCM (0.5 mL). Green bars: de of DP diastereomer; red bars: the total
yield of the diester 3b (sum of diastereomers); triangles: pKa of amines.25,26 Abbreviations: DMA, N,N-dimethylaniline; DMAP, N,N-dimethylaminopyridine; TEA,
triethylamine.

2510
M. Sobkowski et al. / Tetrahedron: Asymmetry 19 (2008) 2508–2518
Figure 3. Diastereomeric excess (bars) of the DP(SP) diastereomer of H-phosphonate diester 3b formed in solvents of various relative permittivities (triangles).26 The reaction
conditions: 0.05 mmol of 1 (B = Ura) + EtOH (3 equiv) + 2,6-lutidine (3 equiv) + PvCl (1.5 equiv) in a given solvent (0.5 mL). For the ionic liquid, [bmim][PF6], the value of
permittivity is not available. Abbreviations: ACN, acetonitrile; DCE, 1,2-dichloroethane; DCM, dichloromethane; DMF, N,N-dimethylformamide; THF, tetrahydrofuran;
[bmim][PF6], 1-n-butyl-3-methylimidazolium hexafluorophosphate.
2.2. Effect of solvents on the stereoselectivity
The use of phenoxyacetyl chloride (PAC-Cl) caused pronounced det-
ritylation (probably due to a relatively high acidity of the released
phenoxyacetic acid, pKa 3.17), and this experiment could be success-
fully repeated with 5 equiv of lutidine.
Several solvents were analyzed with respect to the stereoselec-
tivity of the condensations of ribonucleoside 30-H-phosphonates. In
Figure 3, the solvents were arranged according to their relative
permittivity. The best results, in terms of stereoselectivity (de
71%), were found for a moderately polar DCM.
H-Phosphonate diesters 3b were formed efficiently over the
course of the transesterification of p-nitrophenyl–ribonucleoside
H-phosphonates.22 The stereoselectivity in this reaction was found
to be very low (de ꢀ 10%); however, when the aryl H-phosphonate
diester was reacted with 20,30-O-protected uridine instead of EtOH
(yielding H-phosphonate diester 3a), the stereoselectivity in-
creased significantly (de 72%), that is, to the level only slightly low-
er than that of the reaction promoted by PvCl (cf. Fig. 11).–
Nevertheless, the regular esterification using PvCl as a C.A. was still
more stereoselective and experimentally simpler.
In contrast to the so far presented formation of H-phosphonate
diesters via activation of the H-phosphonate monoester, the con-
densations under Mitsunobu conditions, that is, via activation of
a hydroxylic component with TPP/DIAD tandem,29 appeared to
be a non-stereoselective reaction (de 2%).
2.3. Effect of base concentration on the stereoselectivity
In order to estimate the optimal concentration of the base in the
reaction mixture, four ribonucleoside 30-H-phosphonates
1
(B = AdeBz, CytBz, Guaibuà and Ura; 0.0125 mmol) were reacted with
ethanol (6.8 equiv) in the presence of PvCl (2.5 equiv) in DCM–2,6-
lutidine mixtures (total volume 0.5 mL). Lowering the concentration
of H-phosphonate 1 to 25 mM (instead of standard 100 mM) helped
diminish any possible disturbance due to the protonation of the
amine by acidic species released during the condensation. As shown
in Figure 4, for B = AdeBz, CytBz and Ura, the highest stereoselectivity
was obtained when the concentration of 2,6-lutidine was ca. 10%,
while for B = Guaibu, ca. 3%. The advantage of lower content of an
amine in the case of guanosine H-phosphonate was in line with
the earlier findings of Almer et al.15
2.5. Effects of chiral bases and chiral C.A. on the
stereoselectivity
Chiral amines [(ꢁ)-sparteine and (S)-(ꢁ)-nicotine] or chiral C.A.
[(ꢁ)-camphanoyl chloride] were used in the reactions of H-phos-
phonate 1 with EtOH (under standard conditions) in order to
investigate the influence of this chiral components on the stereose-
lectivity. The obtained de values 70% and 65%, respectively, for the
chiral amines,23 and 67% for the chiral C.A. were very similar to the
analogous reactions without chiral auxiliaries (de 70%, vide supra).
In the experiment with ( )-nicotine the stereochemical outcome of
the condensation (de 64%) was found to be almost identical to that
of enantiopure (S)-(ꢁ)-nicotine. In a supplementary reaction in neat
(S)-(ꢁ)-nicotine the stereoselectivity dropped significantly (de 53%),
analogously as it was observed for the reaction in neat pyridine (de
42%) versus the reaction in a DCM–pyridine mixture (de 62%).17
2.4. Effects of condensing agents (C.A.’s) on the stereoselectivity
The results of investigations on the influence of the kind of a
C.A. on the stereoselectivity are summarized in Figure 5. Acyl chlo-
rides,§ triphosgene and DCC were able to secure fast condensations
(completion before the first 31P NMR spectrum was recorded, i.e.,
in less than 1 min) under standard conditions used in this study.
For the other C.A.’s, the reactions were complete after several min-
utes or hours. No traces of by-products due to bis-activation, P-acyl-
ation or oxidation were noticed in the 31P NMR spectra in either case.
à
In these studies, we did not see any significant differences in stereoselectivity
between reactions of phenylacetyl15 and isobutyryl N-protected guanosine H-
phosphonate 1 (B = GuaPAC and B = Guaibu, respectively).
–
The phenomenon of diverse stereoselectivity of aryl H-phosphonate transesteri-
§
With an exception of 2,4,6-trimethyl- and 2,4,6-triisopropyl-benzoyl chlorides
fication was consistent with the DYKAT mechanism and is discussed in more detail in
Ref. 22.
that appeared to be poor activators of nucleoside H-phosphonates.

M. Sobkowski et al. / Tetrahedron: Asymmetry 19 (2008) 2508–2518
2511
Figure 4. Diastereomeric excess of the DP(SP)-diastereomer of H-phosphonate diester 3b formed in the presence of 2,6-lutidine in various concentrations. Reaction
conditions: 0.0125 mmol of 1 (B = AdeBz, CytBz, Guaibu and Ura) + EtOH (6.8 equiv) + 2,6-lutidine (0.3–100% v/v) in DCM (up to 0.5 mL) + PvCl (2.5 equiv).
Figure 5. Diastereomeric excess of the DP(SP)-diastereomer of H-phosphonate diester 3b formed in the presence of various condensing agents. Reaction conditions:
0.05 mmol of 1 (B = Ura) + EtOH (3 equiv) + 2,6-lutidine (3 equiv) + C.A. (3 equiv) in DCM (0.5 mL). Green bars: acyl chlorides; red bars: other chlorides; blue bars: other C.A.’s.
Abbreviations: p-CN-BzCl, p-cyanobenzoyl chloride; TMBzCl, 3,4,5-trimethoxybenzoyl chloride; Hep-COCl, heptanoyl chloride; ibu-Cl, isobutyryl chloride; p-NO2-BzCl, p-
cyanobenzoyl chloride; Palm-COCl, palmitynoyl chloride; p-tBu-BzCl, 4-tert-butylbenzoyl chloride; p-Cl-BzCl, p-chlorobenzoyl chloride; o-Cl-BzCl, o-chlorobenzoyl chloride;
AdCl, adamantanecarbonyl chloride; PAC-Cl, phenoxyacetyl chloride; (ꢁ)-Camph-Cl, (ꢁ)-camphanoyl chloride; o-TolCl, o-toluoyl chloride; NEP-Cl, neopentylene
chlorophosphate; DPCP, diphenyl chlorophosphate; PyNOP, 6-nitrobenzotriazol-1-y1-oxy-tris(pyrrolidino)phosphonium hexafluorophosphate; Tz*NMM, N-methyl-N-[4,6-
dimethoxy-1,3,5-triazin-2-yl)-morpholinium chloride]28 (without additional base). DCC required an acid catalyst27 added to the reaction mixture in the form of lutidinium
hydrochloride (0.1 equiv).
Another coupling reagent system, 2-chloro-4,6-dimethoxy-
1,3,5-triazine/(ꢁ)-sparteine, which is known to promote an enan-
tioselective peptide bond formation,30 appeared to be unreactive
towards the investigated H-phosphonate monoesters.
cessful. The condensations performed in the presence of anhydrous
magnesium acetate displayed decreased stereoselectivity
(de < 60%), while dimethylaluminium chloride caused a substantial
decomposition of H-phosphonate 1.
Other attempts to increase the stereoselectivity of ribonucleo-
side 30-H-phosphonate condensations by introducing stereospe-
cific interactions, via magnesium ions or dimethylaluminium
ligand,31 which were hypothesized to be capable of establishing
conformational constrains in the mixed anhydride 2, were unsuc-
2.6. Effect of temperature on the stereoselectivity
The influence of temperature on the chemical shifts and the
ratio of the diastereomers of the mixed anhydrides 2 that are

2512
M. Sobkowski et al. / Tetrahedron: Asymmetry 19 (2008) 2508–2518
Table 1
xyl group. The results are shown in Figure 11 in comparison with
those obtained under non-optimized conditions.
Chemical shifts of 31P NMR signals of diastereomers of the mixed anhydride 2 and
diastereomeric excess of 2-DP(RP) at various temperatures in toluene and in
acetonitrile
For three H-phosphonate monoesters of type 1 (B = AdeBz
,
Guaibu or Ura), the diastereoselectivity could be enhanced from
de 50–60% (the initial reaction conditions; Fig. 11, set A) to ca.
70% when the amount of pyridine was reduced (Fig. 11, set B),
and to ca. 80% when pyridine was replaced with 2,6-lutidine
(Fig. 11, set C). A further increase in de to ca. 85% was achieved
when fourfold dilution of the reactants ([1] = 25 mM) was used
along with 7% (v/v) concentration of 2,6-lutidine in DCM (Fig. 11,
set D).
Temperature (°C)
dP [2-DP(RP)]
dP [2-LP(SP)]
D
dP
de (%) (DP)
In toluene
0
1.55
1.47
1.43
1.37
1.27
1.19
1.12
1.06
2.31
2.11
1.95
1.74
1.43
1.15
0.94
0.86
ꢁ0.76
ꢁ0.64
ꢁ0.52
ꢁ0.37
ꢁ0.16
0.04
32
34
31
28
22
30
38
26
10
20
30
40
50
60
70
0.18
0.20
For cytidine H-phosphonate 1 (B = CytBz), the optimization of
the reaction conditions allowed us to increase de to ca. 75%, which
was a significant improvement in comparison to the starting value
of ca. 20%. Using an amidine protection15,32 of the cytosine
exo-amino function [e.g., 4-N-(dimethylamino)methylene or 4-N-
(N-methylpyrrolidin-2-ylidene)], it was possible to enhance the
stereoselectivity to the level of other H-phosphonates (de ca.
82%). Such derivatives are not commercially available, but can be
readily prepared32,33 by the reaction of cytidine with an appropri-
ate N,N-dialkylformamide dimethylacetal.
In acetonitrile
20
35
50
65
2.53
2.47
2.42
2.36
2.72
2.63
2.55
2.44
ꢁ0.19
ꢁ0.16
ꢁ0.13
ꢁ0.08
25
25
25
26
Reaction conditions: 0.05 mmol of
3 equiv) + PvCl (2 equiv) in appropriate solvent (0.5 mL).
1
(B = Ura) + 2,6-lutidine (3.6% v/v;
summarized in Table 1, show continuous changes in the chemical
shifts of the 31P NMR signals of both diastereomers with the con-
stant ratio of their integrals.||
3. Discussion
Temperature dependence of the condensation was investigated
for three solvents: DCM, from ꢁ78 °C to 40 °C (bp); ACN, from
ꢁ46 °C (mp) to 80 °C (bp); and toluene, from ꢁ78 °C to 80 °C
(Fig. 6). The highest stereoselectivity for all solvents investigated
was observed at ca. 20 °C.
The reaction of ribonucleoside H-phosphonates of type 1 with
hydroxylic components (suitably protected nucleosides or alco-
hols) promoted by pivaloyl chloride yields H-phosphonate diesters
of type 3 with the predominant DP(SP) configuration at the phos-
phorous centre (Fig. 1).14–17 It was established that the most plau-
sible mechanism responsible for the observed stereochemistry was
Dynamic Kinetic Asymmetric Transformation (DYKAT) (Fig. 12).22
According to this scenario, the diastereomers of the mixed anhy-
dride 2 exist in a rapid equilibrium, in which the minor diastereo-
mer of 2 (ca. 30%) was identified as having LP(SP)-configuration at
the phosphorous centre. This minor isomer is much more reactive
towards hydroxylic compounds (nucleosides or alcohols) than its
major counterpart (kD ꢂ kL ), yielding the DP(SP) H-phosphonate
2.7. Effect of the substrate concentrations on the
stereoselectivity
In the experiments described in this section, various concentra-
tions and ratios of H-phosphonate monoester 1 (B = Ura) and EtOH
were used for the condensations.
In the 31P NMR spectra of the mixed anhydride intermediate 2
(B = Ura), a downfield shift of the signals of both diastereomers
as a function of the increasing concentration of the reaction mix-
ture was observed along with a decreasing separation of the reso-
nances (Fig. 7). No signal broadening was observed and the ratio of
the diastereomers of 2 remained unchanged in these experiments
(DP:LP = 70:30 in the investigated range of 10–200 mM concentra-
tions of H-phosphonate 2).
p
p
diesters of type 3 as the prevailing products. The depletion in con-
centration of 2-LP(SP) during the esterification is rapidly replen-
ished by a fast equilibration of both diastereomers of 2, catalyzed
probably by pivalate or chloride anions present in the reaction
mixtures.àà
It was also found that the stereoselectivity of the condensation
significantly depends on the kind and excess of the alcohol used,
and on the method of activation of H-phosphonate monoester
1.22 We assumed that variations of these factors affected the rates
of P-epimerization of the diastereomers of the mixed anhydride 2
and their subsequent esterification. This suggested that the rela-
tionship between the reactions involved in the DYKAT mechanism,
which is responsible for the investigated stereoselectivity, might
be susceptible to changes in the reaction conditions, such as tem-
perature, concentration of the reactants, the type of a solvent and a
During the condensations of ribonucleoside H-phosphonate
with EtOH (Table 2), a positive effect of a low excess of both the
hydroxylic (Fig. 8) and H-phosphonic (Fig. 9) components on the
stereoselectivity in the condensation of ribonucleoside H-phospho-
nates was observed. Also the decreased concentration in both reac-
tants yielded an enhanced stereoselectivity for the condensations
(Fig. 10).
2.8. Condensation of H-phosphonates 1 with 20,30-O-dibenzoyl-
uridine
The optimized reaction conditions were validated for the inter-
nucleotide bond formation in which ribonucleoside 30-H-phospho-
nate monoesters of type 1 (B = AdeBz, CytBz, Guaibu and Ura) were
condensed with 20,30-O-dibenzoyl-uridine bearing a free 50 hydro-
The (dimethylamino)methylene protection of cytosine is very labile and was
easily cleaved during the purification process. Thus, H-phosphonate 1 (B = Cytdmm
)
was prepared from
1 ) by deprotection of the cytosine moiety with
(B = CytBz
pyridine–water–ammonia solution, followed by re-protection with N,N-dimethyl-
formamide dimethylacetal.
àà
For the reactions performed in the presence of nucleophilic catalysts, an
equilibrium of diastereomers of P-[nucleophile]+ phosphoammonium adducts was
found to take over the role of the 2-DP ꢀ 2-LP equilibrium shown in Figure 12. This
did not change the general idea of the mechanism and may be neglected in general
considerations. The effect of acid, base and nucleophilic catalysts on the ribonucle-
oside 30-H-phosphonate condensation with alcohols was discussed in detail in a
separate article.23 Among the bases investigated, 2,6-lutidine was found to be the
amine of choice (quantitative yield, de ꢀ 70%) and it was used in further experiments.
||
The minor variations in the ratio of diastereomers of the mixed anhydride 2 could
be attributed to experimental errors. Compound 2 was unstable at higher temper-
atures and its content in the reaction mixture continuously decreased during the NMR
experiment, down to only ca. 10% at 70 °C. Moreover, the 31P NMR signals of the
simultaneously formed H-phosphonate monoester
signals of 2, making precise integration difficult.
1 overlapped partly with the

M. Sobkowski et al. / Tetrahedron: Asymmetry 19 (2008) 2508–2518
2513
Figure 6. Diastereomeric excess of the DP(SP)-diastereomer of H-phosphonate diester 3b formed at various temperatures. Reaction conditions: 0.05 mmol of
1
(B = Ura) + EtOH (3 equiv) + 2,6-lutidine (3.6% v/v; 3 equiv) + PvCl (1.5 equiv) in DCM (0.5 mL).
Figure 7. 31P Chemical shifts of the diastereomers of the mixed anhydride 2 (B = Ura) [ —LP (minor); —DP (major)] in DCM as a function of concentration of 2 (DP + LP).
Table 2
Compilation of reaction conditions used for studies on the effects of the substrates
concentrations
3.1. Effect of solvents on the stereoselectivity
Entry
H-Phosphonate 1 (mM)
EtOH equiv (mM)
Best stereoselectivity
The kind of a solvent may affect stereoselectivity of the con-
densation of ribonucleoside H-phosphonates in several ways.
Firstly, since esterification of the mixed anhydride 2 is believed
to be of SN2(P) type and involves uncharged species, the increase
in polarity of a solvent should enhance the rate of this reaction,
while the reaction with the pivalate anion (P-epimerization)
might be expected to be slowed down. Furthermore, it was plau-
sible that diverse solvation of ribonucleoside 30-H-phosphonate
1 (Fig. 8)
2 (Fig. 9)
3 (Fig. 10)
25
10–200
6.25–200
1.2–150
(1700)
5.0
<3 equiv EtOH
[1] = 10 mM
[1] < 25 mM
base used, the nature of an acid and the type of a condensing agent.
These variables were investigated in this study.

2514
M. Sobkowski et al. / Tetrahedron: Asymmetry 19 (2008) 2508–2518
Figure 8. Diastereomeric excess of the DP(SP)-diastereomer of H-phosphonate diester 3b formed in the presence of different excesses of EtOH. Reaction conditions: 50
of 1 (B = Ura) + EtOH (1.2–150 equiv) + 2,6-lutidine (20 equiv) + PvCl (2 equiv) in DCM (up to total volume of 2.0 mL).
lmol
Figure 9. Diastereomeric excess of the DP(SP)-diastereomer of H-phosphonate diester 3b formed in various concentrations of H-phosphonate 1. Reaction conditions: 5–
100 mol of 1 (B = Ura) + EtOH (50 L, 850 mol) + 2,6-lutidine (100 L, 850 mol) + PvCl (3 equiv with respect of 1) in DCM (350 L). A higher excess of 2,6-lutidine and
EtOH was used in order to maintain pseudo-first order kinetic conditions for all reactions investigated.
l
l
l
l
l
l
molecules in various solvents might create conformational
changes in the ribose ring, altering in that way the spatial envi-
densations, and the reactive species formed during the course of
reactions were probably involved in complex interactions with
the solvents at different stages of the H-phosphonate diester
formation.
ronment of the activated H-phosphonate function. Additionally,
0
the solvent may affect a hypothetical P–HꢃꢃꢃO2 electrostatic inter-
action that can be considered as an underlying source of different
reactivity of the P-epimers of the mixed anhydride 2. In order to
evaluate the possible influence of different solvents on stereo-
selectivity, de was plotted against principal properties (PPs)34 or
against the relative permittivity of the solvents used. Neither of
the PPs parameters showed regularity with respect to the stereo-
selectivity achieved. However, when the solvents were arranged
according to their relative permittivity, a maximum stereoselec-
tivity was found for a moderately polar DCM (de 71%), while less
or more polar solvents gave inferior results (Fig. 3). These changes
in stereochemistry indicated that possibly neither of the solvent
effects mentioned above played a dominant role during the con-
3.2. Effect of base concentration on the stereoselectivity
According to Almer et al., the replacement of neat pyridine with
mixtures of ACN–pyridine had a profound favourable effect only on
stereoselectivity of the condensation of a guanosine derivative 1
(B = GuaPAC), while for the other ribonucleoside H-phosphonates
the effect was rather small or none at all.15 However, when
DCM–pyridine (2.4% v/v) was used as a reaction medium, an im-
proved stereoselectivity was observed for almost all 24 investi-
gated reactions of H-phosphonates 1 (B = AdeBz, CytBz, Guaibu and
Ura) with four nucleosides and ethanol.17

M. Sobkowski et al. / Tetrahedron: Asymmetry 19 (2008) 2508–2518
2515
Figure 10. Diastereomeric excess of the DP(SP) diastereomer of diester 3b formed in various dilutions of reagents. Reaction conditions: 50
l
mol of 1 (B = Ura) + EtOH
(5 equiv) + 2,6-lutidine (3.6% v/v) + PvCl (2 equiv) in DCM (up to total volume of 0.5–5.0 mL).
Figure 11. Diastereomeric excess of the DP(SP)-diastereomer of dinucleoside H-phosphonate 3a under various reaction conditions: 50 l
mol of 1 (B = AdeBz, CytBz, Guaibu or
Ura) + 20,30-di-O-benzoyluridine (1.2 equiv) + PvCl (2.5 equiv) in (A) neat pyridine (0.5 mL); (B) DCM (0.5 mL) containing pyridine (3 equiv); (C) DCM (0.5 mL) containing 2,6-
lutidine (3 equiv); (D) DCM (2.0 mL) containing 2,6-lutidine (7% v/v; 24 equiv).
In this study, we tried to optimize the contents of 2,6-lutidine in
the reaction mixtures to maximize the stereoselectivity. According
to the results shown in Figure 4, for adenosine, cytidine and uridine
H-phosphonates 1 (B = AdeBz, CytBz and Ura), the optimal concen-
tration of 2,6-lutidine was ca. 7–12% (v/v), and for guanosine deriv-
ative (B = Guaibu), ca. 2–4% (v/v). These variations in optimal base
concentration may be attributed to a different reactivity of various
nucleoside H-phosphonates, and thus, different DYKAT parame-
ters. Although the above concentrations of 2,6-lutidine gave the
highest stereoselectivity for individual nucleoside H-phospho-
nates, for practical purposes, it may be more convenient to use a
concentration of ca. 7% (v/v) of this base in DCM for all four H-
phosphonate monoesters 1. At this concentration of 2,6-lutidine,
the stereoselectivity of the condensations was close to its highest
value in all the cases.
3.3. Effect of condensing agents (C.A.’s) on the stereoselectivity
Pivaloyl chloride, which is the most commonly used C.A. for H-
phosphonate diester formation,35 was used as a primary C.A. in our
previous and current studies. However, several other C.A.’s were
also reported to be effective in H-phosphonate condensa-
tions.27,35–41 Since for various C.A.’s, different mechanisms might
be involved or the rates of the DYKAT reactions could be altered,
the kind of C.A. may affect the stereochemical outcome of
reactions.

2516
M. Sobkowski et al. / Tetrahedron: Asymmetry 19 (2008) 2508–2518
Thus, several C.A.’s, including various acyl chlorides and chloro-
(Table 1). Thus, it was concluded that the observed changes in
the chemical shifts did not reflect changes in the equilibrium rates.
The influence of temperature on the stereoselectivity of the
condensations was determined for DCM, ACN and toluene, and
the results are summarized in Figure 6. For all solvents investi-
gated, the maximum of stereoselectivity was observed at ca.
20 °C. Additionally, in DCM and toluene, a minimum stereoselec-
tivity was observed around ꢁ20 °C, while for ACN, the stereoselec-
tivity decreased steadily until the freezing point of the system was
reached. At the moment, we cannot provide a mechanistic expla-
nation for these results. Based on the collected data, the optimal
temperature of 20 °C was used in the further condensation
reactions.
phosphates, as well as representatives of other classes of activating
reagents, such as arylsulfonyl derivatives, onium salts and dii-
mides, were tested with respect to their ability to promote stereo-
selective condensation of ribonucleoside 30-H-phosphonate
monoesters (Fig. 5). The best results were obtained for acyl chlo-
rides, with PvCl being slightly better over the others; this C.A.
was therefore used in further studies. The 1.2 molar excess of PvCl
(relatively to H-phosphonate monoester 1) was found to be suffi-
cient for effective condensation of 1 with ethanol. However, in
order to avoid the detrimental effects of adventitious water, using
2–3 equiv of PvCl is advisable. These variations in the excess of
PvCl were found to have no effect on the stereoselectivity of the
reactions investigated (data not shown).
3.6. Effect of the substrates ratios on the stereoselectivity
3.4. Effects of chiral bases and chiral C.A. on the
stereoselectivity
According to the DYKAT mechanism (Fig. 12), it might be
expected that by decreasing the rates of esterification of the mixed
anhydride 2 and maintaining a high rate of 2-DP ꢀ 2-LP equilib-
rium, the stereoselectivity of the condensation could be improved.
This, in principle, could be achieved by decreasing the excess of an
alcohol used for the reaction, and indeed, it was found that the ste-
reoselectivity was increasing gradually with a decreasing concen-
tration of ethanol, reaching a plateau of de ꢀ 82% at 1.2–3 equiv
of EtOH (Fig. 8). These results were consistent with the previously
observed decrease or reversal in stereoselectivity, when the ester-
ification rate was significantly enhanced by using a large excess of
a reactive alcohol (methanol).22
According to the DYKAT mechanism of asymmetric induction
assumed for the investigated reactions, an application of chiral
axillaries should have little or no effect on the stereoselectivity of
the condensations. Indeed, the stereochemistry of reactions per-
formed in the presence of (ꢁ)-sparteine and (S)-(ꢁ)-nicotine did
not differ significantly from those in which pyridine or ( )-nicotine
was used as bases. Also, a chiral condensing agent, (ꢁ)-camphanoyl
chloride, gave similar stereochemical output as did achiral acyl
chlorides (Fig. 5). Thus, it was concluded that the chiral properties
of the reaction medium and the sense of chirality of the P-leaving
group did not alter the kD /kL ratio (Fig. 12) in the ribonucleoside
p
p
H-phosphonate condensations.
3.7. Effect of concentration of the substrates on the
stereoselectivity
1 + PvCl
In the section above, the positive effect of a low excess of the
hydroxylic component on stereoselectivity in the condensation of
ribonucleoside H-phosphonates was found. The influence of a con-
centration of H-phosphonate monoesters 1 on the stereoselectivity
was less obvious, so at first we investigated the effect of concentra-
tion of 1 on a diastereomeric composition of the intermediate
mixed anhydride 2 formed in situ. In the 31P NMR spectra, only
minor changes in the chemical shifts of the signals from 2
(B = Ura) were observed, and the ratio of its diastereomers re-
mained constant (Fig. 7). Similarly, as for the variable temperature
experiments (vide supra), these changes could not be attributed to
the averaging of the signals due to ligand exchange, as no broaden-
ing of the 31P resonances was observed. However, a larger separa-
tion of the 31P NMR signals of the diastereomers of 2 in diluted
solutions might suggest larger structural differences between the
diastereomers under such reaction conditions.
fast equilibration
2-DP(RP)
2-LP(SP)
major
minor
nucleoside
or alcohol
slow
kLp
fast
kDp
3-DP(SP)
main product
3-LP(RP)
In next experiments, H-phosphonate monoester 1 in various
concentrations (10–200 mM) was condensed with EtOH (1.7 M).
A moderate increase in stereoselectivity (de ꢀ 71%?ꢀ75%) was
observed only at the highest dilution of 1 (10 mM; Fig. 9). How-
ever, when the concentrations of 1 and EtOH were changed simul-
taneously, a distinct increase in stereoselectivity with a decreasing
concentration of the substrates (1 + 5 equiv of EtOH) was found
(from de ꢀ 70% at 100 mM of 1 to de ꢀ 81% at 25 mM of 1;
Fig. 10). Decreasing the concentration further had no effect on
the stereochemistry of the reaction, while completion of the con-
densation became difficult. Therefore, the optimal concentration
of H-phosphonate 1 was established as 25 mM.
Summarizing these model experiments, the reaction conditions
under which the highest yield and stereoselectivity of the reaction
of H-phosphonate monoester 1 with ethanol could be achieved
were found to be 25 mM concentration of H-phosphonate 1 in
DCM, 1.2 equiv of an alcohol, 2,6-lutidine (ca. 7% v/v), PvCl as a
Figure 12. The DYKAT mechanism for the stereoselective formation of the DP(SP)-
diastereomer of H-phosphonate diester 3.
3.5. Effect of temperature on the stereoselectivity
Since the temperature dependence of the equilibrium (2-
DP ꢀ 2-LP) and that of the esterification of diastereomers of the
mixed anhydride 2 might be different, the stereoselectivity might
vary with temperature. We previously found17 that the chemical
shifts of the 31P NMR signals of the individual diastereomers of
the mixed anhydride 2 in toluene were changed to a different ex-
tent with temperature, and above 50 °C the positions of the signals
were inverted. The signals, however, remained sharp in the whole
temperature range investigated and their ratio underwent only
minor variations that could be attributed to experimental errors

M. Sobkowski et al. / Tetrahedron: Asymmetry 19 (2008) 2508–2518
2517
condensing agent (2–3 equiv) and at ambient reaction
temperature.
ing under vacuum (15 min, 0.5 Torr), the flask was filled with air
dried by passing through Sicapent (Merck).
When the above conditions were applied for the condensation
of H-phosphonates 1 with a nucleoside (20,30-O-dibenzoyl uridine),
a practically quantitative yield and an excellent stereoselectivity
were obtained (31P NMR). For 30-H-phosphonates derived from var-
5.2. Standard procedure for condensation of H-phosphonates
of type 1 with ethanol
ious nucleosides, diastereomeric excess of 82–87% for B = AdeBz
,
Nucleoside H-phosphonate 1 (0.05 mmol) was dissolved in
0.5 mL of DCM, and 2,6-lutidine (3 equiv) and EtOH (3 equiv)§§
were added, followed by PvCl (1.5 equiv). 31P NMR spectra were
recorded immediately after mixing the reactants. For details and
variations, see in the text.
Cytpya, Guaibu or Ura, and of ca. 75% for B = CytBz was achieved
(Fig. 11). This was a considerable improvement in comparison to
the de of ca. 50–60% (B = AdeBz, Guaibu or Ura) and 20% for B = CytBz
when the reactions were performed in pyridine.
An application of the developed reaction conditions to the solid
phase stereoselective synthesis of oligoribonucleotide H-phospho-
nates is currently in progress in our laboratory.
5.3. General procedure for condensation of H-phosphonates of
type 1 with 20,30-O-dibenzoyl-uridine
Nucleoside H-phosphonate 1 (0.05 mmol) and 20,30-O-dibenzoyl-
uridine (1.2 equiv, 28 mg) were mixed together, rendered anhy-
drous (vide supra) and dissolved in appropriate solvent mixture
(as described in the text). To this, PvCl (2.5 equiv) was added and
after ca. 10 min of vigorous stirring 31P NMR spectra were recorded.
4. Conclusions
We have investigated the reaction conditions for the condensa-
tions of ribonucleoside 30-H-phosphonates with alcohols and
nucleosides in order to attain the highest stereoselectivity. In this
respect, the most important factors were found to be the type of
a base, concentration of the reactants and the ratio of ribonucleo-
side H-phosphonate and the hydroxylic component. The stereo-
selectivity also appeared to be sensitive to changes in
temperature of the reaction and to polar properties of the medium,
while the influence of chirality of the solvents, bases and condens-
ing agents was insignificant. Acyl chlorides, especially pivaloyl
chloride, were the condensing agents of choice.
The application of the developed conditions to the condensa-
tions of commercially available ribonucleoside H-phosphonates
bearing a standard set of protecting groups, in most cases yielded
a diastereomeric excess of ca. 85% in favour of the DP(SP)-diastereo-
mer of the formed dinucleoside H-phosphonates.
5.4. Variable temperature 31P NMR studies on the
H-phosphonic–pivalic mixed anhydride 2
Nucleoside H-phosphonate 1 (0.05 mmol) was dissolved in
0.5 mL of DCM containing 3 equiv of 2,6-lutidine. PvCl (2 equiv)
was added and the mixture was transferred to the NMR tube.
The temperature of the sample was set to the lowest investigated
value (0 or 20 °C) and increased gradually, recording the spectra
at the desired temperatures.
5.5. Condensation of H-phosphonates of type 1 with ethanol
under the Mitsunobu conditions
The optimized reaction conditions were consistent with the
assumed DYKAT scheme for the asymmetric induction, but the
mechanistic role of several factors remains unclear and requires
further investigations. This study also indicated that by tuning
the steric and electronic properties of the protecting groups in
the H-phosphonate monoesters 1, further increase in the stereo-
selectivity might be possible, and this is a subject of current study
in our laboratory.
To a solution of 0.15 mmol of diethoxytriphenylphosphorane
(prepared in situ according to the procedure of Camp and Jen-
kins46) in DCM (250
lL) a solution of 0.05 mmol of H-phosphonate
1 (B = Ura) in DCM (250 lL) was added under nitrogen atmosphere
at rt and after ca. 10 min a 31P NMR spectrum was recorded.
5.6. 50-O-Dimethoxytrityl-4-N-[N-(dimethylamino)methylene]-
20-O-tert-butyldimethylsilylcytidine 30-H-phosphonate
triethylammonium salt, 1 (B = Cytdmm
)
5. Experimental
5.1. General
H-Phosphonate 1 (B = CytBz, 2.79 g, 3 mmol) was dissolved in
pyridine (3 mL) and 25% aqueous ammonia (10 mL) was added.
The progress of the debenzoylation was monitored with TLC. After
completion of the reaction (5 h), the solvents were evaporated un-
der reduced pressure and the residue obtained was dissolved in
DCM and applied to a short silica gel column using 5% (v/v) MeOH
in DCM followed by 5% (v/v) MeOH + 2% (v/v) TEA in DCM to elute
by-products, and 10% (v/v) MeOH + 3% (v/v) TEA in DCM to elute
Chemicals, instruments and procedures were similar to those
reported earlier.22–24 The yields and diastereomeric excess were
calculated with accuracy of 1.5 percentage points (an average of
three 31P NMR measurements).
1-n-Butyl-3-methylimidazolium hexafluorophosphate ([bmim]-
[PF6]) was rendered anhydrous by triple co-evaporation with dry
acetonitrile. 5,5-Dimethyl-2-oxo-2-chloro-1,3,2-dioxaphosphor-
inane (NEP-Cl)42 and 6-nitrobenzotriazol-1-y1-oxy-tris(pyrrolidi-
no)phosphonium hexafluorophosphate (PyNOP)43 were obtained
according to the published methods. Racemization of (S)-(ꢁ)-nico-
tine was performed according to Ref. 44.
pure 1 (B = CytNH ). Yield: 2.18 g (88%). Rf 0.29 (DCM–MeOH–TEA,
2
85:10:5); 31P NMR (DCM): d = 1.67 (dd, 1JPH = 613.6 Hz,
3JPH = 9.6 Hz); 1H NMR (in CDCl3) d = 7.92 (d, 3J = 7.4 Hz, 1H, 6-H),
7.13–7.39 (m, 9H, aromatic protons), 6.88 (d, 1J = 621 Hz, 1H, P–
H), 6.78 (d, 3J = 8.6 Hz, 4H, aromatic protons), 5.93 (d, 3J = 3.2 Hz,
3
1H, 10-H), 5.40 (d, J = 7.6 Hz, 1H, 5-H), 4.67 (m, 1H, 30-H), 4.38
Nucleoside H-phosphonates 1 (B = AdeBz, CytBz, Guaibu and
(m, 1H, 20-H), 4.35 (m, 1H, 40-H), 3.72 (s, 6H, CH3O), 3.52 (dd,
Ura)45 and (B = Cytpya 32,45
were obtained according to the pub-
)
3
1J = 10.8 Hz, J = 1.5 Hz, 1H, 50-H), (dd, 1J = 10.8 Hz, 3J = 2.6 Hz, 1H,
500-H), 2.95 (q, 3J = 7.3 Hz, 6H, CH3CH2N), 1.23 (t, 3J = 7.3 Hz, 9H,
CH3CH2N), 0.89 (s, 9H, CH3C), 0.11 (2 ꢄ s, 6H, CH3Si).
lished methods. Immediately prior to a reaction, all nucleosidic
derivatives in appropriate amounts were rendered anhydrous by
dissolving in DCM (0.5 mL/0.05 mmol) followed by addition of tol-
uene (3 mL/0.05 mmol) and evaporation of these solvents under
reduced pressure (pyridine was avoided in order not to introduce
traces of a nucleophilic catalyst to the reaction mixture). After dry-
§§
Using EtOH instead of a nucleosidic substrate circumvented possible solubility
problems, while the stereochemical outcome was expected to be very similar to the
condensations with nucleosides.17

2518
M. Sobkowski et al. / Tetrahedron: Asymmetry 19 (2008) 2508–2518
Compound 1 (B = CytNH , 1.65 g, 2 mmol) was dissolved in DMF
(15 mL) after which N,N-dimethylformamide dimethylacetal
(1.5 mL) was added.32,33 After 4 h, the reaction was complete
(TLC analysis) and the reaction mixture was evaporated to dryness
under reduced pressure. Silica gel column chromatography using a
stepwise gradient of MeOH (1–5%) in DCM–TEA (99:1, v/v) as an
eluent afforded the title compound contaminated with ca. 15% of
2
15. Almer, H.; Stawinski, J.; Stromberg, R. Nucleic Acids Res. 1996, 24, 3811–3820.
16. Sobkowski, M.; Jankowska, J.; Stawinski, J.; Kraszewski, A. Nucleosides
Nucleotides Nucleic Acids 2005, 24, 1033–1036.
17. Sobkowski, M.; Jankowska, J.; Stawinski, J.; Kraszewski, A. Nucleosides
Nucleotides Nucleic Acids 2005, 24, 1469–1484.
18. Garegg, P. J.; Lindh, I.; Regberg, T.; Stawinski, J.; Stromberg, R.; Henrichson, C.
Tetrahedron Lett. 1986, 27, 4055–4058.
19. Rozners, E.; Westman, E.; Stromberg, R. Nucleic Acids Res. 1994, 22, 94–99.
20. Sobkowski, M.; Jankowska, J.; Stawinski, J.; Kraszewski, A. Stereochemistry of
Internucleotide Bond Formation by the H-Phosphonate Method. 2.
Transesterification of Aryl Ribonucleoside H-Phosphonate Diesters with
Alcohols. In Collection Symposium Series; Hocek, M., Ed.; Institute of Organic
Chemistry and Biochemistry, Academy of Sciences of the Czech Republic:
Prague, 2005; Vol. 7, pp 183–187.
21. Sobkowski, M.; Jankowska, J.; Stawinski, J.; Kraszewski, A. Nucleosides
Nucleotides Nucleic Acids 2005, 24, 887–890.
22. Sobkowski, M.; Kraszewski, A.; Stawinski, J. Tetrahedron: Asymmetry 2007, 18,
2336–2348.
23. Sobkowski, M.; Stawinski, J.; Kraszewski, A. New. J. Chem., doi:10.1039/
b812780h.
24. Sobkowski, M.; Stawinski, J.; Kraszewski, A. New. J. Chem., in preparation.
25. Perrin, D. D. Dissociation Constants of Organic Bases in Aqueous Solution;
Butterworth: London, 1965.
1 (B = CytNH ) due to on-column decomposition. Yield: 1.45 g
2
(82%). Rf 0.45 (DCM–MeOH–TEA, 85:10:5); 31P NMR (DCM):
1
3
d = 1.76 (dd, JPH = 614.5 Hz, JPH = 9.7 Hz); 1H NMR (in CDCl3)
d = 8.80 (s, 1H, N@CH–N—), d = 8.02 (d, 3J = 7.2 Hz, 1H, 6-H), 7.16–
7.42 (m, 9H, aromatic protons), 6.87 (d, 1J = 621 Hz, 1H, P–H),
6.81 (d, 3J = 8.1 Hz, 4H, aromatic protons), 6.02 (d, 3J = 3.6 Hz, 1H,
3
10-H), 5.67 (d, J = 7.5 Hz, 1H, 5-H), 4.68 (m, 1H, 30-H), 4.44 (t, 1H,
3J = 3.9 Hz, 20-H), 4.38 (m, 1H, 40-H), 3.75 (s, 6H, CH3O), 3.50 (m,
2H, 50-H, 500-H), 3.09 (2 ꢄ s, 6H, NMe2), 2.98 (q, 3J = 7.3 Hz, 6H,
CH3CH2N), 1.25 (t, 3J = 7.3 Hz, 9H, CH3CH2N), 0.88 (s, 9H, CH3C),
0.15 and 0.13 (2 ꢄ s, 6H, CH3Si).
26. CRC Handbook of Chemistry and Physics, 87th ed.; CRC Press LLC: Boca Raton,
2007.
27. Kers, I.; Stawinski, J.; Girardet, J. L.; Imbach, J. L.; Perigaud, C.; Gosselin, G.;
Aubertin, A. M. Nucleosides Nucleotides 1999, 18, 2317–2325.
28. Wozniak, L. A.; Gora, M.; Stec, W. J. J. Org. Chem. 2007, 72, 8584–8587.
29. Mitsunobu, O. Synthesis 1981, 1–28.
30. Kaminski, Z. J.; Kolesinska, B.; Kaminska, J. E.; Gora, J. J. Org. Chem. 2001, 66,
6276–6281.
Acknowledgement
The financial support from Polish Ministry of Science and High-
er Education is gratefully acknowledged.
31. Bongini, A.; Cardillo, G.; Gentilucci, L.; Tomasini, C. J. Org. Chem. 1997, 62,
9148–9153.
References
32. McBride, L. J.; Kierzek, R.; Beaucage, S. L.; Caruthers, M. H. J. Am. Chem. Soc.
1986, 108, 2040–2048.
33. Theisen, P.; McCollum, C.; Andrus, A. Nucleosides Nucleotides 1993, 12, 1033–
1046.
34. Ballistreri, F. P.; Fortuna, C. G.; Musumarra, G.; Pavone, G. ARKIVOC 2002, 54–
64.
35. Froehler, B. C.; Matteucci, M. D. Tetrahedron Lett. 1986, 27, 469–472.
36. Garegg, P. J.; Regberg, T.; Stawinski, J.; Stromberg, R. Chem. Scr. 1985, 25, 280–
282.
1. Lesnikowski, Z. J. Bioorg. Chem. 1993, 21, 127–155.
2. Stec, W. J.; Wilk, A. Angew. Chem., Int. Ed. Engl. 1994, 33, 709–722.
3. Guga, P.; Okruszek, A.; Stec, W. J. Recent Advances in Stereocontrolled
Synthesis of P-chiral Analogues of Biophosphates. In New Aspects in
Phosphorus Chemistry I; Majoral, J. P., Ed.; Springer: Berlin, Heidelberg, 2002;
pp 169–200.
4. Lu, Y. X. Mini-Rev. Med. Chem. 2006, 6, 319–330.
5. Guga, P. Curr. Top. Med. Chem. 2007, 7, 695–713.
6. Iwamoto, N.; Sato, T.; Oka, N.; Wada, T. Nucleic Acids Symp. Ser. 2006, 50, 159–160.
7. Crooke, S. T. Annu. Rev. Med. 2004, 55, 61–95.
37. Garegg, P. J.; Stawinski, J.; Stromberg, R. J. Org. Chem. 1987, 52, 284–
287.
8. Bumcrot, D.; Manoharan, M.; Koteliansky, V.; Sah, D. W. Y. Nat. Chem. Biol.
2006, 2, 711–719.
38. Andrus, A.; Efcavitch, J. W.; McBride, L. J.; Giusti, B. Tetrahedron Lett. 1988, 29,
861–864.
9. Wada, T.; Fujiwara, S.; Sato, T.; Oka, N.; Saigo, K. Nucleic Acids Symp. Ser. 2004,
48, 57–58.
39. Efimov, V. A.; Kalinkina, A. L.; Chakhmakhcheva, O. G. Nucleic Acids Res. 1993,
21, 5337–5344.
10. Sobkowski, M.; Stawinski, J.; Kraszewski, A. Nucleosides Nucleotides Nucleic
Acids 2005, 24, 1301–1307.
40. Padiya, K. J.; Mohe, N. U.; Patil, P. S.; Pal, R. R.; Salunkhe, M. M. J. Chin. Chem. Soc.
2002, 49, 1115–1118.
11. Sobkowski, M.; Stawinski, J.; Kraszewski, A. Nucleosides Nucleotides Nucleic
Acids 2006, 25, 1363–1375.
12. Sobkowski, M.; Stawinski, J.; Kraszewski, A. Nucleosides Nucleotides Nucleic
Acids 2006, 25, 1377–1389.
41. Bhongle, N. N. Methods for H-phosphonate synthesis of oligonucleotides using
triphosgene. Hybridon, Inc., Patent US 5639875, 1997.
42. McConnell, R. L.; Coover, H. W., Jr. J. Org. Chem. 1959, 24, 630–635.
43. Hoeg-Jensen, T.; Olsen, C. E.; Holm, A. J. Org. Chem. 1994, 59, 1257–
1263.
13. Sobkowski, M.; Stawinski, J.; Kraszewski, A.
A Convenient Stereochemical
Notation for P-Chiral Nucleotide Analogs. In Current Protocols in Nucleic Acid
Chemistry, Beaucage, S. L., Bergstrom, D. E., Glick, G. D., Jones R. A., Eds.; John
Wiley & Sons: New York, 2007; pp A.1E.1–A.1E.16.
44. Tsujino, Y.; Shibata, S.; Katsuyama, A.; Kisaki, T.; Kaneko, H. Heterocycles 1982,
19, 2151–2154.
45. Jankowska, J.; Sobkowski, M.; Stawinski, J.; Kraszewski, A. Tetrahedron Lett.
1994, 35, 3355–3358.
14. Almer, H.; Stawinski, J.; Stromberg, R.; Thelin, M. J. Org. Chem. 1992, 57, 6163–
6169.
46. Camp, D.; Jenkins, I. D. J. Org. Chem. 1989, 54, 3045–3049.