Oligonucleotide analogues with integrated bases and backbones: Part 26 - Synthesis, conformational analysis, and association of aminomethylene-linked self-complementary adenosine and uridine dinucleosides with enforced syn-conformation

Article abstract of DOI:10.1002/hlca.201000217

The self-complementary aminomethylene-linked A*[N]U* dinucleosides 23-26 were prepared by reductive coupling of aldehyde 10 and azide 8. The U*[N]A* sequence isomers 19-21 were similarly prepared from aldehyde 14 and azide 3. The substituents at C(6/I) of 23-26 and at C(8/I) of 19-21 strongly favour the syn-conformation. The A*[N]U* dinucleoside 23 associates more strongly than the sequence-isomeric U*[N]A* dinucleoside 19. The A*[N]U* dinucleosides 23 and 24 associate more strongly than the analogues devoid of the substituent at C(6/I), while the U*[N]A* dinucleoside 19 associates less strongly than the analogue devoid of the substituent at C(8/I). While 23 and 24 form cyclic duplexes mostly by Watson - Crick-type base pairing, 25 only forms linear associates. The U*[N]A* dinucleoside 19 forms mostly linear duplexes and higher associates, and 21 forms cyclic duplexes showing both Watson - Crick- and Hoogsteen-type base pairing. The cyclic duplexes of the aminomethylene-linked dinucleosides show both the gg- and gt-orientation of the linker, with the gg-orientation being preferred.

Full text of DOI:10.1002/hlca.201000217

1822
Helvetica Chimica Acta – Vol. 93 (2010)
Oligonucleotide Analogues with Integrated Bases and Backbones
Part 26
Synthesis, Conformational Analysis, and Association of Aminomethylene-Linked
Self-Complementary Adenosine and Uridine Dinucleosides with Enforced
syn-Conformation
by Katja Chiesa, Bruno Bernet, and Andrea Vasella*
Laboratorium fꢀr Organische Chemie, ETH Zꢀrich, Wolfgang Pauli-Strasse 10, CH-8093 Zꢀrich
(e-mail: vasella@org.chem.ethz.ch)
The self-complementary aminomethylene-linked A*[n]U* dinucleosides 23 – 26 were prepared by
reductive coupling of aldehyde 10 and azide 8. The U*[n]A* sequence isomers 19 – 21 were similarly
prepared from aldehyde 14 and azide 3. The substituents at C(6/I) of 23 – 26 and at C(8/I) of 19 – 21
strongly favour the syn-conformation. The A*[n]U* dinucleoside 23 associates more strongly than the
sequence-isomeric U*[n]A* dinucleoside 19. The A*[n]U* dinucleosides 23 and 24 associate more
strongly than the analogues devoid of the substituent at C(6/I), while the U*[n]A* dinucleoside 19
associates less strongly than the analogue devoid of the substituent at C(8/I). While 23 and 24 form cyclic
duplexes mostly by Watson – Crick-type base pairing, 25 only forms linear associates. The U*[n]A*
dinucleoside 19 forms mostly linear duplexes and higher associates, and 21 forms cyclic duplexes showing
both Watson – Crick- and Hoogsteen-type base pairing. The cyclic duplexes of the aminomethylene-
linked dinucleosides show both the gg- and gt-orientation of the linker, with the gg-orientation being
preferred.
Introduction. – Oligonucleotides with integrated bases and backbones (ONIBs)
replace the backbone of nucleic acids by linking elements between adjacent
nucleobases, and form cyclic duplexes and/or linear associates. This was shown by
analysing the association in CDCl3 of partially protected, self-complementary di- and
tetranucleosides of the type U*[x]A(*)1) and A*[x]U(*), possessing a variety of linking
elements [1 – 9]. A conformational analysis of the diastereoisomers of the constitu-
tionally isomeric cyclic duplexes of self-complementary ethylene-, oxymethylene-,
thiomethylene-, and aminomethylene-linked dinucleosides rationalized the propensity
of pairing, i.e., the formation of cyclic duplexes [7]. Pairing of partially protected, self-
complementary dinucleosides in CDCl3 depends on the structure of the linker, the syn-
orientation of the nucleobases, and the conformation of the ribose unit [7]. The
resulting cyclic duplexes form Watson – Crick (WC), reverse-Watson – Crick (rWC),
1
)
Conventions for abbreviated notation: The substitution at C(6) of pyrimidines and C(8) of purines is
denoted by an asterisk (*); for example, A* and U* for thiomethylated adenosine and uridine
derivatives, respectively. The moiety linking C(6)ꢀCH2 or C(8)ꢀCH2 of unit II and C(5’) of unit I is
indicated in square brackets, i.e., [c] for a C-, [o] for an O-, [n] for a N-, and [s] for a S-atom.
ꢁ 2010 Verlag Helvetica Chimica Acta AG, Zꢀrich

Helvetica Chimica Acta – Vol. 93 (2010)
1823
Hoogsteen (H), or reverse-Hoogsteen (rH) H-bonds [10]. Oxymethylene-linked
dinucleosides [5][6] form cyclic duplexes that are characterized by a gg-conformation
about the C(4’)ꢀC(5’) bond. This corresponds to the ap,ap-conformation of the 3-
oxapentylene fragment of these dinucleosides that agrees with the preferred ap,ap-
conformation of diethyl ether [11][12]. In contradistinction, thiomethylene-linked
dinucleosides form cyclic duplexes with a gt-conformation about the C(4’)ꢀC(5’) bond
corresponding to the ap,sc-conformation of the 3-thiapentylene fragment. The
preference for this conformation is in agreement with ab initio calculations for Et2S,
suggesting that it prefers the ap,sc- over the ap,ap-conformation [7][12][13]. Similarly
to the oxymethylene-linked U*[o]A(*) dinucleosides, the self-complementary adeno-
sine- and uridine-derived aminomethylene-linked A*[n]U and U*[n]A pair well, and
adopt preferentially a gg-conformation about the C(4’)ꢀC(5’) bond [9]. These
observations and the failure to prepare the sequence-isomeric A*[o]U(*) dinucleosides
raised our interest in longer aminomethylene-linked oligonucleosides, not least in view
of their solubility in H2O. Their synthesis requires the introduction of a substituted
methylene substituent in unit I, at C(6) of a uridine and at C(8) of an adenosine moiety.
We considered it of interest to investigate the introduction and the effect of such a
substituent at the dinucleoside stage, expecting it to favour the syn-conformation and
thereby to improve pairing.
We describe the synthesis of partially protected adenosine- and uridine-derived
aminomethylene-linked A*[n]U* and U*[n]A* dinucleosides which possess a
substituent at C(6/I) of U and at C(8/I) of A, respectively, their conformational
analysis, and their association.
Results and Discussion. – Synthesis of the U*[n]A* and A*[n]U* Dinucleosides. We
planned to synthesize these dinucleosides by the aza-Wittig reaction [14] of a
mononucleoside aldehyde with the iminophosphorane derived from a 5’-azido-5’-
deoxymononucleoside by treatment with Me3P [15][16], followed by reduction of the
resulting imines [17]. We thus required the azido nucleosides 3 and 8, and the aldehydes
10 and 14 (Schemes 1 and 2). In addition, the ethylamine 7 and the acetamide 9 were
of interest as references for the conformational analysis of the dinucleosides. We
prepared these compounds from the known protected 6-(hydroxymethyl)uridine 1 [7]
(Scheme 1) and the protected 8-formyladenosine 10 [7] (Scheme 2).
The C(6)-substituted p-toluenesulfonate 2 was obtained from the hydroxymethyl
derivative 1 (87%), and transformed into the C(6)-azidomethyl uridine 3 by treatment
with NaN3 in MeCN (82%; Scheme 1). THF also proved a suitable solvent for this
substitution, while the reaction in DMF, DMSO, or HMPT did not lead to the desired
product. The azido derivative 3 proved instable, and the crude product was used for
further transformations.
The C(5’)-substituted p-toluenesulfonate 6 was prepared (86%) from the known
monomethoxytrityl ether 5 [7] that was obtained from 1 and MMTrCl in CH2Cl2,
followed by desilylation [18][19] of the resulting ether 4 (Scheme 1). The azide 8 was
obtained in a yield of 95% from 6 by substitution with NaN3 in DMF in the presence of
excess NaI. Azide 8 was not formed in the presence of only catalytic amounts of NaI, or
by treating 6 with Me3SiN3, while the addition of substoechiometric amounts of 18-
crown-6 led to 8 in a yield of 82%. The ethylamine 7 was prepared by substitution of

1824
Helvetica Chimica Acta – Vol. 93 (2010)
Scheme 1
a) TsCl, 1,4-diazabicyclo[2.2.2]octane (DABCO), CH2Cl2; 87%. b) NaN3, MeCN, 808; 82%. c)
(MMTrCl), EtN(i-Pr)2, CH2Cl2; 88%. d) 1m Bu4NF in THF; 87%. e) TsCl, pyridine; 86%. f) 70% EtNH2
in H2O, DMF, 808; 88%. g) NaN3, NaI, DMF, 808; 95%. h) H2, Pd/C, Ac2O, MeOH; 59%. MMTr¼
(Monomethoxy)trityl (¼(4-methoxyphenyl)diphenylmethyl), TDS ¼ thexyl(dimethyl)silyl (¼dimethyl-
(1,1,2-trimethylpropyl)silyl).
p-toluenesulfonate 6 with EtNH2 in DMF (88%) [9], and the acetamide 9 (59%) by
hydrogenation of 8 in the presence of Pd/C and Ac2O [9].
Reduction [7][20] of aldehyde 10 gave 11 (93%) that was transformed into the
(monomethoxy)trityl ether 12 (85%; Scheme 2). Desilylation of 12 with Bu4NF yielded
91% of the selectively protected alcohol 13 that provided the p-toluenesulfonate 15
(75%) under standard conditions [9]. However, a range of conditions failed to
transform the C(8)-substituted alcohol 13 or the p-toluenesulfonate 15 into the desired
azidodeoxy-adenosine. Neither a Mitsunobu reaction (HN3) [21] of 13 nor a reaction
with diphenylphosphoryl azide (DPPA) [22] afforded the desired product, and
treatment of 15 with NaN3 in DMF in the absence or presence of 18-crown-6 or Bu4NI
led to the N3,5’-cyclo-nucleoside 162) and the hydrolysis product 17. Such hydrolysis
products are well precedented [25][26]. The facile intramolecular nucleophilic
substitution of N(3) of 15 is not surprising, as even anti-configured, 8-unsubstituted
analogues of 15 are readily transformed into cycloadenosine derivatives [23 – 27].
2
)
Compare [23] for the C(8)-unsubstituted, and [24] for the C(8)-unsubstituted and debenzoylated
analogues.

Helvetica Chimica Acta – Vol. 93 (2010)
1825
Scheme 2
a) NaBH4, MeOH; 93%. b) MMTrCl, EtN(i-Pr)2, CH2Cl2; 85%. c) 1m Bu4NF in THF; 91%. d) Dess –
Martinꢂs periodinane, CH2Cl2; 97%. e) TsCl, pyridine; 75%. f) NaN3, DMF; 16% of 16 and 30%
of 17.
We, therefore, decided to prepare the desired dinucleoside from the C(5’)-aldo-
adenosine 14 and the iminophosphorane derived from the azide 3. Oxidation of 13 with
Dess – Martinꢂs periodinane in CH2Cl2 [28] yielded 97% of 14, while oxidation with o-
iodoxybenzoic acid (IBX) in CH2Cl2 or in MeCN [29][30] provided 14 in a yield of only
10%. The aldehyde 14 must be used immediately for the aza-Wittig reaction; a sample
of 14 epimerized within 24 h by 50% to the corresponding a-l-lyxo analogue (cf. [31]
and refs. cit. therein).
The protected U*[n]A* dinucleoside 18 was synthesized by treating the C(6)-
(azidomethyl)-uridine 3 with Me3P, immediately followed by the reaction of the
resulting iminophosphorane 3A with aldehyde 14 and reduction of the resulting imines
to 18 (Scheme 3). The benzamide 18 was debenzoylated to the amine 19. The overall
reaction proceeded in a mediocre yield (51%), reflecting the poor stability of 3. The
1H-NMR spectrum of 19 shows two AB systems corresponding to the C(8/I)ꢀCH2 and
C(6/II)ꢀCH2 groups. C(8/I)ꢀCH2 resonates at 4.48 and 4.40 ppm (Jgem ¼ 12.0 Hz),
while C(6/II)ꢀCH2 resonates at 3.67 and 3.58 ppm (Jgem ¼ 14.4 Hz; see Table 6 in the
Exper. Part). Dinucleoside 19 was desilylated to the trityl ether 20, and detritylated by
treatment with Cl2CHCOOH and Et3SiH [32] to yield 65% of the silyl ether 21.

1826
Helvetica Chimica Acta – Vol. 93 (2010)
Scheme 3
a) Me3P, THF. b) 1. THF; 2. NaBH3CN, AcOH, MeOH. c) MeONa, MeOH; 51% from 3. d) HF ·
pyridine, THF; 73%. e) Et3SiH, Cl2CHCO2H, CH2Cl2; 65%.
The synthesis of the fully protected A*[n]U* dinucleoside 22 was initiated by
treating azide 8 with Me3P, immediately followed by treatment of the resulting
iminophosphorane 8Awith aldehyde 10 (Scheme 4). The resulting imines were directly
reduced with NaBH3CN to yield 22 (75% from 8). Similarly as observed during the
synthesis of the C(6)-unsubstituted A*[n]U dinucleoside [9], formation of the imines
1
required a high concentration of the mononucleosides. The H-NMR spectrum of 22
shows two AB systems, one corresponding to the C(6/I)ꢀCH2 group resonating at 3.99
and 3.92 ppm (Jgem ¼ 12.6 Hz) and the other corresponding to the C(8/II)ꢀCH2 group
resonating at 4.17 and 4.13 ppm (Jgem ¼ 14.1 Hz; see Table 8 in the Exper. Part).
Debenzoylation [7] of 22 yielded 68% of dinucleoside 23 that was desilylated with
(HF)3 · Et3N in THF to the alcohol 24 (70%). Detritylation of 23 with Cl2CHCOOH
and Et3SiH yielded 73% of the silyl ether 25; similarly, 24 yielded 78% of diol 26.
Conformation of the Uridine and Adenosine Mononucleosides. The C(6)-substi-
tuted uridine derivatives 2 and 3 prefer a syn-conformation, as revealed by d(HꢀC(2’))
of 5.12 and 5.23 ppm, respectively. Their substituents at C(6) do not significantly affect
the (N)-conformation of the ribose ring. The gg/gt/tg-rotamer distribution for the p-
toluenesulfonate 2 and the azide 3 shows a preference for the gt-rotamers (J(4’,5’a) þ

Helvetica Chimica Acta – Vol. 93 (2010)
1827
Scheme 4
a) Me3P, THF. b) 1. THF; 2. NaBH3CN, AcOH, MeOH; 75% from 8. c) MeONa, MeOH; 68%. d)
(HF)3 · Et3N, THF; 70%. e) Et3SiH, Cl2CHCO2H, CH2Cl2; 73% of 25; 78% of 26.
J(4’,5’b) ¼ 12.0 – 12.3 Hz; see Table 2 in the Exper. Part). Similarly to 2 and 3, the
monomethoxytrityl ethers 5 – 9 adopt a syn-conformation.
HOꢀC(5’) of 5 forms a partially persistent H-bond to C(2)¼O, leading to a more
extensive population of the gg-conformer (J(4’,5’a) þ J(4’,5’b) ¼ 7.9 Hz) [7] and to a
less pronounced (N)-conformation of the ribose ring of 5 than for the corresponding
azides 3 and 8, p-toluenesulfonates 2 and 6, and acetamide 9. The syn-conformation is
expected on account of the substitution at C(6). Similarly to the C(5’)ꢀOAc and

1828
Helvetica Chimica Acta – Vol. 93 (2010)
C(5’)ꢀSAc analogues [7], 6 shows a preference for the gg-conformation (gg/gt/tg
51:28 :21; calculated according [7][9])3), suggesting that these esters are sterically less
demanding than silyl ethers.
As for the analogous C(6)-unsubstituted N-ethylamine [9], there is no signal of the
imido NH group of 7. The OH signal of the tautomeric hydroxyimino structure of 7 in
CDCl3 is not visible (it is expected at ca. 11 ppm; see compound 25 below), whereas the
NH resonates at 7.63 ppm4). The hydroxyimino structure of 7 is well-evidenced by the
shift of 13C-NMR signals of the uridine C-atoms (see Table 3 in the Exper. Part). A
d(HꢀC(2’)) of 5.03 ppm and an EO-conformation of 7 is due to the intramolecular H-
bond. The EO-conformation leads to a larger distance between HꢀC(2’) and the
nucleobase, and is reflected by an upfield shift of the corresponding signal. J(4’,5’a) and
J(4’,5’b) of < 1.0 Hz document an exclusive population of the gg-conformer.
The preference for the gg-rotamer of the acetamide 9 (J(4’,5’pro-S) ¼ J(4’,5’pro-R) ¼
3.6 Hz; gg/gt/tg 76 :19 :5) may be due to a partially persistent intramolecular H-bond
between the amide NꢀH and the C(2)¼O group. The conformation and, indirectly, the
intramolecular H-bond is also evidenced by the CD spectrum of 9 in CHCl3 (1 mm
solution at 208; data not shown), characterized by a negative Cotton effect with a
minimum at 270 nm [33], similarly as for the C(6)-unsubstituted acetamide [9]. The
C(6)-monomethoxytrityl azide 8 appears to prefer the gt-rotamer (gg/gt/tg 7:61:32).
For this compound, there is no longer a correlation between chemical shift and coupling
constant that formed the basis of the assignment to Hpro-SꢀC(5’) and Hpro-RꢀC(5’)3); we
based the (tentative) gt/tg-rotamer ratio on the coupling constant and on the gauche
effect that favours the gt-conformer.
The alcohol 13 shows an exclusive preference for the gg-conformation (J(4’,5’a) ¼
J(4’,5’b) < 1.0 Hz), the (S)-conformation of the furanose ring (J(1’,2’)/J(3’,4’) ¼ 2.8),
and an upfield shift for HꢀC(2’)) (5.29 ppm), as it is typical for compounds possessing
an intramolecular H-bond from HOꢀC(5’) to N(3) [7]. The C(8)-substituted aldehyde
14 is expected to adopt a syn-conformation (d(HꢀC(2’)) ¼ 5.53 ppm), and a 1:1 (S)/
(N)-equilibrium may evidence an interaction of N(3) with the CHO group. The C(8)-
substituted p-toluenesulfonate 15 is also expected to adopt a syn-conformation, in spite
of the somewhat low value for d(HꢀC(2’)) of 5.47 ppm. The gg/gt/tg rotamer
distribution cannot be assigned due to poorly resolved signals. The N3,5’-cyclo-
nucleosides 16 and 17 are characterized by d(HꢀC(2’)) of 4.88 and 4.70 ppm,
respectively, and by an EO-conformation. The structure of 17 corresponds to a tautomer
of the originally proposed structure, as evidenced by the NH2 group resonating as two
doublets at 10.42 and 7.59 ppm (J ¼ 6 Hz), a cross-peak between these two signals in the
DQF-COSY spectrum, and a downfield shift for the N-benzoyl C¼O group, resonating
at 180.9 ppm, similarly as the corresponding C¼O group of 16. AM1 Calculations
suggest a NH ··· O¼C H-bond between the amino and the N-CHO group of 17. The
1,3,5-oxadiazepane ring of 16 adopts an envelope (O below the ring plane) and that of
The gg/gt/tg ratio is based on the assignment of the 1H-NMR signals to Hpro-SꢀC(5’) and
Hpro-RꢀC(5’). We assumed that the more highly shielded HꢀC(5’), showing the larger J(4’,5’)
coupling, corresponds to Hpro-RꢀC(5’).
3
)
4
)
The triplet at 6.15 ppm of the corresponding C(6)-unsubstituted analogue that we had assigned to
the OH group [9] should be assigned to the NH group.

Helvetica Chimica Acta – Vol. 93 (2010)
1829
17 a chair conformation (O below and N(5) above the ring plane), similarly as the
conformations of closely related cyclo-nucleosides in the solid state [34][35].
Association and Conformation of the U*[n]A* and A*[n]U* Dinucleosides in
CDCl3. The self-association of the U*[n]A* dinucleosides 19 and 21, and of the
A*[n]U* sequence isomers 23 – 25 was investigated by analysing conformational
aspects and the concentration dependence of the chemical shift of HꢀN(3/II) of 19 and
21, and HꢀN(3/I) of 23 – 25 (shift/concentration curves (SCCs)), as it was described
for aminomethylene-linked U*[n]A and A*[n]U analogues [9].
A) U*[n]A* Dinucleosides. The SCC of the tritylated and silylated U*[n]A*
dinucleoside 19 shows a curve progression typical of linear associates and cyclic
duplexes, characterized by a weak bending at low concentration and a continued,
slower increase at higher concentrations (Fig. 1). Unexpectedly, the association is
weaker than that of the C(8)-unsubstituted 19 H [9]. The SCC of 21 shows a curve
progression typical of cyclic duplexes and a minor contribution of linear associates,
characterized by a strong bending at low concentrations and an incomplete plateau
formation at higher concentrations.
The formation of Watson – Crick base-paired cyclic duplexes should lead to a
plateau at 13.3 and 12.75 ppm for C(8/I)-unsubstituted and -substituted U*[x]A(*)
dinucleosides, respectively [1]. The chemical shifts of HꢀN(3/II) at a 30 mm
Fig. 1. Shift/concentration curves (SCCs) of the U*[n]A(*) dinucleosides 19, 19 H, and 21 in CDCl3
solution

1830
Helvetica Chimica Acta – Vol. 93 (2010)
concentration of 19 H (12.27 ppm) and 19 (11.84 ppm) suggest mixtures of linear
associates and of cyclic duplexes possessing mostly Hoogsteen-type base pairs. The
weaker association of 19 than of 19 H is rationalized by an unfavourable steric
interaction of the bulky trityl group in the Hoogsteen-type base-paired cyclic duplexes
of 19. The (incomplete) plateau at ca. 12.1 ppm of 21 suggests a dominant formation of
cyclic duplexes and Hoogsteen-type base pairing.
The SCCs of 19 and 21 (Fig. 1) were analysed numerically by the method proposed
by Gutowsky and Saika [36], including a value of 7.70 ppm for a 0.001 mm solution,
corresponding to the chemical shift of the monoplex, as deduced from d(HꢀN(3)) of
uridine derivatives [7][9]. Inclusion of this value significantly reduced the variance of
Kass (Table 1). The thermodynamic parameters were determined by vanꢀt Hoff
1
analysis5) of the H-NMR spectra [9] recorded of ca. 2 – 5 mm solutions in CDCl3 in
intervals of 108 and in the temperature range of 7 to 508.
Table 1. Association Constants Kass and Extrapolated Chemical Shifts of the Monoplexes (c ¼ 0 mm) and
Duplexes (c ¼1 ) as Calculated from the Concentration Dependence of d(HN(3)) in CDCl3 at 295 K for
the U*[n]A* Dinucleosides 19 and 21, and the A*[n]U* Dinucleosides 23 – 25, and Determination of the
Thermodynamic Parameters of 19, 21, and 23 by vanꢂt Hoff Analysis of the Temperature Dependence of
d(HN(3)) for ca. 2 – 5-mm Solutions in CDCl3 at 7 – 508K
a
b
c
Dinucleoside
Kass
[mꢀ1
dmonoplex
[ppm]
)
dduplex
[ppm]
)
ꢀ DG295
)
ꢀ DH
ꢀ DS
]
[kcal/mol]
[kcal/mol]
[cal/mol · K]
U*[n]A* series
19d)
21d)
386 ꢁ 161
7.81 ꢁ 0.19
7.66 ꢁ 0.10
12.50 ꢁ 0.32
12.29 ꢁ 0.06
3.5
4.9
9.0
12.1
19.5
26.2
4428 ꢁ 613
A*[n]U* series
23d)
24e)
25
5662 ꢁ 562
4768 ꢁ 288
7.63 ꢁ 0.08
9.97 ꢁ 0.03
13.75 ꢁ 0.04
13.14 ꢁ 0.20
5.1
5.0
15.3
34.8
not determined
a) Extrapolated for c ¼ 0. b) Extrapolated for c ¼1 . c) Calculated from Kass
.
d) Including a value of
7.70 ppm for a 0.001-mm solution. e) Including a value of 10.00 ppm for a 0.001-mm solution.
The protected U*[n]A* dinucleoside 19 and the corresponding detritylated alcohol
21 associate more weakly (Kass ¼ 386 and 4428 mꢀ1, resp.; Table 1) than the correspond-
ing U*[o]A* dinucleosides (Kass ¼ 1611 and 6822 mꢀ1, resp. [7]), but more strongly than
the corresponding U*[s]A* dinucleosides (Kass ¼ 227 and 1259 mꢀ1, resp. [7]). In
agreement with the SCC, a ꢀ DH value of 9.0 kcal/mol for 19 evidences the formation
of linear associates and a minor amount only of cyclic duplexes (Table 1). The ꢀ DH
value of 12.1 kcal/mol of 21 is in agreement of the predominant formation of cyclic
duplexes. This corresponds to an average strength of an intermolecular H-bond of ca.
3.0 kcal/mol, and suggests a strong preference for Hoogsteen-type base-pairing, in
5
)
A caveat is in order. Neither the condition of an equilibrium exclusively between monoplex and
cyclic duplex, nor of the dependence of the chemical shift of HꢀN(3) exclusively upon H-bonding
are fulfilled. We give the ꢀ DH and ꢀ DS values to facilitate a semiquantitative comparison of their
contribution to the association.

Helvetica Chimica Acta – Vol. 93 (2010)
1831
agreement with the SCC. In this respect, these dinucleosides behave similarly to the
corresponding oxymethylene analogues.
1H-NMR Data characterizing the conformation for 20 – 30 mm solutions of the
U*[n]A* dinucleosides 19 – 21 in CDCl3 are compiled in Table 6 in the Exper. Part. The
adenosine moiety (unit I) of 19 – 21 adopts a syn-conformation and a ca. 1:1 (S)/(N)-
equilibrium. The J(4’,5’/I) values indicate a 80 :20 gg/gt-orientation of the linker of 19
and a 65 :28 :7 gg/gt/tg-orientation of the linker of 21. This evidences that both the gg-
and the gt-conformers of 21 form cyclic duplexes, while the tg-conformers can form
only linear associates. The uridine moiety (unit II) of 19 – 21 adopts a (N)-
conformation. The HOCH2 group of 20 prefers a gg/gt-orientation (gg/gt/tg
56 :36 :8), while the silyloxymethyl group of 21 adopts a gt/tg-orientation (gg/gt/tg
6 :57:37).
B) A*[n]U* Dinucleosides. The SCCs of the A*[n]U* dinucleosides 23 and 24 show
a curve progression typical of cyclic duplexes and reach a plateau at a concentration of
10 mm (Fig. 2). Their association is stronger than that of the C(6)-unsubstituted
analogues 23 H and 24 H [9] that prefer an anti-orientation of the uridine moiety at
concentrations where they do not yet associate. The SCC of the silylated (but no longer
methoxytritylated) dinucleoside 25 shows a curve progression with a weak bending at
low concentrations (< 5 mm) and a steady increase above a concentration of 5 mm: a
straight line showing no tendency for the formation of a plateau. This may be
rationalized by assuming a weak association of a monoplex to form linear duplexes,
and, conceivably, triplexes involving an intermolecular H-bond of the HOCH2 group.
The monomer cannot be the imido tautomer 25A, as this tautomer would require a
strong curvature of the SCC at low concentrations, tending towards a chemical shift of
ca. 7.7 ppm at 0 mm, as it is typical for the monoplex. This observation strongly suggests
that the monoplex is the hydroxyimino tautomer 25B, as it was already observed for the
monomeric ethylamino derivative 7. HOꢀC(2/I) of the hydroxyimino tautomer 25B is
engaged in a strong intramolecular H-bond to the amino group of the linker and,
therefore, expected to resonate at low field (>10 ppm). Thus, 25 forms an equilibrium
between the hydroxyimino tautomer 25B (dominant at low concentrations) and linear
duplexes possessing the imido structure required for base pairing. The weak association
(estimated Kass ꢂ 100 mꢀ1) indicates that the energy gained by forming a base pair (12 –
15 kcal/mol) is only slightly larger than the energy gained by tautomerisation and
formation of the intramolecular OH ··· N H-bond. The HꢀN(3/I) signal of the diol 26 is
hidden due to coalescence preventing the determination of the SCC. The position of
the plateau of 23 (13.5 ppm) and 24 (13.0 ppm) evidences Watson – Crick-type base
pairs.
As the curve of 25 does not show a plateau, it cannot be analysed numerically by the
method of Gutowsky and Saika. The lowest variance was obtained when a value of 7.7
and 10.0 ppm was included for the SCC of 23 and 24, respectively, at a concentration of
0.001 mm, evidencing that the monomer of 23 adopts completely the imido structure,
whereas the monomer of 24 prefers the hydroxyimino tautomer. The precise role of
C(6/I)ꢀCH2OH of 25 and HOꢀC(5’/II) of 24 in favouring the H-bonded hydroxy-
imino tautomer is not clear, while it certainly increases the polarity of the dinucleoside.
The protected A*[n]U* dinucleoside 23 and the tritylated alcohol 24 associate more
strongly (Kass ¼ 5662 and 4768 mꢀ1, resp.; Table 1) than the C(6/I)-unsubstituted 23 H

1832
Helvetica Chimica Acta – Vol. 93 (2010)
Fig. 2. Shift/concentration curves (SCCs) of the A*[n]U(*) Dinucleosides 23 – 25, 23 H, and 24 H in CDCl3
solution
and 24 H (Kass ¼ 3454 and 2429 mꢀ1, resp. [9]) and the corresponding A*[s]U*
dinucleosides (Kass ¼ 1334 and 658 mꢀ1, resp. [7]).
As the chemical shift of HꢀN(3/I) of monomeric 24 depends strongly on the
position of the equilibrium between the tautomers, we could only perform a vanꢀt Hoff
analysis for 23. The ꢀ DH value for 23 (15.3 kcal/mol; Table 1) agrees well with the
formation of cyclic duplexes by Watson – Crick-type H-bonding (ca. 4.0 kcal/mol per
intermolecular H-bond). A vapour-pressure determination of the apparent molecular
weight of 25 in 1, 5, and 20 mm solutions led to a degree of association of 0.92, 1.38, and
1.67, respectively, confirming the formation of linear duplexes only.
1H- and 13C-NMR data characterizing the conformation for 20 – 30 mm solution of
the A*[n]U* dinucleosides 22 – 26 in CDCl3 are compiled in Tables 8 and 9 in the
Exper. Part. The chemical shifts for uridine C-atoms reveal the imido structure of all
these dinucleosides. The chemical shift of 8.30 – 8.32 ppm for HꢀC(2/II) of the
adenosine moiety of 23 and 24 is in agreement with Watson – Crick-type pairing [7],
while values of 8.19 and 8.08 ppm for HꢀC(2/II) of 25 and 26 hint rather at Hoogsteen-
type base-pairing. The expected syn-conformation of the uridine unit I of 22 – 26 is
evidenced by the downfield shift of HꢀC(2’/I), which is influenced by the (N)/(S)-
equilibrium. Thus, 22, 25, and 26 (d ¼ 5.16 – 5.17 ppm) strongly prefer the (N)-
conformation, while 23 and 24 (d ¼ 5.27 and 5.24 ppm, resp.) adopt a ca. 1:1 (N)/(S)-

Helvetica Chimica Acta – Vol. 93 (2010)
1833
equilibrium. For the calculation of the conformational preference of the linker, we
assume that Hpro-RꢀC(5’) is the HꢀC(5’/I) possessing the larger vicinal coupling, i.e., it
is the more strongly shielded HꢀC(5’/I) of 23, but the more strongly deshielded
HꢀC(5’/I) of 22, 25, and 26. The two HꢀC(5’/I) of 24 are almost isochronous and
show a mean coupling of 4.3 Hz, suggesting similar couplings as observed for 25 (5.5
and 2.8 Hz). The linker of the benzamide 23 adopts a 48 :37:14 gg/gt/tg-orientation,
whereas the linker of 23 – 26 adopts only gg- and gt-orientations (gg/gt 82 :18 (23),
1
58 :42 (24 and 25), and 36 :64 (26)). Considering that the H-NMR spectra were
recorded at a concentration between 20 and 30 mm, when the formation of cyclic
duplexes is nearly complete, the relatively high proportion of the gt-conformer of 23
and 24 must mean that this conformer also forms cyclic duplexes. The linear duplexes of
25 avoid the tg-conformation.
The adenosine moiety (unit II) of the silyloxy derivatives 23 and 25 prefers the (N)-
conformation and is characterized by a ca. 1:1 gt/tg-mixture of conformers. The
adenosine moiety of the alcohols 24 and 26 shows the typical (S)- and the exclusive gg-
conformation resulting from the persistent intramolecular C(5’)ꢀOH ··· N(3) H-bond
[1][34].
Conclusions. – As it is typical also for the type and the strength of the association of
ONIBs possessing a different linker, the aminomethylene-linked dinucleosides depend
on the base sequence, the orientation of the linker relative to the furanose ring, and the
orientation of the nucleobases. However, monoplexes of the aminomethylene-linked
A*[n]U* dinucleosides may form significant amounts of their intramolecularly H-
bonded hydroxyimino tautomer, while their imido tautomers may associate at higher
concentrations and lead to either linear and/or cyclic duplexes. The equilibrium of
tautomers at lower concentrations is strongly affected by the protection of the OH
groups. Although the additional substituent in unit I of the aminomethylene-linked
dinucleosides does lead to a higher population of the syn-conformer, it will not, for
A*[n]U* sequence isomers, automatically favour the formation of cyclic duplexes, as
compared to the A*[n]U analogues. For the first time, we obtained evidence that both
the gg- and the gt-conformers of dinucleosides (21, 23, and 24) form cyclic duplexes.
The U*[n]A(*) dinucleosides pair more strongly than the U*[s]A(*) analogues, but
less well than the U*[o]A(*) analogues [7]. This may reflect a conformational aspect in
that pairing of U*[o]A(*) dinucleosides involves a gg-orientation of the linker, pairing
of the U*[n]A(*) analogues gg- and gt-orientations, and pairing of the U*[s]A(*)
analogues a gt-orientation.
We thank Syngenta AG, Basel, for generous financial support.
Experimental Part
General. See [9].
5’-O-[Dimethyl(1,1,2-trimethylpropyl)silyl]-2’,3’-O-isopropylidene-6-{[(4-methylphenylsulfonyl)-
oxy]methyl}uridine (2). A soln. of 1 [7] (280 mg, 0.61 mmol) in CH2Cl2 (2 ml) was cooled to 08, treated
with DABCO (138 mg, 1.23 mmol) and TsCl (174 mg, 0.92 mmol), stirred for 10 min, and diluted with
sat. NH4Cl soln. After extraction with CH2Cl2, the combined org. layers were dried (MgSO4) and
evaporated. FC (cyclohexane/AcOEt 2 :1 ! AcOEt) gave 2 (322 mg, 87%). Colourless foam. Rf

1834
Helvetica Chimica Acta – Vol. 93 (2010)
(cyclohexane/AcOEt 1:1) 0.52. [a]2D5 ¼ ꢀ1.2 (c ¼ 0.54, CHCl3). IR (ATR): 3182w (br.), 2965w, 2875w,
1709s, 1636w, 1466w, 1427w, 1385m, 1245s, 1209m, 1187m, 1138w, 1091w, 1065m, 1024m, 980w, 960w,
911w, 895w, 866m, 842w. 1H-NMR (300 MHz, CDCl3): see Table 2; additionally, 8.29 (br. s, NH); 7.84 (d,
J ¼ 8.4, 2 arom. H); 7.36 (d, J ¼ 8.1, 2 arom. H); 2.46 (s, Me); 1.59 (sept., J ¼ 6.9, Me2CH); 1.56, 1.34 (2s,
Me2C); 0.85 (d, J ¼ 6.9, Me2CH); 0.83, 0.82 (2s, Me2CSi); 0.07, 0.06 (2s, Me2Si). 13C-NMR (75 MHz,
CDCl3): see Table 3; additionally, 145.81, 132.03 (2s); 129.96 (2d); 128.10 (2d); 113.76 (s, Me2C); 34.13 (d,
Me2CH); 27.31, 25.42 (2q, Me2C); 25.36 (s, Me2CSi); 21.83 (q, Me); 20.38 (q, Me2CSi); 18.53 (q, Me2CH);
ꢀ 3.20 (q, Me2Si). HR-MALDI-MS: 649.2016 (28, [M þ K]þ, C28H42KN2O9SSiþ; calc. 649.2017),
633.2272 (100, [M þ Na]þ, C28H42N2NaO9SSiþ; calc. 633.2278).
Table 2. Selected 1H-NMR Chemical Shifts [ppm] and Coupling Constants [Hz] of the Uridine Monomers
2, 3, and 6 – 9 in CDCl3
2
3
6
7
8
9
HꢀC(5)
CHaꢀC(6)
CHbꢀC(6)
HꢀC(1’)
HꢀC(2’)
HꢀC(3’)
HꢀC(4’)
HaꢀC(5’)
HbꢀC(5’)
J(Ha,Hb)
J(1’,2’)
5.72
4.97
4.90
5.48
5.12
4.76
4.10
3.75
3.70
5.81
4.28
4.28
5.62
5.23
4.81
4.13
3.78
3.74
5.68
3.97
3.92
5.61
5.82
3.98
3.98
5.53
5.03
4.99
4.07
3.99
3.89
5.73
4.04
3.97
5.66
5.20
4.84
4.08
3.60
3.49
5.77
4.03
3.93
5.51
5.17
4.78
4.08
3.72
3.41
5.10
4.76a)
4.23 – 4.20
4.22
4.19
12.3
< 1.0
6.6
b
b
13.2
< 1.0
6.3
)
1.2
6.3
)
12.6
< 1.0
6.3
12.6
2.1
6.6
< 1.0
6.3
J(2’,3’)
J(3’,4’)
J(4’,5’a)
4.5
5.1
4.5
5.1
4.2
3.9
< 1.0
0
4.8
7.8
4.8
3.6
J(4’,5’b)
J(5’a,5’b)
6.9
10.8
7.2
10.8
5.4
10.2
0
12.6
4.8
12.3
3.6
14.1
a) The signal of HꢀC(3’) shows an additional splitting due to virtual coupling. b) Not assigned.
6-(Azidomethyl)-5’-O-[dimethyl(1,1,2-trimethylpropyl)silyl]-2’,3’-O-isopropylideneuridine (3). A
soln. of 2 (500 mg, 0.82 mmol) in dry MeCN (10 ml) was treated with NaN3 (532 mg, 8.2 mmol) and
stirred at 808 for 3 h. The soln. was cooled to 238 and filtered. Evaporation and drying gave 3 (324 mg,
82%). Colourless foam. Rf (AcOEt) 0.85. [a]D25 ¼ ꢀ12.4 (c ¼ 1.08, CHCl3). IR (ATR): 3195w (br.), 3054w
(br.), 2957w, 2867w, 2110m, 1692s, 1626w, 1460m, 1379m, 1329w, 1251m, 1210m, 1158m, 1132m, 1082s,
997w, 927w, 875m, 830s, 777m. 1H-NMR (300 MHz, CDCl3): see Table 2; additionally, 9.03 (br. s, NH);
1.60 (sept., J ¼ 6.9, Me2CH); 1.54, 1.34 (2s, Me2C); 0.86 (d, J ¼ 6.9, Me2CH); 0.85, 0.83 (2s, Me2CSi); 0.08,
0.07 (2s, Me2Si). 13C-NMR (75 MHz, CDCl3): see Table 3; additionally, 113.74 (s, Me2C); 34.15 (d,
Me2CH); 27.28, 25.38 (2q, Me2C); 25.38 (s, Me2CSi); 20.42, 20.37 (2q, Me2CSi); 18.53 (q, Me2CH); ꢀ 3.20
(q, Me2Si). HR-MALDI-MS: 504.2257 (100, [M þ Na]þ, C21H35N5NaO6Siþ; calc. 504.2249).
2’,3’-O-Isopropylidene-6-{[(4-methoxyphenyl)diphenylmethoxy]methyl}-5’-O-(4-methylphenylsulfo-
nyl)uridine (6). A soln. of 5 [7] (3.88 g, 6.62 mmol) in dry pyridine (20 ml) was cooled to 08, treated with
TsCl (1.89 g, 9.93 mmol), and stirred for 6 h at 0 – 238. The mixture was diluted with CH2Cl2 (80 ml),
washed with a 1m H2SO4 (2 ꢃ 70 ml) and sat. NaHCO3 soln. (70 ml), dried (MgSO4), and evaporated. FC
(cyclohexane/AcOEt 3 :1) gave 6 (4.2 g, 86%). Colourless foam. Rf (cyclohexane/AcOEt 1:1) 0.34.
[a]2D5 ¼ þ12.7 (c ¼ 0.32, CHCl3). IR (ATR): 3182w (br.), 3054w, 3024w, 2989w, 2930w, 2843w, 1689s,
1625w, 1607w, 1509m, 1492w, 1448m, 1381m, 1371m, 1300w, 1250m, 1211m, 1175s, 1157m, 1095m, 1063s,
1032m, 972s, 930w, 899w, 877m, 828s, 812m, 782m. 1H-NMR (300 MHz, CDCl3): see Table 2; additionally,

Helvetica Chimica Acta – Vol. 93 (2010)
1835
Table 3. Selected 13C-NMR Chemical Shifts [ppm] of the Uridine Monomers 2, 3, and 6 – 9 in CDCl3
2
3
6
7
8
9
C(2)
C(4)
C(5)
C(6)
CH2ꢀC(6)
C(1’)
C(2’)
C(3’)
C(4’)
C(5’)
149.61
161.71
104.58
147.44
65.35
91.75
83.94
81.80
89.42
63.71
150.01a)
162.04
103.68
149.47a)
51.07
91.70
84.06
81.86
89.52
150.53
163.25
102.86
152.39
62.04
92.22
84.62
81.71
148.88
171.70
107.03
154.62
62.69
92.90
80.22
79.71
84.21
60.46
150.60
163.21
103.06
152.27
62.28
91.93
84.62
82.46
87.21
52.66
151.08
162.68
103.31
152.26
62.29
91.70
83.50
80.71
84.86
40.95
86.54
70.19
63.73
a) Assignments may be interchanged.
8.66 (br. s, NH); 7.74 (d, J ¼ 8.1, 2 arom. H); 7.47 – 7.42 (m, 4 arom. H); 7.37 – 7.23 (m, 10 arom. H); 6.86 (d,
J ¼ 8.7, 2 arom. H); 3.81 (s, MeO); 2.41 (s, Me); 1.40, 1.27 (2s, Me2C). 13C-NMR (75 MHz, CDCl3): see
Table 3; additionally, 158.97, 144.74, 142.99, 142.88, 133.92, 132.69 (6s); 130.23 (4d); 129.54 (2d); 128.11
(2d); 128.01 (2d); 127.98 (2d); 127.43 (4d); 113.82 (s, Me2C); 113.36 (2d); 88.23 (s, Ph2C); 55.19 (q, MeO);
26.90, 25.13 (2q, Me2C); 21.56 (q, Me). HR-MALDI-MS: 779.2053 (47, [M þ K]þ, C40H40KN2O10Siþ; calc.
779.2035), 763.2311 (100, [M þ Na]þ, C40H40N2NaO10Siþ; calc. 763.2296). Anal. calc. for C40H40N2O10S
(740.83): C 64.85, H 5.44, N 3.78; found: C 64.73, H 5.59, N 3.63.
5’-Deoxy-5’-(ethylamino)-2’,3’-O-isopropylidene-6-{[(4-methoxyphenyl)diphenylmethoxy]methyl}-
uridine (7). A soln. of 6 (150 mg, 0.20 mmol) in dry DMF (3 ml) was treated with EtNH2 (70% in H2O,
13.0 ml, 0.18 mmol), and stirred at 808 for 2 h. The soln. was cooled to 238 and evaporated. FC (AcOEt/
MeOH 50 :1 ! 20 :1) gave 7 (108 mg, 88%). Colourless foam. Rf (AcOEt/MeOH 10 :1) 0.29. 1H-NMR
(300 MHz, CDCl3): see Table 2; additionally, 7.68 – 7.58 (br. s, HNꢀC(5’)); 7.45 (br. d, J ꢄ 8.4, 4 arom. H);
7.31 (t, J ꢄ 8.4, 4 arom. H); 7.27 (d, J ¼ 8.7, 2 arom. H); 7.22 (br. t, J ꢄ 7.8, 2 arom. H); 6.83 (d, J ¼ 8.7, 2
arom. H); 3.77 (s, MeO); 3.50 – 3.46 (m, CH2N); 1.23, 1.17 (2s, Me2C); 1.14 (t, J ¼ 7.2, MeCH2). 13C-NMR
(75 MHz, CDCl3): see Table 3; additionally, 158.73, 143.25, 143.14, 134.22 (4s); 130.17 – 127.26 (several
d); 114.87 (s, Me2C); 113.30 (2d); 87.77 (s, Ph2C); 55.26 (q, MeO); 36.72 (t, CH2N); 27.15, 25.18 (2q,
Me2C); 14.67 (q, MeCH2N). HR-MALDI-MS: 652.2431 (16, [M þ K]þ, C35H39KN3O7þ ; calc. 652.2420),
636.2696 (34, [M þ Na]þ, C35H39N3NaO7þ ; calc. 636.2680), 614.2875 (100, [M þ H]þ, C35H40N3Oþ7 ; calc.
614.2861).
5’-Azido-5’-deoxy-2’,3’-O-isopropylidene-6-{[(4-methoxyphenyl)diphenylmethoxy]methyl}uridine
(8). A soln. of 6 (1.02 g, 1.38 mmol) in dry DMF (8 ml) was treated with NaN3 (1.79 g, 27.6 mmol) and
NaI (1.03 g, 6.9 mmol) and stirred at 808 for 6 h. The soln. was cooled to 238, diluted with AcOEt (50 ml),
washed with H2O (5 ꢃ 50 ml), dried (MgSO4), and evaporated to give 8 (800 mg, 95%). Colourless foam.
Rf (AcOEt) 0.81. [a]2D5 ¼ þ12.0 (c ¼ 0.37, CHCl3). IR (ATR): 3195w (br.), 3059w, 2989w, 2934w, 2098m,
1688s, 1607w, 1509m, 1490w, 1447m, 1381m, 1298w, 1251s, 1212m, 1179m, 1157m, 1093m, 1064s, 1033m,
978w, 905m, 876m, 831m. 1H-NMR (300 MHz, CDCl3): see Table 2; additionally, 10.00 (br. s, NH); 7.49 –
7.44 (m, 4 arom. H); 7.37 – 7.22 (m, 8 arom. H); 6.86 (d, J ¼ 7.2, 2 arom. H); 3.81 (s, MeO); 1.44, 1.30 (2s,
Me2C). 13C-NMR (75 MHz, CDCl3): see Table 3; additionally, 158.87, 143.05, 142.89, 133.93 (4s);
130.26 – 127.35 (several d); 113.91 (s, Me2C); 113.34 (2d); 88.29 (s, Ph2C); 55.30 (q, MeO); 27.20, 25.36
(2q, Me2C). HR-MALDI-MS: 650.2006 (60, [M þ K]þ, C33H33KN5O7þ ; calc. 650.2012), 634.2262 (100,
[M þ Na]þ, C33H33N5NaOþ7 ; calc. 634.2272).
5’-Acetamido-5’-deoxy-2’,3’-O-isopropylidene-6-{[(4-methoxyphenyl)diphenylmethoxy]methyl}uri-
dine (9). A suspension of 8 (200 mg, 0.33 mmol), 10% Pd/C (30 mg), and Ac2O (0.04 ml, 0.66 mmol) in
MeOH (5 ml) was stirred under 1 atm of H2 at 248 for 20 h and filtered through Celite. Evaporation and
FC (cyclohexane/AcOEt 2 :1 ! AcOEt) gave 9 (122 mg, 59%). Colourless foam. Rf (AcOEt/MeOH
10 :1) 0.69. [a]2D5 ¼ ꢀ26.6 (c ¼ 0.32, CHCl3). IR (ATR): 3318w (br.), 3186w (br.), 3056w, 2986w, 2927w,

1836
Helvetica Chimica Acta – Vol. 93 (2010)
2830w, 1692s, 1608m, 1538w, 1509m, 1447m, 1373s, 1299w, 1248s, 1210m, 1178m, 1156m, 1091m, 1062s,
978m, 901w, 876w, 830m, 796w. 1H-NMR (300 MHz, CDCl3): see Table 2; additionally, 8.82 (br. s, NH);
7.48 – 7.43 (m, 4 arom. H); 7.38 – 7.24 (m, 8 arom. H); 6.88 – 6.83 (m, 2 arom. H); 6.56 (dd, J ¼ 6.6, 3.0,
AcNH); 3.81 (s, MeO); 1.99 (s, AcN); 1.37, 1.27 (2s, Me2C). 13C-NMR (75 MHz, CDCl3): see Table 3;
additionally, 170.58 (s, MeC¼O); 158.89, 142.96, 142.84, 133.84 (4s); 130.24 – 127.40 (several d); 114.32 (s,
Me2C); 113.35 (2d); 88.30 (s, Ph2C); 55.31 (q, MeO); 27.28, 25.35 (2q, Me2C); 23.15 (q, MeC¼O). HR-
MALDI-MS: 666.2205 (32, [M þ K]þ, C35H37KN3O8þ ; calc. 666.2212), 650.2475 (100, [M þ Na]þ,
C35H37N3NaOþ8 ; calc. 650.2473).
N6-Benzoyl-9-(2,3-O-isopropylidene-b-d-ribo-pentodialdo-1,4-furanosyl)-8-{[(4-methoxyphenyl)di-
phenylmethoxy]methyl}adenine (14). A soln. of 13 [7] (300 mg, 0.42 mmol) in CH2Cl2 (3 ml) was treated
with Dess – Martin periodinane (15 wt.-% in CH2Cl2, 0.8 ml, 0.42 mmol), stirred for 15 min, and treated
with sat. Na2S2O3 soln. After extraction with CH2Cl2, the combined org. layers were dried (MgSO4) and
evaporated to afford 14 (290 mg, 97%). Upon standing for 1 d, 14 epimerized to a 1:1 mixture of 14 and
1
its a-l-lyxo-analogue; the H-NMR spectrum showed only signals of 14, but the 13C-NMR spectrum
(recorded 19 h later) a 1:1 mixture of diastereoisomers. Yellow foam. Rf (cyclohexane/AcOEt 1:1) 0.48.
IR (ATR): 3229w (br.), 3054w, 3024w, 2989w, 2953w, 2931w, 2830w, 1702w, 1608m, 1583m, 1530w, 1509m,
1487m, 1456m, 1448m, 1430w, 1384w, 1374w, 1341w, 1298w, 1250s, 1212m, 1179m, 1156w, 1092m, 1062s,
1032m, 975w, 907m, 880w, 866w, 831w, 797w, 764w, 727s, 704s, 660w, 646w, 631w, 608w. 1H-NMR
(300 MHz, CDCl3): see Table 4; additionally, 8.94 (br. s, NH); 8.00 (d, J ¼ 7.5, 2 arom. H); 7.62 – 7.22 (m,
15 arom. H); 6.85 (d, J ¼ 8.7, 2 arom. H); 3.78 (s, MeO); 1.54, 1.39 (2s, Me2C). 13C-NMR (75 MHz,
CDCl3; 1:1 mixture of diastereoisomers): see Table 5; additionally, 164.30 (s, NC¼O); 158.76, 143.18/
143.10 (2s), 134.11/133.49 (2s); 132.77 (br. d); 130.52 – 127.09 (several d); 113.72 (s, Me2C); 113.37/113.20
(2d); 88.23 (s, Ph2C); 55.38/55.14 (2q, MeO); 26.71, 25.34/25.27 (3q, Me2C). HR-MALDI-MS: 750.2332
(14, [M þ K]þ, C41H37KN5O7þ ; calc. 750.2325), 734.2590 (100, [M þ Na]þ, C41H37N5NaOþ7 ; calc.
734.2585), 712.2777 (26, [M þ H]þ, C41H38N5O7þ ; calc. 712.2766).
N6-Benzoyl-2’,3’-O-isopropylidene-8-{[(4-methoxyphenyl)diphenylmethoxy]methyl}-5’-O-(4-methyl-
phenylsulfonyl)adenosine (15). A soln. of 13 (500 mg, 0.7 mmol) in CH2Cl2 (3 ml) was cooled to 08,
treated with DABCO (236 mg, 2.10 mmol) and TsCl (267 mg, 1.40 mmol), stirred for 10 min, and treated
Table 4. Selected 1H-NMR Chemical Shifts [ppm] and Coupling Constants [Hz] of the Adenosine
Monomers 14 – 17 in CDCl3
14
15
8.62
4.49
4.43
6.22
5.47
5.03
16
7.82
17a)
HꢀC(2)
CHaꢀC(8)
CHbꢀC(8)
HꢀC(1’)
HꢀC(2’)
HꢀC(3’)
HꢀC(4’)
HaꢀC(5’)
HbꢀC(5’)
J(Ha,Hb)
J(1’,2’)
J(2’,3’)
J(3’,4’)
J(4’,5’a)
J(4’,5’b)
J(5’a,5’b)
8.62
4.64
4.54
6.53
5.53
5.32
4.48
9.32
–
8.49
4.37
4.16
5.98
4.70
4.68
4.64
4.95
3.07
12.2
0
4.56
4.11
5.87
4.88
4.51
4.75
4.50
4.12
4.23 – 4.13
4.40 – 4.30
4.40 – 4.30
12.3
12.3
11.7
0
< 1.0
6.3
< 1.0
0
1.2
6.3
3.3
5.7
0
2.1
2.7
14.4
5.6
0
2.4
1.6
b
)
)
)
b
b
–
–
14.0
a) Same numbering as for 14 – 16. Assignments based on a DQF-COSY and a HSQC spectrum. b) Not
assigned.

Helvetica Chimica Acta – Vol. 93 (2010)
1837
Table 5. Selected 13C-NMR Chemical Shifts [ppm] of the Adenosine Monomers 14 – 17 in CDCl3
14a)
15
16b)
17c)
C(2)
C(4)
C(5)
C(6)
152.21, 152.07
151.14
121.92
152.10
151.37
122.04
151.62
149.12
59.36
90.07
83.81
81.60
142.93
137.31
121.37
149.81
146.05
58.34
91.03
85.03
80.30
86.09
55.98
163.56
137.49
125.54
160.99
142.60
59.01
90.97
85.60
82.26
85.80
151.41 (br.)
149.13 (br.)
59.63 (br.)
93.19 (br.)
85.16, 84.94
84.21, 83.97
91.63, 91.53
199.79, 199.67
C(8)
CH2ꢀC(8)
C(1’)
C(2’)
C(3’)
C(4’)
C(5’)
84.69
69.28
44.83
a) 1 : 1 Mixture of 14 and its a-l-lyxo-analogue. b) Assignment based on comparison with [38][39].
c) Same numbering as for 14 – 16. Assignments based on a HSQC spectrum and on comparison with data
in [40].
with sat. NH4Cl soln. After extraction with CH2Cl2, the combined org. layers were dried (MgSO4) and
evaporated. FC (cyclohexane/AcOEt 3 :1 ! AcOEt) gave 15 (456 mg, 75%). Colourless foam. Rf
(cyclohexane/AcOEt 1:1) 0.63. [a]2D5 ¼ ꢀ19.4 (c ¼ 0.40, CHCl3). IR (ATR): 3054w, 2986w, 2927w,
1698w, 1607m, 1579m, 1537w, 1508m, 1488m, 1447m, 1360m, 1338m, 1297w, 1248s, 1188m, 1175s, 1156m,
1
1094m, 1066s, 1031m, 978s, 863w, 827m, 798m. H-NMR (300 MHz, CDCl3): see Table 4; additionally,
9.11 (br. s, NH); 8.00 (d, J ¼ 7.5, 2 arom. H); 7.62 – 7.47 (m, 8 arom. H); 7.39 – 7.22 (m, 9 arom. H); 7.11 (d,
J ¼ 8.4, 2 arom. H); 6.85 (d, J ¼ 8.7, 2 arom. H); 3.76 (s, MeO); 2.34 (s, Me); 1.48, 1.33 (2s, Me2C).
13C-NMR (75 MHz, CDCl3): see Table 5; additionally, 164.35 (s, NC¼O); 158.77, 144.68 (2s); 143.10 (2s);
134.09, 133.59 (2s); 132.70 (d); 132.23 (s); 130.40 – 127.22 (several d); 114.46 (s, Me2C); 113.29 (2d); 88.07
(s, Ph2C); 55.28 (q, MeO); 27.20, 25.53 (2q, Me2C); 21.69 (q, Me). HR-MALDI-MS: 906.2573 (24, [M þ
K]þ, C48H45KN5O9Sþ; calc. 906.2570), 890.2814 (100, [M þ Na]þ, C48H45N5NaO9Sþ; calc. 890.2830),
868.3004 (36, [M þ H]þ, C48H46N5O9Sþ; calc. 868.3011).
Attempted Azidation of 15. A soln. of 15 (100 mg, 0.11 mmol) in dry DMF (2 ml) was treated with
NaN3 (63 mg, 1.15 mmol) and stirred for 4 h at 808. The soln. was cooled to 238, diluted with AcOEt
(5 ml), washed with H2O (5 ꢃ 5 ml), dried (MgSO4), and evaporated. FC (cyclohexane/AcOEt 3 :1 !
AcOEt) gave 16 (12 mg, 16%) and 17 (23 mg, 30%).
Data of 3,5’-Anhydro-N6-benzoyl-2’,3’-O-isopropylidene-8-{[(4-methoxyphenyl)diphenylmethoxy]-
methyl}adenosine (16). Rf (cyclohexane/AcOEt 1:1) 0.73. IR (ATR): 3100 – 2900w, 1640w (sh), 1610m,
1600w (sh), 1509w, 1447w, 1405w, 1381w, 1309w, 1272w, 1250m, 1210w, 1178w, 1159w, 1132w, 1095w,
1052m, 975w, 905s, 874w, 859w, 834w, 725s, 712s, 677w, 646m. 1H-NMR (300 MHz, CDCl3): see Table 4;
additionally, 10.42 (br. d, J ¼ 6.0, NH); 8.16 (d, J ¼ 6.9, 2 arom. H); 7.50 – 7.20 (m, 15 arom. H); 6.84 (d,
J ¼ 8.6, 2 arom. H); 3.80 (s, MeO); 1.47, 1.20 (2s, Me2C). 13C-NMR (75 MHz, CDCl3): see Table 5;
additionally, 180.66 (s, PhC¼O); 158.96, 143.41, 143.14, 135.37, 133.97 (5s); 131.94 (d); 130.43 (2d);
129.64 (2d); 128.16 – 127.07 (several d); 113.58 (s, Me2C); 113.34 (2d); 87.92 (s, Ph2C); 55.28 (q, MeO);
26.15, 24.73 (2q, Me2C). HR-MALDI-MS: 696.2817 (100, [M þ H]þ, C41H40N5O6þ ; calc. 696.2822),
273.1267 (33, CPh2(C6H4OMe)þ; calc. 273.1274).
Data of N5,5’-Anhydro-4-(Nꢂ-benzoylcarbamimidoyl)-5-formamido-1-(2,3-O-isopropylidene-b-d-ri-
bofuranosyl)-2-{[(4-methoxyphenyl)diphenylmethoxy]methyl}-1H-imidazole (17). Yellow foam. Rf (cy-
1
clohexane/AcOEt 1:1) 0.41. H-NMR (400 MHz, CDCl3; assignments based on a DQF-COSY and a
HSQC spectrum): see Table 4; additionally, 10.42 (br. d, J ¼ 6.0, NH); 8.03 (d, J ¼ 6.8, 2 arom. H); 7.59
(br. d, J ¼ 6.0, NH); 7.43 – 7.15 (m, 15 arom. H); 6.78 (d, J ¼ 8.8, 2 arom. H); 3.72 (s, MeO); 1.43, 1.24 (2s,
Me2C). 13C-NMR (100 MHz, CDCl3; assignments based on a HSQC spectrum): see Table 5; additionally,
180.92 (s, PhC¼O); 158.96, 143.50, 143.28, 134.44, 132.25 (5s); 131.84 (d); 130.49 (2d); 129.49 (2d);

1838
Helvetica Chimica Acta – Vol. 93 (2010)
128.41 – 127.97 (several d); 127.30, 127.29 (2d); 113.37 (2d); 113.06 (s, Me2C); 87.95 (s, Ph2C); 55.25 (q,
MeO); 26.24, 24.68 (2q, Me2C). HR-MALDI-MS: 714.2913 (100, [M þ H]þ, C41H40N5O6þ ; calc.
714.7851), 273.1265 (33, CPh2(C6H4OMe)þ; calc. 273.1274).
5’-O-[Dimethyl(1,1,2-trimethylpropyl)silyl]-2’,3’-O-isopropylideneuridine-6-methyl-(61 ! 5’-N)-5’-
amino-5’-deoxy-2’,3’-O-isopropylidene-8-{[(4-methoxyphenyl)diphenylmethoxy]methyl}adenosine (19).
A soln. of 3 (200 mg, 0.41 mmol) in THF (1.5 ml) was treated dropwise with 1m Me3P in THF
(0.46 ml, 0.46 mmol), and stirred for 24 h at 238 (TLC: complete conversion of 3). The mixture was
treated with a soln. of 14 (168 mg, 0.41 mmol) in THF (1 ml), stirred for 24 h at 238, and evaporated. A
soln. of the residue in AcOEt (10 ml) was washed with H2O (3 ꢃ 10 ml), dried (MgSO4), and evaporated.
A soln. of the residue in MeOH (3 ml) was treated with NaBH3CN (103 mg, 1.64 mmol) and AcOH
(0.1 ml, 1.64 mmol), and stirred for 20 h at 238. The mixture was diluted with AcOEt, washed with a sat.
NaHCO3 soln. and brine, dried (MgSO4), and evaporated. A soln. of the residue (crude 18) in MeOH
(1 ml) was treated with a soln. of MeONa (221 mg, 4.10 mmol) in MeOH (2 ml) and stirred for 24 h at
248. After evaporation, a soln. of the residue in AcOEt (5 ml) was washed with a sat. NH4Cl soln. and
brine, dried (MgSO4), and evaporated. FC (cyclohexane/AcOEt/acetone 4 :2 :1 ! 2 :2 :1) gave 19
(220 mg, 51%). Colourless foam. Rf (cyclohexane/AcOEt/acetone 1:1:1) 0.32. IR (ATR): 3323w, 3191w,
2984w, 2956w, 2931w, 2870w, 2835w, 1689s, 1644m, 1606m, 1579w, 1510w, 1487w, 1463w, 1447m, 1379m,
1329w, 1299w, 1251m, 1212m, 1180w, 1156m, 1129w, 1063s, 1034m, 977w, 908m, 871w, 862w, 829s, 799w,
777m, 766m, 731s, 705m, 662w, 651w, 647w. 1H-NMR (400 MHz, CDCl3; assignments based on a DQF-
COSY spectrum): see Table 6; additionally, 11.20 (s, HꢀN(3/II)); 7.50 (d, J ¼ 8.0, 4 arom. H); 7.39 (d, J ¼
8.8, 2 arom. H); 7.33 – 7.22 (m, 6 arom. H); 6.84 (d, J ¼ 8.8, 2 arom. H); 6.40 (br. s, H2NꢀC(6/I)); 3.82 –
3.75 (br. s, HNꢀC(5’/I)); 3.79 (s, MeO); 1.58 (sept., J ¼ 6.9, Me2CH); 1.54, 1.43, 1.36, 1.26 (4s, 2 Me2C);
0.83 (d, J ¼ 6.9, Me2CH); 0.80, 0.79 (2s, Me2CSi); 0.04, 0.03 (2s, Me2Si). 13C-NMR (100 MHz, CDCl3): see
Table 7; additionally, 158.97, 143.72, 143.64, 134.75 (4s); 130.60 (2d); 128.59 (2d); 128.56 (2d); 128.08
(4d); 127.30 (2d); 114.16, 113.45 (2s, 2 Me2C); 113.35 (2d); 89.67 (s, Ph2C); 55.34 (q, MeO); 34.21 (d,
Me2CH); 27.67, 27.50, 25.91, 25.73 (4q, 2 Me2C); 25.39 (s, Me2CSi); 20.45 (q, Me2CSi); 18.55 (q, Me2CH);
ꢀ 3.18 (q, Me2Si). HR-MALDI-MS: 1085.4553 (50, [M þ K]þ, C55H70KN8O11Siþ; calc. 1085.4565),
1069.4845 (100, [M þ Na]þ, C55H70N8NaO11Siþ; calc. 1069.4826).
2’,3’-O-Isopropylideneuridine-6-methyl-(61 ! 5’-N)-5’-amino-5’-deoxy-2’,3’-O-isopropylidene-8-{[(4-
methoxyphenyl)diphenylmethoxy]methyl}adenosine (20). In a polyethylene flask, a soln. of 19 (100 mg,
0.1 mmol) in THF (1 ml) was treated with HF · pyridine (86 ml, 1 mmol) and stirred for 10 min at 248. The
mixture was treated with 1m NaOH soln. and extracted with AcOEt. The combined org. phases were
washed with brine, dried (MgSO4), and evaporated. FC (CH2Cl2/MeOH 60 :1 ! 20 :1) gave 20 (66 mg,
73%). Colourless foam. Rf (CH2Cl2/MeOH 10 :1) 0.70. IR (ATR): 3334w, 3192w, 2989w, 2933w, 1693s,
1639m 1607m, 1579m, 1512m, 1490w, 1443m, 1379m, 1330m, 1297m, 1251m, 1212m, 1180m, 1155m, 1063s,
1032s, 978m, 901w, 863m, 832m, 794m, 765m, 731m, 704s. 1H-NMR (400 MHz, CDCl3; assignment based
on a DQF-COSY spectrum): see Table 6; additionally, 12.16 – 11.96 (br. s, HꢀN(3/II)); 7.50 (d, J ¼ 7.6, 4
arom. H); 7.38 (d, J ¼ 9.2, 2 arom. H); 7.32 – 7.20 (m, 6 arom. H); 6.84 (d, J ¼ 8.8, 2 arom. H); 6.55 (br. s,
H2NꢀC(6/I)); 3.79 (s, MeO); 1.55, 1.44, 1.37, 1.35 (4s, 2 Me2C); signals for HNꢀC(5’/I) and HOꢀC(5’/II)
not visible. 13C-NMR (100 MHz, CDCl3): see Table 7; additionally, 158.88, 143.62, 143.57, 134.64 (4s);
130.52 (2d); 128.48 (4d); 127.99 (4d); 127.21 (2d); 114.17, 113.96 (2s, 2 Me2C); 113.36 (2d); 88.08 (s,
Ph2C); 55.25 (q, OMe); 27.56, 27.45, 25.80, 25.47 (4q, 2 Me2C). HR-MALDI-MS: 943.3387 (60, [M þ K]þ,
C47H52KN8Oþ11 ; calc. 943.3387), 927.3655 (100, [M þ Na]þ, C47H52N8NaOþ11 ; calc. 927.3648), 905.3802 (82,
[M þ H]þ, C47H53N8O1þ1 ; calc. 905.3828).
5’-O-[Dimethyl(1,1,2-trimethylpropyl)silyl]-2’,3’-O-isopropylideneuridine-6-methyl-(61 ! 5’-N)-5’-
amino-5’-deoxy-8-(hydroxymethyl)-2’,3’-O-isopropylideneadenosine (21). A soln. of 19 (100 mg,
0.1 mmol) in CH2Cl2 (2 ml) was treated with Et3SiH (0.19 ml, 1.2 mmol) and Cl2CHCO2H (66 ml,
0.8 mmol), and stirred for 10 min at 248. The mixture was treated with Amberlite IRA-68 (20 – 50 mesh;
free base), stirred for 30 min, filtered, and evaporated. FC (CH2Cl2/MeOH 80 :1 ! 30 :1) gave 21 (50 mg,
65%). Colourless solid. Rf (CH2Cl2/MeOH 10 :1) 0.61. [a]2D5 ¼ ꢀ3.5 (c ¼ 0.2, CHCl3). IR (ATR): 3334w,
3194w, 2953w, 2928w, 2864w, 1690s, 1643m, 1607m, 1579w, 1454m, 1375m, 1329m, 1295m, 1251m, 1211m,
1
1157m, 1132m, 1073s, 986m, 935m, 860m, 829s, 798m, 776s. H-NMR (400 MHz, CDCl3; assignments
based on a DQF-COSY spectrum): see Table 6; additionally, 11.63 (br. s, HꢀN(3/II)); 6.77 (br. s,

Helvetica Chimica Acta – Vol. 93 (2010)
1839
Table 6. Selected 1H-NMR Chemical Shifts [ppm] and Coupling Constants [Hz] of the U*[n]A*
Dinucleosides 19 – 21 in CDCl3
19a)
20a)
21a)
19a)
20a)
21a)
Adenosine unit (I)
HꢀC(2/I)
8.26
4.48
4.40
6.04
5.66
5.34
4.27
3.04
2.93
8.27
4.48
4.42
6.06
5.65
5.28
4.29
2.98
2.98
8.18
4.93
4.81
6.22
5.59
5.30
4.39
3.05
2.88
CHaꢀC(8/I)
CHbꢀC(8/I)
HꢀC(1’/I)
J(Ha,Hb/I)
J(1’,2’/I)
J(2’,3’/I)
12.0
3.6
6.4
11.6
3.2
6.4
14.8
3.2
6.4
HꢀC(2’/I)
HꢀC(3’/I)
J(3’,4’/I)
3.6
3.2
3.6
12.0
ca. 3.6
ca. 3.6
ca. 3.6
3.6
3.6
4.4
12.0
HꢀC(4’/I)
J(4’,5’a/I)
J(4’,5’b/I)
J(5’a,5’b/I)
HaꢀC(5’/I)
b
HbꢀC(5’/I)
)
Uridine unit (II)
HꢀC(5/II)
5.72
3.67
3.58
6.16
5.32
4.86
4.13
3.82 – 3.72
3.82 – 3.72
5.70
3.64
3.59
5.97
5.30
5.08
4.16
3.83
3.76
5.67
3.66
3.59
6.11
5.31
4.81
4.13
3.78
3.73
CHaꢀC(6/II)
CHbꢀC(6/II)
HꢀC(1’/II)
HꢀC(2’/II)
HꢀC(3’/II)
HꢀC(4’/II)
HaꢀC(5’/II)
HbꢀC(5’/II)
J(Ha,Hb/II)
J(1’,2’/II)
J(2’,3’/II)
14.4
< 1.0
6.0
14.4
1.2
6.0
14.4
< 1.0
6.4
J(3’,4’/II)
4.4
3.6
2.8
4.8
12.0
4.4
5.2
7.2
10.4
b
J(4’,5’a/II)
J(4’,5’b/II)
J(5’a,5’b/II)
)
)
b
b
)
a) Assignments based on a DQF-COSY spectrum. b) Not assigned.
Table 7. Selected 13C-NMR Chemical Shifts [ppm] of the U*[n]A* Dinucleosides 19 – 21 in CDCl3
19
Adenosine unit (I)
20
21
19
20
21
Uridine unit (II)
C(2/II)
C(4/II)
C(5/II)
C(6/II)
C(2/I)
C(4/I)
C(5/I)
C(6/I)
154.16
150.78
119.08
155.71
149.05
59.43
90.18
82.72
81.26
153.54
150.48
118.89
155.69
148.97
59.28
90.20
82.79
80.67
85.10
154.08
151.31
117.99
155.47
150.20
56.91
89.63
82.83
81.43
151.17
163.64
102.92
152.48
51.08
151.89
163.50
103.28
152.37
50.07
151.49
163.75
103.04
152.50
50.23
C(8/I)
CH2ꢀC(6/II)
CH2ꢀC(8/I)
C(1’/I)
C(2’/I)
C(3’/I)
C(4’/I)
C(5’/I)
C(1’/I)
C(2’/I)
C(3’/I)
C(4’/I)
C(5’/I)
91.48
84.41
82.52
88.01
64.28
91.43
84.15
81.50
87.91
62.83
91.37
84.44
82.28
89.54
64.09
84.68
50.46
84.84
50.77
50.85
H2NꢀC(6/I)); 1.58 (sept., J ¼ 6.9, Me2CH); 1.63, 1.54, 1.39, 1.34 (4s, 2 Me2C); 0.83 (d, J ¼ 6.9, Me2CH);
0.81, 0.80 (2s, Me2CSi); 0.06, 0.05 (2s, Me2Si); HNꢀC(5’/I) and OH not visible. 13C-NMR (75 MHz,
CDCl3): see Table 7; additionally, 114.29, 113.30 (2s, 2 Me2C); 34.11 (d, Me2CH); 27.60, 27.44, 25.85, 25.63
(4q, 2 Me2C); 25.34 (s, Me2CSi); 20.42, 20.38 (2q, Me2CSi); 18.56, 18.52 (2q, Me2CH); ꢀ 3.17 (q, Me2Si).
HR-MALDI-MS: 797.3643 (44, [M þ Na]þ, C35H54N8NaO10Siþ; calc. 797.3624), 775.3807 (100, [M þ H]þ,
C35H55N8O10Siþ; calc. 775.3805).

1840
Helvetica Chimica Acta – Vol. 93 (2010)
N6-Benzoyl-5’-O-[dimethyl(1,1,2-trimethylpropyl)silyl]-2’,3’-O-isopropylideneadenosine-8-methyl-
(81 ! 5’-N)-5’-amino-5’-deoxy-2’,3’-O-isopropylidene-6-{[(4-methoxyphenyl)diphenylmethoxy]methyl}-
uridine (22). A soln. of 8 (1.57 g, 2.57 mmol) in THF (15 ml) was treated dropwise with 1m Me3P in THF
(3.08 ml, 3.08 mmol), and stirred for 24 h at 238 (TLC: complete conversion of 8). The mixture was
treated with a soln. of 10 (1.49 g, 2.57 mmol) in THF (10 ml), stirred for 48 h at 238, and evaporated. A
soln. of the residue in AcOEt (20 ml) was washed with H2O (3 ꢃ 20 ml), dried (MgSO4), and evaporated.
A soln. of the residue in MeOH (10 ml) was treated with NaBH3CN (646 mg, 10.28 mmol) and AcOH
(0.59 ml, 10.28 mmol), and stirred for 20 h at 238. The mixture was diluted with AcOEt, washed with sat.
NaHCO3 soln. and brine, dried (MgSO4), and evaporated. FC (CH2Cl2/MeOH 100 :1 ! 30 :1) gave 22
(2.22 g, 75%). Yellow foam. Rf (CH2Cl2/MeOH 10 :1) 0.43. IR (ATR): 3186w, 3050w, 2957w, 2927w,
2857w, 1695s, 1609m, 1583w, 1530w, 1509m, 1488m, 1449m, 1381m, 1324w, 1298w, 1251s, 1212m, 1179m,
1157m, 1083s, 1068s, 1034m, 977w, 908m, 871m, 830s, 797w, 778m, 761m, 728s, 706s, 646w, 631w. 1H-NMR
(300 MHz, CDCl3): see Table 8; additionally, 9.80 – 9.50 (br. s, HꢀN(3/I)); 9.28 (br. s, HNꢀC(6/II)); 7.98
(d, J ¼ 7.2, 2 arom. H); 7.55 (t, J ¼ 7.2, 1 arom. H); 7.48 – 7.42 (m, 6 arom. H); 7.36 – 7.23 (m, 8 arom. H);
6.84 (d, J ¼ 8.7, 2 arom. H); 3.78 (s, MeO); 2.05 – 1.95 (br. s, HNꢀC(5’/I)); 1.61, 1.42, 1.38, 1.26 (4s,
2 Me2C); 1.54 (sept., J ¼ 6.9, Me2CH); 0.82 (d, J ¼ 6.9, Me2CH); 0.77, 0.76 (2s, Me2CSi); ꢀ 0.04, ꢀ 0.05
(2s, Me2Si). 13C-NMR (75 MHz, CDCl3): see Table 9; additionally, 164.81 (s, NC¼O); 158.82, 143.04,
142.93, 133.97, 133.63 (5s); 132.57 (d); 130.32, 130.15 (2d); 128.54 – 127.39 (several d); 113.80, 113.74 (2s,
2 Me2C); 113.41 (2d); 88.20 (s, Ph2C); 55.15 (q, MeO); 34.11 (d, Me2CH); 27.29, 27.25, 25.53, 25.48 (4q,
2 Me2C); 25.26 (s, Me2CSi); 20.42, 20.25 (2q, Me2CSi); 18.62, 18.44 (2q, Me2CH); ꢀ 3.34, ꢀ 3.37 (2q,
Me2Si). HR-MALDI-MS: 1069.4834 (44, [M þ Na]þ, C55H70N8NaO11Siþ; calc. 1069.4831), 1047.5024
(100, [M þ H]þ, C55H71N8O11Siþ; calc. 1047.5012).
5’-O-[Dimethyl(1,1,2-trimethylpropyl)silyl]-2’,3’-O-isopropylideneadenosine-8-methyl-(81 ! 5’-N)-
5’-amino-5’-deoxy-2’,3’-O-isopropylidene-6-{[(4-methoxyphenyl)diphenylmethoxy]methyl}uridine (23).
A soln. of 22 (90 mg, 0.08 mmol) in MeOH (1 ml) was treated with a soln. of MeONa (42 mg,
0.80 mmol) in MeOH (2 ml), stirred for 48 h at 248, and evaporated. A soln. of the residue in AcOEt
(5 ml) was washed with a sat. NH4Cl soln. and brine, dried (MgSO4), and evaporated. FC (CH2Cl2/
MeOH 100 :1 ! 60 :1) gave 23 (55 mg, 68%). Yellow foam. Rf (CH2Cl2/MeOH 10 :1) 0.47. [a]2D5 ¼ ꢀ89.7
(c ¼ 0.36, CHCl3). IR (ATR): 3327w, 3182w, 2956w, 2865w, 1695m, 1638m, 1605m, 1576w, 1510w, 1447w,
1373m, 1329w, 1298w, 1251m, 1212m, 1156w, 1067s, 869w, 828s, 799w, 766m, 703w, 630w. 1H-NMR
(300 MHz, CDCl3): see Table 8; additionally, 13.40 (s, HꢀN(3/I)); 7.53 (br. s, H2NꢀC(6/II)); 7.45 (d, J ¼
7.2, 4 arom. H); 7.37 – 7.22 (m, 8 arom. H); 6.85 (d, J ¼ 9.0, 2 arom. H); 3.71 (s, MeO); 1.90 (br. s,
HNꢀC(5’/I)); 1.59, 1.39, 1.29, 1.26 (4s, 2 Me2C); 1.54 (sept., J ¼ 6.9, Me2CH); 0.82 (d, J ¼ 6.9, Me2CH);
0.78, 0.77 (2s, Me2CSi); ꢀ 0.02, ꢀ 0.06 (2s, Me2Si). 13C-NMR (75 MHz, CDCl3): see Table 9; additionally,
158.89, 143.22, 143.11, 134.15 (4s); 130.22 (2d); 128.12 (8d); 127.33 (2d); 113.60, 113.24 (2s, 2 Me2C);
113.36 (2d); 88.10 (s, Ph2C); 55.14 (q, MeO); 33.98 (d, Me2CH); 27.21, 27.11, 25.39, 25.10 (4q, 2 Me2C);
25.10 (s, Me2CSi); 20.17 (q, Me2CSi); 18.37, 18.33 (2q, Me2CH); ꢀ 3.57 (q, Me2Si). HR-MALDI-MS:
1069.4834 (44, [M þ Na]þ, C55H70N8NaO11Siþ; calc. 1069.4831), 1047.5024 (100, [M þ H]þ,
C55H71N8O11Siþ; calc. 1047.5012).
2’,3’-O-Isopropylideneadenosine-8-methyl-(81 ! 5’-N)-5’-amino-5’-deoxy-2’,3’-O-isopropylidene-6-
{[(4-methoxyphenyl)diphenylmethoxy]methyl}uridine (24). In a polyethylene flask, a soln. of 23
(120 mg, 0.11 mmol) in THF (2 ml) was treated with (HF)3 · Et3N (0.18 ml, 1.1 mmol), and stirred for
24 h at 248. The mixture was treated with 1m NaOH and extracted with AcOEt. The combined org. phases
were washed with brine, dried (MgSO4), and evaporated. FC (CH2Cl2/MeOH 60 :1 ! 20 :1) gave 24
(70 mg, 70%). Colourless solid. Rf (CH2Cl2/MeOH 10 :1) 0.47. M.p. 155 – 1578. [a]2D5 ¼ ꢀ146.4 (c ¼ 0.2,
CHCl3). IR (ATR): 3330w, 3177w, 2984w, 2933w, 2830w, 1695s, 1618m, 1576w, 1509m, 1447m, 1381m,
1334w, 1301w, 1252m, 1214m, 1179w, 1155m, 1066s, 1033m, 971w, 942w, 901w, 852m, 830m, 797w, 765w,
736w, 701m, 657w, 630w. 1H-NMR (500 MHz, CDCl3; assignments based on a DQF-COSY and a HSQC
spectrum): see Table 8; additionally, 12.70 (br. s, HꢀN(3/I)); 7.70 – 7.50 (br. s, H2NꢀC(6/II)); 7.48 – 7.43
(m, 4 arom. H); 7.36 – 7.23 (m, 8 arom. H); 6.82 (d, J ¼ 8.6, 2 arom. H); 3.80 (s, MeO); 1.65, 1.39, 1.34, 1.25
(4s, 2 Me2C); signals for HOꢀC(5’/II) and HNꢀC(5’/I)) not visible. 13C-NMR (125 MHz, CDCl3;
assignments based on a HSQC spectrum): see Table 9; additionally, 159.04, 143.32, 143.20, 134.14 (4s);
130.39 (2d); 128.24 (4d); 128.14 (2d); 128.11 (2d); 127.46 (2d); 113.96, 113.63 (2s, 2 Me2C); 113.46 (2d);

Helvetica Chimica Acta – Vol. 93 (2010)
1841
Table 8. Selected 1H-NMR Chemical Shifts [ppm] and Coupling Constants [Hz] of the A*[n]U*
Dinucleosides 22 – 26 in CDCl3
22
23
24a)
25a)
26
Uridine unit (I)
HꢀC(5/I)
CHaꢀC(6/I)
CHbꢀC(6/I)
HꢀC(1’/I)
HꢀC(2’/I)
HꢀC(3’/I)
HꢀC(4’/I)
HaꢀC(5’/I)
HbꢀC(5’/I)
J(Ha,Hb/I)
J(1’,2’/I)
5.67
3.99
3.92
5.52
5.17
4.90
4.07
3.01
2.96
12.6
< 1.0
6.3
5.4
5.4
5.88
5.78
5.38
5.90
4.58
4.42
5.59
5.16
4.92
4.26
3.07
2.95
14.4
< 1.0
5.7
4.8
7.5
4.03
3.93
5.37
5.27
5.11
4.31
3.11
3.02
12.6
3.6
6.6
3.0
2.7
4.02
3.95
5.51
5.24
5.04
4.15
2.98
2.98
4.68
4.38
6.21
5.16
5.00
4.26
3.14
2.94
13.8
< 1.0
6.3
6.2
5.5
12.7
2.7
6.6
3.6
J(2’,3’/I)
J(3’,4’/I)
J(4’,5’a/I)
J(4’,5’b/I)
b
)
)
)
b
c
3.9
12.0
3.6
12.3
2.8
12.8
1.8
12.6
J(5’a,5’b/I)
Adenosine unit (II)
HꢀC(2/II)
CHaꢀC(8/II)
CHbꢀC(8/II)
HꢀC(1’/II)
HꢀC(2’/II)
HꢀC(3’/II)
HꢀC(4’/II)
HaꢀC(5’/II)
HbꢀC(5’/II)
J(Ha,Hb/II)
J(1’,2’/II)
8.73
8.32
8.30
4.19
3.94
6.10
5.36
5.17
4.50
3.97
3.78
8.19
8.08
4.17
4.13
6.40
5.88
5.15
4.27
3.65
3.55
4.19
3.95
6.46
5.96
5.14
4.26
3.75
3.59
4.09
3.99
6.41
5.83
5.14
4.23
3.56
3.49
4.16
4.04
6.23
5.33
5.16
4.46
3.94
3.77
14.4
< 1.0
6.3
14.1
1.8
6.3
14.4
5.1
5.9
13.9
1.3
6.2
14.0
4.8
5.7
J(2’,3’/II)
J(3’,4’/II)
J(4’,5’a/II)
J(4’,5’b/II)
J(5’a,5’b/II)
2.7
6.6
6.3
10.5
2.7
6.6
7.2
10.5
1.1
3.2
6.8
6.3
10.5
< 1.5
< 1.5
< 1.5
12.3
< 1.5
< 1.5
c
)
a) Assignments based on a DQF-COSY and a HSQC spectrum. b) HaꢀC(5’/I) and HbꢀC(5’/I) are
nearly isochronous, showing a mean coupling of 4.3 Hz. c) Not assigned.
55.27 (q, MeO); 27.73, 27.27, 25.37, 25.24 (4q, 2 Me2C). HR-MALDI-MS: 943.3387 (60, [M þ K]þ,
C47H52KN8Oþ11 ; calc. 943.3387), 927.3655 (100, [M þ Na]þ, C47H52N8NaOþ11 ; calc. 927.3653), 905.3802 (82,
[M þ H]þ, C47H53N8O1þ1 ; calc. 905.3828).
5’-O-[Dimethyl(1,1,2-trimethylpropyl)silyl]-2’,3’-O-isopropylideneadenosine-8-methyl-(81 ! 5’-N)-
5’-amino-5’-deoxy-6-(hydroxymethyl)-2’,3’-O-isopropylideneuridine (25). A soln. of 23 (150 mg,
0.14 mmol) in CH2Cl2 (2 ml) was treated with Et3SiH (0.27 ml, 1.68 mmol) and Cl2CHCO2H (93 ml,
1.12 mmol), and stirred for 10 min at 248. The mixture was treated with Amberlite IRA-68 (20 – 50 mesh;
free base), stirred for 30 min, filtered, and evaporated. FC (CH2Cl2/MeOH 80 :1 ! 30 :1) gave 25 (79 mg,
73%). Colourless foam. Rf (CH2Cl2/MeOH 10 :1) 0.43. [a]2D5 ¼ ꢀ29.4 (c ¼ 0.33, CHCl3). IR (ATR):

1842
Helvetica Chimica Acta – Vol. 93 (2010)
Table 9. Selected 13C-NMR Chemical Shifts [ppm] of the A*[n]U* Dinucleosides 22 – 26 in CDCl3
22
23
24a)
25a)
26
Uridine unit (I)
C(2/I)
C(4/I)
C(5/I)
C(6/I)
151.88
162.58
103.02
152.30
62.30
91.71
83.84
81.73
89.84
50.97
151.69
163.13
103.61
152.77
62.16
92.51
82.97
80.94
88.41
51.57
151.77
163.05
103.75
152.58
62.39
88.30
83.17
81.45
151.99
165.25
102.85
154.63
62.26
91.27
84.32
80.88
87.63
50.14
151.21
164.08
102.30
154.23
61.46
91.42
83.13
81.47
CH2ꢀC(6/I)
C(1’/I)
C(2’/I)
C(3’/I)
C(4’/I)
C(5’/I)
86.75
51.44
85.80
50.26
Adenosine unit (II)
C(2/II)
C(4/II)
C(5/II)
C(6/II)
152.15
151.72
121.98
153.76
150.41
46.74
91.48
83.13
82.24
87.87
62.91
152.18
150.99
117.80
155.34
150.01
46.21
90.26
82.53
81.66
152.19
150.83
118.56
155.94
149.51
46.33
92.53
82.89
81.77
151.42
151.26
118.53
155.02
150.02
46.67
89.69
83.33
82.22
88.37
63.00
151.34
150.70
118.73
155.33
149.20
46.17
92.06
84.40
82.09
88.95
63.16
C(8/II)
CH2ꢀC(8/II)
C(1’/II)
C(2’/II)
C(3’/II)
C(4’/II)
C(5’/II)
88.10
63.05
85.87
63.41
a) Assignments based on a HSQC spectrum.
3329w, 3195w, 2956w, 2940w, 2857w, 1698s, 1640m, 1603m, 1576w, 1455m, 1374m, 1330m, 1251m, 1211m,
1157m, 1066s, 969w, 871m, 828s, 799m, 777m, 660w. 1H-NMR (500 MHz, CDCl3; assignments based on a
DQF-COSYand a HSQC spectrum): see Table 8; additionally, 11.70 – 11.50 (br. s, HꢀN(3/I)); 7.70 – 7.45
(br. s, H2NꢀC(6/II)); 1.9 – 1.5 (br. s, OH, HNꢀC(5’/I)); 1.62, 1.57, 1.45, 1.34 (4s, 2 Me2C); 1.52 (sept., J ¼
6.9, Me2CH); 0.80 (d, J ¼ 6.9, Me2CH); 0.75, 0.74 (2s, Me2CSi); ꢀ 0.06, ꢀ 0.07 (2s, Me2Si). 13C-NMR
(125 MHz, CDCl3; assignments based on a HSQC spectrum): see Table 9; additionally, 114.38, 113.62 (2s,
2 Me2C); 34.08 (d, Me2CH); 27.49, 27.20, 25.51, 25.45 (4q, 2 Me2C); 25.21 (s, Me2CSi); 20.30, 20.24 (2q,
Me2CSi); 18.46, 18.41 (2q, Me2CH); ꢀ 3.48, ꢀ 3.49 (2q, Me2Si). HR-MALDI-MS: 797.3659 (10, [M þ
Na]þ, C35H54N8NaO10Siþ; calc. 797.3630), 775.3817 (100, [M þ H]þ, C35H55N8O10Siþ; calc. 775.3810).
Anal. calc. for C35H54N8O11Si (774.94): C 54.25, H 7.02, N 14.46; found: C 53.96, H 7.06, N 14.08.
2’,3’-O-Isopropylideneadenosine-8-methyl-(81 ! 5’-N)-5’-amino-5’-deoxy-6-(hydroxymethyl)-2’,3’-
O-isopropylideneuridine (26). A soln. of 24 (70 mg, 0.08 mmol) in CH2Cl2 (1 ml) was treated with Et3SiH
(0.15 ml, 0.93 mmol) and Cl2CHCO2H (0.05 ml, 0.62 mmol), and stirred for 10 min at 248. The mixture
was treated with Amberlite IRA-68 (20 – 50 mesh; free base), stirred for 30 min, filtered, and evaporated.
FC (CH2Cl2/MeOH 50 :1 ! 10 :1) gave 26 (38 mg, 78%). Colourless foam. Rf (CH2Cl2/MeOH 10 :1) 0.17.
IR (ATR): 3318m (br.), 3189m (br.), 2987m, 2933m, 2848w, 1680s, 1647s, 1609m, 1576w, 1454m, 1381s,
1333m, 1303w, 1261w, 1214m, 1158m, 1101s, 1083s, 1068s, 1020w, 993w, 969w, 945w, 908w, 873w, 857w,
1
827w, 797w, 760w, 691w, 654w, 619w. H-NMR (300 MHz, CDCl3): see Table 8; additionally, 7.19 (br. s,
H2NꢀC(6/II)); 1.67, 1.55, 1.43, 1.32 (4s, 2 Me2C); signals for HꢀN(3/I) and 2 OH not visible. 13C-NMR
(75 MHz, CDCl3): see Table 9; additionally, 114.00, 113.91 (2s, 2 Me2C); 27.82, 27.43, 25.42 (2 C) (3q,
2 Me2C). HR-MALDI-MS: 671.2188 (25, [M þ K]þ, C27H36KN8O1þ0 ; calc. 671.2186), 655.2446 (41, [M þ
Na]þ, C27H36N8NaO1þ0 ; calc. 655.2447), 633.2624 (100, [M þ H]þ, C27H37N8O1þ0 ; calc. 633.2627).

Helvetica Chimica Acta – Vol. 93 (2010)
1843
REFERENCES
[1] X. Zhang, B. Bernet, A. Vasella, Helv. Chim. Acta 2006, 89, 2861.
[2] X. Zhang, B. Bernet, A. Vasella, Helv. Chim. Acta 2007, 90, 792.
[3] X. Zhang, B. Bernet, A. Vasella, Helv. Chim. Acta 2007, 90, 864.
[4] X. Zhang, B. Bernet, A. Vasella, Helv. Chim. Acta 2007, 90, 891.
[5] A. J. Matthews, P. K. Bhardwaj, A. Vasella, Chem. Commun. 2003, 950.
[6] A. J. Matthews, P. K. Bhardwaj, A. Vasella, Helv. Chim. Acta 2004, 87, 2273.
[7] A. Ritter, D. Egli, B. Bernet, A. Vasella, Helv. Chim. Acta 2008, 91, 673.
[8] B. Bernet, Z. Johar, A. Ritter, B. Jaun, A. Vasella, Helv. Chim. Acta 2009, 92, 2596.
[9] K. Chiesa, A. Shvoryna, B. Bernet, A. Vasella, Helv. Chim. Acta 2010, 93, 668.
[10] W. Saenger, ꢃPrinciples of Nucleic Acid Structureꢂ, Springer-Verlag, New York, 1984, p. 556.
[11] I. Kanesaka, R. G. Snyder, H. L. Strauss, J. Chem. Phys. 1986, 84, 395.
[12] P. Vansteenkiste, E. Pauwels, V. Van Speybroeck, M. Waroquier, J. Phys. Chem. A 2005, 109,
9617.
´
[13] D. F. Plusquellic, R. D. Suenram, B. Mate, J. O. Jensen, A. C. Samuels, J. Chem. Phys. 2001, 115,
3057.
[14] F. Palacios, C. Alonso, D. Aparicio, G. Rubiales, J. M. de los Santos, Tetrahedron 2007, 63,
523.
[15] A. Jayanthi, V. K. Gumaste, A. R. A. S. Deshmukh, Synlett 2004, 979.
´
´
[16] P. Molina, J. Alcantara, C. Lopez-Leonardo, Tetrahedron 1996, 52, 5833.
[17] R. C. Reynolds, N. Bansal, J. Rose, J. Friedrich, W. J. Suling, J. A. Maddry, Carbohydr. Res. 1999, 317,
164.
[18] M. C. Pirrung, S. W. Shuey, D. C. Lever, L. Fallon, Bioorg. Med. Chem. Lett. 1994, 4, 1345.
[19] T. D. Nelson, R. D. Crouch, Synthesis 1996, 1031.
[20] W. Czechtizky, A. Vasella, Helv. Chim. Acta 2001, 84, 594.
[21] O. Mitsunobu, Synthesis 1981, 1.
[22] F. Liu, D. J. Austin, Tetrahedron Lett. 2001, 42, 3153.
[23] A. Hampton, T. Sasaki, B. Paul, J. Am. Chem. Soc. 1973, 95, 4404.
[24] J. Hollmann, E. Schlimme, Liebigs Ann. Chem. 1984, 98.
[25] S. David, J.-C. Fischer, Bull. Soc. Chim. Fr. 1972, 3610.
[26] S. Dimitrijevich, J. P. H. Verheyden, J. G. Moffatt, J. Org. Chem. 1979, 44, 400.
[27] V. M. Clark, A. R. Todd, J. Zussman, J. Chem. Soc. 1951, 2952.
[28] D. B. Dess, J. C. Martin, J. Org. Chem. 1983, 48, 4155.
[29] M. Frigerio, M. Santagostino, S. Sputore, G. Palmisano, J. Org. Chem. 1995, 60, 7272.
[30] S. Hanessian, G. Huang, C. Chenel, R. Machalani, O. Loiseleur, J. Org. Chem. 2005, 70, 6721.
[31] G. H. Jones, M. Taniguchi, D. Tegg, J. G. Moffatt, J. Org. Chem. 1979, 44, 1309.
[32] V. T. Ravikumar, A. H. Krotz, D. L. Cole, Tetrahedron Lett. 1995, 36, 6587.
[33] D. W. Miles, M. J. Robins, R. K. Robins, M. W. Winkley, H. Eyring, J. Am. Chem. Soc. 1969, 91, 824.
[34] I. Leban, B. Golankiewicz, J. Zeidler, G. Giester, J. Kobe, Acta Crystallogr., Sect. C: Cryst. Struct.
Commun. 2002, 58, o133.
[35] S. Sugio, K. Kitamura, H. Mizuno, A. Wakahara, M. Ikehara, K.-I. Tomita, Acta Crystallogr., Sect. C:
Cryst. Struct. Commun. 1984, 40, 1608.
[36] H. S. Gutowsky, A. Saika, J. Chem. Phys. 1953, 21, 1688.
[37] A. Gaudemer, F. Nief, R. Pontikis, J. Zylber, Org. Magn. Reson. 1977, 10, 135.
´
´
´
´
[38] K. Meszarosova, A. Holy, M. Maojꢄdkova, Collect. Czech. Chem. Commun. 2000, 65, 1109.
[39] R. Kupfer, S. Meier, E.-U. Wꢀrthwein, Synthesis 1984, 688.
[40] J. Uzawa, K. Anzai, Can. J. Chem. 1985, 63, 3537.
Received June 3, 2010