Oligonucleotides containing 7-deaza-2′-deoxyxanthosine: Synthesis, base protection, and base-pair stability

Article abstract of DOI:10.1002/hlca.200690251

Oligonucleotides incorporating 7-deaza-2′-deoxyxanthosine (3) and 2′-deoxyxanthosine (1) were prepared by solid-phase synthesis using the phosphoramidites 6-9 and 16 which were protected with allyl, diphenylcarbamoyl, or 2-(4-nitrophenyl)ethyl groups. Amo

Full text of DOI:10.1002/hlca.200690251

2794
Helvetica Chimica Acta – Vol. 89 (2006)
Oligonucleotides Containing 7-Deaza-2’-deoxyxanthosine: Synthesis, Base
Protection, and Base-Pair Stability
by Frank Seela* and Khalil I. Shaikh
Laboratorium für Organische und Bioorganische Chemie, Institut für Chemie, Universität Osnabrück,
Barbarastrasse 7, D-49069 Osnabrück, and
Laboratory of Bioorganic Chemistry and Chemical Biology, Center for Nanotechnology (CeNTech),
Heisenbergstrasse 11, D-48149 Münster
(phone: +49(0)541 969 2791; fax: +49(0)541 969 2370; e-mail: Frank.Seela@uni-osnabrueck.de)
Oligonucleotides incorporating 7-deaza-2’-deoxyxanthosine (3) and 2’-deoxyxanthosine (1) were
prepared by solid-phase synthesis using the phosphoramidites 6–9 and 16 which were protected with
allyl, diphenylcarbamoyl, or 2-(4-nitrophenyl)ethyl groups. Among the various groups, only the 2-(4-
nitrophenyl)ethyl group was applicable to 7-deazaxanthine protection being removed with 1,8-diazabicy-
clo[5.4.0]undec-7-ene (DBU) by b-elimination, while the deprotection of the allyl residue with Pd0 cata-
lyst or the diphenylcarbamoyl group with ammonia failed. Contrarily, the allyl group was found to be an
excellent protecting group for 2’-deoxyxanthosine (1). The base pairing of nucleoside 3 with the four can-
onical DNA constituents as well as with 3-bromo-1-(2-deoxy-b-D-erythro-pentofuranosyl)-1H-pyra-
zolo[3,4-d]pyrimidine-4,6-diamine (4) within the 12-mer duplexes was studied, showing that 7-deaza-
2’-deoxyxanthosine (3) has the same universal base-pairing properties as 2’-deoxyxanthosine (1). Con-
trary to the latter, it is extremely stable at the N-glycosylic bond, while compound 1 is easily hydrolyzed
under slightly acidic conditions. Due to the pKa values 5.7 (1) and 6.7 (3), both compounds form mono-
anions under neutral conditions (95% for 1; 65% for 3). Although both compounds form monoACHTERaUNG nions at
pH 7.0, pH-dependent Tm measurements showed that the base-pair stability of 7-deaza-2’-deoxyxantho-
sine (3) with dT is pH-independent. This indicates that the 2-oxo group is not involved in base-pair for-
mation.
Introduction. – The 2’-deoxyxanthosine (1) and 2’-deoxyoxanosine (2) are the
major products of the spontaneous nitrosative deamination of 2’-deoxyguanosine [1]
(purine numbering is used throughout the discussion section). Both nucleosides are sus-
ceptible to N-glycosylic-bond hydrolysis, thereby forming abasic sites in the DNA helix.
The 2’-deoxyxanthosine (1) is sensitive to acidic conditions [2] but is relatively stable at
physiological pH [3][4]. Compound 1 has been prepared by enzymatic transglycosyla-
tion [2][5] or nitrosative deamination of dG under alkaline conditions [6]. An enzy-
matic synthesis of 1 from 2’-deoxyisoinosine has been reported by our laboratory [7].
Contrary to the labile xanthine nucleoside 1, the glycosylic bond of 7-deaza-2’-deoxy-
xanthosine (3), which was first synthesized in our laboratory, is resistant to ꢀdepurina-
tionꢁ even under strong acidic conditions [8]. Both, 2’-deoxyxanthosine (1) and 7-deaza-
2’-deoxyxanthosine (3), form oligonucleotide duplexes and triplexes [9–12]. Also, 2’-
deoxyxanthosine (1) forms four-stranded (quartets) and/or six-stranded structures in
aqueous solution [13]. As the recognition sites of the nucleosides 1 and 3 show the
same donor–acceptor pattern as thymidine (dT), base pairs are formed with comple-
ꢂ 2006 Verlag Helvetica Chimica Acta AG, Zürich

Helvetica Chimica Acta – Vol. 89 (2006)
2795
mentary nucleosides which normally pair with dU or dT. However, pairing with dA or
related purine-2,6-diamine nucleosides generates purine–purine pairs which can distort
the DNA-helix structure but can contribute better stacking interactions induced by an
increased surface area.
It has been reported that 2’-deoxyxanthosine (1) might expand the genetic code
when it pairs with a pyrimidine-2,4-diamine C-nucleoside [14]. However, experimental
data show that the enzymatic incorporation of a pyrimidine-2,4-diamine C-nucleoside
triphosphate on a DNA template containing compounds 1 or 3 is slow compared to the
canonical nucleotides [14]. Therefore, the expansion of the genetic code system by a
new base pair containing compound 1 or 3 is limited due to polymerases acceptance
[14]. Nevertheless, compound 1 is valuable in the structural investigation of DNA as

2796
Helvetica Chimica Acta – Vol. 89 (2006)
it is the major deamination product formed from 2’-deoxyguanosine, and compound 3 is
useful for DNA-triplex and -quartet formation. Also, compounds 1 and 3 can be
applied to oligonucleotide diagnostics and therapeutics as well as for primer probing.
Having this in mind, efforts were made to prepare oligonucleotides containing 7-
deaza-2’-deoxyxanthosine (3) or 2’-deoxyxanthosine (1). Solid-phase oligonucleotide
synthesis has already been performed with 3 by using the protocol of phosphonate
chemistry. In this case, the base of the 7-deaza-2’-deoxyxanthosine phosphonate 5
stayed unprotected [12][14]. As oligonucleotide synthesis by using phosphonate
chemistry is now rarely undertaken, phosphoramidite building blocks which can be
employed in solid-phase synthesis are needed.
For the first time, we now report on a base-protected 7-deaza-2’-deoxyxanthosine
phosphoramidite (see 9) which allowed multiple incorporations into an oligonucleotide
chain with coupling yields identical to those of the canonical nucleosides. The corre-
sponding unprotected phosphoramidite 6 does not fulfill these necessities as coupling
yields were low, by-products were formed, and multiple incorporations were almost
impossible [15]. The applicability of allyl, diphenylcarbamoyl, and 2-(4-nitrophenyl)-
ethyl protecting groups to block the 2,6-dioxo functions of compound 3 were compared.
Allyl protection was also applied to 2’-deoxyxanthosine (1) which has been protected
earlier with one or two 2-(4-nitrophenyl)ethyl residues as protecting groups
[4][9][10]. The hybridization properties of oligonucleotide duplexes containing 1 and
3 with the canonical DNA bases as well as with the 3-bromo-1-[2-deoxy-b-D-erythro-
pentofuranosyl)-1H-pyrazolo[3,4-d]pyrimidine-4,6-diamine (4) were also studied.
Results and Discussion. – 1. Nucleobase Protection and Phosphoramidite Synthesis.
Different from the large number of amino-protecting groups used in oligonucleotide
synthesis, the number of oxo-protecting groups used for the nucleobase protection is
limited. This results from the observation that in some of the cases, e.g., for 2’-deoxygua-
nosine, oxo-group protection is not required when phosphoramidite chemistry is
employed. However, already in the case of 2’-deoxyisoguanosine, it was necessary to
protect the 2-oxo-function, otherwise the preparation of a phosphoramidite was diffi-
cult, and multiple incorporations of the modified nucleoside were impossible [16]. In
the first set of experiments, the allyl group was selected as a protecting group for 7-
deaza-2’-deoxyxanthosine (3). The O-allyl protection of nucleosides has already been
reported by Hayakawa, Noyori, and co-worker [17]. The O-allyl protecting group is sta-
ble under standard conditions applied to oligonucleotide synthesis and deprotection.
As it is sensitive to the palladium(0) complexes in the presence of nucleophiles but sta-
ble in ammonia, it represents an orthogonal protecting group among the common pro-
tecting groups used in oligonucleotide chemistry. Phosphoramidites with allyl protec-
tion were already prepared for 2’-deoxyguanosine (dG) and 2’-deoxythymidine (dT)
and were used successfully in coupling reactions performed in solution or on solid-
phase [17][18]. Eschenmoser and co-workers used the 2,6-di-O-allyl-protected phos-
phoramidite of hexopyranosylxanthine nucleoside in oligonucleotide solid-phase syn-
thesis [19]. The same protecting group is now employed for the protection of the 2,6-
dioxo functions of nucleoside 3. For that, 2,4-dichloro-7-[2-deoxy-3,5-di-O-(p-tol-
uoyl)-b-D-erythro-pentofuranosyl]pyrrolo[2,3-d]pyrimidine (10a) was prepared [20].
It was treated with 1M NaOCH2CH=CH2 in allyl alcohol at 508 furnishing 2,6-bis(allyl-

Helvetica Chimica Acta – Vol. 89 (2006)
2797
Scheme 1
oxy) compound 11a in 84% yield. Then the 4,4’-dimethoxytrityl ((MeO)2ACHTRETUGN r) residue
was introduced with 4,4’-dimethoxytrityl chloride in pyridine yielding derivative 12a.
Subsequent treatment of 12a with 2-cyanoethyl diisopropylphosphoramidochloridite
afforded phosphoramidite 7 (Scheme 1).
Phosphoramidite 7 was successfully employed in solid-phase synthesis resulting in
coupling yields being compatible with normal phosphoramidites (see Sect. 2). How-
ever, the deblocking of the allyl group from the otherwise deprotected oligonucleotide
failed. As it was not clear whether this was due to the 7-deazapurine system, the allyl
protecting group was also used for the 2’-deoxyxanthosine (1) protection. The
same protocol as performed with 3 was also applied to compound 1. In this case,
2,6-dichloro-9-[2-deoxy-3,5-di-O-(p-toluoyl)-b-D-erythro-pentofuranosyl]purine (10b)
[20] was used as precursor. Allylation of 10b was performed in a similar way as descri-
bed for 11a affording 11b. Then, compound 11b was treated with 4,4’-dimethoxytrityl
chloride/pyridine furnishing the 4,4’-dimethoxytrityl derivative 12b. Subsequent treat-
ment with 2-cyanoethyl diisopropylphosphoramidochloridite afforded phosphorami-
dite 8. When this building block was used in solid-phase oligonucleotide synthesis in
the same way as described for phosphoramidite 7, the same high coupling yields
were realized with 8 as observed with 7. However, contrary to the 7-deazapurine con-
taining oligonucleotides, the deprotection of the allyl residues from the xanthine

2798
Helvetica Chimica Acta – Vol. 89 (2006)
oligomers progressed smoothly furnishing the deprotected oligonucleotides (more
information is given in Sect. 2). Therefore, the diphenylcarbamoyl (dpc) group was
chosen next for the protection of compound 3.
The dpc group was used earlier as oxo-protecting group for 7-deaza-2’-deoxyguano-
sine and 7-deaza-2’-deoxyisoguanosine [21][22]. It can be removed under standard con-
ditions (25% aqueous NH3 ACHTREsUGN olution, 608). Treatment of 3 with diphenylcarbamic chlo-
ride in pyridine afforded compound 13a (74%) together with tris(diphenycarbamoyl)
derivative 13b (6%) (Scheme 2). Different from 7-deaza-2’-deoxyguanosine or 7-
deaza-2’-deoxyisoguanosine where the reaction with diphenylcarbamic chloride occur-
red at the oxo group, the dpc residues of 13a became attached to the ring N-atoms. This
was confirmed by 13C-NMR spectroscopy (Table 1). The chemical shifts of C(2) and
C(6) of compound 13a (purine numbering) appeared at similar positions as those of
the nucleoside 3 or the N(1), N(3)-methylated derivative 13c (=7-(2-deoxy-b-D-
erythro-pentofuranosyl)-1,7-dihydro-1,3-dimethyl-2H-pyrrolo[2,3-d]pyrimidine-2,4(3H)-
dione) [23] (13a: (d151.1(C(2)), d157.6(C(6)); 3: d150.4(C(2)), d159.5(C(6)); 13c:
d151.5(C(2)), d158.1(C(6)), while for the O-allyl-protected compound 11a the C(2)
and C(6) signals are shifted downfield (d160.1(C(2)) 162.7(C(6))) (see Table 1).
Hence, the dpc residues of compound 13a are connected to the ring N-atoms. As in
related nucleosides such as 2’-deoxyuridine or 6-aza-2’-deoxyuridine the protection
of the lactam N-atoms by acyl protecting groups was successfully employed in oligonu-
cleotide synthesis [24], the synthetic route with compound 13a was continued. The lat-
ter was treated with (MeO)
ative 14 in 74% yield. To circumvent the side reaction of 3 with dpc–Cl on OH groups,
the 5’-hydroxy group of 3 was protected first with the (MeO)2ATr residue resulting in the
(MeO)2ATr derivative 15. Then compound 15 was treated with dpc–Cl affording com-
2A
CTHREUNG
CTHREUNG
pound 14 (95% yield). Subsequently, the phosphoramidite 16 was prepared by using
14 and 2-cyanoethyl diisopropylphosphoramidochloridite (Scheme 2). Finally, the
dpc-protected phosphoramidite 16 was employed in solid-phase oligonucleotide syn-
thesis which resulted in coupling yields similar to those of the allyl-protected phosphor-
amidite 7 (see Sect. 2). However, also in this case, we were not able to obtain a clean
deprotection product from the oligonucleotide (25% aqueous NH3 solution, 608).
Due to the difficulties encountered during the removal of the allyl and dpc groups
from 7-deazaxanthine oligonucleotides, the 2-(4-nitrophenyl)ethyl (npe) group was
selected for the protection of the oxo functionalities of 3. It is a b-eliminating group
requiring a strong base (DBU=1,8-diazabicyclo[5.4.0]undec-7-ene) to achieve depro-
tection. Earlier, this group was used by Pfleiderer and co-workers for the protection of
2’-deoxyguanosine [25]. The first solid-phase oligonucleotide synthesis with 2’-deoxy-
xanthosine (1) which was performed by Eritja and co-workers used mono npe-pro-
tected, i.e., 6-O-[2-(4-nitrophenyl)ethyl]-protected phosphoramidite [9]. As this
resulted in low coupling yields and by-products, Herdewijn and co-workers prepared
the 2,6-bis-O-npe-protected phosphotriester which was applied to oligonucleotide sol-
ution synthesis [26]. Later, Battersby and co-workers reported the synthesis of the 2,6-
bis-O-npe-protected phosphoramidite of 1 in which the 2-oxo group was protected with
the npe residue by using npe–I and Ag2CO3 (Kçnigs–Knorr conditions) [10]. Also, the
2,6-bis-O-npe-protected phosphoramidite of 1 was prepared by the Mitsunobu reaction
[4]. The Mitsunobu conditions were now used to synthesize the bis-O-npe-protected

Helvetica Chimica Acta – Vol. 89 (2006)
2799
Scheme 2
phosphoramidite 9 (Scheme 3). For this, 7-deaza-2’-deoxyguanosine (c7Gd) [27] was
chosen as a starting material. In the initial step, the 3’,5’-hydroxy groups of c7Gd were
blocked by acetylation with Ac2O in pyridine affording compound 17 (88% yield).
Next, the 2-amino group of 17 was protected with the 4,4’-dimethoxytrityl group to
give 18. When the Mitsunobu reaction was performed with 18 and 2-(4-nitrophenyl)e-
thanol in the presence of DEAD/PPh3 in dioxane, the (MeO)2ACHTRETUGN r residue was cleaved
from the 2-amino function, and compound 19 was isolated in only 23% yield. There-
fore, the Mitsunobu reaction was directly applied to the 3’,5’-di-O-acetyl-protected
compound 17 in THF yielding the 6-O-protected nucleoside 19 (55%). The amino
group of compound 19 was transformed to an OH group with NaNO2/AcOH in
H2O/acetone furnishing compound 20. Next, a second Mitsunobu reaction with 20
gave the fully protected nucleoside 21. Treatment of 21 with methanolic NH
the 2,6-bis-O-[2-(4-nitrophenyl)ethyl]-protected 7-deaza-2’-deoxyxanthosine 22.
Then, compound 22 was protected with the (MeO)2ATr residue to give the derivative
3 ACHTREUyNG ielded
CTHREUNG
23. Subsequent treatment of 23 with 2-cyanoethyl diisopropylphosphoramidochloridite
afforded the phosphoramidite 9. The position of the npe residue of compound 22 was
established by 13C-NMR spectroscopy. According to Table 1, the chemical-shifts of C(2)
and C(6) of the npe-protected compound 22 appear at almost the same d as observed
for the di-O-allyl nucleoside 11a (22: d160.2(C(2)), d162.8(C(6)); 11a: d160.1(C(2)),
162.7(C(6))). Hence the npe residues of 22 protect the oxo functions.
To compare the applicability of the npe-protected phosphoramidite 9 in the oligo-
nucleotide synthesis, we prepared the 2,6-O-unprotected phosphoramidite 6 as well.

2800
Helvetica Chimica Acta – Vol. 89 (2006)
Table 1. 13C-NMR Chemical Shifts of 2’-Deoxyxanthosine and 7-Deaza-2’-deoxyxanthosine Derivativesa)
Compoundb)c) C(2)d) C(4)d) C(4a) C(6) C(7a) C(5) C(1’) C(2’) C(3’) C(4’) C(5’)
C(2)d) C(6)d) C(5)
C(8) C(4)
C(7)
f
f
f
f
f
f
f
f
f
f
11be)
12be)
11a
12a
13a
13b
3h)
13ci)
14
15
17
18
19
159.9
159.8
160.1
160.0
151.1
151.0
150.4
151.5
151.1
151.1
152.8
157.6
159.6
160.3
160.3
160.2
160.2
160.4
160.5
162.7
162.6
157.6
157.7
159.5
158.1
158.0
159.6
158.6
158.3
162.3
163.3
162.9
162.8
162.8
116.8 140.9 153.0
–
–
83.3
83.7
)
)
)
)
)
)
)
)
)
)
70.6
70.7
71.3
70.8
70.8
70.7
70.8
70.2
70.6
70.0
74.6
74.5
74.6
74.5
74.3
71.0
70.7
87.7
86.0
87.2
85.3
87.5
84.2
87.3
87.4
85.5
85.2
82.3
83.0
82.2
82.7
83.1
87.2
85.3
61.6
64.3
61.9
64.3
61.7
65.8
61.4
61.4
64.0
63.6
63.8
63.6
63.8
63.7
63.7
61.9
64.2
g
117.2
)
152.8
100.3 122.6 153.0
99.0 82.8
98.9 83.0
99.7 83.1
99.9 82.9
g
100.4
107.3
107.2
)
)
)
152.9
141.5
141.4
g
g
99.4 117.6 138.0 103.1 85.5
101.0 117.7 138.0 103.5 84.7
g
107.3
)
141.5
99.9 82.7
99.0 115.9 138.9 103.7 83.1
100.1 116.3 150.9 102.9 80.8
100.5 116.8 149.8 102.6 80.7
35.7
36.4
35.5
35.6
35.5
97.1 119.0 154.7
99.7 80.7
99.9 80.9
j
20
21
22
23
98.7 120.5
)
99.5 122.6 153.2 100.5 80.9
99.0 122.5 153.0 100.2 82.7
99.0 122.6 152.9 100.4 82.9
f
)
)
f
a) Measured in (D6)DMSO at 258. b) First heading row=systematic numbering. c)Second heading row=
f
purine numbering. d) Tentative. e) Only purine numbering is used. ) Superimposed by (D6)DMSO.
g) Overlapped with aromatic signals. h) [8] i) [23] j) Not detected.
For this compound 15 was treated with 2-cyanoethyl diisopropylphosphoramidochlor-
idite in the presence of diisopropyl(ethyl)amine which afforded the phosphoramidite 6
(Scheme 4).
1
The structures of all new compounds were confirmed by H-, 13C-, and 31P-NMR
spectroscopy as well as by elemental analyses (see Table 1 and Exper. Part). The 13C-
NMR chemical-shift assignment was made by means of gated-decoupled 13C-NMR
spectra as well as by comparison with the other related 7-deazapurine nucleosides
[8][23]. The assignments for C(2) and C(6) are tentative. The assignment of C(1’)
1
and C(4’) are based on the difference of the J(C,H) coupling constants, which are
larger for C(1’) than for C(4’) [28].
2. Synthesis and Characterization of Oligonucleotides 26–32. For the 12-mer-duplex
studies based on the reference duplex 24·25 (see Sect. 3), the oligonucleotides 26–40
were needed. Oligonucleotide syntheses were carried out at a 1-mmol scale following
the synthesis protocol for phosphoramidite chemistry. The phosphoramidites 6–9
and 16 were used together with those of the phosphoramidites of canonical DNA con-
stituents as well as that of the pyrazolo[3,4-d]pyrimidine nucleoside 4 [29]. The cou-
pling yields of the oligonucleotide samples 26–32 are displayed in Table 2. From this
it is apparent that the coupling steps of the modified bases proceed with a similar effi-
ciency as those of the canonical phosphoramidites. The syntheses were performed with
the (MeO)2CAHTRETUNG r-on mode. The following protocols were employed: i) allyl protection, ii)
npe-protection, and iii) diphenylcarbamoyl protection. The deallylation was performed
on the solid-support-bound oligonucleotide with [Pd0(PPh3)4] in the presence of PPh3,

Helvetica Chimica Acta – Vol. 89 (2006)
2801
Scheme 3
i) Ac2O, pyridine, overnight, r.t. ii) (MeO)
diazenedicarboxylate), 2-(4-nitrophenyl)ethanol, THF or dioxane, 1 h, r.t. iv) 30% aq. AcOH/acetone
2 :1, NaNO2, 2 h, r.t. v) NH3/MeOH, 2 h, r.t. vi) (MeO)2A
TrꢀCl, pyridine, 1 h, r.t. vii) 2-Cyanoethyl
diisopropylphosphoramidochloridite, (i-Pr)2ANEt, CH2Cl2, 30 min, r.t.
2A
CTHREUNG
CTHREUNG

2802
Helvetica Chimica Acta – Vol. 89 (2006)
Scheme 4
Table 2. Coupling Yields, Deprotection Conditions, and Molecular Masses of Oligonucleotides Obtained
with Various Phosphoramidites
Oligonucleotide
Phosphoramidite Coupling yield Deprotection
[M+H+]
[%]
protocol
calc. found
Pd0, r.t.
3695
)
a
e
[5’-d(TAG G3C AA3 ACT)] (26=30) 7 )
97, 98
in mixture
Pd0, r.t.
Pd0, r.t.
NH3, 608
3696 3696
a
[5’-d(TAG G1C AA1 ACT)] (27)
[5’-d(TAG GTC A1T ACT)] (28)
[5’-d(TAG G3C AA3 ACT)] (29=30) 16 b)
8 )
95, 94
95
98, 85
a
8 )
3662 3661
e
3695
)
in mixture
c
[5’-d(TAG G3C AA3 ACT)] (30)
[5’-d(TAG GTC A3T ACT)] (31)
[5’-d(TAG G3C AAT ACT)] (32)
9 )
100, 96
96
60
DBU in py
DBU in py
NH3, 608
3695 3695
3660 3660
3669 3670
c
9 )
d
6 )
d
a) Allyl-protected b) dpc-protected. c) npe-protected. ) Unprotected. e) See text.
and HCOOH/butylamine 1:1 as scavenger in anhydrous THF [17][18] (for details, see
Exper. Part). The excess of the reagent was removed with 5M aq. N,N-diethyldithiocar-
bamate (=diethylcarbamodithioate=DDTC). For the npe-group removal, the solid-
supported oligonucleotides were treated with 0.5M DBU in pyridine. The solid-sup-
port-bound oligonucleotides obtained by the allyl or the npe deprotection were then
deprotected in 25% aqueous NH3
applied to the dpc-protected oligonucleotides. The (MeO)
were purified by reversed-phase HPLC. The (MeO)2ATr-residues were removed with
ACHTREUNG
CTHREUNG
CTHREUNG
2.5% dichloroacetic acid in CH2Cl2 followed by reversed-phase HPLC (RP-18) purifi-
cation.
The allyl protocol i) was successfully employed for the oligonucleotides 27 and 28
containing 2’-deoxyxanthosine (1) while the deprotection of the O-allyl-protected 7-
deaza-2’-deoxyxanthosine residues failed (! 26 in mixture). As the 7-deazapurine
nucleobases have a significantly lower redox potential than the corresponding purine
compounds, destructive oxidation by a PdII procedure during the deallylation is the
most likely process causing this failure. The npe deprotection protocol ii) was success-
fully applied for the oligonucleotides 30 and 31 containing 7-deaza-2’-deoxyxanthosine

Helvetica Chimica Acta – Vol. 89 (2006)
2803
(3) as it was reported for 2’-deoxyxanthosine [4][10]. The dpc deprotection protocol iii)
which was applied only to oligonucleotides containing 7-deaza-2’-deoxyxanthosine (3)
failed as the protecting groups could not be removed from the protected precursor of
oligomer 29.
Next the oligonucleotides were characterized by MALDI-TOF mass spectrometry
(Table 2). Correct masses (reflector mode) were found for the oligonucleotides 27 and
28 containing 2’-deoxyxanthosine (1) prepared via allyl protection (Table 2). In both
oligonucleotides, apart from the main peaks, mass peaks with low intensity were
observed which corresponds to the release of one or two xanthine bases and the addi-
tion of one or two Na+ ions. The depurination of the xanthine moiety is expected as 2’-
deoxyxanthosine is susceptible to acidic conditions (matrix used for MALDI-TOF: 3-
hydroxypicolinic acid).
The mixture obtained from the allyl removal expected to give oligonucleotide 26
containing compound 3 showed masses of 3694, 3713, and 3731 Da. While the low
mass corresponds to the deprotected molecule 26, the higher masses represent mole-
cules which were oxidized (addition of 19 and 37 Da with respect to the deprotected
26). The complex reaction mixtures obtained by the deprotection of the dpc groups
expected to give oligomer 29 containing compound 3 gave MALDI-TOF spectra
with masses of the deprotected oligomer 29 as well as of partially deprotected com-
pounds. Correct masses were obtained for oligonucleotides 30 and 31 containing com-
pound 3 prepared by the npe protection. No additional peaks caused by ꢀdepurinationꢁ
or oxidation were detected. This results from the high glycosylic-bond stability of com-
pound 3 under acidic conditions. Similar observations have been already reported for
oligonucleotides incorporating 7-deaza-2’-deoxyadenosine or 7-deaza-2’-deoxyguano-
sine residues which are also not susceptible to ꢀdepurinationꢁ [30]. As these nucleosides
could be used as surrogates for the canonical purine compounds in enzyme catalyzed
DNA incorporation, the sensitivity of MALDI-TOF spectroscopy was significantly
increased by using these modified analogs [30]. Thus, the 2-(4-nitrophenyl)ethyl
(npe) group proves to be an ideal protecting group for the protection of 2,6-dioxo func-
tions of 7-deaza-2’-deoxyxanthosine (3), while the allyl group represents an orthogonal
protecting group for 2’-deoxyxanthosine (1) and not for 7-deaza-2’-deoxyxanthosine
(3).
Also, the composition of the oligonucleotides containing 1 and 3 was determined by
reversed-phase HPLC (RP-18) after tandem hydrolysis with snake-venom phospho-
diesterase followed by alkaline phosphatase in 0.1M Tris·HCl buffer (pH 8.3) (Fig. 1,
Exper. Part). According to Fig. 1 the canonical as well as the modified residues were
detected. From the peak area, the nucleoside ratio was determined which was in agree-
ment with the composition of the oligonucleotides. On the lipophilic reversed-phase
column, 7-deaza-2’-deoxyxanthosine (3) shows a lower mobility than that of 2’-deoxy-
xanthosine (1), which corresponds to its increased hydrophobic character.
3. Base Pairing of Oligonucleotides. Next, the duplex stability of the oligonucleoti-
des was investigated. Oligonucleotides containing 2’-deoxyxanthosine (1) have been
reported, and their base pairing was discussed [9][10]. As nothing is known about
the duplex stability of oligonucleotides containing 7-deaza-2’-deoxyxanthosine (3),
we studied this behavior with respect to the 12-mer duplex 5’-d(TAG GTC AAT
ACT) (24)·3’-d(ATC CAG TTA TGA) (25) which is used as reference in most of

2804
Helvetica Chimica Acta – Vol. 89 (2006)
Fig. 1. Reversed-phase HPLC profiles after the enzymatic hydrolysis of a) oligonucleotide 28 contain-
ing 1 and b) oligonucleotide 30 containing 3 with snake-venom phosphodiesterase and alkaline phos-
phatase in 0.1M Tris·HCl buffer (pH 8.3). Eluent: A, flow rate 0.7 ml/min (A=0.1M (Et3ACHTERNUNG H)OAc (pH
7.0)/MeCN 95 :5).
our studies [29]. In the first strand, 24, one dA residue located near the centre of the
molecule was replaced by compounds 3 or by 1 for comparison. The second strand,
25, was modified in such a way that the four canonical bases were located opposite
to compound 1 or 3. The Tm values are given in Table 3.
According to Table 3, the incorporation of the nucleosides 1 or 3 opposite to the
four canonical nucleosides resulted in a destabilization of the duplex structure com-
pared to unmodified duplex 24·25. All Tm values are now in a similar range. Neverthe-
less, the melting temperatures are higher when 7-deaza-2’-deoxyxanthosine (3) is
located opposite to dT or dG or dA than to dC. A similar behavior was observed for
oligonucleotide duplexes containing 2’-deoxyxanthosine (1), except that 1 shows a
lower Tm opposite to dA (Table 3). This melting behavior fulfils the requirements of
universal nucleosides. The Tm values of duplexes containing a classical universal
nucleoside, namely 5-nitro-1H-indole 2’-deoxy-b-D-ribofuranoside (N) which solely
contributes base stacking to duplex stability are shown for comparison [31][32].
Other universal nucleosides, namely 2’-deoxyinosine as well as 7-deaza-2’-deoxyinosine
[33] which are related to compounds 1 and 3 show a similar behavior. However, the
base pairs of the 2’-deoxyinosines with dC are always stronger than those with the
other canonical residues [33]. The Tm decrease of duplexes within universal base
pairs formed by 2’-deoxyxanthosine (1) and 7-deaza-2’-deoxyxanthosine (3) compared
to 2’-deoxyinosine [33] can be explained by the spatial demand of the 2-oxo group.
Related observation were already made on dA·dT base pairs substituted with halogen
atoms at the 2-position.
Possible base pairs of 7-deaza-2’-deoxyxanthosine with dG, dA, and dC (motif I, II,
and III resp.) are depicted in Fig. 2. Pairing motifs for 2’-deoxyxanthosine (1) have

Helvetica Chimica Acta – Vol. 89 (2006)
2805
Table 3. Tm Values and Thermodynamic Data of Oligonucleotide Duplexes Containing 2’-Deoxy-
xanthosine (1) and 7-Deaza-2’-deoxyxanthosine (3)a)
Duplexes
Tm
[8]
DTm
[8]
DG8310
[kcal/mol]
DS8
[cal/K mol]
DH8
[kcal/mol]
5’-d(TAG GTC AAT ACT)-3’ (24)
3’-d(ATC CAG TTA TGA)-5’ (25)
47
0
ꢀ10.6
ꢀ253
ꢀ89
5’-d(TAG GTC A1T ACT)-3’ (28)
3’-d(ATC CAG TTA TGA)-5’ (33)
5’-d(TAG GTC A1T ACT)-3’ (28)
3’-d(ATC CAG TGA TGA)-5’ (34)
5’-d(TAG GTC A1T ACT)-3’ (28)
3’-d(ATC CAG TAA TGA)-5’ (35)
5’-d(TAG GTC A1T ACT)-3’ (28)
3’-d(ATC CAG TCA TGA)-5’ (36)
42
40
39
36
ꢀ5
ꢀ7
ꢀ9.1
ꢀ8.6
ꢀ8.2
ꢀ7.2
ꢀ232
ꢀ209
ꢀ226
ꢀ196
ꢀ81
ꢀ73
ꢀ78
ꢀ68
ꢀ8
ꢀ11
5’-d(TAG GTC A3T ACT)-3’ (31)
3’-d(ATC CAG TTA TGA)-5’ (33)
5’-d(TAG GTC A3T ACT)-3’ (31)
3’-d(ATC CAG TGA TGA)-5’ (34)
5’-d(TAG GTC A3T ACT)-3’ (31)
3’-d(ATC CAG TAA TGA)-5’ (35)
5’-d(TAG GTC A3T ACT)-3’ (31)
3’-d(ATC CAG TCA TGA)-5’ (36)
42
40
43
38
-5
ꢀ7
ꢀ4
ꢀ9
ꢀ10.3
ꢀ10.0
ꢀ10.7
ꢀ9.3
ꢀ208
ꢀ348
ꢀ206
ꢀ226
ꢀ75
ꢀ118
ꢀ75
ꢀ80
5’-d(TAG GTC ANT ACT)-3’ (37)
3’-d(ATC CAG TTA TGA)-5’ (33)
5’-d(TAG GTC ANT ACT)-3’ (37)
3’-d(ATC CAG TGA TGA)-5’ (34)
5’-d(TAG GTC ANT ACT)-3’ (37)
3’-d(ATC CAG TAA TGA)-5’ (35)
5’-d(TAG GTC ANT ACT)-3’ (37)
3’-d(ATC CAG TCA TGA)-5’ (36)
42b)
41b)
44b)
45b)
ꢀ5
ꢀ6
ꢀ3
ꢀ2
ꢀ8.7
ꢀ8.2
ꢀ8.9
ꢀ9.6
ꢀ173
ꢀ163
ꢀ181
ꢀ210
ꢀ63
ꢀ59
ꢀ65
ꢀ75
5’-d(TAG GTC A1T ACT)-3’ (28)
3’-d(ATC CAG T4A TGA)-5’ (38)
5’-d(TAG GTC A3T ACT)-3’ (31)
3’-d(ATC CAG T4A TGA)-5’ (38)
5’-d(TAG G1C AA1 ACT)-3’ (27)
3’-d(ATC C4G TT4 TGA)-5’ (39)
5’-d(TAG G3C AA3 ACT)-3’ (30)
3’-d(ATC C4G TT4 TGA)-5’ (39)
5’-d(TAG GTC AAT ACT)-3’ (40)
3’-d(ATC C4G TT4 TGA)-5’ (39)
47
46
49
49
59
0
ꢀ1
ꢀ11.6
ꢀ10.9
ꢀ11.9
ꢀ12.1
ꢀ13.9
ꢀ237
ꢀ192
ꢀ211
ꢀ199
ꢀ85
ꢀ71
ꢀ77
ꢀ74
+2
+2
c
c
+12
)
)
a) Measured in 100 mM NaCl, 10 mM MgCl2, and 10 mM Na-cacodylate (pH 7.0) with 5 mM single-strand
concentration. b)Measured in 1M NaCl, 100 mM MgCl2, and 60 mM Na-cacodylate (pH 7.0) with 5 mM sin-
gle-strand concentration, N=5-nitro-1 H-indole 2’-deoxy-b-D-ribofuranoside [32]. c) Not reported [29].
already been reported [9][34]. Similar modes are expected for 7-deaza-2’-deoxyxantho-
sine (3). However, as we observed a significant Tm increase when compound 3 pairs
with dA (Table 3), a change from the Watson–Crick to the Hoogsteen mode might be
the reason for this behavior (Fig. 2, motif IV).

2806
Helvetica Chimica Acta – Vol. 89 (2006)
Fig. 2. Base-pair motifs for compounds 3 (X=CH) or 1 (X=N) with dG, dC, and dA
The oxo groups play an important role in the properties and functions of 2’-deoxy-
xanthosine (1) or its 7-deaza analogue 3. The excellent review of Shugar and co-work-
ers has already discussed many aspects of the special properties of the xanthine and the
xanthosine molecule [34]. While all canonical nucleosides are neutral under physiolog-
ical conditions, the xanthosine exists predominantly as monoanion. The pKa value
refers to the deprotonation of HꢀN(3) leading to a mesomeric stabilized 2-oxo
anion of compound 1. According to its pKa ACHTERoUNG f 5.7, ca. 95% of the 2’-deoxyxanthosine
molecules are monoanions at pH 7.0. Thus it was suggested that not the neutral mole-
cule but the monoanion contributes to base pairing. The situation becomes even more
complicated as two different motifs can be formed due to the donor and acceptor pat-
tern of the molecules involved in base pairing. Fig. 3 shows possible base-pair motifs for
2’-deoxyxanthosine (1) as well as for 7-deaza-2’-deoxyxanthosine (3) with dT within
duplex DNA.
Earlier, we determined a pKa value of 7.2 for 7-deaza-2’-deoxyxanthosine (3). A
recent experiment changed this to 6.7. This value is 1 pKa unit higher than that of 1
(pKa 5.7), with the consequence that significantly less molecules are monoanions at
pH 7.0 in the case of 3. The single-crystal X-ray structure of the neutral 7-deaza-2’-
deoxyxanthosine (3) has been reported, which was crystallized under slightly acidic
conditions [35]. Consequently, we measured the Tm values of the duplex 31·33 contain-
ing one 3·dT base pair at three different pH values (pH 5.5, 7.0, and 8.0) under the con-
ditions described in Table 3. The Tm value of 31·33 (428) at pH 7.0 was not changed
either under alkaline or acidic conditions. This is in contrast to the results observed
for a related molecule with a similar pKa value (6-aza-2’-deoxyuridine, pKa 6.8). A sig-
nificant increase of the Tm value was observed for oligonucleotide-duplex formation
when changing from neutral to acidic conditions [24]. However, the 2-oxo group

Helvetica Chimica Acta – Vol. 89 (2006)
2807
Fig. 3. Base-pair motifs for compounds 3 (X=CH) or 1 (X=N) with dT
might not be directly involved in H-bonding as shown in the motifs VI and VIII (Fig. 3);
thus, this behavior becomes understandable.
Oligonucleotides containing the nucleosides 1 and 3 were also hybridized with those
incorporating the diamine nucleoside 4 (3-bromo-1-(2-deoxy-b-D-erythro-pentofurano-
syl)-1H-pyrazolo[3,4-d]pyrimidine-4,6-diamine) [29]. A single incorporation of 1
against 4 in the place of the dA·dT pair stabilized the duplex 28·38 to the level of
the unmodified duplex 24·25 (Tm 478) and hence, base pair 1·4 is as stable as the
dA·dT base pair. Two modifications enhanced the stability to 498 (duplex 27·39) show-
ing the nearest neighbor-influence on the base-pair stability. Also, a similar experiment
was carried out with 7-deaza-2’-deoxyxanthosine (3). The corresponding duplex 31·38
showed a similar Tm (468). Also two incorporations of 3 in duplex 30·39 resulted in the
same Tm ACHTREiUGN ncrease (498) as that observed for 1. This Tm increase results mainly from the
Br-substitution and not from the presence of the additional 2-amino function. This was
already verified earlier in related duplex structures [29]. The higher stability of the
dA·dT base pair including that of a ꢀpyrazolo-dAꢁ with dT than that of dA with com-
pound 1 or 3 might result from a distortion of the helix axis by the purine·purine pair.
On the other hand, the larger surface area of ꢀpurineꢁ residues can cause better stacking.
It stays to proof if duplexes containing the base pair motif IX (Fig. 4) exclusively are
more stable than those formed by dA·dT pairs.
Conclusions. – Among various groups (allyl, diphenylcarbamoyl, 2-(4-nitrophen-
ACHTREUNGyl)ethyl) studied for the oxo-group protection of the phosphoramidite of 7-deaza-2’-
deoxyxanthosine (3) only the 2-(4-nitrophenyl)ethyl (npe) group was found to be effec-
tive. It was removed from CPG-bound oligonucleotide by DBU in pyridine. The allyl
group which works efficiently for the protection of 2’-deoxyxanthosine (1) caused dif-
ficulties in the case of 7-deaza-2’-deoxyxanthosine (3) due to the oxidation of the 7-de-
azapurine base in the presence of Pd-catalyst. Although it was possible to incorporate

2808
Helvetica Chimica Acta – Vol. 89 (2006)
Fig. 4. Base-pair motifs suggested for compounds 3 (X=CH) or 1 (X=N) with 4 and dT with 4
one 7-deaza-2’-deoxyxanthosine residue without base protection in a moderate yield
(see 32 in Table 2), multiple incorporations failed. Oligonucleotides containing several
7-deaza-2’-deoxyxanthosine (3) residues required oxo-group protecting, preferentially
by the npe residue. Compound 3 can act as universal nucleoside as it forms almost
equally stable base pairs with the canonical DNA constituents as reported for 2’-deoxy-
ACHTREUNGxanthosine (1) or the corresponding 2’-deoxyinosine derivatives. According to its pKa
value (6.7), a significant amount of 3 forms a monoanion at pH 7.0 in the monomeric
state as well as when it is part of the oligonucleotide chain. Surprisingly, no change of
the duplex stability was observed when the pH value was changed from 5.5 to 8.0. As
the related 6-aza-2’-deoxyuridine (pKa =6.8) shows a significant Tm change, it is con-
cluded that the 2-oxo group of nucleoside 3 does not participate in the base pairing.
The nucleosides 1 and 3 pair with 3-bromo-1-(2-deoxy-b-D-erythro-pentofuranosyl)-
1H-pyrazolo[3,4- d]pyrimidine-4,6-diamine (4) resulting in duplexes which are as stable
as those containing dA·dT pairs. This results from the influence of the Br-substituents
and not from the additional amino group as it was already demonstrated with com-
pounds incorporating base pairs of 4 with dT [29]. It is expected that duplexes are fur-
ther stabilized when the nucleoside 3 bears 7-halogen or 7-alkynyl substituents [36].
Experimental Part
1. General. All chemicals were purchased from Aldrich, Sigma, or Fluka (Sigma-Aldrich Chemie
GmbH, Deisenhofen, Germany). Solvents were of laboratory grade. TLC: aluminium sheets, silica gel
60 F254, 0.2-mm layer (VWR, Germany). Column flash chromatography (FC): silica gel 60 (VWR, Ger-
many) at 0.4 bar; sample collection with an UltroRac-II fractions collector (LKB Instruments, Sweden).
Melting temp.: Cary-1/3-UV/VIS spectrophotometer (Varian, Australia) equipped with a Cary thermo-
electrical controller; continuous temp. measurement in the reference cell with a Pt-100 resistor. UV Spec-
tra: U-3200 spectrometer (Hitachi, Tokyo, Japan); lmax (e) in nm. NMR Spectra: Avance-250 or AMX-500
spectrometers (Bruker, Karlsruhe, Germany), at 250.13 MHz for 1H and 13C; d in ppm relative to Me
4ACHTRESUNG i
as internal standard or external 85% H3PO4 for 31P, J values in Hz. Elemental analyses were performed by
the Mikroanalytisches Laboratorium Beller (Gçttingen, Germany).
2. Synthesis, Deprotection, and Purification of the Oligonucleotides. The oligonucleotide syntheses
were carried out in an ABI-392-08-DNA synthesizer (Applied Biosystems, Weiterstadt, Germany) on a
1-mmol scale following the synthesis protocol for 3’-cyanoethyl phosphoramidites (user manual for the
392-DNA synthesizer, Applied Biosystems, Weiterstadt, Germany). The coupling efficiency was always
higher than 95%. The following deprotection protocols were used for the deprotection of the oligonu-
cleotides.

Helvetica Chimica Acta – Vol. 89 (2006)
2809
i) Deprotection of O-Allyl-Protected Oligonucleotides with a Pd0 Complex Followed by Treatment
with 25% Aqueous NH3 Solution. After the completion of the oligonucleotide synthesis, the columns
were washed with Ar-flushed THF and disassembled. The supported oligonucleotides were treated at
r.t. with a THF soln. of [Pd0(PPh3)4] (ca. 3 equiv./allylic group), PPh3 (ca. 25 equiv./allylic group), and
excess of allyl scavenger HCOOH/butylamine 1:1 (1.5 ml) for 30 min at r.t. [17][18][37]. Then the mix-
ture was centrifuged, and the soln. was discarded. The supports were washed with THF, acetone, 5M aq.
DDTC, and finally with distilled H2O (2”each). The deprotection of other base-protecting groups was
achieved by 25% NH3 soln. as described below.
ii) Deprotection of O-npe-Protected Oligonucleotides with DBU in Pyridine Followed by Treatment
with 25% Aqueous NH3 Solution. Solid-supported oligonucleotides were treated with 0.5M DBU in pyr-
idine at r.t. for 24 h [9][25]. The solvent was evaporated in a Speed-Vac evaporator, and the residue was
treated with 25% aq. NH3 soln. under standard conditions.
The oligonucleotides obtained by the treatment with Pd0 or DBU were cleaved from the solid sup-
port and treated with 25% aq. NH3 soln. at 608 for 14–16 h. The oligonucleotides without 2’-deoxyxan-
thosine (1) and 7-deaza-2’-deoxyxanthosine (3) residues were deprotected only in 25% aq. NH3 soln. at
608 for 14–16 h. The completely deprotected oligomers were purified by HPLC. Purification of 5’-dime-
thoxytrityl oligomers was performed by reversed-phase HPLC (RP-18) with the following solvent gradi-
ent system: A=0.1M (Et3ACHTREUNGN H)OAc (pH 7.0)/MeCN 95 :5; B=MeCN; 3 min 20% B in A, 15 min 50% B
in A, and 25 min 20% B in A, with a flow rate of 0.8 ml/min. The soln. was dried and treated with 2.5%
CHCl2COOH/CH2Cl2 for 5 min at 08 to remove the 4,4’-dimethoxytrityl residues. The detritylated
oligomers were purified by reversed-phase HPLC with the gradient: within 20 min 0 ! 20% B in A,
with a flow rate of 0.8 ml/min. The oligomers were desalted (RP-18, silica gel) and lyophilized on a
Speed-Vac evaporator to yield colorless solids which were frozen at ꢀ248.
The enzymatic hydrolysis of the oligonucleotides was performed as described by Seela and Becher
[29] with snake-venom phosphodiesterase (EC 3.1.15.1, Crotallus adamanteus) and alkaline phosphatase
(EC 3.1.3.1, Escherichia coli from Roche Diagnostics GmbH, Germany) in 0.1M Tris·HCl buffer (pH 8.3),
which was carried out on reversed-phase HPLC (0–20 min A, A=0.1M (Et3ACHTRENUNG H)OAc (pH 7.0)/MeCN
95 :5). Quantification of the constituents was made on the basis of the peak areas, which were divided
by the extinction coefficients of the nucleosides ((e260): dT 8800, dC 7300, dA 15400, dG 11700, 1 8400,
3 4800). The molecular masses of the oligonucleotides were determined by MALDI-TOF Biflex-III
mass spectrometer (Bruker-Saxonia, Leipzig, Germany) with 3-hydroxypicolinic acid (3-HPA) as a
matrix.
7-[2-Deoxy-b-D-erythro-pentofuranosyl]-2,4-bis(prop-2-enyloxy)-7H-pyrrolo[2,3-d]pyrimidine
(11a). A suspension of 10a [20] (500 mg, 0.92 mmol) in 1M sodium prop-2-enolate (50 ml) was stirred at
508 for 3 h. The soln. was cooled to r.t. and applied on the top of the column (silica gel, 2.5”15 cm,
CH2Cl2 (! prop-2-en-1-ol), then CH2Cl2/Me
2ACHTRECUNG O 8 :2): 11a (240 mg, 84%) Colorless solid. TLC (silica
gel, CH2Cl2/Me2ACO 8 :2): Rf 0.28. UV (MeOH): 259 (5000), 271 (5000). H-NMR ((D6)DMSO): 2.22
1
CTHREUNG
(m, Haꢀ(2’)); 2.49 (m, Hbꢀ(2’)); 3.51 (m, CH2(5’)); 3.81 (m, HꢀC(4’)); 4.35 (m, HꢀC(3’)); 4.83–4.99
(m, 2 CH2, OHꢀC(5’)); 5.22–5.44 (m, 2=CH2, OHꢀC(3’)); 6.07 (m, 2=CH); 6.47 (m, HꢀC(1’), Hꢀ
C(5)); 7.40 (d, J=3.71, HꢀC(6)). Anal. calc. for C
17ACHTREUNHG 21ACHTREUNNG 3AHCTREUNOG 5 ACTHRE(UNG 347.37): C 58.78, H 6.09, N 12.10; found:
C 58.92, H 6.04, N 12.02.
7-[2-Deoxy-5-O-(4,4’-dimethoxytrityl)-b-D-erythro-pentofuranosyl]-2,4-bis(prop-2-enyloxy)-7H-pyr-
rolo[2,3-d]pyrimidine (12a). Compound 11a (790 mg, 2.27 mmol) was dried by co-evaporation with anh.
pyridine (2”10 ml) and dissolved in anh. pyridine (30 ml). To this soln. was added 4,4’-dimethoxytrityl
chloride (1.149 mg, 3.405 mmol) while stirring, and stirring was continued for 3 h. The mixture was
then partitioned between H2O and CH2Cl2, the org. layer washed with H2O and brine (each 2”50 ml),
the solvent evaporated and the residue subjected to FC (silica gel, 3”12 cm, CH2Cl2/Me
2ACHTREUCNG O 95 :5):
12a (880 mg, 59%). Colorless solid. TLC (silica gel, CH2Cl2/Me2ACO 97.5 :2.5): Rf 0.41. UV (MeOH):
CTHREUNG
1
273 (11500). H-NMR ((D6)DMSO): 2.23 (m, HaꢀC(2’)); 2.59 (m, HbꢀC(2’)); 3.14 (m, CH2(5’)); 3.71
(2s, 2 MeO); 3.94 (m, HꢀC(4’)); 4.38 (m, HꢀC(3’)); 4.80 (m, CH2); 4.97 (d, J=5.46, CH2); 5.25–5.45
(m, 2=CH2, OHꢀC(3’)); 6.08 (m, 2=CH); 6.46 (m, HꢀC(1’), HꢀC(5)); 6.78–6.84 (m, arom. H);
7.21–7.25 (m, arom. H); 7.34–7.37 (m, HꢀC(6), arom. H). Anal. calc. for C
38ACHTUERHNG 39ACHTREUNGN3AHCTREOUNG 7 ACHTER(UNG 649.73): C
70.25, H 6.05, N 6.47; found: C 70.18, H 5.94, N 6.39.

2810
Helvetica Chimica Acta – Vol. 89 (2006)
7-[2-Deoxy-5-O-(4,4’-dimethoxytrityl)-b-D-erythro-pentofuranosyl]-2,4-bis(prop-2-enyloxy)-7H-pyr-
rolo[2,3-d]pyrimidine 3’-(2-Cyanoethyl Diisopropylphosphoramidite) (7). To a soln. of 12a (150 mg, 0.23
mmol) in anh. CH2Cl2 (5 ml), diisopropylethylamine ((i-Pr)2ACHTREUENG tN; 67 ml, 0.39 mmol) and 2-cyanoethyl di-
isopropylphosphoramidochloridite (55.5 ml, 0.25 mmol) were added under Ar. After stirring for 30 min,
5% aq. NaHCO3 soln. (25 ml) was added, and the mixture was extracted with CH2Cl2 (2”25 ml). The org.
layer was dried (Na2SO4), the solvent evaporated, and the residue subjected to FC (silica gel, 1.5”7 cm,
CH2Cl2/Me
2ACHRTUNEGCO 95 :5): 7 (140 mg, 71%). Colorless foam. TLC (silica gel, CH2Cl2/Me2ACHTRECUGN O 97.5 :2.5): Rf
0.71. 31P-NMR (CDCl3): 149.63; 149.96.
2’-Deoxy-2,6-bis-O-(prop-2-enyloxy)xanthosine (11b). As described for 11a, with 10b [20] (500 mg,
0.92 mmol) and 1M sodium prop-2-enolate (50 ml): 11b (210 mg, 75%). Colorless solid. TLC (silica gel,
1
CH2Cl2/MeOH 9 :1): Rf 0.47. UV (MeOH): 244 (8100), 265 (12200). H-NMR ((D6)DMSO): 2.25 (m,
HaꢀC(2’)); 2.73 (m, HbꢀC(2’)); 3.53 (m, CH2(5’)); 3.85 (m, HꢀC(4’)); 4.41 (m, HꢀC(3’)); 4.85–5.02
(m, 2 CH2, OHꢀC(5’)); 5.25–5.44 (m, 2=CH2, OHꢀC(3’)); 6.09 (m, 2=CH); 6.31 (m, HꢀC(1’)); 8.37
(s, HꢀC(8)). Anal. calc. for C
16CATHERUNHG 20ACHTERNUNG 4ACHTRENUOG 5 ACHTRE(UNG 348.35): C 55.17, H 5.79, N 16.08; found: C 55.20, H 5.90, N
16.02.
2’-Deoxy-5’-O-(4,4’-dimethoxytrityl)-2,6-bis-O-(prop-2-enyloxy)xanthosine (12b). As described for
12a, with 11b (686 mg, 1.96 mmol) and 4,4’-dimethoxytrityl chloride (800 mg, 2.36 mmol): 12b (850 mg,
1
66%). Colorless solid. TLC (silica gel, CH2Cl2/MeOH 95 :5): Rf 0.27. UV (MeOH): 265 (19700). H-
NMR ((D6)DMSO): 2.34 (m, Haꢀ(2’)); 2.88 (m, Hbꢀ(2’)); 2.90–3.27 (m, CH2(5’)); 3.70, 3.72 (2s, 2
MeO); 4.00 (m, HꢀC(4’)); 4.48 (m, HꢀC(3’)); 4.67 (m, CH2); 5.02 (d, J=2.61, CH2); 5.23–5.47 (m,
2=CH, OHꢀC(3’)); 6.01, 6.13 (m, 2=CH); 6.36 (t, J=6.09, 6.34, HꢀC(1’)); 6.71–6.79 (m, arom. H);
7.17–7.31 (m, arom. H); 8.28 (s, HꢀC(8)). Anal. calc. for C
37ACHTUERHNG 38ACHTERUGNN 4ACHRTENUOG 7 ACHTER(UNG 650.72): C 68.29, H 5.89, N
8.61; found: C 68.38, H 6.00, N 8.76.
2’-Deoxy-5’-O-(4,4’-dimethoxytrityl)-2,6-bis-O-(prop-2-enyloxy)xanthosine 3’-(2-Cyanoethyl Diiso-
propylphosphoramidite) (8). As described for 7, with 12b (300 mg, 0.46 mmol), anh. CH2Cl2 (5 ml), (i-
Pr)2ACHTREUENG tN (134 ml, 0.78 mmol), and 2-cyanoethyl diisopropylphosphoramidochloridite (111 ml, 0.50
mmol). Workup with 5% aq. NaHCO3 soln. (30 ml) and CH2Cl2 (2”30 ml), followed by FC, as described
CO 95 :5): Rf 0.60. 31P-NMR
for 7:8 (339 mg, 86%). Colorless foam. TLC (silica gel, CH2Cl2/Me2ACHTREUNG
(CDCl3): 150.02; 150.19.
7-(2-Deoxy-b-D-erythro-pentofuranosyl)-1,3-bis(diphenylcarbamoyl)-1,7-dihydro-2H-pyrrolo[2,3-d]-
pyrimidine-2,4(3H)-dione (13a) and 7-[2-Deoxy-5-O-(diphenylcarbamoyl)-b-D-erythro-pentofuranosyl]-
1,3-bis(diphenylcarbamoyl)-1,7-dihydro-2H-pyrrolo[2,3-d]pyrimidine-2,4(3H)-dione (13b).
To
a
soln. (7-(2-deoxy-b-D-erythro-pentofuranosyl)-1,7-dihydro-2H-pyrrolo[2,3-d]pyrimidine-2,
2AEtN (400 ml, 2.33 mmol)
4(3H)-dione) [8] (3; 200 mg, 0.75 mmol) in anh. pyridine (5 ml) was (i-Pr) CTHREUNG
and diphenylcarbamic chloride (490 mg, 2.11 mmol) while stirring. After 100 min, 5% aq. NaHCO3
soln. (60 ml) was added, and the mixture was extracted with CH2Cl2 (3”50 ml). The combined org.
phase was dried (Na2SO4), the solvent evaporated, and the residue co-evaporated with toluene (3”50
ml) and subjected to FC (silica gel, 10”3 cm, CH2Cl2/Me2ACHTREUCNG O 85 :15 (100 ml), then CH2Cl2/MeOH 9 :1
(400 ml)): less polar 13b (37 mg, 6%) and more polar 13a (364 mg, 74%). Data of 13b: Colorless
foam. TLC (silica gel, CH2Cl2/MeOH 9 :1): Rf 0.4. 1H-NMR ((D6)DMSO): 2.12–1.97 (m, CH2(2’));
4.01 (m, HꢀC(4’)); 4.32–4.24 (m, HꢀC(3’), CH2(5’)); 5.47 (d, J=4.2, OHꢀC(3’)); 6.46 (t, J=6.8, Hꢀ
C(1’)); 6.70 (d, J=3.8, HꢀC(5)); 7.12 (d, J=3.8, HꢀC(6)); 7.54–7.22 (m, arom. H).
Data of 13a: TLC (silica gel, CH2Cl2/MeOH 9 :1): Rf 0.2. UV (MeOH): 226 (48000). 1H-NMR
((D6)DMSO): 2.50 (br. s, CH2(2’)); 3.54 (m, CH2(5’)); 3.84 (br. s, HꢀC(4’)); 4.38 (br. s, HꢀC(3’)); 4.96
(t, J=5.2, OHꢀC(5’)); 5.35 (d, J=3.9, OHꢀC(3’)); 6.54 (t, J=6.7, HꢀC(1’)); 6.77 (d, J=3.5, Hꢀ
C(5)); 7.51–7.32 (m, arom. H); 7.86 (d, J=3.6, HꢀC(6)). Anal. calc. for C37H31N5O7 (657.67): C 67.57,
H 4.75, N 10.65; found: C 67.37, H 4.91, N 10.30.
7-[2-Deoxy-5-O-(4,4’-dimethoxytrityl)-b-D-erythro-pentofuranosyl]-1,3-bis(diphenylcarbamoyl)-1,7-
dihydro-2H-pyrrolo[2,3-d]pyrimidine-2,4(3H)-dione (14). Method A: Compound 15 (380 mg, 0.67 mmol)
was dissolved in anh. pyridine (44 ml) while stirring at r.t. To this soln., (i-Pr)2ACHTREEUNG tN (325 ml, 1.87 mmol) and
diphenylcarbamic chloride (432 mg, 1.86 mmol) was added. The mixture was stirred for 2.5 h at r.t. and
then poured into ice-cold aq. of 5% NaHCO3 soln. (80 ml). The mixture was extracted with CH2Cl2

Helvetica Chimica Acta – Vol. 89 (2006)
2811
(3”50 ml), the combined org. phase dried (Na2SO4), the solvent evaporated, and the residue co-evapo-
rated with toluene (50 ml”3) and subjected to FC (silica gel, 10”3 cm, CH2Cl2/Me2ACO 98 :2 (250 ml),
then CH2Cl2/Me2ACO 95 :5 (320 ml): 14 (608 mg, 95%). Colorless foam.
Method B: To a soln. of 13a (50 mg, 0.08 mmol) in anh. pyridine (6.4 ml) was added 4,4’-dimethoxy-
CTHREUNG
CTHREUNG
ACHTREUNG
7-[2-Deoxy-5-O-(4,4’-dimethoxytrityl)-b-D-erythro-pentofuranosyl)-1,7-dihydro-2H-pyrrolo[2,3-d]-
pyrimidin-2,4-(3H)-dione (15). Compound 3 [8] was dried by repeated co-evaporation of anh. pyridine
and dissolved in anh. pyridine (65 ml). To this suspension, 4,4’-dimethoxytrityl chloride (1.25 g, 3.69
mmol) was added and stirred for 2.5 h at r.t. The reaction was quenched by adding MeOH (3 ml). To
this soln., 5% aq. NaHCO3 soln. was added (100 ml). The mixture was extracted with CH2Cl2 (3”50
ml), the org. phase dried (Na2SO4), the solvent evaporated, the residue dissolved in CH2Cl2, and the
soln. adsorbed on silica gel (10 g) and subjected to FC (silica gel, CH2Cl2/MeOH 98 :2 (350 ml), then
CH2Cl2/MeOH 9 :1 (150 ml)): 15 (800 mg, 75%). Colorless foam. TLC (silica gel, CH2Cl2/MeOH 9 :1):
Rf 0.5. UV (MeOH): 235 (27400), 273 (9300). 1H-NMR ((D6)DMSO): 2.27 (m, Haꢀ(2’)); 2.42–2.39
(m, Hbꢀ(2’)); 3.09 (m, CH2(5’)); 3.73 (s, 2 MeO); 3.87 (m, HꢀC(4’)); 4.30 (m, HꢀC(3’)); 5.33 (d,
J=4.3, OHꢀC(3’)); 6.24 (t, J=5.9, HꢀC(1’)); 6.28 (d, J=3.5, HꢀC(5)); 6.79 (d, J=3.5, HꢀC(6));
6.87–6.83 (m, arom. H); 7.36–7.20 (m, arom. H); 10.68 (s, NH); 11.71 (s, NH). Anal. calc. for
C32H31N3O7 (569.60): C 67.48, H 5.49, N 7.38; found: C 67.29, H 5.61, N 7.32.
7-[2-Deoxy-5-O-(4,4’-dimethoxytrityl)-b-D-erythro-pentofuranosyl]-1,3-bis(diphenylcarbamoyl)-1,7-
dihydro-2H-pyrrolo[2,3-d]pyrimidine-2,4(3H)-dione 3’-(2-Cyanoethyl Diisopropylphosphoramidite)
(16). As described for 7, with 14 (200 mg, 0.21 mmol), CH2Cl2 (10 ml), (i-Pr)2ACHTEREUNG tN (49 ml, 0.28 mmol),
and 2-cyanoethyl diisopropylphosphoramidochloridite (60 ml, 0.27 mmol) for 20 min. Workup with 5%
aq. NaHCO3 soln. (30 ml) and CH2Cl2 (3”20 ml) as described for 7. FC (silica gel, 7”1.5 cm, CH2Cl2/
Me
2A
2ACHTRECUNG O 95 :5 (25 ml)) afforded 16 (230 mg, 95%). Colorless foam.
TLC (silica gel, CH2Cl2/Me CTHREUNG
CO 95 :5): Rf 0.4. 31P-NMR (CDCl3): 149.84; 150.01.
7-[2-Deoxy-5-O-(4,4’-dimethoxytrityl)-b-D-erythro-pentofuranosyl]-1,7-dihydro-2H-pyrrolo[2,3-d]-
pyrimidin-2,4(3H)-dione 3’-(2-Cyanoethyl Diisopropylphosphoramidite) (6). As described for 7, with 15
(390 mg, 0.68 mmol), anh. THF (5 ml) instead of CH2Cl2, (i-Pr)
noethyl diisopropylphosphoramidochloridite (200 ml, 0.90 mmol) for 20 min. After evaporation (no
extraction), the residue was subjected to FC (silica gel, 10”1.5 cm, CH2Cl2/Me2ACO 9 :1 (20 ml), then
CH2Cl2/Me2ACO 4 :1 (20 ml), and CH2Cl2/Me2A
2ACHTREEUGN tN (357 ml, 2.05 mmol), and 2-cya-
CTHREUNG
C
G
CO 1:1): Rf 0.7. 31P-NMR
zone afforded 6 (150 mg, 28%). Colorless foam. TLC (silica gel, CH2Cl2/Me2ACHTREUNG
(CDCl3): 149.57; 149.98.
2-Amino-7-(3,5-di-O-acetyl-2-deoxy-b-D-erythro-pentofuranosyl)-3,7-dihydro-4H-pyrrolo[2,3-d]pyr-
imidin-4-one (17). To a suspension of 7-deaza-2’-deoxyguanosine (c7Gd) [27] (2 g, 7.51 mmol) in anh. pyr-
idine (35 ml) was added Ac2O (3 ml, 31.7 mmol) and the mixture was stirred overnight at r.t. The reaction
was quenched by additon of MeOH (1 ml). Then the mixture was poured into 5% aq. NaHCO3 soln. (100
ml) and extracted with CH2Cl2 (3”100 ml), the combined org. phase dried (Na2SO4), the solvent evapo-
rated, and the residue subjected to FC (silica gel, 5”12 cm, CH2Cl2/MeOH 9 :1): 17 (2.31 g, 88%). Color-
1
less solid. TLC (CH2Cl2/MeOH 9 :1): Rf 0.42. UV (MeOH): 259 (14600). H-NMR ((D6)DMSO): 2.05,
2.07 (2s, 2 MeCO); 2.33 (m, Haꢀ(2’)); 2.68 (m, Hbꢀ(2’)); 4.18 (m, HꢀC(4’), CH2(5’)); 5.21 (d, J=5.25,
HꢀC(3’)); 6.03 (m, HꢀC(1’), HꢀC(5), NH2); 6.90 (d, J=4.12, HꢀC(6)); 10.42 (s, NH). Anal. calc. for
C15CATHERUNGH18ACHTRUENNG 4ACHTREUGON 6 ACHTRE(UNG 350.33): C 51.43, H 5.18, N 15.99; found: C 51.50, H 5.14 N 15.86.

2812
Helvetica Chimica Acta – Vol. 89 (2006)
7-(3,5-Di-O-acetyl-2-deoxy-b-D-erythro-pentofuranosyl)2-[(4,4’-dimethoxytrityl)amino]-3,7-dihydro-
4H-pyrrolo[2,3-d]pyrimidin-4-one (18). To a soln. of 17 (2 g, 5.70 mmol) in anh. pyridine was added 4,4’-
dimethoxytrityl chloride (2.03 g, 6.0 mmol). After stirring overnight at r.t., the mixture was poured into
5% aq. NaHCO3 soln. (100 ml) and extracted with CH2Cl2 (3”150 ml). The combined org. phase was
dried (Na2SO4), the solvent evaporated, and the residue subjected to FC (silica gel, 5”12 cm, CH2Cl2/
(CH3)2ACHTRECUGN O 98 :2, then 95 :5): 18 (3.4 g, 91%). Colorless foam. TLC (CH2Cl2/MeOH 95 :5): Rf 0.51. UV
(MeOH): 268 (15800), 285 (14600). 1H-NMR ((D6)DMSO): 1.78 (m, Haꢀ(2’)); 2.02, 2.08 (2s, 2
MeCO); 2.16 (m, Hbꢀ(2’)); 3.71 (s, 2 MeO); 4.06 (m, HꢀC(4’), CH2(5’)); 5.02 (d, J=5.35, HꢀC(3’));
5.56 (dd, J=5.57, 5.42, HꢀC(1’)), 6.24 (d, J=3.55, HꢀC(5)); 6.73 (d, J=3.57, HꢀC(6)); 6.73–7.33 (m,
arom. H); 7.42 (s, NH); 10.40 (s, NH). Anal. calc. for C36ACHTREUGNH36ACHTREUGNN 4ACHTREUNOG 8 ACHTER(UNG 652.69): C 66.25, H 5.56, N 8.58;
found: C 66.31, H 5.62, N 8.60.
2-Amino-7-(3,5-di-O-acetyl-2-deoxy-b-D-erythro-pentofuranosyl)-4-O-[2-(4-nitrophenyl)ethyl]-7H-
pyrrolo[2,3-d]pyrimidine (19). Method A: A soln. of 18 (2 g, 3.06 mmol) in abs. dioxane (10 ml) was
evaporated. The residue was dissolved in abs. dioxane (20 ml), and to this soln. were added PPh3 (6.4
g, 24.8 mmol), 2-(4-nitrophenyl)ethanol (4.12 g , 24.8 mmol) and DEAD (4.23 g, 24.8 mmol). The mix-
ture was stirred for 1 h at. r.t. The solvent was evaporated and the residue subjected to FC (silica gel,
5”12 cm, Et2ACHTREOUNG ). The main fraction was rechromatographed: 19 (350 mg, 23%). Yellow foam.
Method B: To a soln. of 17 (3.5 g, 10.0 mmol) in anh. THF (75 ml) were added PPh3 (8.4 g, 33.03
mmol), DEAD (5.82 g, 33.03 mmol), and 2-(4-nitrophenyl)ethanol (5.61 g, 33.03 mmol). After stirring
for 1 h at. r.t., the mixture was worked up as described for Method A: 19 (2.75 g, 55%). Yellow foam.
1
TLC (Et
2A
MeCO); 2.34 (m, Haꢀ(2’)); 2.75 (m, Hbꢀ(2’)); 3.21 (t, J=6.45, 6.75, CH2); 4.18 (m, HꢀC(4’),
CH2(5’)); 4.59 (t, J=6.40, 6.55, OCH2); 5.23 (d, J=5.45, HꢀC(3’)); 6.31 (m, HꢀC(1’), HꢀC(5), NH2);
7.07 (d, J=3.70, HꢀC(6)); 7.60 (d, J=8.50, arom. H); 8.15 (d, J=8.5, arom. H). Anal. calc. for C
23ACHTREHUNG 25-
ACHTREUNGN5ACHTREUNGO CHTRE(UNG 499.47): C 55.31, H 5.05, N 14.02; found: C 55.40, H 5.00, N 13.96.
8 A
7-(3,5-Di-O-acetyl-2-deoxy-b-D-erythro-pentofuranosyl)-4-O-[2-(4-nitrophenyl)ethyl]-1,7-dihydro-
2H-pyrrolo[2,3-d]pyrimidin-2-one (20). To a soln. of 19 (3 g, 6.0 mmol) in 30% aq. AcOH/Me2ACO 2 :1,
CTHREUNG
(375 ml) was added NaNO2 (10 g, 140 mmol) in H2O (25 ml). After stirring for 2 h at r.t., the solvent
was evaporated and the residue subjected to FC (silica gel, 5”12 cm, CH2Cl2/MeOH 99 :1, then 97:3):
20 (2.25 g, 75%). Slightly yellow foam. TLC (CH2Cl2/MeOH 95 :5): Rf 0.38. UV (MeOH): 277
1
(17000). H-NMR ((D6)DMSO): 2.04, 2.08 (2s, 2 MeCO); 2.40 (m, Haꢀ(2’)); 2.73 (m, Hbꢀ(2’)); 3.23
(t, J=6.32, 6.42, CH2); 4.22 (m, HꢀC(4’), CH2(5’)); 4.65 (t, J=6.28, 6.40, OCH2); 5.24 (d, J=5.20, Hꢀ
C(3’)); 6.39 (m, HꢀC(1’), HꢀC(5)); 7.23 (d, J=2.82, HꢀC(6)); 7.60 (d, J=8.62, arom. H); 8.16 (d,
J=6.90, arom. H); 11.43 (s, NH). Anal. calc. for C23ACHTREUNGH24CAHTREUNNG 4ACHTRENUOG 9 ACHTRE(UNG 500.46): C 55.20, H 4.83, N 11.20; found:
C 55.38, H 4.85, N 11.15.
7-(3,5-Di-O-acetyl-2-deoxy-b-D-erythro-pentofuranosyl)-2,4-bis-O-[2-(4-nitrophenyl)ethyl]-7H-pyr-
rolo[2,3-d]pyrimidine (21). As described for 19 (Method B), with 20 (1.6 g, 3.19 mmol), anh. THF (50
ml), PPh3 (3.4 g, 13.0 mmol), DEAD (2.26 g, 13.0 mmol), and 2-(4-nitrophenyl)ethanol (2.17 g, 13.0
O): Rf 0.5. UV (MeOH): 272 (32400). 1H-NMR
mmol): 21 (1.7 g, 82%). Yellow foam. TLC (Et2ACHTREUNG
((D6)DMSO): 2.01, 2.08 (2s, 2 MeCO); 2.41 (m, Haꢀ(2’)); 2.88 (m, Hbꢀ(2’)); 3.22 (m, 2 CH2); 4.20
(m, HꢀC(4’), CH2(5’)); 4.58 (t, J=6.27, 6.25, OCH2); 4.67 (t, J=6.14, 6.42, OCH2); 5.31 (d, J=5.45,
HꢀC(3’)); 6.43 (m, HꢀC(1’), HꢀC(5)); 7.35 (d, J=3.67, HꢀC(6)); 7.59 (m, arom. H); 8.16 (m, arom.
H). Anal. calc. for C31CAHTRUENGH31ACHTRUENGN5ACHTREUGON 11 ACTHER(UNG 649.60): C 57.32, H 4.81, N 10.78; found: C 57.40, H 4.90, N 10.66.
7-(2-Deoxy-b-D-erythro-pentofuranosyl)-2,4-bis-O-[2-(4-nitrophenyl)ethyl]-7H-pyrrolo[2,3-d]pyri-
midine (22). Compound 21 (1.5 g, 2.31 mmol) was dissolved in sat. NH3/MeOH and stirred for 3 h at r.t.
The solvent was evaporated and the residue subjected to FC (silica gel, 5”12 cm, CH2Cl2/MeOH 95 :5):
CO, 8 :2): Rf 0.45. UV (MeOH): 272 (26400). 1H-NMR
22 (1.19 g, 91%). Yellow foam. TLC (CH2Cl2/Me2ACHTREUNG
((D6)DMSO): 2.14 (m, Haꢀ(2’)); 2.44 (m, Hbꢀ(2’)); 3.21 (m, 2 CH2); 3.49 (m, CH2(5’)); 3.80 (m, Hꢀ
C(4’)); 4.33 (m, HꢀC(3’)); 4.57 (t, J=6.27, 6.18, OCH2); 4.66 (t, J=6.11, 6.18, OCH2); 4.92 (t, J=4.74,
5.12, OHꢀC(5’)); 5.29 (d, J=3.07, OHꢀC(3’)); 6.39 (d, J=2.66, HꢀC(5)); 6.46 (t, J=7.04, 6.60, Hꢀ
C(1’)); 7.38 (d, J=2.90, HꢀC(6)); 7.60 (m, arom. H); 8.15 (m, arom. H). Anal. calc. for C
27ACHTREUNGH27AHCERTUNGN5ACHTREUONG 9
ACHTREUNG(565.53): C 57.34, H 4.81, N 12.38; found: C 57.40, H 4.90, N 12.27.

Helvetica Chimica Acta – Vol. 89 (2006)
2813
7-[2-Deoxy-5-O-(4,4’-dimethoxytrityl)-b-D-erythro-pentofuranosyl]-2,4-bis-O-[2-(4-nitrophenyl)-
ACHTREUNG
on silica gel (10 g), and subjected to FC (silica gel, 3”12 cm, CH2Cl2/Me2CAHTRECUGN O 95 :5): 23 (243 mg, 79%).
Slightly yellow foam. TLC (CH2Cl2/MeOH 95 :5): Rf 0.75. UV (MeOH): 272 (34100). 1H-NMR
((D6)DMSO): 2.26 (m, Haꢀ(2’)); 2.56 (m, Hbꢀ(2’)); 3.12–3.22 (m, 2 CH2, CH2(5’)); 3.69 (s, 2 MeO);
3.92 (m, HꢀC(4’)); 4.37 (m, HꢀC(3’)); 4.52 (m, OCH2); 4.66 (t, J=6.16, 6.41, OCH2); 5.36 (d, J=4.02,
OHꢀC(3’)); 6.37 (d, J=3.61, HꢀC(5)); 6.46 (t, J=6.46, 6.77, HꢀC(1’)); 6.76–6.82 (m, arom. H);
7.18–7.35 (m, HꢀC(6), arom. H); 7.56 (m, arom. H); 8.13 (m, arom. H). Anal. calc. for C
48ACHTREUNHG 45AHCERTUNNG 5ACHTREUOGN 11
ACHTREUNG
ACHTREUNG
CTHREUNG
CTHREUNG
77%). Colorless foam. TLC (silica gel, CH2Cl2/Me
2ACHTREUNG
149.96.
We thank Mr. T. Wiglenda for his valuable contribution, Dr. P. Leonard for helpful comments, and
Ms. S. Budow and Mrs. P. Chittepu for measuring the NMR spectra. Financial support by Roche Diagnos-
tics GmbH (Penzberg, Germany) is gratefully acknowledged.
REFERENCES
[1] T. Suzuki, R. Yamaoka, M. Nishi, H. Ide, K. Makino, J. Am. Chem. Soc. 1996, 118, 2515.
[2] M. Friedkin, J. Am. Chem. Soc. 1952, 74, 112.
[3] V. Vongchampa, M. Dong, L. Gingipalli, P. Dedon, Nucleic Acids Res. 2003, 31, 1045.
[4] G. E. Wuenschell, T. R. OꢁConnor, J. Termini, Biochemistry 2003, 42, 3608.
[5] C. M. D. S. Stefano, E. Magri, Enzymologia 1972, 43, 423; J. Holguin-Hueso, R. Cardinaud, FEBS
Lett. 1972, 20, 171.
[6] R. C. Moschel, L. K. Keefer, Tetrahedron Lett. 1989, 30, 1467.
[7] F. Seela, Y. Chen, M. Sauer, Nucleosides Nucleotides 1998, 17, 39.
[8] F. Seela, H. Driller, U. Liman, Liebigs Ann. Chem. 1985, 312.
[9] R. Eritja, D. M. Horowitz, P. A. Walker, J. P. Ziehler-Martin, M. S. Boosalis, M. F. Goodman, K. Ita-
kura, B. E. Kaplan, Nucleic Acids Res. 1986, 14, 8135.
[10] S. C. Jurczyk, J. Horlacher, K. G. Devined, S. A. Benner, T. R. Battersby, Helv. Chim. Acta 2000, 83,
1517.
[11] Y. Ueno, A. Shibata, A. Matsuda, Y. Kitade, Bioorg. Med. Chem. 2004, 12, 6581.
[12] J. F. Milligan, S. H. Krawczyk, S. Wadwani, M. D. Matteucci, Nucleic Acids Res. 1993, 21, 327.
[13] K. B. Roy, J. Frazier, H. T. Miles, Biopolymers 1979, 18, 3077.
[14] M. J. Lutz, H. A. Held, M. Hottiger, U. Hübscher, S. A. Benner, Nucleic Acids Res. 1996, 24, 1308.
[15] H. Rosemeyer, F. Seela, J. Chem. Soc., Perkin Trans. 2 2002, 746.
[16] F. Seela, C. Wei, Z. Kazimierczuk, Helv. Chim. Acta 1995, 78, 1843; F. Seela, C. Wei, Helv. Chim.
Acta 1997, 80, 73.
[17] Y. Hayakawa, M. Hirose, R. Noyori, J. Org. Chem. 1993, 58, 5551.
[18] N. Spinelli, A. Meyer, Y. Hayakawa, J.-L. Imbach, J.-J. Vasseur, Eur. J. Org. Chem. 2002, 49.
[19] K. Groebke, J. Hunziker, W. Fraser, L. Peng, U. Diederichsen, K. Zimmermann, A. Holzner, C.
Leumann, A. Eschenmoser, Helv. Chim. Acta 1998, 81, 375.
[20] Z. Kazimierczuk, H. B. Cottam, G. R. Revankar, R. K. Robins, J. Am. Chem. Soc. 1984, 106, 6379.
[21] H. Li, X. Peng, P. Leonard, F. Seela, Bioorg. Med. Chem. 2006, 14, 4089.
[22] F. Seela, C. Wei, Helv. Chim. Acta 1999, 82, 726.

2814
Helvetica Chimica Acta – Vol. 89 (2006)
[23] F. Seela, U. Kretschmer, J. Heterocycl. Chem. 1990, 27, 479.
[24] F. Seela, X. Peng, H. Li, P. Chittepu, K. I. Shaikh, J. He, Y. He, I. Mikhailopulo, Collection Sympo-
sium Series 2005, 7, 1.
[25] F. Himmelsbach, B. S. Schulz, T. Trichtinger, R. Charubala, W. Pfleiderer, Tetrahedron 1984, 40, 59.
[26] A. Van Aerschot, M. Mag, P. Herdewijn, H. Vanderhaeghe, Nucleosides Nucleotides 1989, 8, 159.
[27] H.-D. Winkeler, F. Seela, J. Org. Chem. 1983, 48, 3119.
[28] F. Seela, W. Bussmann, Nucleosides Nucleotides 1985, 4, 391.
[29] F. Seela, G. Becher, Nucleic Acids Res. 2001, 29, 2069.
[30] K. Schneider, B. T. Chait, Nucleic Acids Res. 1995, 23, 1570; C. W. Siegert, A. Jacob, H. Kçster, Anal.
Biochem. 1996, 243, 55.
[31] D. Loakes, D. M. Brown, S. Linde, F. Hill, Nucleic Acids Res. 1995, 23, 2361.
[32] F. Seela, A. Jawalekar, Helv. Chim. Acta 2002, 85, 1857.
[33] F. Seela, C. Mittelbach, Nucleosides Nucleotides 1999, 18, 425.
[34] E. Kulikowska, B. Kierdaszuk, D. Shugar, Acta Biochim. Pol. 2004, 51, 493.
[35] F. Seela, T. Wiglenda, H. Rosemeyer, H. Eickmeier, H. Reuter, Angew. Chem., Int. Ed. 2002, 41, 603.
[36] F. Seela, K. Shaikh, Helv. Chim. Acta 2004, 87, 1325.
[37] Y. Hayakawa, H. Kato, M. Uchiyama, H. Kajino, R. Noyori, J. Org. Chem. 1986, 51, 2400.
Received June 22, 2006