13C-31P couplings in the study of conformational properties of some diastereomeric nucleoside-3′-thiophosphate derivatives

Article abstract of DOI:10.1081/NCN-59691

Synthesis and stereochemical characterization of enantiomerically pure nucleoside-3-phosphorothioate esters and salts are reported. Vicinal carbon phosphorus couplings reflect different predominance of the ε conformation in the isomeric (Rp and Sp) esters

Full text of DOI:10.1081/NCN-59691

This article was downloaded by: [University of Colorado at Boulder Libraries]
On: 04 January 2015, At: 06:05
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
Nucleosides, Nucleotides and Nucleic Acids
Publication details, including instructions for authors and subscription information:
http://www.tandfonline.com/loi/lncn20
13C-31P Couplings in the Study of Conformational
Properties of Some Diastereomeric Nucleoside-3′-
Thiophosphate Derivatives
Eszter Gács-Baitz a , Gyula Sági a & Mária Kajtár-Peredy a
a Institute of Structural Chemistry, Chemical Research Center , Hungarian Academy of
Sciences , H-1025 Budapest, P.O.Box 17, Pusztaszeri út 59-67, Budapest, H-1525, Hungary
Published online: 24 Jun 2011.
To cite this article: Eszter Gács-Baitz , Gyula Sági & Mária Kajtár-Peredy (2005) 13C-31P Couplings in the Study of
Conformational Properties of Some Diastereomeric Nucleoside-3′-Thiophosphate Derivatives, Nucleosides, Nucleotides and
Nucleic Acids, 24:4, 247-257, DOI: 10.1081/NCN-59691
To link to this article: http://dx.doi.org/10.1081/NCN-59691
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

Nucleosides, Nucleotides, and Nucleic Acids, 24 (4):247–257, (2005)
Copyright D Taylor & Francis, Inc.
ISSN: 1525-7770 print/ 1532-2335 online
DOI: 10.1081/NCN-200059691
13C-31P COUPLINGS IN THE STUDY OF CONFORMATIONAL
PROPERTIES OF SOME DIASTEREOMERIC
NUCLEOSIDE-3’-THIOPHOSPHATE DERIVATIVES
5
Eszter G´acs-Baitz, Gyula S´agi, and M´aria Kajt´ar-Peredy
Institute of Structural
Chemistry, Chemical Research Center, Hungarian Academy of Sciences, Budapest, Hungary
5
Synthesis and stereochemical characterization of enantiomerically pure nucleoside-3’-phosphor-
othioate esters and salts are reported. Vicinal carbon phosphorus couplings reflect different
predominance of the e conformation in the isomeric (Rp and Sp) esters, while for the salts the et
conformation prevails in both stereoisomers. The influence of solvent and temperature on the
conformational preferences is also described.
1
Keywords Modified Nucleotides, H, 13C, 31P NMR 13C-31P Coupling Constants, Solution
Conformation
INTRODUCTION
It is now generally recognized that a correlation exists between the
stereochemical properties and the biological activity of modified oligonucleotides.
The structural and biochemical consequences of diastereoisomerism have been
extensively studied by Steck and coworkers.[1] In particular, the role of the
configuration of the phosphorothioate group in the conformational preferences was
described by Cosstick and Eckstein.[2] In most cases, the NMR methods proved to
be extremely useful in correlating molecular conformation with biological activity.
Introduction of various modifications at the phosphorus atom results in chiral
phosphate derivatives (Rp and Sp configuration). Conformational preferences
about the C3’–O3’ bond differ for the Rp and Sp stereoisomers, which are best
reflected by the systematic differences of the vicinal carbon–phosphorus coupling
3
values (DJ = JC4’,P-3JC2’,P) found for the diastereomers.[3–9] The assessment of
Received 10 December 2004, accepted 10 March 2005.
Financial support from NKFP Medichem2 (Nr.:1/A/005/2004) is gratefully acknowledged.
Address correspondence to Eszter G´acs-Baitz, Institute of Structural Chemistry, Chemical Research Center,
Hungarian Academy of Sciences, H-1025 Budapest, P.O. Box 17, Pusztaszeri u´t 59-67, Budapest H-1525, Hungary;
Fax: 36-1-325-7554; E-mail: egacs@chemres.hu
247
Order reprints of this article at www.copyright.rightslink.com

248
E. Ga´cs-Baitz, G. Sa´gi, and M. Kajta´r-Peredy
conformational equilibrium inferred from these data together with the results from
NOE and TROESY experiments provided the configurational assignment of some
diastereomeric nucleotide derivatives.
The DJ values are influenced also by factors other than the absolute
configuration. Thus the nature and size of the pendant groups at the phospho-
rus atom, the presence of the 5’-protecting group, the aromatic solvent effect, and
changes in temperature all affect the conformational properties of the modi-
fied nucleotides.
In the cases studied by us, the conformational equilibrium about the C3’–O3’
bond was primarily affected by two competing interactions of the phosphorus
substituents with the nucleobase and the C5’-ODMT-protected sugar unit.
1. Similar to the stacking interaction of two nucleobases in the dinucleotides, a
phenyl ring in the side-chain of a P-modified nucleotide is also prone to a
stacking-like interaction with the nucleobase.[6] The steric arrangements of the
stacked states, depending also on the nature of the other P-substituents, differ for
the stereoisomers.
2. Sizeable substituents at the phosphorus aim at avoiding the steric proximity with
the sugar unit and the C5’-protecting group. As a result, the predominant
conformation notably differs in the stereoisomers, as reflected by their DJ
values.[7]
In earlier papers we described the NMR characterization of some modified
nucleotides, where one of the substituents at the phosphorus was S-Me. One of
those isomer pairs (3a, 3b)[9] was synthesized from salts 2a and 2b. In this article,
we compare the conformational properties of diastereomeric 2a, 2b with 3a, 3b
and consider the effect of the introduction of a methyl group to the sulfur atom on
the chemical shifts and coupling constant values. We also reported for 3a and 3b
that the temperature dependence of the DJ values differed from that of some P-
modified parent compounds studied earlier. This difference, according to our
supposition, was connected with the altered stacking-like interactions. It seemed
reasonable, therefore, to observe the temperature-dependent changes of the DJ
values also in 2a and 2b, where one of the steric factors was absent and only the
stacking-like interaction remained. To our knowledge, this is the first case when
results of NMR experiments were reported for conformational characterization of
diastereomeric nucleotide derivatives in ionic states.
EXPERIMENTAL
Materials
N4-Benzoyl-5’-O-dimethoxytrityl-2’-deoxycytidine and 2-(4-nitrophenyl) ethanol
were purchased from Pharma Waldhof GmbH and Fluka, respectively. Pyridine

13C-31P Couplings in Diastereomeric Nucleoside-3’-Thiophosphate Derivatives
249
was distilled from P2O5 and stored over molecular sieves. Pivaloyl chloride was
freshly distilled before use.
Chromatography
For silica gel column chromatographic separations Kieselgel 60 (particle size:
0.04–0.63 mm) was used. Solvent systems applied were the following: (A): EtOAc-
MeOH (20%)-Et3N (1%), (B): EtOAc-MeOH (5%)-Et3N (1%), (C): EtOAc-Et3N (1%).
MS Spectrometry
ESI mass spectra were recorded with a Perkin Elmer SCIEX, API 2000 tandem
mass spectrometer equipped with electrospray ion source both in positive and
negative ion mode.
NMR Spectroscopy
High-resolution 1H, 13C, and 31P magnetic resonance experiments were carried
out on a Varian Unity Inova 400 NMR spectrometer equipped with a 5-mm probe
in CDCl3 and C6D6 solutions. The 1H NMR spectra were acquired with 64 scans, 4
KHz spectral width and 32 K data points, giving a digital resolution of 0.125 Hz.
The 13C spectra were accumulated with 2000–7000 scans using a spectral width of
ca 14 KHz, 64 K data points, and WALTZ decoupling. The FIDs were zero-filled to
128 K, giving a digital resolution of 0.11 Hz. The proton resonances were assigned
with the help of homonuclear decoupling experiments. The assignment of the
carbon resonances were inferred from two-dimensional heterocorrelated experi-
ments by a program incorporated in the Varian software package. The chemical
shifts are expressed on the d scale relative to internal TMS. The 13C-31P–coupling
constant values were directly determined from the proton-decoupled 13C spectra
and given in Hz. Temperature-dependent 1H, 13C, and 31P experiments were run in
the 17–65°C range in C6D6 solutions.
(RP)- and (SP)-N4-Benzoyl-5’-O-dimethoxytrityl-2’-deoxycyti-
dine-3’-O-monothio-phosphate-O-(2-(4-nitrophenyl)ethyl) ester
triethylammonium salts (2a and 2b). Coupling of N4-benzoyl-5’-O-
dimethoxytrityl-2’-deoxycytidine-3’-H-phosphonate DBU salt[10] (1)/4.88 g, 5.75
mmol with 2-(4-nitrophenyl)ethanol/0.84 g, 5.0 mmol followed by sulfurization of
the H-phosphonate diester intermediates was carried out according to the
literature.[5,11] The crude P-diastereomeric mixture of the nucleoside-3’-thiophos-
phate-O-esters obtained was dissolved in CH2Cl2-Et3N (1%)/6.0 mL and applied to a
silica gel column (320 g) packed in system (C). It was first eluted with a linear
gradient of systems (B) ! (A) 800–800 mL, finally with system (A) 500 mL. The
appropriate fractions were combined and evaporated to give the required P-
diastereomerically pure RP and SP isomers (2a and 2b) as pale yellow solid foams,
which were then dissolved in EtOAc and added dropwise to tenfold volumes of

250
E. Ga´cs-Baitz, G. Sa´gi, and M. Kajta´r-Peredy
cold hexane. The white precipitates thus obtained were filtered and dried in vacuo
over paraffin shavings. Thus we isolated 2.31 g = 2.33 mmol of (RP)-N4-benzoyl-5’-
O-dimethoxytrityl-2’-deoxycytidine-3’-O-monothiophosphate-O-(2-(4-nitropheny-
l)ethyl) ester triethyl-ammonium salt (2a) as white powder. Yield: 46.7%, Rf (A):
0.20. ESI MS m/z (%): 102.0 (100) [M + H]+ (triethylamine). For the free acid: 879.2
(14) [M + H]+, 901.0 (12) [M + Na]+, C45H43N4O11PS requires 878.23. The pure,
higher moving SP isomer (2b) was also obtained in the same form. Yield: 1.60
g = 1.62 mmol (32.4%), Rf (A): 0.24. Its MS data were identical with those of 2a.
The absolute configurations of 2a and 2b follow from those of 3a and 3b, studied
earlier.[9]
(RP)- and (SP)- N4-Benzoyl-5’-O-dimethoxytrityl-2’-deoxycyti-
dine-3’-O-monothio-phosphate-O-(2-(4-nitrophenyl)ethyl)-S-meth-
yl triesters (3a and 3b). 2a 33 mg, 33.3 mmol was dissolved in dry pyridine
3 mL then MeI 21 mL, 335 mmol was added and the mixture was stirred overnight at
ambient temperature. Since TLC indicated incomplete reaction the next morning
additional MeI 11 mL, 176 mmol and Et3N 28 mL, 202 mmol were added and the
stirring was continued for further 6 h. Although the mixture still contained some
starting material, it was worked up on the following way: CH2Cl2 12 mL was added
and the diluted solution was washed with satd. aq. NaHCO3 4 mL and brine 4 mL.
The organic phase was separated, dried with Na2SO4, filtered, and evaporated to
dryness. Pyridine traces were removed by repeated evaporations with toluene
2 ꢀ 10 mL and the residue was purified by silica gel column chromatography using
linear gradient of systems (C) ! (B) 200–200 mL as eluent. The pure fractions
were combined and evaporated to give 3a as semisolid residue. Yield: 18.3
mg = 20.3 mmol (61%), Rf (B): 0.33, ESI MS m/z (%): 893.2 (9) [M + H]+, 915.2 (100)
[M + Na]+, 931.0 (10) [M + K]+, C46H45N4O11PS requires 892.24. Starting from 2b
the corresponding pure SP S-methyl triester (3b) was prepared and isolated in the
same way. Yield: 19.5 mg = 21.6 mmol (65%), Rf (B): 0.36. Its MS data were
essentially identical with those of 3a.
RESULTS AND DISCUSSION
Synthesis
N4-Benzoyl-5’-O-dimethoxytrityl-2’-deoxycytidine-3’-H-phosphonate-1,8-diazabi-
cyclo [5.4.0] undec-7-ene (DBU) salt (1) was prepared from the protected
nucleoside (N4-Bz-5’-DMT-dC) according to the literature.[10]
Coupling of 1 with 2-(4-nitrophenyl)ethanol (see Scheme 1) was accomplished
applying standard H-phosphonate nucleotide synthesis protocol[11] resulting in
the corresponding N4-benzoyl-5’-dimethoxytrityl-2’-deoxycytidine-3’-H-phospho-
nate-O-(2-(4-nitrophenyl)ethyl) esters as a P-diastereomeric mixture. It was
subsequently oxidized with elemental sulfur to give the chemically much more

13C-31P Couplings in Diastereomeric Nucleoside-3’-Thiophosphate Derivatives
251
SCHEME 1 a, i. 2-(4-Nitrophenyl)ethanol, pivaloyl chloride, pyridine, 25°C, 30 min. ii. Sulfur, pyridine-CS2,
25°C, 16 h. iii. Silica gel chromatography. b, i. MeI, pyridine, Et3N, 25°C, 24 h. ii. Silica gel chromatography.
Scheme: Synthesis of N4-Bz-5’-DMT-dC-3’-monothiophosphate-O-(p-nitrophenylethyl)-S-methyl triesters.
stable N4-benzoyl-5’-dimethoxytrityl-2’deoxycytidine-3’-O-monothiophosphate-O-(2-
(4-nitrophenyl)ethyl) esters (2a and 2b). Separation of the P-isomers by silica gel
column chromatography was carried out at this stage providing the pure RP (2a)
and SP (2b) diastereomers in good overall yield. Probably due to electron
withdrawing effect of the 2-(4-nitrophenyl)ethyl group, the S-methylations, even with
large excess of MeI, proceeded slowly and gave the required S-methyl triesters (3a
and 3b) with significantly lower yields compared to analogous cases when there
were no electron-withdrawing substituents on the thiophosphate moiety.[5,9]
NMR Study
Earlier, the configurational determination of the P-isomers was based on the
phosphorus chemical shifts. Inspection of Table 1 reveals that these data hardly
differ in the stereoisomers, thus stereochemical assignment cannot be based on
them. Similarly, minor differences were observed between the isomers in the other
parameters collected in the table, such as carbon chemical shifts and proton-
phosphorus couplings. Larger deviations were observed when comparing the esters
(3a, 3b) and salts (2a, 2b), since the above parameters reflect the changes of the
TABLE 1 Selected NMR Parameters Reflecting Changes of Electrondensity Around the Phosphorusa
dP (ppm)b
dC3’ (ppm)c
dOCH2 (ppm)c
3JH3’ P
,
(Hz)
3JCH2 P
,
(Hz)
2JOCH2 P
, (Hz)
2a
2b
3a
3b
57.93
58.20
29.75
29.77
75.95
75.64
77.46
77.66
65.73
65.78
67.18
67.16
8.6
8.5
8.0
7.8
8.2
8.3
7.4
7.5
5.6
5.5
6.1
5.9
aIn CDCl3 solutions.
bRelative to external H3PO4.
cRelative to internal TMS.

252
E. Ga´cs-Baitz, G. Sa´gi, and M. Kajta´r-Peredy
TABLE 2 1H Chemical Shifts (dppm)a and Proton–Proton Couplings ( J, Hz) of Diastereomers 2--3a,b
2a
2b
3a
3b
H1’
H2’a
6.28 (6.5 + 5.9)
2.93 (ꢁ14.0 +
5.9 + 3.1)
6.28 (6.3 + 5.9)
2.89 (ꢁ14.0 +
5.9 + 3.1)
6.32 (6.8 + 6.1)
2.93 (ꢁ14.3 +
6.1 + 3.0)
6.29 (6.9 + 6.0)
2.92 (ꢁ14.3 +
6.0 + 2.9)
H2’b
H3’
2.28 (ꢁ14.0 +
2.26 (ꢁ14.0 +
2.34 (ꢁ14.3 +
2.33 (ꢁ14.3 + 6.9 +
6.5 + 6.3)
6.3 + 6.2)
6.8 + 6.1)
6.1 + 1.7)
5.25 (6.3 + 3.1 +
2.9 + 8.6)
5.15 (6.2 + 3.1 +
3.4 + 8.5)
5.18 (6.1 + 3.0 +
3.1 + 8.0)
5.14 (6.1 + 2.9 +
3.0 + 7.8)
H4’
H5’
4.37 (2.9 + 3.4 + 3.4)
3.35 (3.4)
4.45 (3.4 + 3.2 + 3.2)
3.47 (3.2)
4.38 (3.1 + 3.4 + 3.4)
3.43 (ꢁ10.6 + 3.4)
3.49 (ꢁ10.6 + 3.4)
3.13 (m)
4.33 (m)
7.25 (7.6)
4.39 (3.0 + 3.4 + 3.2)
3.45 (ꢁ10.7 + 3.4)
3.48 (ꢁ10.7 + 3.2)
3.13 (m)
4.34 (m)
7.26 (7.5)
Ph-CH2
OCH2
C5-H
C6-H
S-Me
3.08 (m)
4.16 (m)
7.30 (7.5)
8.16 (7.5)
—
3.05 (m)
4.18 (m)
7.30 (7.5)
8.18 (7.5)
—
8.13 (7.6)
2.12 (15.5)
8.14 (7.5)
2.18 (15.4)
aIn CDCl3, relative to TMS.
electron density around the phosphorus. In agreement with literature data,[12,13] the
increased electron density of the molecules in ionic states results in mighty increase
in the phosphorus chemical shift and smaller, but still characteristic changes in the
other NMR parameters of Table 1. Other spectral parameters, which were found to
be characteristic for the conformational properties of the sugar ring and for the
configuration at the phosphorus, are listed in Tables 2 and 3.
The proton–proton coupling constants are known to provide information
about the preferred conformation of the sugar ring (S and N, see Figure 1). From
the sum of the H2’a couplings, with the help of an empirical expression[14] the
percentage of the S-type (C2’-endo) conformer can be estimated. We have found
for all of our modified 5’-protected mono- and dinucleotides that in solution the S-
type sugar conformation is predominant. The S-character had the largest value for
mononucleotides where stacking-like interactions were absent (94–99%).[4] For all
TABLE 3 13C Chemical Shifts (dppm) of the Pertinent Carbons and 13C-31P Couplings ( J, Hz) of
Diastereomers 2--3a,ba
2a
2b
87.63
3a
87.08
3b
87.35
C1’
87.60
C2’
C3’
C4’
C5’
41.29 (3.4)
75.95 (broad)
85.90 (6.6)
63.43
40.91 (3.1)
75.64 (broad)
86.13 (7.1)
63.37
40.90 (3.1)
77.46 (5.6)
85.19 (7.2)
62.73
40.61 (5.1)
77.66 (5.5)
85.65 (5.3)
62.79
S-CH3
Ph-CH2
O-CH2
DJ
—
—
12.58 (4.8)
36.52 (7.4)
67.18 (6.1)
4.1
12.60 (4.8)
36.51 (7.5)
67.16 (5.9)
0.2
36.86 (8.2)
65.73 (5.6)
3.2
36.91 (8.3)
65.58 (5.5)
4.0
aIn CDCl3 solutions, at 30°C.

13C-31P Couplings in Diastereomeric Nucleoside-3’-Thiophosphate Derivatives
253
FIGURE 1 Equilibrium between the two extreme sugar puckers, north and south.
four molecules presented here, the calculated percentage of S-conformations are
definitely lower (74–79%) and are close to those of the natural dinucleotides.[15] The
lower value indicates that the conformational equilibrium is somewhat biased
toward the N-puckered form. The conformational distribution of the pentose,
however, does not show any meaningful changes on going from ionic to ester-type
molecules. Furthermore, no significant differences can be seen in the proton
parameters of the stereoisomers (Table 2).
Conformational preferences about the various bonds in nucleotides and
deoxynucleotides are known to be interdependent,[16] consequently, the confor-
mational variations in the sugar ring pucker may influence the torsional preferences
about the C3’–O3’ bond. These characteristics can be inferred from the proton-
decoupled 13C spectra of the compounds (Table 3).
Applying a Karplus-type relationship, Lankhorst and coworkers[15] have
demonstrated for ribonucleotides and deoxyribonucleotides that the vicinal
13C-31P couplings reflect the conformational preferences about the C3’–O3’ bond.
3
3
In phosphate-modified nucleotides, the difference of the JC4’,P and JC2’,P vicinal
coupling values (DJ parameter) was introduced[3] to provide information about the
preferred et (trans) and eꢁ (gauche-) conformations (Figure 2). Accordingly, large
positive DJ value refers to prevalent et conformation, while an increase of the eꢁ
conformation in the conformational equilibrium reduces the DJ values. Negative DJ
values are indicative of the preferred eꢁ conformation about the C3’–O3’ bond. It
was found for all diastereomer pairs studied by us that different chirality at the
FIGURE 2 Newman projection about the C3’–O3’ bond.

254
E. Ga´cs-Baitz, G. Sa´gi, and M. Kajta´r-Peredy
phosphorus was always accompanied by characteristic conformational differences
about the C3’–O3’ bond, as reflected by their DJ values.[3–9]
In our earlier studies, the largest differences in the DJ values of the
diastereomers were found for mononucleotides with S-Me and Se-Me substituents
(DDJ = 5.2 and 7.0 Hz, respectively), where the other substituent on the phosphorus
was a methyl group, thus stacking, or stacking-like, interactions could not be
formed.[4] On the contrary, we supposed for 3a and 3b that the phenyl ring in the
side chain may take part in a stacking-like interaction with the nucleobase.[9] As a
result, the restricted rotation about the C3’–O3’ bond is reflected by a smaller but
still characteristic DJ difference in the diastereomers (DDJ = 3.9 Hz in CDCl3
solution). These conformational differences in the isomers were sufficient to identify
the absolute configuration at the phosphorus by T-ROESY experiments.[9] Since 3a
and 3b were derived from 2a and 2b, respectively, considering also that the CIP
priorities of the functional groups have not changed (Sꢁ > nucleoside 3’-
O > O(CH2)2pNO2PH > = O), the configuration of this latter isomeric pair directly
follows from the esters of known configurations (2a = Rp, 2b = Sp).
In salts 2a and 2b one interaction, namely that of the S-Me group with the
protected sugar unit, is absent and solely the stacking interaction remains. In the
absence of the conformational perturbance caused by the different configuration of
the S-Me group, the extent of the stacking in the isomers 2a and 2b becomes
rather similar and differs from that of the corresponding esters (Figure 3). As a
result, the preferred conformation about the C3’–O3’ bond is predominantly et in
both isomers, as reflected by their DJ values (3.2 and 4.0 Hz, respectively, in CDCl3
solution). This steric arrangement can be correlated with that of the natural
dinucleotides for which the vicinal couplings at room temperature reflect the
predominancy of the et conformation (Figure 4). The DJ value (3.3 Hz) calculated
from the respective vicinal 13C-31P couplings of 2’-deoxycytidilyl(3’ ! 5’)-2’-
deoxycytidine (CpC)[15] is in good agreement with that of 2a and 2b.
Other similarities can be found in the temperature dependence of 2a and 2b
when compared with the natural dinucleotides, where higher temperature induced
destacking and detachment of the interacting nucleobases. As a result, the vicinal
FIGURE 3 Changes in conformational preferences on going from esters 3a,3b to salts 2a,2b.

13C-31P Couplings in Diastereomeric Nucleoside-3’-Thiophosphate Derivatives
255
FIGURE 4 Schematic representation of base–base and base–phenyl overlaps in the stacked states.
13C-31P couplings reflected an increase of the eꢁ conformer population.[15]
According to our calculations, the decrease in DJ of CpC is 1.8 Hz in 79°C
temperature range.
Due to their tendency to decompose at elevated temperatures, the temperature-
dependent measurements of compounds 2a, 2b, 3a, and 3b were accomplished
in deuterobenzene solutions. The aromatic solvent effect and change of
temperature are both known to influence the conformational preferences;
consequently, the DJ values. The solvation in benzene enhanced the extent to
which the nucleobase and the phenyl ring in the side-chain overlap. The changes in
the DJ values demonstrate that the more extensive overlap in C6D6 solution induces
TABLE 4 Temperature Dependence of DJ Valuesa
17°C
30°C
45°C
60°C
2a
2b
4.6
3.1b
4.5
3.2
4.3
2.5
4.0
2.1
25°C
35°C
50°C
65°C
3a
3b
5.3
1.0b
5.1
1.1
4.8
1.0
4.4
1.0
aIn C6D6, in Hz.
bCalculated from broad signals.

256
E. Ga´cs-Baitz, G. Sa´gi, and M. Kajta´r-Peredy
opposite shift in the conformational equilibrium of the isomers 2a and 2b relative
to their DJ values in CDCl3 (Table 4).
Elevation of temperature over the 17–65°C temperature range and consequent
destacking caused increase of population in the eꢁ conformer of 2a and 2b, and a
similar trend was found for 3a (Table 4). The temperature-induced behavior is in
line with that found in natural dinucleotides and differs from that observed in some
modified derivatives studied earlier, where the change of DJ values reflected shifts
toward the et conformation.[6,7] 3b presents here the only exception, where the DJ
values remained practically invariant over the temperature range. This insensitivity
to temperature changes reflects that in this stereoisomer the steric control imposed
on the eꢁ–et equilibrium by the S-Me group may have a more significant role than
the tendency to destacking.
The above temperature dependence of compounds 2a, 2b, and 3a in
comparison with other modifed nucleotides[6,7] may be attributable to numerous
factors. However, the type of the nucleobases cannot be as significant as we
supposed earlier,[9] since in mononucleotides 2a, 2b, and 3a the heating caused
the same, anticlockwise rotation about the C3’–O3’ bond as in the natural
dinucleotides. Furthermore, the presence of the bulky 5’-ODMT group cannot be
responsible for the different temperature dependence, either, since it was present in
all of the cases we studied earlier. An important structural feature of the previously
studied thiophosphoric-acid-O,O,O-triester derivatives[6,7] was that the double-
bonded heteroatom was a more voluminous and less electronegative sulfur than the
oxygen atom in the present compounds and the natural dinucleotides. In order to
demonstrate such heteroatom effects, a more systematic examination is in progress.
CONCLUSION
The present study shows that salts 2a and 2b have a comparable distribution
of e conformers. Comparison of diagnostic NMR data of 2a and 2b with those of
unmodified CpC demonstrated that neither P-Oꢁ! P-Sꢁ exchange nor the
presence of a p-nitrophenyl-ethyl moiety instead of a second nucleoside unit may
not markedly affect the conformational preferences about the C3’–O3’ bond. As a
consequence, the predominant conformations are et in both stereoisomers. For 2a
and 2b, the change in the preferred e conformation upon elevation of temperature
also resembles that found for unmodified dinucleotides. One can thus anticipate
that replacement of a second nucleoside moiety to a side-chain carrying a phenyl
group does not induce substantial changes in the stacking ability. Contrary to this,
our study here has shown that methylation at the sulfur causes sizeable differences
in the conformational equilibrium of 3a and 3b, as reflected by their DJ values.
This can be interpreted by the stacked states, which differ for the stereoisomers.
Differences are also evident in the temperature-dependent conformational changes
of the two isomeric esters, which can be attributed to the significant role of the S-Me
substituent on the stereochemical constraints.

13C-31P Couplings in Diastereomeric Nucleoside-3’-Thiophosphate Derivatives
257
REFERENCES
1. Steck, W.J.; Wilk, A. Stereocontrolled synthesis of oligo-nucleoside-phosphorothioates. Angew. Chem., Int.
Ed. Engl. 1994, 33, 709–722.
2. Cosstick, R.; Eckstein, F. Synthesis of D(CG)octamers containing alternating phosphoro-thioate linkages.
Effect of the phosphorothioate group on the B-Z transition. Biochemistry 1985, 24, 3630–3638.
3. G´acs-Baitz, E.; To¨mo¨sko¨zi, I.; Kajt´ar-Peredy, M. Configurational determination and conformational properties
of diastereomeric nucleoside-phosphorothioates. Tetrahedron: Asymmetry 1996, 7, 2447–2452.
4. G´acs-Baitz, E.; Wozniak, L.A.; Kajt´ar-Peredy, M. Conformational properties of diastereomeric 5’-ODMT-
2’deoxythymidine 3’-O-(S-methyl-methanephosphono-thiolate)s and (Se-methyl-methanephophono-seleno-
late)s in solution as studied by NMR methods. Chirality 2000, 12, 675–680.
5. Machytka, D.; S´agi, G.; Kajt´ar-Peredy, M.; G´acs-Baitz, E. Synthesis and NMR characterization of
diastereomeric CPSMe derivatives. Nucleosides Nucleotides 2000, 19, 903–915.
6. Kajt´ar-Peredy, M.; To¨mo¨sko¨zi, I.; G´acs-Baitz, E. Synthesis and stereochemical characterization of
diastereomeric nucleoside-phosphorothioselenoates by NMR methods. Nucleosides Nucleotides Nucleic
Acids 2001, 20, 1615–1623.
7. G´acs-Baitz, E.; Kajt´ar-Peredy, M. Solution structure of diastereomeric thymidine 3’-O(N,N-diisopropyl-O-b-
cyanoethyl-phosphoramidothioate)s by NMR spectroscopy. Chirality 2002, 14, 814–818.
8. Machytka, D.; G´acs-Baitz, E.; Tegyey, Z. Extension of the applicability of DJ values for the configurational
assignment of diastereomeric phosphate-modified dideoxy-nucleotides. Nucleosides Nucleotides 1998, 17,
2311–2322.
9. G´acs-Baitz, E.; Kajt´ar-Peredy, M.; S´agi, G. NMR study of some diastereomeric nucleoside-3’-thiophosphate
derivatives in solution. Anal. Bioanal. Chem. 2004, 379, 56–59.
10. Froehler, B.C.; Mateucci, M.D. Synthesis of DNA via deoxynucleoside H-phosphonate intermediates. Nucleic
Acids Res. 1986, 14, 5399–5407.
11. Seela, F.; Kretschmer, U. Stereochemistry of oxidation of diastereoisomeric d(TpA) phosphonates with
sulphur and iodine[18O] water. J. Chem. Soc., Chem. Commun. 1990, 1154–1159.
12. Gorenstein, D.G.Phosphorus-31 NMR: Principles and Applications. Academic Press: Orlando, FL, 1984; 7–
36.
13. Glemarec, C.; Reynolds, R.C.; Crooks, P.A.; Maddry, J.A.; Akhtar, M.S.; Montgomery, J.A.; Secrist, J.A., III;
Chattopadhyaya, J. Conformational studies of thymidine dimers containing sulfonate and sulfonamide
linkages by NMR spectroscopy. Tetrahedron 1993, 49, 2287–2298.
14. Wijmenga, S.S.; Mooren, M.M.V.; Hilbers, C.W.NMR of Nucleic Acids, from Spectrum to Structures.
Oxford University Press: Oxford, 1993; 266–267.
15. Lankhorst, P.P.; Haasnoot, C.A.G.; Erkelens, C.; Westerink, H.P.; van der Marel, G.A.; van Boom, J.H.;
Altona, C. Carbon-13 NMR in conformational analysis of nucleic acid fragments. The torsion angle
distribution about the C3’–O3’ bond in DNA constituents. Nucleic Acids Res. 1985, 13, 927–942.
16. Cheng, D.M.; Sarma, R.H. Intimate details of the conformational characteristics of deoxyribonucleoside
monophosphates in aqueous solution. J. Am. Chem. Soc. 1977, 99, 7333–7348.