Oligonucleotide analogues with a 'nucleobase-including backbone': Part 10 - Design, synthesis, and association of ether-linked dimers

Article abstract of DOI:10.1002/hlca.200490204

The dinucleoside analogues 24, 25, 28-30, and 33 associate in CDCl ₃ solution. Association constants, as determined from the concentration-dependent chemical shift for H-N(3) of the uridine moiety and from thermodynamic parameters, range from 265 M^-1 (33) to 3220 M ^-1 (30). The association of 31 in CDCl₃ is too strong to be determined (concentration independent δ5(H-N(3)) of ca. 12.8 ppm) and the fully deprotected dimer 32 proved insufficiently soluble in CDCl ₃. This observation strongly evidences that structural differentiation of oligonucleotides and their analogues into backbone and nucleobases is not required for pairing. The dinucleotide analogues were prepared by O-alkylation of C(8)-unsubstituted or of C(8)-oxymethylated, partially protected adenosines by the C(6)-mesyloxy- or C(6)-halomethylated uridines 20-22, followed by partial or total deprotection.

Full text of DOI:10.1002/hlca.200490204

Helvetica Chimica Acta Vol. 87 (2004)
2273
Oligonucleotide Analogues with a −Nucleobase-Including Backbone×
Part 10
Design, Synthesis, and Association of Ether-Linked Dimers
by Andrew John Matthews, Punit Kumar Bhardwaj, and Andrea Vasella*
Laboratorium f¸r Organische Chemie, ETH-Hˆnggerberg, HCI, CH-8093 Z¸rich
The dinucleoside analogues 24, 25, 28 30, and 33 associate in CDCl3 solution. Association constants, as
determined from the concentration-dependent chemical shift for HÀN(3) of the uridine moiety and from
thermodynamic parameters, range from 265 mÀ1 (33) to 3220 mÀ1 (30). The association of 31 in CDCl3 is too
strong to be determined (concentration independent d(HÀN(3)) of ca. 12.8ppm) and the fully deprotected
dimer 32 proved insufficiently soluble in CDCl3. This observation strongly evidences that structural
differentiation of oligonucleotides and their analogues into backbone and nucleobases is not required for
pairing. The dinucleotide analogues were prepared by O-alkylation of C(8)-unsubstituted or of C(8)-
oxymethylated, partially protected adenosines by the C(6)-mesyloxy- or C(6)-halomethylated uridines 20 22,
followed by partial or total deprotection.
Introduction. The most-conspicuous property of nucleic acids is their ability to
form stable duplexes by selective base pairing, the sequence of bases leading to the
structural diversity required for their biological role. In keeping with the central nature
of this role, the structure of nucleic acids has been modified by changing the bases, the
(deoxy)ribose moiety, and the phosphodiester linker, while maintaining the differ-
entiation of nucleobase and backbone. This constant leads to the question of whether a
dedifferentiation of backbone and nucleobase is compatible with pairing. We have
attempted to answer this question by designing and synthesising oligonucleotide
analogues that no longer differentiate between backbone and base [1].
We first prepared protected phosphodiester-linked decamers 1 and 2, derived from
C(6)-hydroxymethylated uridine dU*1) and C(8)-hydroxymethylated adenosine dA*,
respectively [2 4]. Whilst decamer 1 was readily deprotected, the dA* decamer 2
decomposed, presumably by facile elimination of the phosphate group at CH2C(8),
leading to the stabilised cation A.
A second-generation of neutral, ethynediyl-linked analogues is illustrated by the
adenosine tetramer 3 [5 7] and uridine hexamer 4 [1][8] (cf. Fig. 2). The tetramer 3
proved poorly water-soluble, whilst the hexamer 4 showed the limits of the synthetic
methodology. The pentamer corresponding to 4 did not undergo hetero-pairing with a
1
)
Conventions for abbreviated notation: The substitution at C(6) of pyrimidines and C(8) of purines is
denoted by an asterisk (*); for example, dU* and dA* for the hydroxymethylated uridine and adenosine 2-
deoxyribose derivatives, and U* and A* for the corresponding ribose derivatives. The moiety linking
C(6)CH2 or C(8)CH2 and C(5') of the previous unit is indicated in square brackets, such as -[O]- for
oxygen.
¹ 2004 Verlag Helvetica Chimica Acta AG, Z¸rich

2274
Helvetica Chimica Acta Vol. 87 (2004)
complementary RNA heptamer. Modeling suggested that an anti-conformation of
these analogues is a prerequisite for pairing, whilst NMR analysis of an adenosine
dimer showed that a syn-conformation is preferred [9].

Helvetica Chimica Acta Vol. 87 (2004)
2275
These results led us to design a third generation of analogues capable of duplex
formation with bases in a syn-conformation. This is known to occur in Z-DNA [10][11],
whose biological roles are drawing much interest [12]. A Me substituent at C(8) of
guanine stabilises the syn-form of short oligonucleotides in a variety of sequences [13].
Modeling studies2) suggested that oxymethylene-bridged oligomers (Fig. 1) should
form duplexes with a right-handed B-helix (seven residues per turn), maintaining a
parallel orientation of WatsonÀCrick base pairs at a distance of 2.96 ä, comparable to
the distance between base pairs in RNA duplexes (2.6 3.3 ä [10]).
Fig. 1. Oxymethylene-bridged analogues derived from U* and A*
Our initial goal was to prepare uridine U*-[O]-U*1), adenosine A*-[O]-A*, and
self-complementary U*-[O]-A* and A*-[O]-U* dimers, and to analyse their pairing
properties.
Results and Discussion. 1. Monomer Synthesis. We have briefly communicated the
preparation of the (4,4'-dimethoxytrityloxy)methyl adenosine 9 (Scheme 1) and the
hydroxymethylated uridine 15 [15] (cf. Scheme 2), and we now provide details.
N6-Benzoyl-2',3'-O-isopropylideneadenosine (5) [16] was silylated to 6, formylated
[2][17], and then reduced to the C(8)-hydroxymethylated 7 (71% from 5). This
intermediate was, on the one hand, 4,4'-dimethoxytritylated to 8 and then desilylated to
the alcohol 9 (84% from 7). On the other hand, sequential deprotection of 7 by N-
debenzoylation to 10 (96%), desilylation to 11 (87%), and deisopropylidenation
yielded 8-(hydroxymethyl)adenosine (12; 92%).
The alcohols 5, 9, and 11 form a strong intramolecular H-bond from HOÀC(5') to
N(3), as evidenced by d(HOÀC(5')) in CDCl3 (5.8 6.7 ppm), a small J(5'a,OH)
(ꢀ2.2 Hz), a large J(5'b,OH) value (ꢁ10.3 Hz), and small J(4',5'a) and J(4',5'b) values
(ꢀ 2.2 Hz) indicating anti-periplanar HÀC(4') and C(5')ÀO bonds (see [5]). This H-
bond requires a syn-conformation, with the sugar adopting predominantly the (S)-
conformation in CDCl3 ((S)/(N) 77:23 to 83 :17 as deduced from J(1',2')/J(3',4') in
Table 1; [18]). A bifurcated intramolecular H-bond of HO(5') to N(3) and O(4')
characterised by a O(5')H ¥¥¥ N(3) and O(5')H ¥¥¥ O(4') distances of 1.92 and 2.48ä is
also found in the solid state of 113) (Fig. 2).
2
)
)
Maruzen model studies and Macromodel calculations using the Amber* force-field [14].
The crystallographic data have been deposited with the Cambridge Crystallographic Data Centre as
deposition No. CCDC-216413 (11) and No. CCDC-216414 (18). Copies of the data can be obtained, free of
charge, on application to the CCDC, 12 Union Road, Cambridge CB2 1EZ, UK (fax: ‡ 44(1223)336033;
e-mail: deposit@ccdc.cam.ac.uk).
3

2276
Helvetica Chimica Acta Vol. 87 (2004)
Scheme 1
DMTrˆ 4,4'-Dimethoxytrityl
a) (i-Pr)3SiCl (TIPSCl), 1H-imidazole, 4-(dimethylamino)pyridine (DMAP), CH2Cl2; 96%. b) Lithium
diisopropylamide (LDA), THF, then DMF, then AcOH, À 708, then EtOH, NaBH4, 238; 74%. c) DMTrCl,
EtN(i-Pr)2, DMAP, CH2Cl2; 94%. d) Bu4NF (TBAF), THF; 89% of 9; 87% of 11. e) NH3, H2O/MeOH; 96%.
f) HCO2H/H2O 4 :1; 92%.
As a rule, the syn-conformation of 5'-O-protected adenosines is revealed by a
downfield shift for HÀC(2') and an upfield shift for C(2') [19]. The significant
downfield shift for HÀC(2') (Dd(HÀC(2')) ˆ 0.40 0.49 ppm), and upfield shift for
C(2') (Dd(C(2')) ˆ À1.4 to À 1.8ppm) of the C(8)-substituted silyl ethers 7, 8, and 10,
as compared to the corresponding signals for 6, evidence that C(8)-substitution leads
predominantly to a syn-conformation (Table 1). The J(4',5'a) and J(4',5'b) values for 7,
8, and 10 (5.6 6.5 Hz; see Table 4 in Exper. Part) indicate some steric interaction
between the nucleobase and the TIPSOÀC(5') (TIPS ˆ (i-Pr)3Si); the confomers
possessing anti-periplanar HÀC(4') and HÀC(5') bonds are favoured. The (S)/(N)
conformations of the fully protected derivatives 6 and 8 are equally populated in CDCl3
(J(1',2')/J(3',4') ˆ 47:53), whilst the methanols 7 and 10 slightly prefer the (N)-
conformation (J(1',2')/J(3',4') ˆ 39 :61). The isopropylidenated silyl ethers 6 8 and 10

Helvetica Chimica Acta Vol. 87 (2004)
2277
Fig. 2. Crystal structures of the diols 11 and 183)
show a somewhat decreased ring pucker (J(1',2') ‡ J(3',4') ˆ 5.3 5.8Hz; see [20]), as
compared to the isopropylidenated alcohols 5, 9, and 11 (6.0 6.5 Hz). De-isopro-
pylidenation increases the ring pucker (9.1 Hz for 12).
Table 1. 1H- and 13C-NMR Shift Differences for HÀC(2') and C(2') of the Adenosine-Derived Monomers 7 11
as Compared to 6, and the Uridine-Derived Monomers 13 and 15 18 as Compared to 14, and Ratio of (S)/(N)-
Conformers of these Nucleosides, as Deduced from J(1',2')/J(3',4')
Nucleoside
Dd(HÀC(2')) [ppm]
À 0.10
Dd(C(2')) [ppm]
À 1.5
J(1',2') [Hz]
J(3',4') [Hz]
(S)/(N)
5
6
7
8
9
10
11
13
14
15
16
17
18
5.0
2.5
2.2
3.1
5.0
2.2
5.0
3.1
2.83.4
1.2
1.2
2.847 : 53
3.4
47 : 53
1.5
3.4
1.0
3.4
45 : 55
4.5
4.5
81 : 19
a
a
)
)
0.40
0.49
À 1.6
À 1.82.7
À 2.0
À 1.4
À 2.0
0.7
39 : 61
À 0.03
77 : 23
39 : 61
83 : 17
48: 52
0.41
À 0.04
0.34
b
b
)
)
0.50
0.50
0.53
0.55
À 1.3
À 1.3
À 2.2
À 1.7
21 : 79
21 : 79
39 : 61
33 : 67
1.2
2.5
2.2
3.9
4.4
a) d(HÀC(2')) ˆ 5.32 and d(C(2')) ˆ 84.6 ppm. b) d(HÀC(2')) ˆ 4.69 and d(C(2')) ˆ 85.4 ppm.
The C(6)-hydroxymethylated 15 was prepared from the protected uridine 13 [21] by
silylation to 14 [1], formylation [2][22], and reduction (67% from 13; Scheme 2). The
intermediate 15 was, on the one hand, 4,4'-dimethoxytritylated to 16 (95%), which was
desilylated to the alcohol 17 (89%) and, on the other hand, desilylated to the diol 18
(89%) and further deisopropylidenated to 6-(hydroxymethyl)uridine (19; 90%).

2278
Helvetica Chimica Acta Vol. 87 (2004)
Scheme 2
DMTrˆ 4,4'-Dimethoxytrityl
a) TIPSCl, 1H-imidazole, DMAP, CH2Cl2; 95%. b) LDA, THF, then DMF, then AcOH, À 708, then EtOH,
NaBH4, 238; 70%. c) DMTrCl, EtN(i-Pr)2, DMAP, CH2Cl2; 95%. d) TBAF, THF; 89% of 17; 89% of 18.
e) HCO2H/H2O 4 :1; 90%.
In polar solvents, like DMSO, uridines prefer the anti-conformation [23]. In apolar
solvents, an intramolecular O(5')H ¥¥¥ OˆC(2) H-bond may favour the syn-conforma-
tion [24]. The persistence of this H-bond may be derived from the J(5'a,OH) and
J(5'b,OH) values [25]. MM3* Calculation and the Karplus equation of Fraser et al. [26]
suggest J(5'a,OH) and J(5'b,OH) values of 0.5 and 11.5 Hz, respectively, for the
intramolecularly H-bonded species of 13. The CDCl3 1H-NMR spectrum of a saturated
solution of the sparingly soluble 13 shows broad signals of NH, HOÀC(5'), and
HbÀC(5') due to soluteÀsolute interactions. However, sharp signals for HOÀC(5'),
HaÀC(5'), and HbÀC(5') of a one-third-saturated solution of 13 allowed us to assign
J(5'a,OH) ˆ 3.6 and J(5'b,OH) ˆ 7.5 Hz (see Table 6 in the Exper. Part), evidencing a
partly persistent intramolecular H-bond. With 0.5 and 11.5 Hz as limiting values, 35%
of 13 are engaged in an intramolecular H-bond. Substitution at C(6) enhances the
persistence of the H-bond only slightly (42% for 17; J(5'a,OH) ˆ 3.1, J(5'b,OH) ˆ
7.8Hz), whilst broad signals for H ÀC(5') and the OH groups prevented determination
of the J(H,OH) values of 18. In agreement with the enhanced persistence of the
intramolecular H-bond, HOÀC(5') of 17 is shifted downfield (13: 2.62, 17: 3.07 ppm).
A bifurcated intramolecular H-bond of HO(5') to OˆC(2) and O(4') is present in the
solid state of 18 (Fig. 2; O(5')H ¥¥¥ OˆC(2) ˆ 1.71 and O(5')H ¥¥¥ O(4') ˆ 2.27 ä).
The chemical shift of HÀC(2') is a good indicator for the syn/anti equilibrium of
uridines; 6-unsubstitued uridines prefer the anti-conformation [23]. HÀC(2') of

Helvetica Chimica Acta Vol. 87 (2004)
2279
RO(5')-protected derivatives of 13 (R ˆ Me, Allyl, Pr, Bn, and Ac) resonates in CDCl3
solution at 4.95 5.04 ppm [27][28]. Sterically demanding protecting groups at O(5')
disfavour the syn-conformer. Indeed, HÀC(2') of the TIPS-protected derivative 14 and
its TBS analogue [29] resonate at the most-upfield position of O(5')-protected
derivatives of 13 (4.69 4.71 ppm). Presumably, these two silyl ethers prefer nearly
exclusively the anti-conformation. Substitution at C(6) leads to a downfield shift for
HÀC(2') of 15 18 (Dd ˆ 0.50 0.55 ppm) and to an upfield shift for C(2') (Dd ˆ 1.3
2.2 ppm), evidencing a syn-conformation (Table 1). Apparently, the intramolecular H-
bond of 17 and 18 has, at best, only a weak influence on the chemical shifts of HÀC(2')
and C(2'). HÀC(2') of isopropylidenated uridines possessing sterically more-demand-
ing alkyl substituents at C(6) resonates at 5.14 5.21 ppm (Dd ˆ 0.45 0.52 ppm) [30],
suggesting that such 6-alkylated uridines prefer exclusively the syn-conformation.
HÀC(2') of 13 is shifted downfield by only 0.34 ppm indicating a syn/anti ratio of
65 :35. Thus, the syn-conformation of 13 and 17 is distinctly more highly populated than
suggested by the persistence of the intramolecular H-bond (65 and 100 vs. 35 and 42%,
resp.). The partly persistent intramolecular C(5')ÀOH ¥¥¥ OˆC(2) H-bond of 13, 17,
and 18 is also evidenced by small J(4',5'a) and J(4',5'b) values (2.3 5.3 Hz; see Table 6
in the Exper. Part). Larger J(4',5'a) and J(4',5'b) values of the silyl ethers 15 and 16
(5.7 6.9 Hz) evidence a steric interaction between the nucleobase and TIPSOÀC(5');
the conformer with anti-periplanar HÀC(4') and HÀC(5') bonds is favoured. Solutions
of the C(6)-unsubstituted 13 and 14 in CDCl3 are characterised by a ca. 1:1 (S)/(N)
equilibrium (J(1',2')/J(3',4') ˆ 48:52 and 45 :55, resp., cf. Table 1), whilst the C(6)-
substituted silyl ethers 15 and 16 adopt predominantly the (N)-conformation (J(1',2')/
J(3',4') ˆ 21:79). The 6-substituted alcohols 17, 18, and 19 have a smaller preference for
the (N)-conformer (J(1',2')/J(3',4') ˆ 39 :61, 33 :67, and 40 :60, resp.). The isopropy-
lidenated compounds show a similar ring puckering (J(1',2') ‡ J(3',4') ˆ 5.7 6.6 Hz),
which is increased by deisopropylidenation (10.3 Hz for 19).
2. Dimer Synthesis. We first prepared the self-complementary U*-[O]-A and U*-
[O]-A* dimers 23 32 (Schemes 3 and 4). The synthesis of 27 32 was briefly
communicated [15]. The electrophiles 20 22 required for the ether formation were
prepared from the alcohol 15 by mesylation (Ms2O) to form the unstable mesylate 20
(90%), chlorination to 21 (via 20, using MsCl, 95%), and bromination with PBr3 to 22
(67%). O-Alkylation of the C(8)-unsubstituted adenosine derivative 5 (Scheme 1)
gave the U*-[O]-A dimer 23 (45% from 20; 62% from 21; 60% from 22). N-
Debenzoylation of 23 gave the amine 24 (98%), which was desilylated to the alcohol 25
(80%), and further deprotected [7] to yield 93% of the fully deprotected 26.
The C(8)-substituted adenosine derivative 9 (Scheme 1) was O-alkylated by 21 to
the U*-[O]-A* dimer 27 (63%; Scheme 4). N-Debenzoylation of 27 gave the amine 28
(98%). Detritylation or desilylation of 28 led to the monoalcohols 29 (85%) and 30
(71%), respectively, which were further deprotected to the same diol 31 (86 and 81%,
resp.). Hydrolysis yielded the completely deprotected dimer 32 (89%).
The 1H-NMR spectra of the debenzoylated dimers 24, 25, and 28 31 in CDCl are
3
characterised by a concentration-dependent downfield shift of the uridine HÀN(3),
evidencing an intermolecular H-bond.
The syn-conformation of the adenosine unit (unit I) of the U*-[O]-A dimers 23 25
in CDCl3 is more highly populated than the syn-conformation of 6, as evidenced by

2280
Helvetica Chimica Acta Vol. 87 (2004)
Scheme 3
a) Ms2O, Et3N, CH2Cl2; 90% of 20. b) MsCl, DMAP, pyridine; 95% of 21. c) CaH2, Br3P, Et 2O; 67% of 22.
d) NaH, 5, DMF/THF 2 :1; 45% from 20; 62% from 21; 60% from 22. e) NH3, H2O/MeOH; 98%. f) TBAF,
THF; 80%. g) HCO2H/H2O 4 :1; 93%.
Dd(HÀC(2')) ˆ 0.34 0.43 ppm and Dd(C(2')) ˆ À0.4 ppm (Table 2). This probably
reflects the weaker steric hindrance by the uridinemethoxy moiety than by the
TIPSO group. As expected, the U*-[O]-A* dimers 27 30 show an even stronger
preference for a syn-conformation (Dd(HÀC(2')) ˆ 0.38 0.71 ppm, Dd(C(2')) ˆ À1.0
to À 1.2 ppm). The steric repulsion between the nucleobase and the uridinemethoxy
moiety is stronger in 27 31 than in 23 25, as evidenced by larger J(4',5'a) and J(4',5'b)
values (3.4 5.7 vs. 2.7 3.3 Hz; see Table 8 in the Exper. Part). Unit I prefers a (N)-
conformation, and debenzoylation increases this preference (J(1',2')/J(3',4') ˆ 43 :57
and 33 :67 for 23 and 27, resp., as compared to 23 :77 to 32 :68for 24, 25, and 28 30). In
the isopropylidenated dimers, the ring pucker in unit I dimers is larger for U*-[O]-A*
(J(1',2') ‡ J(3',4') ꢂ 5 6.4 Hz) than for U*-[O]-A (ca. 5 5.7 Hz)4). Deisopropylide-
nation increases the ring pucker (9.0 and 10.1 Hz for 26 and 32, resp.).
The uridine unit (unit II) of the U*-[O]-A dimers 23 25 and of the U*-[O]-A*
dimers 27 30 prefers
a
syn-conformation (Dd(HÀC(2')) ˆ 0.46 0.70 ppm,
Dd(C(2')) ˆ À1.0 to À 1.3 ppm relative to 14; Table 2). The J(4',5'a) and J(4',5'b)
values for the silyl ethers 23, 24, and 27 29 (5.3 7.1 Hz; see Table 8 in the Exper. Part)
4
)
There appears to be a correlation between the population of the syn-conformation and ring pucker in
unit I, with an increase in the syn-population leading to an increase of ring pucker. This correlation can
also be seen for the adenosine silyl ethers 7, 8, and 10 as compared to 6. No correlation is seen for the
uridine monomers 15 and 16, as compared to 14.

Helvetica Chimica Acta Vol. 87 (2004)
2281
Scheme 4
DMTrˆ 4,4'-Dimethoxytrityl
a) NaH, DMF/THF 2 :1; 63%. b) NH3, H2O/MeOH; 97%. c) HCO2H, MeNO2; 85% of 29; 86% of 31.
d) TBAF, THF; 71% of 30; 81% of 31. e) HCO2H/H2O 4 :1; 89%.
Table 2. 1H- and 13C-NMR Shift Differences for HÀC(2') and C(2') of the U*-O-A Dimers 23 25, and the U*-
O-A* Dimers 27 30 in CDCl3, as Compared to the Adenosine-Derived Monomer 6 and the Uridine-Derived
Monomer 14, and the Ratio of (S)/(N)-Conformers of these Nucleosides, as Deduced from J(1',2')/J(3',4')
Nucleoside
6
14
23
24
25
27
28
29
30
Adenosyl unit I
Dd(HÀC(2')) [ppm]
Dd(C(2')) [ppm]
J(1',2') [Hz]
a
)
)
0.43
À 0.4
1.6
0.34
À 0.4
1.1
0.36
1.2b)
1.0
0.380.56
À 1.2
0.71
À 1.0
0.43
a
À 1.2
À 1.0
2.5
2.1
1.3
< 1.0
1.2
J(3',4') [Hz]
(S)/(N)
2.8
47: 53
2.1
2.3
43 : 57
2.2
32 : 6831 : 69
4.2
3.83.3
3.7
23 : 77
33 : 67
25 : 75
24 : 76
Uridyl unit II
Dd(HÀC(2')) [ppm]
Dd(C(2')) [ppm]
J(1'2') [Hz]
c
)
)
0.47
À 1.2
0.59
0.57
0.4b)
1.2
0.52
0.580.71
À 1.0
< 1.0
4.5
0.60
c
À 1.0
< 1.0
4.4
À 1.3
< 1.0
4.5
À 1.1
À 1.3
1.9
4.5
2.81.0
3.4
< 1.0
4.7
J(3',4') [Hz]
4.3
4.5
(S)/(N)
45 : 55
19 : 81
19 : 81
21 : 79
18 : 82
18 : 82
18 : 82
30 : 70
a) d(HÀC(2')) ˆ 5.32 and d(C(2')) ˆ 84.6 ppm. b) From
d(C(2')) ˆ 85.4 ppm.
a
CD3OD spectrum. c) d(HÀC(2')) ˆ 4.69 and

2282
Helvetica Chimica Acta Vol. 87 (2004)
reflect a steric interaction between the nucleobase and TIPSOÀC(5'), whilst lower
values for the alcohols 25 and 30 (2.4 4.2 Hz) are in keeping with a partly persistent
intramolecular O(5')H ¥¥¥ OˆC(2) H-bond. Unfortunately, HOÀC(5'/II) of 25 and 30
could not be assigned in the CDCl3 1H-NMR spectra. Unit II of 23 25 and 27 29
shows a larger preference for a (N)-conformation than unit I (cf. Table 2). This is not so
for the alcohol 30 (30 :70), on account of the intramolecular H-bond. The
isopropylidenated U*-[O]-A and U*-[O]-A* dimers show similar ring puckering in
unit II (J(1',2') ‡ J(3',4') ꢂ 5.0 5.7 Hz), with the exception again of 30 (6.4 Hz).
Deisopropylidenation increases the ring pucker (9.0 and 9.9 Hz for 26 and 32, resp.).
We abstained from preparing U*-[O]-U dimers, as base treatment of 2',3'-O-
isopropylideneuridine (13) leads to an intramolecular b-addition [31]. The C(6)-
substituted alcohol 17 (Scheme 2), however, prefers a syn-conformation preventing
such a b-addition, and O-alkylation of 17 with the mesylate 20, chloride 21, or bromide
22 gave the protected U*-[O]-U* dimer 33 (29% from 20; 35% from 21; 50% from 22;
Scheme 5).
Scheme 5
DMTrˆ 4,4'-Dimethoxytrityl
a) NaH, 17, DMF/THF 2 :1; 29% from 20; 35% from 21; 50% from 22.
1
The H-NMR spectra of the U*-[O]-U* dimer 33 in CDCl3 is characterised by a
concentration-dependent downfield shift of HÀN(3) of units I and II, evidencing
intermolecular H-bonds. The syn-conformation of both units in CDCl3 is more highly
populated than the syn-conformation of 14, as evidenced by Dd(HÀC(2')) ˆ 0.49
0.52 ppm and Dd(C(2')) ˆ À1.0 to À 1.2 ppm. The large J(4',5') values of 6.5 7.5 Hz
reflect a steric interaction between the nucleobase and the substituent at O(5'); the
conformers possessing antiperiplanar HÀC(4') and HÀC(5') bonds are favoured.
Units I and II both prefer the (N)-conformation (J(1',2')/J(3',4') ˆ 18 :82 and 19 :81,
resp.).
The hypothesis that the strand breaks observed upon the attempted deprotection of
the dA* decamer 2 are due to facile elimination of the phosphate group at CH2ÀC(8),
and generation of the stabilised cation A [2] suggested to prepare the A*-[O]-A* dimer
by additionÀelimination. The trichloroacetimidate 36 (86%) was thus prepared from

Helvetica Chimica Acta Vol. 87 (2004)
2283
Scheme 6
Thexyl ˆ 1,1,2-trimethylpropyl
a) 34, Cl3CCN, 1,8-diazabicyclo[5.4.0]undec-7-ene (DBU), CH2Cl2; 86% of 36. b) 35, Ms2O, Et3N, CH2Cl2; 8 5%
of 37. c) 35, MsCl, Et3N, LiCl, CH2Cl2; 74% of 38.
the alcohol 34, and the unstable mesylate 37 (50 90%) and the chloride 38 (74%) were
obtained from the alcohol 35 (Scheme 6)5).
The trichloroacetimidate 36 did, however, not react with the alcohols 5 or 9
(Scheme 1) in the presence of catalytic amounts of MsOH, CSA (ˆcamphorsulfonic
acid), BF3 ¥ OEt2, or TMSOTf in CH2Cl2, cyclohexane, Et2O, THF, MeCN, or N,N-
dimethylpivalamide. With a molar equivalent of these acids, or a stronger acid such as
TfOH or HBF4, increasing the temperature, or prolonging the reaction time led to
deprotection and side reactions. Addition of NaH or lithium tetramethylpiperidide to
deprotonate the N(6)-benzamido group and force the elimination led to decomposition
of the trichloroacetimidate 36. Similarly, base-promoted O-alkylation of the alcohols 5
or 9 with the mesylate 37 or the chloride 38 led to decomposition of the electrophile;
the alcohol was recovered. Some N6-alkylation products were observed, as evidenced
1
by H-NMR and MS.
3. Association of the Dimers. Association constants Ka and the thermodynamic
parameters DH8, DS8 and DG8 in CDCl3 were calculated for the U*-[O]-A dimers 24
and 25, and for the U*-[O]-A* dimers 28 30 from the concentration and temperature
dependence of d(HÀN(3)) (Fig. 3 and Table 3). The association constant Ka in CDCl3
was also calculated for the U*-[O]-U* dimer 33. The Dd(HÀN(3)) value of the diol 31
in CDCl3 was almost concentration-independent (12.96 12.65 ppm for 33 1 mm),
evidencing a strong association. The fully deprotected dimers 26 and 32 were not
sufficiently soluble in CDCl3 to determine their association. The d(HÀN(3)) value of
the dimers is about the same as for the monomers in (D5)pyridine and (D6)DMSO.
These results support the contention that the structural differentiation between
nucleobases and backbone is not required for association.
The Ka values compare favourably with that determined for the interaction of 3',5'-
di-O-acetyl-2'-deoxyuridine with 3',5'-di-O-acetyl-2'-deoxyadenosine (70 mÀ1 [32]) and
for the interaction of 9-ethyladenine and 1-cyclohexyluracil (110 Æ 9 mÀ1 [33]). The
5
)
We thank Anne Viger for the experiments with the thexyldimethylsilyl ether 35 (A.V. Ph.D. thesis, in
preparation).

2284
Helvetica Chimica Acta Vol. 87 (2004)
Fig. 3. a) Dependence of d(HÀN(3)) of the U*-O-A* dimers 28 30 upon the concentration. b) Van×t Hoff
correlation for the U*-O-A* dimer 28.
Table 3. Association Constants and Thermodynamic Parameters for the U*-O-A Dimers 24 and 25, the U*-O-A*
Dimers 28 30, and Association Constant and DG for the U*-O-U* Dimer 33
Dimer
Ka [mÀ1
]
À DH [kcal/mol]
À DS [e.u.]
À DG [kcal/mol]a)
24
25
28
29
30
33
1890
2500
970
280
3220
265
22
13
16
22
24
62
30
40
64
65
4.0
4.6
4.0
3.3
4.7
3.3
4.4
4.2
3.9
3.0
5.3
a) The first value is calculated from Ka, the second from DH and DS.
high Ka values for 25 and 30, and the inability to dissociate 31 appreciably in CDCl3
correlate with a downfield shift of HOÀC(5') as compared to 13 (Dd ca. 1.0 ppm). This
downfield shift evidences an intramolecular C(5')OÀH ¥¥¥ OˆC(2) H-bond that may,
in its turn, be correlated with the observation that the dimers 25, 30, and 31 possessing a
HOÀC(5'/II) group have higher association constants than the corresponding silyl
ethers 24, 28, and 29. This higher association constant is, thus, interpreted as the result
of a more-favourable pairing conformation (−preorganisation×) rather than a reduced
steric hindrance of the alcohols. The data for the silyl ethers 28 and 29 highlight the
contribution of the lipophilic DMTr group. Substitution at C(8) of the adenine unit
increases the Ka value for the HOÀC(5'/II) dimer 30, but decreases the Ka value for the
silyl ether 29.
Cross-peaks between the H-bonded imino HÀN(3) and adenine HÀC(2) and
HÀC(8) in a 2D-NOESY experiment on the associated dimer 24 suggest both
WatsonÀCrick- and Hoogsteen-type base pairing [32][33]. A cross-peak between the
H-bonded imino HÀN(3) and adenine HÀC(2) in a 2D-NOESY experiment on the
associated dimer 28 suggests WatsonÀCrick-type base pairing, but this experiment can
not detect Hoogsteen-type base pairing, as there is no HÀC(8). It has been suggested
that a HÀC(2) downfield shift implies WatsonÀCrick H-bonds, and that a HÀC(8)
downfield shift implies interactions of the Hoogsteen-type [34][35]. There is a small
upfield shift for HÀC(8) of the adenine unit for the associated dimer 24 (Dd ˆ

Helvetica Chimica Acta Vol. 87 (2004)
2285
0.11 ppm) and for the associated dimer 25 (Dd ˆ 0.14 ppm) as compared to the
monomer 2',3'-O-isopropylidene-5'-O-(triethylsilyl)adenosine (d ˆ 8.07 ppm [6]). The
chemical shift for HÀC(2) does not change (d ˆ 8.37 8.38 ppm).
Although the thermodynamic parameters for the association listed in Table 3
suggest that there are two groups of dimers (24, 29, and 30 vs. 25 and 28) and evidence
enthalpy/entropy compensation [36], there is no obvious correlation with structural
parameters; such a correlation requires the synthesis and study of the association of
further −structurally dedifferentiated× di- and oligonucleosides.
We thank the ETH-Z¸rich and F. Hoffmann-La Roche AG, Basel, for generous support.
Experimental Part
General. THF was distilled from Na/benzophenone. CH2Cl2, DMF, pyridine, diisopropylamine (iPr2NH),
and ethyl(diisopropyl)amine (EtNiPr2) were distilled from CaH2. Reactions were run under Ar or N2. Qual.
TLC: precoated silica-gel plates (Merck silica gel 60 F254); detection by spraying with −mostain× (400 ml of 10%
aq. H2SO4, 20 g of (NH4)6Mo7O24 ¥ H2O, 0.4 g of Ce(SO4)2) and heating. Workup: org. phases were dried
(Na2SO4), filtered, and processed as indicated. Flash chromatography (FC): silica gel Merck 60 (0.04
0.063 mm). Optical rotations: 1-dm cell at 258 and 589 nm. FT-IR: 1 2% in the indicated solvent or solid
state (ATR). 1H- and 13C-NMR: at 300, 400, or 500 MHz and at 75, 100, or 125 MHz, resp. MS: matrix-assisted
laser desorption ionization time-of-flight mass spectrometry (MALDI-TOF-MS), indol-3-acrylic acid (IAA),
0.05m in THF or a-cyano-4-hydroxycinnamic acid (CCA) 0.05m in MeCN/EtOH/H2O, and 2,5-dihydroxyben-
zoic acid (DHB) 0.05m in THF for high-resolution (HR) MALDI-MS.
General Procedure for NMR Studies. NMR Spectra were recorded at 295 K on a Varian Gemini 300
spectrometer (300 MHz) in CDCl3 passed through basic aluminum oxide immediately prior to use. Experiments
started at the highest indicated concentration, with stepwise replacement of 0.2 ml of the 0.7 ml soln. with 0.2 ml
of pure CDCl3. The data were analysed graphically [37] and by nonlinear least-squares fitting [38].
Thermodynamic parameters were determined by van×t Hoff analysis. The uridyl d(HÀN(3)) was monitored
between 508 and À 308 at a fixed concentration (between 20 80% of saturation).
N6-Benzoyl-2',3'-O-isopropylidene-5'-O-(triisopropylsilyl)adenosine (6).
A soln. of 5 [16] (10.0 g,
24.3 mmol), 1H-imidazole (3.34 g, 48.6 mmol), and DMAP (0.30 g, 2.4 mmol) in CH2Cl2 (155 ml) was treated
dropwise at 238 with TIPSCl (10.84 ml, 48.6 mmol), stirred for 3 h, washed with H2O (2 Â 70 ml) and brine (1 Â
70 ml), dried (MgSO4), and adsorbed onto silica (25 g). FC (silica gel (500 g); AcOEt/cyclohexane 1 :1) gave 6
(13.24 g, 96%). White solid. Rf (AcOEt) 0.56. M.p. 60 628. [a]2D5 ˆ À34.6 (c ˆ 1, CHCl3). IR (CHCl3): 3411w,
3012m, 2961m, 2947m, 2895w, 2869m, 1710m, 1613s, 1586m, 1505w, 1456m, 1385m, 1261s, 1092s, 1014m, 8 8 m1 ,
808w. 1H-NMR (300 MHz, CDCl3): see Table 4; additionally, 9.34 (br. s, NH); 8 .8 1s,(HÀC(2)); 8.21 (s,
HÀC(8)); 8.04 8.00 (m, 2 arom. H); 7.61 7.46 (m, 3 arom. H); 1.64, 1.41 (2s, Me2C); 1.03 0.98( m,
(Me2CH)3Si). 13C-NMR (75 MHz, CDCl3): see Table 5; additionally, 164.53 (s, CˆO); 133.63 (s); 132.55 (d);
128.65 (2d); 127.81 (2d); 114.27 (s, Me2C); 27.28, 25.44 (2q, Me2C); 17.98( q, (Me2CH)3Si); 11.93 (d,
‡
‡
(Me2CH)3Si). HR-MALDI-MS: 568.2955 (100, [M ‡ H] , C29H42N5O5Si ; calc. 568.2956), 590.2766 (80, [M ‡
‡
‡
Na] , C29H41N5NaO5Si ; calc. 590.2777).
N6-Benzoyl-8-(hydroxymethyl)-2',3'-O-isopropylidene-5'-O-(triisopropylsilyl)adenosine (7). A soln. of
iPr2NH (14.2 ml, 100.3 mmol) in THF (57 ml) was cooled to À 788, treated dropwise with 1.6m BuLi in hexane
(62.7 ml, 100.3 mmol), stirred for 15 min, warmed to 08, stirred for 15 min, cooled to À 788, and treated
dropwise with a soln. of 6 (10.65 g, 20.07 mmol) in THF (81 ml) over 10 min, stirred for 1 h, treated with DMF
(38.8 ml, 501.4 mmol), stirred for 2.5 h, treated with AcOH (14.3 ml), allowed to warm to 238, diluted with
EtOH (142 ml), treated slowly with NaBH4 (2.34 g, 60.2 mmol), and stirred for 25 min. After evaporation, the
residue was dissolved in CH2Cl2 (200 ml), washed with H2O (2 Â 100 ml) and brine (1 Â 100 ml), dried
(MgSO4), and evaporated. FC (silica gel (500 g); AcOEt/cyclohexane 1:1 ! AcOEt) gave 7 (8.88 g, 74%).
White solid. Rf (AcOEt) 0.46. M.p. 93 958. [a]2D5 ˆ À10.8( c ˆ 1, CHCl3). IR (CHCl3): 3610w, 3410w (br.),
1
3013m, 2961m, 2947m, 2894m, 1709m, 1615s, 1591m, 1479s, 1462s, 1430m, 1258s, 1092s, 1075s, 8 8 m3 . H-NMR
(300 MHz, CDCl3): see Table 4; additionally, 9.50 (br. s, NH); 8.01 7.97 (m, 2 arom. H); 7.53 7.39 (m, 3 arom.
H); 5.85 (br. s, OH); 1.59, 1.42 (2s, Me2C); 1.00 0.90 (m, (Me2CH)3Si). 13C-NMR (75 MHz, CDCl3): see

2286
Helvetica Chimica Acta Vol. 87 (2004)
Table 4. Selected 1H-NMR Chemical Shifts [ppm] and Coupling Constants [Hz] of the Adenosine Monomers
5
11 and 36 38 in CDCl3, and of 12 in CD3OD
a
a
a
5 [5]
6a)
7
)
8
)
9
10
11
)
12
36
37
38
HÀC(1')
HÀC(2')
HÀC(3')
HÀC(4')
HaÀC(5')
HbÀC(5')
HÀC(2)
HÀC(8)
CHaÀC(8)
CHbÀC(8)
J(1',2')
J(2',3')
J(3',4')
J(4',5'a)
J(4',5'b)
J(5'a,5'b)
J(5'a,OH)
J(5'b,OH)
J(Ha,Hb)
5.95
5.22
5.10
4.54
3.98
3.80
8.76
8.08
6.22
5.32
5.05
4.43
3.96
3.84
8.81
8.21
6.29
5.72
5.12
4.25
3.77
3.65
8.65
6.27
5.81
5.11
4.23
3.91
3.74
8.77
6.16
5.29
5.12
4.484.27
4.01
6.25
5.73
5.13
6.10
5.28
5.04
6.13
4.93
4.35
6.386.22
5.88
5.13
4.29
3.79
6.32
4.30
3.76
3.64
8.79
5.77
5.06
5.83
5.13
4.49
4.19
4.26
3.83
3.70
8.24
3.93
3.77
8.10
3.89
3.73
8.12
3.75
3.62
8.81
3.84
8.77
3.67
8.85
4.96
4.90
2.2
4.54
4.49
2.7
4.51
4.45
5.0
4.94
4.88
2.2
4.85
4.80
5.0
4.90
4.80
7.5
5.71
5.64
2.1
5.57
5.57
2.7
4.92
4.92
2.4
5.0
5.9
1.2
2.5
6.2
2.83.4
4.4
4.4
11.1
6.5
6.5
5.9
6.5
5.6
5.3
6.5
6.3
6.3
3.1
1.5
3.4
1.0
1.6
< 1.0
< 1.0
12.810.810.811.9
< 2.0
3.3
2.2
2.2
3.3
5.4
5.86.0
3.5
1.9
6.2
5.6
10.4
6.5
6.2
10.3
1.2
2.1
5.9
5.9
5.7
6.0
2.2
12.5
2.2
10.9
5.7
12.810.812.0
1.8
10.9
12.1
10.3
16.0
b
15.9
12.1
15.5
14.5
13.1
)
15.3
a) Assignment based on selective homodecoupling experiments. b) Not determined.
Table 5. Selected 13C-NMR Chemical Shifts [ppm] of the Adenosine Monomers 5 11 and 36 38 in CDCl3, and
12 in (D6)DMSO
5 [5]
6
7
8
9
10
89.84
83.25
81.42
87.48
63.02
11
91.74
82.65
81.53
85.70
63.05
12
88.76
72.42
71.06
86.60
62.32
36
90.20
83.18
81.44
87.60
63.03
37
90.41
82.81
81.43
87.27
62.63
38
90.42
83.24
81.59
87.65
62.83
C(1')
C(2')
C(3')
C(4')
C(5')
C(2)
C(4)
C(5)
C(6)
C(8)
94.11
83.07
81.55
86.20
63.26
91.38
84.60
81.26
87.40
63.49
89.85
83.00
81.45
87.50
62.94
90.52
82.83
82.14
87.43
63.42
92.34
82.58
81.27
85.58
63.31
152.19 152.70 152.12 152.25 151.99
150.11 149.51 148.71 149.02 149.73
124.20 123.10 121.03 121.78 122.41
152.78152.14 151.96 153.02 153.29 152.58
150.37 149.29 149.73 149.72 149.97 150.51
117.84 118.18 118.12 121.66 122.08 122.26
150.20 151.02 151.90 151.93 151.37a) 154.90 155.13 156.18152.14 152.24 152.58
142.39 141.45 154.39 151.93 151.43a) 151.84 150.93 151.13 148.74 147.19 149.77
CH2ÀC) (8
57.4859.4859.52
57.26
57.14
56.5862.55
62.33
36.73
a) Assignment may be interchanged.
Table 5; additionally, 164.75 (s, CˆO); 133.32 (s); 132.47 (d); 128.48 (2d); 127.72 (2d); 114.17 (s, Me2C); 27.18,
‡
25.41 (2q, Me2C); 17.85 (q, (Me2CH)3Si); 11.84 (d, (Me2CH)3Si). HR-MALDI-MS: 598.3060 (100, [M ‡ H] ,
‡
‡
‡
C30H44N5O6Si ; calc. 598.3062), 620.2873 (60, [M ‡ Na] , C30H43N5NaO6Si ; calc. 620.2883). Anal. calc. for
C30H43N5O6Si (597.79): C 60.28, H 7.25, N 11.72: found: C 60.37, H 7.39, N 11.35.
N6-Benzoyl-8-{[(bis(4-methoxyphenyl)](phenyl)methyl}-2',3'-O-isopropylidene-5'-O-(triisopropylsilyl)-
adenosine (8). A soln. of 7 (2.00 g, 3.28mmol), EtN iPr2 (1.13 ml, 6.56 mmol), and DMAP (0.041 g, 0.33 mmol)
in CH2Cl2 (30 ml) was cooled to 08, treated with DMTrCl (2.27 g, 6.56 mmol), stirred at 508 for 3 h, cooled to r.t.,
washed with H2O (3 Â 10 ml), brine (1 Â 10 ml), dried (MgSO4), and evaporated. FC (silica gel (180 g); AcOEt/
cyclohexane 2 :1) gave 8 (2.78g, 94%). Yellow foam. Rf (AcOEt/cyclohexane 2 :1) 0.36. M.p. 137 1398 (dec.).

Helvetica Chimica Acta Vol. 87 (2004)
2287
[a]2D5 ˆ À16.6 (c ˆ 1, CHCl3). IR (CHCl3): 3408w, 3064w, 3014m, 2961m, 2944m, 2867m, 1709m, 1613s, 1589m,
1510s, 1463s, 1253s, 1177m, 1094s, 1071s, 1035m, 8 8w3, 8 2m9 . 1H-NMR (300 MHz, CDCl3): see Table 4;
additionally, 9.08(br. s, NH); 8.05 8.00 (m, 2 arom. H); 7.65 7.19 (m, 12 arom. H); 6.87 6.80 (m, 4 arom. H);
3.77 (s, 2 MeO); 1.51, 1.39 (2s, Me2C); 1.05 0.95 (m, (Me2CH)3Si). 13C-NMR (75 MHz, CDCl3): see Table 5;
additionally, 164.31 (s, CˆO); 158.57 (2s); 143.80 (s); 134.86 (2s); 133.84 (s); 132.57 (d); 130.08(4 d); 128.76
(2d); 128.01 (2d); 127.89 (2d); 127.70 (2d); 126.69 (d); 113.94 (s, Me2C); 113.22 (4d); 87.82 (s, Ph3C); 55.24 (q, 2
MeO); 27.35, 25.66 (2q, Me2C); 17.96 (q, (Me2CH)3Si); 11.96 (d, (Me2CH)3Si). HR-MALDI-MS: 303 (65,
‡
‡
(MeO2)Ph3C), 618(100), 922.4173 (86, [ M ‡ Na] , C51H61N5NaO8Si ; calc. 922.4189).
N6-Benzoyl-8-{[bis(4-methoxyphenyl)](phenyl)methyl}-2',3'-O-isopropylideneadenosine (9). A soln. of 8
(2.00 g, 2.20 mmol) in THF (35 ml) was treated dropwise with 1m Bu4NF ¥ 3 H2O in THF (4.4 ml, 4.40 mmol) and
stirred at 238 for 4 h. Evaporation and FC (silica gel (180 g); AcOEt/cyclohexane 2 :1) gave 9 (1.46 g, 89%).
White solid. Rf (AcOEt) 0.38. M.p. 186 1888. [a]2D5 ˆ À46.5 (c ˆ 1, CHCl3). IR (CHCl3): 3611w, 3407w, 3287w
(br.), 3067w, 3015m, 2960w, 2939w, 2874w, 2840w, 1711m, 1612s, 1592m, 1510s, 1463m, 1430m, 1253s, 1178m,
1
1082s, 1036m, 8 29w. H-NMR (300 MHz, CDCl3): see Table 4; additionally, 9.23 (br. s, NH); 8.04 8.00 (m, 2
arom. H); 7.63 7.19 (m, 12 arom. H); 6.85 6.80 (m, 4 arom. H); 5.80 (br. d, J ˆ 1.8, 10.9, HOÀC(5')); 3.77 (s, 2
MeO); 1.37, 1.32 (2s, Me2C). 13C-NMR (75 MHz, CDCl3): see Table 5; additionally, 164.30 (s, CˆO); 158.58
(2s); 143.78( s); 134.81 (s); 134.77 (s); 133.55 (s); 132.66 (d); 130.01 (4d); 128.71 (2d); 127.93 (2d); 127.89 (2d);
127.75 (2d); 126.98( d); 114.03 (s, Me2C); 113.22 (4d); 87.94 (s, Ar3C); 55.22 (q, 2 MeO); 27.61, 25.36 (2q, Me2C).
‡
‡
‡
HR-MALDI-MS: 744.3034 (3, [M ‡ H] , C42H42N5O8
;
calc. 744.3034), 766.2840 (100, [M ‡ Na] ,
‡
C42H41N5NaO8 ; calc. 766.2855).
8-(Hydroxymethyl)-2',3'-O-isopropylidene-5'-O-(triisopropylsilyl)adenosine (10). A soln. of 7 (500 mg,
0.83 mmol) in MeOH (16.8 ml) was treated at 238 with 25% aq. NH3 (12.8ml, 83 mmol), warmed to 60 8, stirred
for 2 h, cooled to r.t., and evaporated. FC (silica gel (50 g); CH2Cl2/MeOH 9 :1) gave 10 (392 mg, 96%). White
solid. Rf (AcOEt) 0.25. M.p. 143 1468. [a]D25 ˆ À15.3 (c ˆ 1, CHCl3). UV (MeOH): 261 (12500). IR (CHCl3):
3524w, 3495w, 3413w (br.), 3016m, 2961m, 2945m, 2867m, 1635s, 1592w, 1462w, 1375m, 1239w, 1087s, 8 8m3.
1H-NMR (300 MHz, CDCl3): see Table 4; additionally, 6.12 (br. s, NH2); 5.85 (br. s, OH); 1.61, 1.40 (2s, Me2C);
1.00 0.90 (m, (Me2CH)3Si). 13C-NMR (75 MHz, CDCl3): see Table 5; additionally, 114.24 (s, Me2C); 27.13,
‡
25.34 (2q, Me2C); 17.78( q, (Me2CH)3Si); 11.79 (d, (Me2CH)3Si). HR-MALDI-MS: 494.2801 (60, [M ‡ H] ,
‡
‡
‡
C23H40N5O5Si ; calc. 494.2799), 516.2620 (100, [M ‡ Na] , C23H39N5NaO5Si ; calc. 516.2620).
8-(Hydroxymethyl)-2',3'-O-isopropylideneadenosine (11). A soln. of 10 (300 mg, 0.60 mmol) and AcOEt
(34 ml, 0.60 mmol) in THF (4.8ml) was treated dropwise with 1 m Bu4NF ¥ 3 H2O in THF (1.19 ml, 1.19 mmol),
stirred at 238 for 3 h, and evaporated. A soln. of the residue in CH2Cl2 (20 ml) was washed with H2O (2 Â 10 ml)
and brine (1 Â 10 ml), dried (MgSO4), and evaporated. FC (silica gel (25 g); AcOEt/MeOH 9 :1) gave 11
(176 mg, 87%). White solid. Recrystallization in AcOEt gave colourless crystals submitted for X-ray analysis. Rf
(AcOEt/MeOH 9 :1) 0.17. M.p. 212 2148. [a]D25 ˆ À40.8( c ˆ 1, CHCl3). UV (MeOH): 259 (9300). IR (CHCl3):
3526w, 3412w, 3218w (br.), 3017m, 2994w, 2942w, 2870w, 1637s, 1592w, 1476w, 1454w, 1420w, 1375m, 1103m,
1082m, 8 52m. 1H-NMR (300 MHz, CDCl3): see Table 4; additionally, 6.65 (br. d, J ˆ 10.3, HOÀC(5')); 6.48(br.
s, NH2); 5.78(br. s, HOCH2ÀC(8)); 1.64, 1.37 (2s, Me2C). 13C-NMR (75 MHz, CDCl3): see Table 5; additionally,
‡
‡
113.99 (s, Me2C); 27.68, 25.47 (2q, Me2C). HR-MALDI-MS: 338.1464 (100, [M ‡ H] , C14H19N5O5 ; calc.
‡
‡
338.1465), 360.1282 (50, [M ‡ Na] , C14H19N5NaO5 ; calc. 360.1286). Anal. calc. for C14H19N5O5 (337.33): C
49.85, H 5.68, N 20.76, O 23.71: found: C 50.09, H 5.76, N 20.97, O 23.50.
8-(Hydroxymethyl)adenosine (12). A suspension of 11 (100 mg, 0.29 mmol) in 80% aq. HCO2H soln.
(2.5 ml) was stirred for 6 h at 08 and evaporated. The residue was dissolved in H2O (2 ml), adsorbed onto (silica
gel (1.0 g) and taken to dryness. FC (silica gel (15 g); AcOEt/MeOH/H2O 7:2 :1) gave 12 (79 mg, 92%). White
solid. Rf (AcOEt/MeOH/H2O 7:2 :1) 0.28. M.p. > 2008 (dec.). [a]2D5 ˆ À41.7 (c ˆ 0.33, MeOH/H2O 2 :1). UV
(MeOH): 259 (8300). IR (ATR): 3421m, 3317m, 3162m (br.), 2925m, 2852m, 1654s, 1607m, 1594m, 1491w,
1454w, 1423w, 1374w, 1323m, 1312m, 1256w, 1124w, 1067s, 98 7w. 1H-NMR (300 MHz, CD3OD): see Table 4.
‡
‡
13C-NMR (75 MHz, (D6)DMSO): see Table 5. HR-MALDI-MS: 298.1140 (32, [M ‡ H] , C11H16N5O5 ; calc.
‡
‡
298.1152), 320.0954 (3, [M ‡ Na] , C11H15N5NaO5 ; calc. 320.0973).
i
6-(Hydroxymethyl)-2',3'-O-isopropylidene-5'-O-(triisopropylsilyl)uridine (15). A soln. of Pr2NH (53 ml,
370.7 mmol) in THF (210 ml) was cooled to À 788, treated dropwise with 1.6m BuLi in hexane (232 ml,
370.7 mmol), stirred for 15 min, warmed to 08, stirred for 15 min, cooled to À 788, treated dropwise with a soln.
of 14 [1] (33.0 g, 74.15 mmol) in THF (300 ml) over 10 min, stirred for 1 h, treated with DMF (144 ml,
1854 mmol), stirred for 2.5 h, treated with AcOH (54 ml), allowed to warm to 238, diluted with EtOH (524 ml),
treated slowly with NaBH4 (8.7 g, 222 mmol), stirred for 25 min, and evaporated. A soln. of the residue in
CH2Cl2 (500 ml) was washed with H2O (2 Â 250 ml) and brine (1 Â 250 ml), dried (MgSO4), and evaporated. FC

2288
Helvetica Chimica Acta Vol. 87 (2004)
(silica gel (2 Â 500 g); AcOEt/cyclohexane 1:1 ! AcOEt) gave 15 (24.48g, 70%). White solid. Rf (AcOEt/
cyclohexane 2 :1) 0.23. M.p. 77 798. [a]2D5 ˆ 21.5 (c ˆ 1, CHCl3). UV (MeOH): 259 (9300). IR (CHCl3): 3606w,
3391w (br.), 3031w, 2946m, 2894w, 2869m, 1698s (br.), 1627w, 1463m, 1383m, 1159w, 1135w, 1094m, 1067m,
1
974w, 8 8w.1 H-NMR (300 MHz, CDCl3): see Table 6; additionally, 9.39 (br. s, HÀN(3)); 3.53 (br. s,
HOCH2ÀC(6)); 1.55, 1.33 (2s, Me2C); 1.10 1.00 (m, (Me2CH)3Si). 13C-NMR (75 MHz, CDCl3): see Table 7;
additionally, 113.67 (s, Me2C); 27.24, 25.35 (2q, Me2C); 17.90 (q, (Me2CH)3Si); 11.97 (d, (Me2CH)3Si). HR-
‡
‡
MALDI-MS: 493.2355 (100, [M ‡ Na] , C22H38N2NaO7Si ; calc. 493.2348). Anal. calc. for C22H38N2O7Si
(470.63): C 56.15, H 8.14, N 5.95: found: C 56.15, H 8.24, N 6.00.
Table 6. Selected 1H-NMR Chemical Shifts [ppm] and Coupling Constants [Hz] of the Uridine Monomers 13
18 and 20 22 in CDCl3, and 19 in CD3OD
13a)
14 [1]
15
16
17
18
19
20
21
22
HÀC(1')
HÀC(2')
HÀC(3')
HÀC(4')
HaÀC(5')
HbÀC(5')
HÀC(5)
HÀC(6)
CHaÀC(6)
CHbÀC(6)
J(1',2')
J(2',3')
J(3',4')
J(4',5'a)
J(4',5'b)
J(5'a,5'b)
J(5'a,OH)
J(5'b,OH)
J(5,6)
5.56
5.05
4.97
4.30
3.93
3.81
5.67
7.687.68
5.99
4.69
4.85
4.27
4.00
3.88
5.67
5.81
5.19
4.83
4.184.07
3.94
5.60
5.19
4.78
5.51
5.22
5.00
5.77
5.24
4.96
5.47
4.74
4.31
5.88
5.23
4.82
5.82
5.25
4.84
5.86
5.28
4.85
4.12
4.23 84.139.8
3.89
4.19
4.21
3.86
3.80
5.76
3.86
3.75
5.81
3.79
3.66
5.88
3.87
3.85
5.58
3.87
3.85
5.82
3.87
3.86
5.80
3.89
5.80
3.82
5.82
4.57
4.50
1.2
6.3
4.5
5.7
6.9
11.1
4.03
3.97
2.5
4.04
3.94
2.2
6.6
3.9
3.1
4.0
12.1
3.1
7.8
4.55
4.46
4.1
6.5
4.4
2.83.1
5.3
4.53
4.47
1.2
6.2
6.2
5.3
5.6
11.811.2
5.12
5.04
1.2
6.5
4.4
5.4
4.44
4.32
1.2
6.2
4.3
4.26
4.17
3.1
6.5
3.4
2.3
3.6
12.3
3.6
7.5
8.1
2.81.2
6.2
3.4
2.5
3.4
6.3
4.5
5.7
6.9
6.3
4.4
5.3
7.2
5.1
7.2
b
11.2
)
12.5
10.9
12.1
b
)
)
b
8.1
J(5,CH2)
J(Ha,Hb)
< 1.0
14.9
< 1.0
12.9
< 1.0
12.9
< 1.0
14.5
1.0
15.0
< 1.0
13.4
< 1.0
13.2
< 1.0
11.5
a) Sparingly soluble in CDCl3; spectrum of a third-saturated solution. Additional signals: 8.20 (br. s, NH); 2.62
(dd, J ˆ 3.6, 7.5 HOÀC(5')). b) Not determined.
Table 7. Selected 13C-NMR Chemical Shifts [ppm] of the Uridine Monomers 13 in CD3OD, 14 18 and 20 22 in
CDCl3, and 19 in (D6)DMSO
13 [1]
14 [1]
15
91.06
84.05
81.76
89.51
64.34
163.84
150.32
101.09
155.43
60.55
16
91.95
84.10
82.14
89.44
64.45
163.32
150.33
102.35
152.80
62.07
17
92.22
83.21
80.26
87.28
62.72
162.70
18
91.44
83.66
80.47
87.94
62.55
163.74
151.28149.98150.26
101.39
155.63
60.54
19
92.06
71.77
70.24
85.12
62.52
164.68
20
91.96
84.00
81.85
89.67
64.25
162.46
21
22
C(1')
C(2')
C(3')
C(4')
C(5')
C(2)
C(4)
C(5)
C(6)
CH2ÀC(6)
94.4
86.1
82.4
88.6
63.3
166.6
152.5
102.9
144.2
91.7
85.4
80.0
86.9
63.5
163.7
150.4
102.5
140.9
91.6891.79
84.16
81.83
89.82
64.29
162.66
150.15
104.58
150.54
41.06
84.18
81.89
89.90
64.39
162.14
151.08150.95
103.04
152.40
62.13
99.86
157.56
59.53
104.52
147.74
38.52
104.52
151.05
26.41

Helvetica Chimica Acta Vol. 87 (2004)
2289
6-{[Bis(4×-methoxyphenyl)](phenyl)methyl}-2',3'-O-isopropylidene-5'-O-(triisopropylsilyl)uridine (16). A
soln. of 15 (10.0 g, 20.82 mmol), EtNiPr2 (7.2 ml, 41.65 mmol), and DMAP (0.26 g, 2.08mmol) in CH 2Cl2
(133 ml) was cooled to 08, treated with DMTrCl (14.55 g, 41.65 mmol), stirred at 408 for 1 h, washed with H2O
(3 Â 60 ml) and brine (1 Â 60 ml), dried (MgSO4), and evaporated. FC (silica gel (400 g); AcOEt/cyclohexane
1:2) gave 16 (15.21 g, 95%). Yellow foam. Rf (AcOEt/cyclohexane 1:2) 0.21. M.p. 65 688 (dec.). [a]2D5 ˆ À2.0
(c ˆ 1, CHCl3). IR (CHCl3): 3393w, 3064w, 3011m, 2944m, 2868w, 2840w, 1697s, 1609m, 1585w, 1510s, 1465m,
1384m, 1251s, 1179m, 1067m, 1038m, 978w, 8 8 2w, 8 32m. 1H-NMR (300 MHz, CDCl3): see Table 6; additionally,
8.86 (br. s, HÀN(3)); 7.50 7.20 (m, 9 arom. H); 6.90 6.80 (m, 4 arom. H); 3.80 (s, 2 MeO); 1.43, 1.31 (2s,
Me2C); 1.10 1.00 (m, (Me2CH)3Si). 13C-NMR (75 MHz, CDCl3): see Table 7; additionally, 158.66 (2s); 143.55
(s); 134.64, 134.52 (2s); 129.86 (4d); 127.91 (2d); 127.84 (2d); 127.11 (d); 113.29 (s, Me2C); 113.29, 113.24 (4d);
87.92 (s, Ar3C); 55.23 (q, 2 MeO); 27.27, 25.54 (2q, Me2C); 17.95 (q, (Me2CH)3Si); 12.01 (d, (Me2CH)3Si). HR-
‡
‡
MALDI-MS: 795.3658([ M ‡ Na] , C43H56N2NaO9Si ; calc. 795.3655).
6-{[Bis(4-methoxyphenyl)](phenyl)methyl}-2',3'-O-isopropylideneuridine (17). A soln. of 16 (5.00 g,
6.4 mmol) in THF (52 ml) was treated dropwise with 1m Bu4NF ¥ 3 H2O in THF (12.8ml, 12.8mmol) and
stirred at 238 for 2 h. Evaporation and FC (silica gel (100 g); AcOEt/cyclohexane 1 :1) gave 17 (3.51 g, 89%).
White solid. Rf (AcOEt/cyclohexane 1:1) 0.29. M.p. 132 1358. [a]2D5 ˆ À21.6 (c ˆ 1, CHCl3). IR (CHCl3):
3608w, 3480w (br.), 3390w, 3030w, 3012w, 2937w, 2839w, 1695s (br.), 1608m, 1584w, 1510m, 1457w, 1385w,
1
1251m, 1177m, 1070w, 1035m, 977w, 8 35w. H-NMR (300 MHz, CDCl3): see Table 6; additionally, 9.32 (br. s,
HÀN(3)); 7.50 7.20 (m, 9 arom. H); 6.90 6.80 (m, 4 arom. H); 3.80 (s, 2 MeO); 3.07 (dd, J ˆ 3.1, 7.8,
HOÀC(5')); 1.37, 1.29 (2s, Me2C). 13C-NMR (75 MHz, CDCl3): see Table 7; additionally, 158.70 (2s); 143.50 (s);
134.55, 134.44 (2s); 129.89 (4d); 127.95 (2d); 127.80 (2d); 127.16 (d); 113.94 (s, Me2C); 113.33, 113.28(4 d); 88.00
‡
‡
(s, Ar3C); 55.26 (q, 2 MeO); 27.29, 25.33 (2q, Me2C). HR-MALDI-MS: 639.2305 ([M ‡ Na] , C34H36N2NaO9
;
calc. 639.2321).
6-(Hydroxymethyl)-2',3'-O-isopropylideneuridine (18). A soln. of 15 (400 mg, 0.84 mmol) and AcOEt
(97 ml, 1.68mmol) in THF (6.8ml) was treated dropwise with 1 m Bu4NF ¥ 3 H2O in THF (1.68ml, 1.68mmol),
stirred at 238 for 3 h, and evaporated. A soln. of the residue in CH2Cl2 (20 ml) was washed with H2O (2 Â 10 ml)
and brine (1 Â 10 ml), dried (MgSO4), and evaporated. FC (silica gel (25 g); AcOEt) gave 18 (235 mg, 89%).
White solid. Recrystallization in AcOEt gave colourless crystals submitted for X-ray analysis. Rf (AcOEt) 0.22.
M.p. 192 1958. [a]D25 ˆ À6.7 (c ˆ 0.5, CHCl3). UV (MeOH): 258(8300). IR (CHCl 3): 3603w, 3400w (sh), 3388w
(br.), 3019s, 2931w, 2857w, 1699s (br.), 1627w, 1602w, 1457w, 1384w, 1101w, 1068m. 1H-NMR (300 MHz,
CDCl3): see Table 6; additionally, 9.26 (br. s, HÀN(3)); 3.45 3.30 (br. s, 2 OH); 1.56, 1.34 (2s, Me2C). 13C-NMR
(75 MHz, CDCl3): see Table 7; additionally, 114.20 (s, Me2C); 27.29, 25.34 (2q, Me2C). HR-MALDI-MS:
‡
‡
337.1003 (28, [M ‡ Na] , C13H18N2NaO7 ; calc. 337.1014). Anal. calc. for C13H18N2O7 (314.29): C 49.68, H 5.77, N
8.91, O 35.63: found: C 49.58, H 5.85, N 8.92, O 35.62.
6-(Hydroxymethyl)uridine (19). An 80% aq. soln. of HCO2H (2.5 ml) was treated with 18 (100 mg,
0.31 mmol), stirred at 08 for 14 h, and evaporated. A soln. of the residue in H2O (2 ml) was adsorbed onto silica
gel (1.0 g) and evaporated. FC (silica gel (15 g); AcOEt/MeOH/H2O 7:2 :1) gave 19 (77 mg, 90%). White solid.
Rf (AcOEt/MeOH/H2O 7:2 :1) 0.31. M.p. 173 1758. [a]2D5 ˆ À33.1 (c ˆ 0.33, MeOH/H2O 2 :1). UV (MeOH):
251 (7900). IR (ATR): 3364m, 3299m (br.), 3126w, 2919w, 2857w, 1719s, 1685s, 1655s, 1620m, 1481w, 1447w,
1
1390m, 1339m, 1270m, 1102s, 1072s, 1031s, 98 3m, 940m, 8 8 m9, 8 28s, 764s. H-NMR (300 MHz, CD3OD): see
Table 6. 13C-NMR (75 MHz, CD3OD): see Table 7.
2',3'-O-Isopropylidene-6-{[(methylsulfonyl)oxy]methyl}-5'-O-(triisopropylsilyl)uridine (20). A soln. of 15
(250 mg, 0.53 mmol) and Et3N (0.3 ml) in CH2Cl2 (10 ml) was cooled to 08, treated with Ms2O (193 mg,
1.06 mmol), stirred for 1 h, washed with H2O (2 Â 25 ml) and brine (1 Â 25 ml), dried (MgSO4), and evaporated.
FC (silica gel (5.0 g); AcOEt/cyclohexane 1:1) gave 20 (263 mg, 90%). White foam. Rf (AcOEt) 0.58. M.p. 51
538. [a]2D5 ˆ 9.3 (c ˆ 1, CHCl3). IR (CHCl3): 3384w, 3084w, 2944m, 2891m, 2867m, 1703s, 1459m, 1376m, 1354m,
1178w, 1090m, 1067m, 965w, 8 8 2w. 1H-NMR (300 MHz, CDCl3): see Table 6; additionally, 9.75 (br. s, HÀN(3));
3.15 (s, MsO); 1.53, 1.33 (2s, Me2C); 1.10 1.00 (m, (Me2CH)3Si). 13C-NMR (75 MHz, CDCl3): see Table 7;
additionally, 113.82 (s, Me2C); 38.52 (q, MsO); 27.24, 25.46 (2q, Me2C); 17.95 (q, (Me2CH)3Si); 11.97 (d,
‡
‡
(Me2CH)3Si). HR-MALDI-MS: 571.2117 (100, [M ‡ Na] , C23H40N2NaO9SSi ; calc. 571.2124).
6-(Chloromethyl)-2',3'-O-isopropylidene-5'-O-(triisopropylsilyl)uridine (21). soln. of 15 (982 mg,
A
2.07 mmol) and DMAP (24 mg, 0.2 mmol) in pyridine (16.7 ml) was cooled to 08, treated dropwise with MsCl
(0.33 ml, 4.13 mmol), warmed to 238, stirred for 1 h, treated with CH2Cl2 (50 ml), washed with H2O (2 Â 25 ml)
and brine (1 Â 25 ml), dried (MgSO4), and evaporated. FC (silica gel (25 g); AcOEt) gave 21 (952 mg, 95%).
1
Yellow foam. Rf (AcOEt) 0.66. H-NMR (300 MHz, CDCl3): see Table 6; additionally, 10.18(br. s, HÀN(3));

2290
Helvetica Chimica Acta Vol. 87 (2004)
1.54, 1.33 (2s, Me2C); 1.10 1.00 (m, (Me2CH)3Si). 13C-NMR (75 MHz, CDCl3): see Table 7; additionally, 113.64
(s, Me2C); 27.28, 25.39 (2q, Me2C); 17.91 (q, (Me2CH)3Si); 11.99 (d, (Me2CH)3Si).
6-(Bromomethyl)-2',3'-O-isopropylidene-5'-O-(triisopropylsilyl)uridine (22). A soln. of 15 (500 mg,
1.04 mmol) and CaH2 (90 mg, 2.08mmol) in Et 2O (11 ml) was cooled to 08, treated with PBr3 (160 ml,
1.26 mmol), stirred in the dark for 2 h, treated with H2O (10 ml), and extracted with CH2Cl2 (3 Â 10 ml). The
org. layer was washed with H2O (2 Â 10 ml) and brine (1 Â 10 ml), dried (MgSO4), and evaporated. FC (silica
gel (15 g); cyclohexane/AcOEt 2 :1) gave 22 (370 mg, 67%). White solid. Rf (cyclohexane/AcOEt 2 :1) 0.38.
1H-NMR (300 MHz, CDCl3): see Table 6; additionally, 9.38(br. s, HÀN(3)); 1.56, 1.35 (2s, Me2C); 1.10 0.98
(m, (Me2CH)3Si). 13C-NMR (75 MHz, CDCl3): see Table 7; additionally, 113.68( s, Me2C); 27.34, 25.47 (2q,
‡
Me2C); 17.99 (q, (Me2CH)3Si); 12.04 (d, (Me2CH)3Si). HR-MALDI-MS: 557.1485 (54, [M ‡ Na] ,
C22H3781BrN2NaO6Si ; calc. 557.1481), 555.1504 (53, [M ‡ Na] , C22H3779BrN2NaO6Si ; calc. 555.1502).
2',3'-O-Isopropylidene-5'-O-(triisopropylsilyl)uridine-6-methyl-(61 ! 5')-N6-benzoyl-2',3'-O-isopropylidene-
adenosine (23). A soln. of 5 (358mg, 0.86 mmol) in DMF (6.7 ml) was cooled to 0 8, treated with 60% NaH in oil
(172 mg, 4.30 mmol), stirred for 30 min, treated dropwise with a soln. of 21 (510 mg, 0.98mmol) in THF
(7.0 ml), stirred for 2 h, treated with NH4Cl soln. (10 ml), and extracted with CH2Cl2 (3 Â 10 ml). The org. layer
was washed with H2O (2 Â 10 ml) and brine (1 Â 10 ml), dried (MgSO4), and evaporated. FC (silica gel (100 g);
AcOEt/cyclohexane 1:1 ! AcOEt) gave 23 (459 mg, 62%). White solid. Rf (AcOEt) 0.22. M.p. 133 1368.
[a]2D5 ˆ ‡26.5 (c ˆ 1, CHCl3). IR (CHCl3): 3390w, 2996m, 2944m, 2867m, 1704s, 1610s, 1586m, 1516w, 1480w,
1456s, 1383s, 1157m, 1092s, 1072s, 8 8m1. 1H-NMR (500 MHz, CDCl3): see Table 8; additionally, 11.14 (br. s,
HÀN(3/II)); 10.06 (br. s, HNÀC(6/I)); 8.14 8.12 (m, 2 arom. H); 7.54 7.40 (m, 3 arom. H); 1.63, 1.51, 1.44, 1.33
(4s, 2 Me2C); 1.10 0.95 (m, (Me2CH)3Si). 13C-NMR (75 MHz, CDCl3): see Table 9; additionally, 165.28( s,
CˆO/I); 133.03 (s); 132.48( d); 128.49 (2d); 128.37 (2d); 113.77, 113.45 (2s, 2 Me2C); 27.33, 27.17, 25.78, 25.52
‡
‡
‡
‡
(4q, 2 Me2C); 17.91 (q, (Me2CH)3Si); 11.95 (d, (Me2CH)3Si). HR-MALDI-MS: 886.3772 (100, [M ‡ Na] ,
‡
‡
‡
C42H57N7NaO11Si ; calc. 886.3785), 864.3985 (36, [M ‡ H] , C42H58N7O11Si ; calc. 864.3964).
2',3'-O-Isopropylidene-5'-O-(triisopropylsilyl)uridine-6-methyl-(61 ! 5')-2',3'-O-isopropylideneadenosine
(24). A soln. of 23 (218mg, 0.25 mmol) in MeOH (4.0 ml) was treated at 23 8 with 25% aq. NH3 (3.85 ml,
25 mmol), stirred at 608 for 2 h, cooled to r.t., and evaporated. FC (silica gel (15 g); CH2Cl2/MeOH 9 :1) gave 24
(187 mg, 99%). White solid. Rf (CH2Cl2/MeOH 9 :1) 0.33. M.p. 126 1298. [a]2D5 ˆ ‡49.5 (c ˆ 1, CHCl3). UV
(MeOH): 259 (23000). IR (CHCl3): 3484w, 3185w, 2993m, 2944m, 2866m, 1717s (br.), 1642s, 1600m, 1471m,
1384s, 1158m, 1090s, 1048m, 8 8 m1 , 8 33w. 1H-NMR (500 MHz, CDCl3): see Table 8; additionally, 13.53 (br. s,
HÀN(3/II)); 6.94 (br. s, H2NÀC(6/I)); 1.61, 1.54, 1.43, 1.37 (4s, 2 Me2C); 1.01 0.93 (m, (Me2CH)3Si). 13C-NMR
(75 MHz, CDCl3): see Table 9; additionally, 113.52, 113.38(2 s, 2 Me2C); 27.42, 27.19, 26.02, 25.65 (4q, 2 Me2C);
‡
‡
17.89 (q, (Me2CH)3Si); 11.96 (d, (Me2CH)3Si). HR-MALDI-MS: 782.3509 (100, [M ‡ Na] , C35H53N7NaO10Si ;
‡
‡
calc. 782.3523), 760.3699 (38, [M ‡ H] , C35H54N7O10Si ; calc. 760.3702).
2',3'-O-Isopropylideneuridine-6-methyl-(61 ! 5')-2',3'-O-isopropylideneadenosine (25). A soln. of 24
(220 mg, 0.29 mmol) in THF (2.3 ml) was treated with TBAF ¥ 3 H2O (186 mg, 0.57 mmol), stirred at 238 for
3 h, and evaporated. FC (silica gel (50 g); CH2Cl2/MeOH 9 :1) gave 25 (138mg, 80%). White solid. Rf (CH2Cl2/
MeOH 8:2) 0.51. M.p. 175 178 8. UV (MeOH): 258(19500). IR (ATR): 3338 w (br.), 3265w, 3207w (br.),
2988w, 2925w, 2857w, 2810w, 1693s (br.), 1648m, 1601m, 1573w, 1474w, 1374m, 1209s, 1156m, 1090s, 1068s, 8 67m.
1H-NMR (300 MHz, CDCl3): see Table 8; additionally, 13.02 (br. s, HÀN(3/II)); 6.82 (br. s, H2NÀC(6/I)); 1.62,
1.57, 1.44, 1.40 (4s, 2 Me2C). 13C-NMR (75 MHz, CD3OD): see Table 9; additionally, 114.90, 114.78(2 s, 2 Me2C);
‡
‡
27.70, 27.50, 26.15, 25.79 (4q, 2 Me2C). HR-MALDI-MS: 432.2 (100), 604.2360 (45, [M ‡ H] , C26H34N7O10 ; calc.
‡
‡
604.2368), 626.2177 (21, [M ‡ Na] , C26H33N7NaO10 ; calc. 626.2189).
Uridine-6-methyl-(61 ! 5')-adenosine (26). An 80% aq. soln. of HCO2H (0.9 ml) was cooled to 08, treated
with 25 (24 mg, 0.04 mmol), stirred for 6 h, diluted with H2O (5 ml), adsorbed onto silica gel (1.5 g), and dried.
FC (silica gel (15 g); AcOEt/MeOH/H2O 7:2 :1) gave 26 (19 mg, 93%). White solid. Rf (AcOEt/MeOH/H2O
7:2 :1) 0.32. M.p. 177 1808 (dec.). IR (ATR): 3452m, 3408m (br.), 3282m, 3202m (br.), 3148m, 2963w, 2918w,
2882w, 2754w (br.), 2708w, 1886w, 1705s, 1673s, 1654s, 1607m, 1573w, 1473w, 1445m, 1419m, 1382m, 1300m,
1119s, 1102m, 1076s, 1045s, 8 54m, 767m. 1H-NMR (300 MHz, (D6)DMSO): see Table 8; additionally, 11.36 (br.
s, HÀN(3/II)); 7.28(br. s, H2NÀC(6/I)); 5.49 (d, J ˆ 6.0, OH); 5.28( d, J ˆ 4.8, OH); 5.14 (d, J ˆ 5.1, OH); 4.98
(d, J ˆ 6.3, OH); 4.71 (t, J ˆ 5.7, HOÀC(5'/II)). 13C-NMR (75 MHz, (D6)DMSO): see Table 9. HR-MALDI-
‡
‡
‡
‡
MS: 524.1730 (100, [M ‡ H] , C20H26N7O10, calc. 524.1742), 546.1564 (80, [M ‡ Na] , C20H25N7NaO10 ; calc.
546.1563).
2',3'-O-Isopropylidene-5'-O-(triisopropylsilyl)uridine-6-methyl-(61 ! 5')-N6-benzoyl-8-{[bis(4-methoxyphen-
yl)](phenyl)methyl}-2',3'-O-isopropylideneadenosine (27). A soln. of 9 (647 mg, 0.86 mmol) in DMF (6.7 ml)
was cooled to 08, treated with 60% NaH in oil (103 mg, 2.58mmol), stirred for 30 min, treated with a soln. of 21

Helvetica Chimica Acta Vol. 87 (2004)
2291
Table 8. Selected 1H-NMR Chemical Shifts [ppm] and Coupling Constants [Hz] for the Dimers 23 25 and 27
30 in CDCl3, 26 and 31 in (D6)DMSO, and 32 in D2O
23a)
24a)
25
26b)
27a)
28a)
29
30b)
6.32
5.30
31b)
32
HÀC(1'/I)
HÀC(2'/I)
HÀC(3'/I)
HÀC(4'/I)
HaÀC(5'/I)
HbÀC(5'/I)
HÀC(2/I)
HÀC(8/I)
CHaÀC(8/I)
CHbÀC(8/I)
J(1',2')
J(2',3')
J(3',4')
J(4',5'a)
J(4',5'b)
J(5'a,5'b)
J(Ha,Hb/I)
6.186.20
5.75
5.19
4.61
3.74
3.74
8.79
8.23
6.17
5.90
6.33
6.22
8 65.083
5.19
6.33
5.75
6.19
5.59
6.06
5.66
5.26
4.57
3.75
3.70
8.38
7.96
5.684.62
5.22
4.584.11
3.85
3.73
8.37
5.70
4.33
4.26
5.34
3.65
3.62
8.37
5.40
3.69
8.29
5.32
5.10
4.66
3.98
8.11
4.31
4.42
4.34
3.64
4.28 4.28
3.70
3.79
3.71
8.27
8.16
3.77
3.683.63
8.81
3.76
3.55
8.34
3.69
8.12
7.93
4.60
4.53
2.1
6.5
4.2
4.55
4.41
1.3
6.3
3.83.3
4.6
3.4
10.5
11.815.6
4.97
4.92
< 1.0
4.51
4.43
1.2
6.2
3.5
5.6
5.6
11.1
4.67
4.67
2.2
4.90
4.90
5.3
-
1.6
6.0
2.1
3.0
3.0
1.1
6.0
2.3
2.83.3
2.7
1.0
6.3
2.2
4.0
2.7
10.5
5.1
4.2
3.9
5.4
4.0
5.9
6.2
5.3
3.7
5.7
4.5
4.8
5.3
2.2
5.3
4.84.9
10.5
12.1
c
)
10.4
10.6
10.5
10.9
11.2
9.7
c
11.8
)
HÀC(1'/II)
HÀC(2'/II)
HÀC(3'/II)
HÀC(4×/II)
HaÀC(5'/II)
HbÀC(5'/II)
HÀC(5/II)
CHaÀC(6/II)
CHbÀC(6/II)
J(1',2')
5.57
5.16
4.71
4.12
3.78
3.76
5.41
4.30
3.93
1.0
5.67
5.285.26
4.83
4.17
3.82
3.82
5.40
4.36
3.86
< 1.0
6.3
5.46
4.52
5.03
4.22
3.84
3.84
5.34
4.32
4.05
1.2
5.39
5.21
4.11
3.71
3.60
3.48
5.77
4.47
4.47
3.9
5.66
5.27
4.81
4.14
3.84
3.80
5.62
4.44
4.22
< 1.0
6.3
5.75
5.40
4.87
4.15
3.85
3.83
5.36
4.44
4.03
< 1.0
6.3
5.69
5.63
5.685.41
5.29
5.184.63
4.83
4.16
3.84
3.84
5.40
4.33
4.00
5.09
4.13
3.83
3.77
5.385.70
4.39
4.16
1.9
4.73
3.93
3.55
3.47
5.75
4.47
4.25
< 1.0
6.5
4.33
3.89
3.82
3.77
4.44
4.44
3.1
< 1.0
5.9
J(2',3')
6.4
6.3
6.3
6.4
6.5
J(3',4')
4.3
4.4
4.5
5.1
4.5
4.5
4.7
4.5
4.7
6.8
J(4',5'a)
5.4
6.4
4.2
3.3
5.3
5.4
6.2
2.4
6.0
4.7
J(4',5'b)
J(5'a,5'b)
J(Ha,Hb/II)
6.6
10.5
12.3
6.4
)
11.812.5
4.2
)
5.4
12.0
7.0
10.5
13.4
7.1
10.5
12.5
6.2
)
3.6
6.0
12.0
9.7
2.8
10.3
c
c
c
11.4
13.7 13.4
13.5
13.9
a) Assignment based on a DQFCOSY.GP spectrum. b) Assignment based on selective homodecoupling
experiments. c) Not determined.
(510 mg, 1.03 mmol) in THF (6.8ml), stirred for 2 h, treated with NH 4Cl soln. (10 ml), and extracted with
CH2Cl2 (3 Â 10 ml). The org. layer was washed with H2O (2 Â 10 ml) and brine (1 Â 10 ml), dried (MgSO4), and
evaporated. FC (silica gel (113 g); AcOEt/cyclohexane 1:1) gave 27 (652 mg, 63%). White solid. Rf (AcOEt)
0.47. M.p. 124 1268. [a]2D5 ˆ À2.6 (c ˆ 1, CHCl3). IR (CHCl3): 3391w, 3007m, 2943m, 2867m, 1703s (br.), 1612s,
1590w, 1509s, 1462s, 1427m, 1384m, 1157m, 1070s, 1036m, 8 8m2, 8 31m. 1H-NMR (500 MHz, CDCl3): see
Table 8; additionally, 10.10 (br. s, HÀN(3/II)); 9.64 (br. s, HNÀC(6/I)); 8.07 8.04 (m, 2 arom. H); 7.59 7.46
(m, 4 arom. H); 7.41 7.37 (m, 4 arom. H); 7.31 7.20 (m, 4 arom. H); 7.06 6.81 (m, 4 arom. H); 3.77 (s, 2 MeO);
1.54, 1.52, 1.40, 1.34 (4s, 2 Me2C); 1.10 1.00 (m, (Me2CH)3Si). 13C-NMR (75 MHz, CDCl3): see Table 9;
additionally, 164.93 (s, CˆO/I); 158.55 (2s); 143.81 (s); 134.82, 134.77 (2s); 133.39 (s); 132.52 (d); 130.07 (4d);
128.53 (2d); 128.15 (2d); 128.00 (2d); 127.84 (2d); 126.96 (d); 114.36, 113.46 (2s, 2 Me2C); 113.20 (4d); 87.85 (s,
Ar3C); 55.22 (q, 2 MeO); 27.35, 27.35, 25.69, 25.55 (4q, 2 Me2C); 17.96 (q, (Me2CH)3Si); 11.20 (d, (Me2CH)3Si).
‡
‡
‡
HR-MALDI-MS: 303 (100, DMTr ), 1218.499 (10, [M ‡ Na] , C64H77N7NaO14Si ; calc. 1218.5198).

2292
Helvetica Chimica Acta Vol. 87 (2004)
Table 9. Selected 13C-NMR Chemical Shifts [ppm] for the Dimers 23, 24, and 27 30 in CDCl3, 25 and 31 in
CD3OD, 26 in (D6)DMSO, and 32 in D2O
23a)
24a)
25
93.20
26
91.68
27a)
28a)
29
89.90
30
90.08
31
91.17
32
91.91
C(1'/I)
93.01
92.01
89.92
90.00
C(2'/I)
C(3'/I)
C(4'/I)
C(5'/I)
84.21
81.73
86.38
71.39
84.22
81.71
86.94
71.43
85.79
82.90
87.88
72.30
71.19
69.72
82.78
70.39
83.43
81.69
85.79
70.55
83.57
81.67
86.74
69.80
83.44
81.25
87.02
70.53
83.61
80.57
87.17
69.46
84.50
82.95
87.30
71.50
71.56
69.16
83.09
69.71
C(2/I)
C(4/I)
C(5/I)
C(6/I)
151.80 152.82 153.86 152.52 151.75 151.91 152.04 151.81 153.42 151.97
150.37 149.03 150.15 151.14 149.54 150.10 150.19 149.88 151.41 149.65
124.18 119.09 120.10 118.88 122.76 118.45 117.60 118.41 118.91 117.31
151.55 155.66 157.06 155.83 152.45 155.61 155.44 155.53 156.76 154.58
141.77 138.55 140.90 139.19 151.75 148.81 151.87 148.84 152.95 152.48
C(8/I)
CH2ÀC(8/I)
C(1'/II)
C(2'/II)
C(3'/II)
C(4'/II)
C(5'/II)
C(2/II)
C(4/II)
C(5/II)
C(6/II)
CH2ÀC(6/II)
59.60
91.37
84.12
82.04
89.51
64.32
59.27
91.39
84.37
82.26
89.65
64.50
57.14
91.69
84.29
82.13
89.73
64.44
59.14
91.38
84.08
81.69
88.10
62.87
58.41
92.83
85.74
83.07
90.30
63.79
56.56
88.48
71.66
69.88
83.09
61.37
91.87
84.21
82.09
89.94
64.34
92.01
84.42
82.33
89.96
64.52
92.68
85.79
83.11
90.55
63.74
87.24
73.22
70.39
84.71
61.95
163.00 164.11 165.30 163.34 162.59 164.07 164.52 163.55 165.60 165.05
150.02 149.15 152.18149.29 150.39 150.24 150.30 150.13 152.90 151.15
104.05 104.79 104.91 102.09 102.93 103.71 103.77 103.90 103.93 102.77
149.50 151.02 152.49 151.61 151.06 150.74 150.99 151.61 152.52 151.97
69.06
69.49
70.08
67.94
68.62
68.08
69.04
68.92
69.86
68.33
a) Assignment based on a DQFCOSY.GP and a HSQC.GP spectrum.
2',3'-O-Isopropylidene-5'-O-(triisopropylsilyl)uridine-6-methyl-(61 ! 5')-8-{[bis(4-methoxyphenyl)](phen-
yl)methyl}-2',3'-O-isopropylideneadenosine (28). A soln. of 27 (439 mg, 0.33 mmol) in MeOH (6.8ml) was
treated at 238 with 25% aq. NH3 (5.15 ml, 33 mmol), stirred at 608 for 2 h, cooled to r.t., and evaporated. FC
(silica gel (110 g); AcOEt) gave 28 (384 mg, 97%). White solid. Rf (AcOEt) 0.36. M.p. 134 1368. [a]2D5 ˆ ‡24.4
(c ˆ 1, CHCl3). IR (CHCl3): 3488w, 3185w, 2993m, 2943m, 2866m, 2840w, 1712s, 1636m, 1608m, 1509s, 1446m,
1383m, 1157m, 1068s, 1036m, 8 8 m2 , 8 31m. 1H-NMR (500 MHz, CDCl3): see Table 8; additionally, 13.09 (br. s,
HÀN(3/II)); 8.37 7.49 (m, 2 arom. H); 7.47 7.38( m, 4 arom. H); 7.32 7.21 (m, 3 arom. H); 6.90 (br. s,
H2NÀC(6/I)); 6.86 6.83 (m, 4 arom. H); 3.79 (s, 2 MeO); 1.55, 1.55, 1.42, 1.41 (4s, 2 Me2C); 1.01 0.96 (m,
(Me2CH)3Si). 13C-NMR (75 MHz, CDCl3): see Table 9; additionally, 158.50 (2s); 144.00 (s); 135.08(2 s); 130.02
(4d); 128.10 (2d); 127.81 (2d); 126.90 (d); 113.59, 113.26 (2s, 2 Me2C); 113.17 (4d); 87.50 (s, Ar3C); 55.23 (q, 2
MeO); 27.36, 27.36, 25.76, 25.76 (4q, 2 Me2C); 17.95 (q, (Me2CH)3Si); 11.99 (d, (Me2CH)3Si). HR-MALDI-MS:
‡
‡
‡
303 (100, DMTr ), 1114.492 (3, [M ‡ Na] , C57H73N7NaO13Si ; calc. 1114.4936). Anal. calc. for C57H73N7O13Si
(1092.33): C 62.68, H 6.74, N 8.98: found: C 62.88, H 6.94, N 8.82.
2',3'-O-Isopropylidene-5'-O-(triisopropylsilyl)uridine-6-methyl-(61 ! 5')-8-(hydroxymethyl)-2',3'-O-isopro-
pylideneadenosine (29). A soln. of 28 (170 mg, 0.15 mmol) in MeNO2 (0.4 ml) was treated with HCO2H (0.1 ml,
2.60 mmol), stirred at 238 for 1.5 h, diluted with CH2Cl2 (10 ml), washed with sat. NaHCO3 soln. (5 ml) and
brine (2 Â 5 ml), dried (MgSO4), and evaporated. FC (silica gel(80 g); AcOEt/MeOH 95 :5) gave 29 (103 mg,
85%). White solid. Rf (CH2Cl2/MeOH 9 :1) 0.34. M.p. 142 1448. [a]2D5 ˆ ‡38.7 (c ˆ 0.5, CHCl3). UV (MeOH):
260 (28500). IR (CHCl3): 3484w, 3409w, 3335w, 3198w (br.), 3018w, 2995w, 2943m, 2894w, 2867m, 1702s (br.),
1641m, 1604w, 1447m, 1383s, 1157m, 1089s, 1070s, 8 8m2, 8 66m. 1H-NMR (300 MHz, CDCl3): see Table 8;
additionally, 13.33 (br. s, HÀN(3/II)); 7.15 (br. s, H2NÀC(6/I)); 5.24 (br. s, HOCH2ÀC(8/I); 1.61, 1.56, 1.42, 1.42
(4s, 2 Me2C); 1.01 0.94 (m, (Me2CH)3Si). 13C-NMR (75 MHz, CDCl3): see Table 9; additionally, 113.61, 113.39
(2s, 2 Me2C); 27.33, 27.28, 25.67, 25.60 (4q, 2 Me2C); 17.95 (q, (Me2CH)3Si); 11.99 (d, (Me2CH)3Si). HR-
‡
‡
‡
MALDI-MS: 790.3822 (35, [M ‡ H] , C36H56N7O11Si , calc. 790.3808), 812.3629 (100, [M ‡ Na] ,
‡
C36H55N7NaO11Si ; calc. 812.3629).

Helvetica Chimica Acta Vol. 87 (2004)
2293
2',3'-O-Isopropylideneuridine-6-methyl-(61 ! 5')-8-{[bis(4-methoxyphenyl)](phenyl)methyl}-2',3'-O-iso-
propylideneadenosine (30). A soln. of 28 (165 mg, 0.16 mmol) in THF (1.2 ml) was treated with 1m TBAF ¥ 3
H2O in THF (0.3 ml, 0.30 mmol), stirred at 238 for 4 h, and evaporated. FC (silica gel (50 g); AcOEt) gave 30
(114 mg, 81%). White solid. Rf (CH2Cl2/MeOH 9 :1) 0.21. M.p. 143 1478 (dec.). [a]2D5 ˆ ‡8.9 (c ˆ 0.5, CHCl3).
IR (CHCl3): 3486w, 3409w, 3330w (br.), 3192w, 3015m, 2958w, 2930m, 2856w, 2842w, 1704s (br.), 1637m, 1607m,
1
1509m, 1446m, 1383m, 1376m, 1300w, 1252s, 1177m, 1156m, 1103m, 1070s, 1038m, 8 32m. H-NMR (300 MHz,
CDCl3): see Table 8; additionally, 13.43 (br. s, HÀN(3/II)); 7.52 7.20 (m, 9 arom. H); 6.91 (br. s, H2NÀC(6/I));
6.87 6.82 (m, 4 arom. H); 3.79 (s, 2 MeO); 1.55, 1.54, 1.42, 1.40 (4s, 2 Me2C); HOÀC(5'/II) hidden by the signals
at 3.84 3.72. 13C-NMR (75 MHz, CDCl3): see Table 9; additionally, 158.52 (2s); 143.96 (s); 135.03 (2s); 130.06
(4d); 128.08 (2d); 127.85 (2d); 126.91 (d); 113.99, 113.67 (2s, Me2C); 113.21 (4d); 87.54 (s, Ar3C); 55.25 (q, 2
‡
‡
MeO); 27.40, 27.40, 25.85, 25.52 (4q, 2 Me2C). HR-MALDI-MS: 303 (100, DMTr ), 958.3604 (12, [M ‡ Na] ,
‡
C48H53N7NaO13 ; calc. 958.3601).
2',3'-O-Isopropylideneuridine-6-methyl-(61 ! 5')-8-(hydroxymethyl)-2',3'-O-isopropylideneadenosine (31).
a) From 29. A soln. of 29 (66 mg, 0.08mmol) in THF (1.0 ml) was treated with 1 m TBAF in THF (0.16 ml,
0.16 mmol), stirred at 238 for 4 h, and evaporated. FC (silica gel (25 g); AcOEt) gave 31 (41 mg, 81%).
b) From 30. A soln. of 30 (90 mg, 0.10 mmol) in MeNO2 (0.4 ml) was treated with HCO2H (0.1 ml,
2.60 mmol), stirred at 238 for 1 h, diluted with CH2Cl2 (10 ml), washed with sat. NaHCO3 soln. (5 ml) and brine
(2 Â 5 ml), dried (MgSO4), and evaporated. FC (silica gel (15 g); CH2Cl2/MeOH 9 :1 ! 8:2) gave 31 (64 mg,
86%). White solid. Rf (CH2Cl2/MeOH 9 :1) 0.17. M.p. 176 1798. [a]2D5 ˆ À31.0 (c ˆ 1, MeOH). UV (MeOH):
260 (20800). IR (ATR): 3333w (br.), 3198w, 2987w, 2924w, 2853w, 1685s (br.), 1644m, 1606m, 1577w, 1447m,
1374m, 1330w, 1300w, 1259w, 1211m, 1157w, 1065s (br.), 969w, 8 63m, 8 31m. 1H-NMR (300 MHz, (D6)DMSO):
see Table 8; additionally, 11.45 (br. s, HÀN(3/II)); 7.30 (br. s, H2NÀC(6/I)); 5.79 (t, J ˆ 5.3, HOCH2ÀC(8/I));
4.79 (t, J ˆ 5.9, HOÀC(5'/II)); 1.52, 1.44, 1.31, 1.26 (4s, 2 Me2C). 13C-NMR (75 MHz, CD3OD): see Table 9;
additionally, 115.00, 114.76 (2s, 2 Me2C); 27.71, 27.58, 25.92, 25.75 (4q, 2 Me2C). HR-MALDI-MS: 462.2 (100),
‡
‡
‡
‡
634.2460 (49, [M ‡ H] , C27H36N7O11 ; calc. 634.2474), 656.2289 (14, [M ‡ Na] , C27H35N7NaO11 ; calc.
656.2295).
Uridine-6-methyl-(61 ! 5')-8-(hydroxymethyl)adenosine (32). An 80% aq. soln. of HCO2H (1.25 ml) was
cooled to 08, treated with 31 (50 mg, 0.08mmol), stirred for 8h at 23 8, diluted with H2O (5 ml), adsorbed onto
silica gel (0.5 g) and taken to dryness. FC (silica gel (15 g); AcOEt/MeOH/H2O 7 :2 :1) gave 32 (39 mg, 89%).
White solid. Rf (AcOEt/MeOH/H2O 7:2 :1) 0.27. M.p. 177 1798 (dec.). [a]2D5 ˆ À8.6 (c ˆ 1, H2O). IR (ATR):
3322s (br.), 3192s (br.), 2935m, 2863m, 1653s (br.), 1578m, 1447m, 1388m, 1331m, 1294m, 1250w, 1102s, 1044s,
1023s (br.), 904w, 8 52m, 796w. 1H-NMR (300 MHz, D2O): see Table 8. 13C-NMR (75 MHz, D2O): see Table 9.
‡
‡
‡
HR-MALDI-MS: 554.1848 (13, [M ‡ H] , C21H28N7O11
;
calc. 554.1848), 576.1668 (14, [M ‡ Na] ,
‡
C21H27N7NaO11 ; calc. 576.1669).
2',3'-O-Isopropylidene-5'-O-(triisopropylsilyl)uridine-6-methyl-(61 ! 5')-6-{[bis(4-methoxyphenyl)](phen-
yl)methyl}-2',3'-O-isopropylideneuridine (33). A soln. of 17 (196 mg, 0.31 mmol) in DMF (2.4 ml) was cooled to
08, treated with 60% NaH in oil (37 mg, 0.94 mmol), stirred for 30 min, treated with a soln. of 22 (170 mg,
0.31 mmol) in THF (2.5 ml), stirred for 1 h, treated with NH4Cl soln. (10 ml), and extracted with CH2Cl2 (3 Â
10 ml). The org. layer was washed with H2O (2 Â 10 ml) and brine (1 Â 10 ml), dried (MgSO4), and evaporated.
FC (silica gel (30 g); AcOEt/cyclohexane 2 :1) gave 33 (167 mg, 50%). Yellow solid. Rf (AcOEt) 0.44. M.p.
125 1288. IR (CHCl3): 3389w, 3019m, 2941m, 2867m, 1698s (br.), 1608w, 1509w, 1456w, 1383m, 1252m, 1176w,
1156w, 1088m, 1069m (br.), 1036w, 8 8w0, 8 33w. 1H-NMR (300 MHz, CDCl3, assignment based on selective
homodecoupling experiments): 10.35 (d, J ˆ 1.9, HÀN(3/I)); 9.84 (d, J ˆ 2.2, HÀN(3/II)); 7.47 6.82 (m, 13
arom. H); 5.84 (d, J ˆ 1.9, HÀC(5/I)); 5.72 (d, J ˆ 2.2, HÀC(5/II)); 5.71 (br. s, HÀC(1'/I)); 5.68(br. s, HÀC(1'/
II)); 5.22 (br. d, J ˆ 6.5, HÀC(2'/I)); 5.19 (br. d, J ˆ 6.2, HÀC(2'/II)); 4.88 (dd, J ˆ 4.7, 6.5, HÀC(3'/I)); 4.82 (dd,
J ˆ 4.4, 6.2, HÀC(3'/II)); 4.51, 4.34 (2d, J ˆ 14.3, CH2ÀC(6/II)); 4.19 (dt, J ˆ 4.7, 6.5, HÀC(4'/I)); 4.15 (dt, J ˆ
4.4, 6.5, HÀC(4'/II)); 3.02, 3.96 (2d, J ˆ 12.5, CH2ÀC(6/I)); 3.84 (d, J ˆ 6.5, 2 HÀC(5'/II)); 3.80 (s, 2 MeO);
3.80 3.72 (m, 2 HÀC(5'/I)); 1.51, 1.44, 1.31, 1.31 (4s, 2 Me2C); 1.10 1.00 (m, (Me2CH)3Si). 13C-NMR (75 MHz,
CDCl3): 163.43, 163.19 (2s, 2 C(2)); 158.67 (2s); 152.14, 151.64 (2s, 2 C(4)); 150.64, 150.33 (2s, 2 C(6)); 143.50 (s);
134.58, 134.44 (2s); 129.90 (4d); 127.90 (4d); 127.15 (d); 113.70, 113.35 (2s, 2 Me2C); 113.29, 113.22 (4d); 103.18,
102.81 (2d, 2 C(5)); 92.11, 91.22 (2d, 2 C(1')); 89.66 (d, C(4'/II)); 88.01 (s, Ar3C); 87.42 (d, C(4'/I)); 84.43, 84.24
(2d, 2 C(2')); 82.01, 81.68 (2d, 2 C(3')); 70.77 (t, C(5'/I)); 68.59 (t, CH2ÀC(6/II)); 64.46 (t, C(5'/II)); 62.33 (t,
CH2ÀC(6/I)); 55.25 (q, 2 MeO); 27.33, 27.23, 25.50 (2 C) (3q, 2 Me2C); 17.97 (q, (Me2CH)3Si); 12.01 (d,
‡
‡
‡
(Me2CH)3Si). HR-MALDI-MS: 303 (100, DMTr ), 1091.465 (5, [M ‡ Na] , C56H72N4NaO15Si ; calc.
1091.466).

2294
Helvetica Chimica Acta Vol. 87 (2004)
N6-Benzoyl-2',3'-O-isopropylidene-8-[(trichloroacetimidoxy)methyl]-5'-O-(triethylsilyl)adenosine (36). A
soln. of 34 (500 mg, 0.88 mmol) and trichloroacetonitrile (0.1 ml, 0.99 mmol) in CH2Cl2 (10 ml) was cooled to
08, treated with DBU (0.15 ml, 0.99 mmol), stirred for 1 h, and evaporated. FC (silica gel (20 g); cyclohexane/
AcOEt 8:2) gave 36 (530 mg, 86%). Yellow solid. Rf (AcOEt) 0.69. 1H-NMR (200 MHz, CDCl3): see Table 4;
additionally, 9.11 (br. s, HNÀC(6)); 8.64 (br. s, HNˆC); 8.05 7.90 (m, 2 arom. H); 7.64 7.40 (m, 3 arom. H);
1.60, 1.42 (2s, Me2C); 0.84 (t, J ˆ 7.4, (MeCH2)3Si); 0.50 (q, J ˆ 7.4, (MeCH2)3Si). 13C-NMR (50 MHz, CDCl3):
see Table 5; additionally, 164.45 (s, CˆO); 153.78( s, CˆNH); 133.66 (s); 132.77 (d); 128.87 (2d); 127.79 (2d);
114.26 (s, Me2C); 91.30 (s, Cl3C); 27.15, 25.41 (2q, Me2C); 6.55 (q, (MeCH2)3Si)); 4.11 (t, (MeCH2)3Si)).
N6-Benzoyl-5'-O-[(dimethyl)(1,1,2-trimethylpropyl)silyl]-2',3'-O-isopropylidene-8-{[(methylsulfonyl)oxy]-
methyl}-5'-O-(thexyldimethylsilyl)-adenosine (37). A soln. of 35 (100 mg, 0.17 mmol) and Et3N (62 ml) in
CH2Cl2 (1.2 ml) was cooled to 08, treated with Ms2O (60 mg, 3.4 mmol), stirred for 1 h, diluted with CH2Cl2
(10 ml), washed with H2O (2 Â 5 ml) and brine (1 Â 5 ml), dried (MgSO4), and evaporated to give 37 (95 mg,
1
85%). White foam. Rf (cyclohexane/AcOEt 1:1) 0.38. H-NMR (300 MHz, CDCl3): see Table 4; additionally,
9.11 (br. s, NH); 8.05 7.90 (m, 2 arom. H); 7.64 7.40 (m, 3 arom. H); 3.12 (s, MsO); 1.62, 1.41 (2s, Me2C); 1.57
(sept., J ˆ 6.9, Me2CH); 0.83 (d, J ˆ 6.9, Me2CH); 0.78( s, Me2CSi); 0.00 (s, Me2Si). 13C-NMR (75 MHz, CDCl3):
see Table 5; additionally, 164.48( s, CˆO); 133.41 (s); 132.91 (d); 128.89 (2d); 127.80 (2d); 114.53 (s, Me2CO2);
38.57 (q, MsO); 33.97 (d, Me2CH); 27.15, 25.36 (2q, Me2CO2); 25.07 (s, Me2CSi); 20.50 (q, Me2CH); 18.47 (q,
Me2CSi); À 3.33 (q, Me2Si).
N6-Benzoyl-8-(chloromethyl)-5'-O-[(dimethyl)(1,1,2-trimethylpropyl)silyl]-2',3'-O-isopropylideneadeno-
sine (38). A soln. of 35 (500 mg, 0.86 mmol) and Et3N (0.24 ml) in CH2Cl2 (5 ml) was cooled to 08, treated with
MsCl (0.13 ml, 1.7 mmol), warmed to 238, stirred for 1 h, treated with LiCl (360 mg, 8.57 mmol), stirred for 1 h,
diluted with CH2Cl2 (25 ml), washed with H2O (2 Â 10 ml) and brine (1 Â 10 ml), dried (MgSO4), and
evaporated. FC (AcOEt/cyclohexane 1 :4) gave 38 (380 mg, 74%). White foam. Rf (cyclohexane/AcOEt 1 :1)
0.47. 1H-NMR (300 MHz, CDCl3): see Table 4; additionally, 8.94 (br. s, NH); 8.05 7.90 (m, 2 arom. H); 7.64
7.40 (m, 3 arom. H); 1.64, 1.42 (2s, Me2C); 1.57 (sept., J ˆ 6.9, Me2CH); 0.84 (d, J ˆ 6.9, Me2CH); 0.80 (s,
Me2CSi); 0.04 (s, Me2Si). 13C-NMR (75 MHz, CDCl3): see Table 5; additionally, 164.94 (s, CˆO); 133.56 (s);
133.15 (d); 129.06 (2d); 128.22 (2d); 114.66 (s, Me2CO2); 34.25 (d, Me2CH); 27.43, 25.63 (2q, Me2CO2); 25.47 (s,
‡
Me2CSi); 20.46 (q, Me2CH); 18.65 (q, Me2CSi); À 3.28( q, Me2Si). MALDI-MS: 602 (15, [M ‡ H] ), 624 (25,
‡
[M ‡ Na] ).
REFERENCES
¬
[1] S. Eppacher, N. Solladie, B. Bernet, A. Vasella, Helv. Chim. Acta 2000, 83, 1311.
[2] W. Czechtizky, A. Vasella, Helv. Chim. Acta 2001, 84, 594.
[3] W. Czechtizky, A. Vasella, Helv. Chim. Acta 2001, 84, 1000.
[4] W. Czechtizky, X. Daura, A. Vasella, W. van Gunsteren, Helv. Chim. Acta 2001, 84, 2132.
[5] H. Gunji, A. Vasella, Helv. Chim. Acta 2000, 83, 1331.
[6] H. Gunji, A. Vasella, Helv. Chim. Acta 2000, 83, 2975.
[7] H. Gunji, A. Vasella, Helv. Chim. Acta 2000, 83, 3229.
¬
[8] S. Eppacher, N. Solladie, A. Vasella, in preparation.
[9] P. K. Bhardwaj, A. Vasella, Helv. Chim. Acta 2002, 85, 699.
[10] W. Saenger, −Principles of Nucleic Acid Structure×, Springer-Verlag, Berlin, 1984.
[11] A. Herbert, A. Rich, J. Biol. Chem. 1996, 271, 11595.
[12] S. Wˆlfl, C. Martinez, A. Rich, J. A. Majzoub, Proc. Natl. Acad. Sci. U.S.A. 1996, 93, 3664.
[13] H. Sugiyama, K. Kawai, A. Matsunaga, K. Fujimoto, I. Saito, H. Robinson, A. H.-J. Wang, Nucleic Acids
Res. 1996, 24, 1272.
[14] F. Mohamadi, N. G. J. Richards, W. C. Guida, R. Liskamp, C. Caufield, M. Lipton, G. Chang, T.
Hendrickson, W. C. Still, J. Comput. Chem. 1990, 11, 440.
[15] A. J. Matthews, A. Vasella, Chem. Commun. 2003, 950.
[16] S. Chladek, J. Smrt, Collect. Czech. Chem. Commun. 1964, 29, 214.
[17] H. Hayakawa, K. Haraguchi, H. Tanaka, T. Miyasaka, Chem. Pharm. Bull. 1987, 35, 72.
[18] D. B. Davies, Progr. NMR Spectrosc. 1978, 12, 135.
[19] L. Dudycz, R. Stolarski, R. Pless, D. Shugar, Z. Naturforsch. C: Biosci. 1978, 34, 359.
[20] C. Altona, M. Sundaralingam, J. Am. Chem. Soc. 1973, 95, 2333.
[21] P. A. Levene, R. S. Tipson, J. Biol. Chem. 1934, 106, 113.

Helvetica Chimica Acta Vol. 87 (2004)
[22] H. Tanaka, T. Miyasaka, Chem. Pharm. Bull. 1981, 29, 3565.
2295
[23] H. Rosemeyer, G. Toth, B. Golankiewicz, Z. Kazimierczuk, W. Bourgeois, U. Kretschmer, H. P. Muth, F.
Seela, J. Org. Chem. 1990, 55, 5784.
[24] D. Plochocka, A. Rabczenko, D. B. Davies, J. Chem. Soc., Perkin Trans. 2 1981, 8 2.
[25] B. Bernet, A. Vasella, Helv. Chim. Acta 2000, 83, 2055.
[26] R. R. Fraser, M. Kaufman, P. Morand, G. Govil, Can. J. Chem. 1969, 47, 403.
[27] M. Bessodes, J. Shamsazar, K. Antonakis, Synthesis 1988, 560.
[28] S. J. Danishefsky, S. L. DeNinno, S. Chen, L. Boisvert, M. Barbachyn, J. Am. Chem. Soc. 1989, 111, 5810.
[29] E. Trevisiol, E. Defrancq, J. Lhomme, A. Laayoun, P. Cros, Tetrahedron 2000, 56, 6501.
[30] H. Tanaka, H. Hayakawa, T. Miyasaka, Tetrahedron 1982, 38, 2635; A. Kittaka, H. Kato, H. Tanaka, Y.
Nonaka, M. Amano, K. T. Nakamura, T. Miyasaka, Tetrahedron 1999, 55, 5319.
[31] A. L. Pogolotti, D. V. Santi, −Bioorganic Chemistry, Vol. 1: Enzyme Action×, Academic Press, New York,
1977, p. 277 311.
[32] A. Dunger, H. H. Limbach, K. Weisz, J. Am. Chem. Soc. 2000, 122, 10109.
[33] G. M. Nagel, S. Hanlon, Biochemistry 1972, 11, 823.
[34] R. A. Newmark, C. R. Cantor, J. Am. Chem. Soc. 1968, 90, 5010.
[35] O. F. Schall, G. W. Gokel, J. Am. Chem. Soc. 1994, 116, 6089.
[36] J. D. Dunitz, Chem. Biol. 1995, 2, 709; E. Grunwald, L. L. Comeford, Protein-Solvent Interact. 1995, 421
443.
[37] J. S. Chen, R. B. Shirts, J. Phys. Chem. 1985, 89, 1643; J. S. Chen, F. Rosenberger, Tetrahedron Lett. 1990, 31,
3975.
[38] B. R. Peterson, Ph.D. Thesis, University of California, 1994.
Received March29, 2004