Indole in DNA: Comparison of a nucleosidic with a non-nucleosidic DNA base substitution

Article abstract of DOI:10.1002/ejoc.200800863

The synthetic incorporation of indole as an artificial DNA base into oligonucleotides by two different structural approaches is described. For both types of modification, the indole moiety is attached through the C-3 position to the oligonucleotides. As a

Full text of DOI:10.1002/ejoc.200800863

FULL PAPER
DOI: 10.1002/ejoc.200800863
Indole in DNA: Comparison of a Nucleosidic with a Non-Nucleosidic DNA
Base Substitution
Janez Barbaric,[a] Claudia Wanninger-Weiß,[a] and Hans-Achim Wagenknecht*[a]
Keywords: Nitrogen heterocycles / DNA / Oligonucleotides / Nucleosides
The synthetic incorporation of indole as an artificial DNA
base into oligonucleotides by two different structural ap-
proaches is described. For both types of modification, the in-
dole moiety is attached through the C-3 position to the oligo-
nucleotides. As a mimic of natural nucleosides, the indole
nucleoside of β-2Ј-deoxyribofuranoside (In) was synthesized.
The corresponding In-modified duplexes were compared
with duplexes that contained the indole group connected
through (S)-3-amino-1,2-propanediol as an acyclic linker be-
tween the phosphodiester bridges of the oligonucleotides.
This linker was tethered to the C-3 position of the indole het-
erocycle either directly (InЈЈ) or by a carbamate function (InЈ).
The melting temperatures of the corresponding indole-modi-
fied DNA duplexes were measured and compared. Interest-
ingly, not only the InЈ and InЈЈ modifications but also the nat-
ural-like In base surrogate destabilize the DNA duplex
strongly. This result supports our approach to apply the acy-
clic glycol linker to incorporate aromatic molecules as artifi-
cial DNA base substitutions. The major advantage of acyclic
glycol linkers [such as the applied (S)-3-amino-1,2-propane-
diol] is that the corresponding modifications are synthetically
more easily and readily accessible, as it avoids the prepara-
tion of the nucleosidic bond and the separation and purifica-
tion of the α- and β-anomers.
(© Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
Germany, 2009)
trap for charge transport studies in DNA owing to its low
oxidation potential and the characteristic transient absorp-
tion of the corresponding radical or radical cation. Those
charge transfer experiments have been performed with in-
dole either as part of tryptophan in short DNA-binding
peptides, mainly Lys-Trp-Lys,[17,18] or DNA-binding pro-
teins.[19] Experiments with indole connected through the N-
1 position as an artificial DNA base[20,21] revealed that the
aromatic N–H group of indole is crucial for an efficient
trapping of positive charges. The indole radical cation is
very acidic (pKa ≈ 4) and couples the charge transfer with a
deprotonation step.[22] Hence, the C-nucleoside with indole
attached either by the C-3 or the C-2 position of the hetero-
cycle is more desirable for charge transfer studies than the
corresponding N-1 nucleoside.
Excluding the benzene part of indole, there are in prin-
ciple three alternative positions for the attachment of the
heterocycle to DNA: (i) The synthesis of the β-nucleoside
of indole attached through the nitrogen in the 1-position to
the anomeric centre of 2Ј-deoxyribofuranoside or ribofur-
anoside has been published several times.[6–9] The resulting
artificial DNA base as part of oligonucleotides or nucleo-
tide triphosphates has been used in DNA base pairing and
stacking experiments and to probe DNA polymerase-cata-
lyzed primer extension.[9–11] (ii) The synthesis of the C-nu-
cleoside bearing the indole heterocycle attached through its
C-2 position is found rarely in the literature.[12–14] This nu-
cleoside has been used mainly as an unnatural base pair
bearing the indole N–H group as a specific minor-groove
Introduction
The DNA helix is a supramolecular assembly that con-
sists of a π-stacked one-dimensional array of four aromatic
heterocycles that are attached as nucleosides to 2Ј-deoxyri-
bofuranosides and held in place by two antiparallel phos-
phodiester backbones. The four natural DNA bases ade-
nine, cytosine, guanine and thymine vary only slightly in
polarity or stacking ability in comparison to the structural
diversity of organic molecules that potentially could be in-
corporated into oligonucleotides as substitutions or surro-
gates for DNA bases. Simple aromatic hydrocarbons and
aromatic heterocycles were among the first replacements for
DNA bases without hydrogen-bonding capabilities.[1–4]
Universal base analogues are molecules that can pair
equally well with all four natural DNA bases,[1–4] like, for
instance, 5-nitroindole.[5] As a result of the structural simi-
larity with the two natural purine bases, the unsubstituted
indole represents an important artificial DNA base that iso-
sterically substitutes guanine or adenine in DNA.[6–16] In-
dole-containing nucleosides and oligonucleotides are also
target molecules for bioanalytical and medicinal applica-
tions.[6–16]
Beside the significance for chemical biology and bioor-
ganic chemistry, indole represents a very promising charge
[a] Institute for Organic Chemistry, University of Regensburg,
Universitätsstr. 31, 93053 Regensburg, Germany
Fax: +49-941-943-4617
E-mail: achim.wagenknecht@chemie.uni-regensburg.de
364
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2009, 364–370

Comparison of a Nucleosidic with a Non-Nucleosidic DNA Base Substitution
hydrogen-bond donor.[12] (iii) The synthesis of the C-nucleo- tre of 1 and the C-3 position of 1-phenylsulfonylindole (2).
side attached through the C-3 position of the indole hetero- This reaction was performed through Lewis acid promoted
cycle is known, as this position exhibits the highest reactiv- electrophilic substitution at –15 °C. In this case, the Lewis
ity towards electrophilic aromatic substitution.[8,15,16] How- acid BF3·OEt2 catalyzes the formation[16] of para-toluoyl-
ever, the resulting indole C-nucleoside has not yet been in- protected C-nucleoside 3α/3β in 35% isolated yield. The an-
corporated into oligonucleotides.
omeric mixture 3α/3β was separated by flash chromatog-
Recently, we chose a different approach in order to study raphy to afford pure 3β in 26% yield and pure 3α in 9%
1
the stacking interactions of indole by incorporation as an yield. The assignment of the two anomers is based on H
artificial DNA base at specific sites in duplex DNA.[23] The NMR spectroscopy. The multiplicity of the H-1Ј signal (3β)
2Ј-deoxyribofuranoside moiety was replaced by (S)-3- shows a doublet of doublets (J = 5.2 Hz, J = 10.4 Hz) at δ
amino-1,2-propanediol as an acyclic linker that is tethered = 5.43 ppm and is typical for the β-isomer.[1] The NOE
to the C-3 position of the indole heterocycle. We used a studies of compound 3β show interactions between H-2 and
similar synthetic approach for the incorporation of ethid- H-4 of the indole moiety and H-1Ј and H-2αЈ. Thus, these
ium,[24] perylene bisimide,[25] phenothiazine[26] and thiazole contacts do not provide clear evidence for the β-anomer.
orange[27] as DNA base surrogates. Avoiding the labile gly- However, the strong NOE contacts between H-1Ј and
cosidic bond, this approach provides the necessary chemical H-2α Ј and H-1Ј and H-4Ј in the sugar part could only be
stability for the preparation of DNA modifications by auto- explained by the existence of H-1αЈ that is part of β-anomer
mated phosphoramidite chemistry.[28] Similar propanediol 3β. The stereoselectivity of the C–C bond formation de-
linkers have been used by others to prepare glycol nucleic pends on the temperature. Because 3β is the thermodynami-
acid (GNA),[29] twisted intercalating nucleic acids cally controlled reaction product, high temperatures yield
(TINA)[30] and alkyne-modified oligonucleotides for the more β-product (at –15 °C 3β/3α = 4:1). The deprotection
click-Huisgen cycloaddition as a postsynthetic modifica- of the 3Ј- and 5Ј-OH functions of 3β was carried out by
tion.[31]
standard treatment with NaOMe in dry MeOH[32] to give
4 in 83% yield. Nucleoside 4 was deprotected with KOH in
the presence of 18-crown-6 to the completely unprotected
indole nucleoside 5. The low yield of the last step was due
to the decomposition of the product that was observed un-
der the strongly basic conditions. Furthermore, the follow-
ing tritylation of the 5Ј-OH-function of 5 could not be car-
ried out in high yields. Hence, for the later synthetic incor-
poration into oligonucleotides it turned out to be useful
to first introduce the DMT-protecting group to the 5Ј-OH
function of 4 and subsequently desulfonylate the NH func-
tion of 6. By using this synthetic route, 7 could be obtained
in 55% yield for both steps. The synthesis of In-phos-
phoramidite 8 was finished by using standard pro-
cedures.[23]
For the non-nucleosidic InЈ and InЈЈ modifications the S
configuration of the acyclic linker was chosen to mimic the
stereochemical situation at the 3Ј-position of natural 2Ј-de-
oxyribofuranosides. We showed recently that the synthesis
of the InЈЈ-DNA building block could be accomplished in
a three-step route.[23] However, the secondary amine func-
tion of the linker system in the InЈЈ-type conjugate needs to
be protected with a trifluoroacetyl group to avoid irrevers-
ible acetylation at this position during the capping steps of
the DNA synthesizer cycle. Especially, this reaction turned
out to be a real bottleneck of the synthetic route, as it was
difficult to remove the trifluoroacetate from the hydroxy
function of the linker molecule while keeping it at the amine
functionality. To prevent this protection step and to there-
fore simplify the synthesis we chose a carbamate function
to link the indole moiety and the acyclic linker for InЈ-DNA
Results and Discussion
Herein, we present the synthetic routes for two different
structural approaches to incorporate indole as an artificial
DNA base into oligonucleotides through attachment to the
C-3 position of the heterocycle. The natural-like C-nucleo-
side (In) containing the indole moiety tethered to the β-
2Ј-deoxyfuranoside is compared with the non-nucleosidic
alternative bearing (S)-3-amino-1,2-propanediol as an acy-
clic linker between the phosphodiester bridges (InЈ)
(Scheme 1). In contrast to our previously published InЈЈ,[23]
the indole group of InЈ is tethered to (S)-3-amino-1,2-pro-
panediol through a carbamate function that facilitates the
synthesis of the corresponding DNA building block, be-
cause protection of the N–H group of the linker part is no
longer necessary (Scheme 1).
Scheme 1. Structural variations to incorporate indole as an artifi-
cial base into oligonucleotides.
The precursor for the synthesis of In-DNA building building block 12 (Scheme 3). Because of the low basicity of
block 8 is 1-O-methyl-3,5-di-O-toluoyl-2-desoxyribose (1), the NH of the carbamate group, protection is not necessary
which can be synthesized according to the literature during the oligonucleotide synthesis. DMT-protected (S)-3-
(Scheme 2).[33] The key step of the synthetic route to 8 rep- amino-1,2-propanediol 9 as the precursor was synthesized
resents the C–C bond formation between the anomeric cen- according to the literature.[24] The hydroxy function of com-
Eur. J. Org. Chem. 2009, 364–370
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
365

J. Barbaric, C. Wanninger-Weiß, H.-A. Wagenknecht
FULL PAPER
Scheme 3. Synthesis of InЈ-phosphoramidite 12. Reagents and con-
ditions: (a) CDI (1.1 equiv.), DMF, room temp., 3 d, 69%; (b) 2-
cyanoethyl-N,N-diisopropylchlorophosphoramidite
(1 equiv.),
EtN(iPr)2 (3.6. equiv.), CH2Cl2, room temp., 30 min, 96%.
compared the In-modified and InЈ-modified duplexes with
the previously published InЈЈ-modified ones (DNA5,
DNA6) and the completely unmodified duplexes DNA7
and DNA8 bearing a G instead of any indole-containing
artificial base pair.
Scheme 2. Synthesis of In-phosphoramidite 8. Reagents and condi-
tions: (a) BF3·OEt2 (6.9 equiv.), CH2Cl2, –15 °C, 1.5 h, 9% (3α),
26% (3β); (b) NaOMe (3 equiv.), MeOH, 6 h, room temp., 83%;
(c) 18-crown-6 (1.5 equiv.), KOH (33 equiv.), MeOH, 1,4-dioxane,
2 h, room temp., 15%; (d) DMTCl (1.3 equiv.), NEt3 (3 equiv.),
pyridine, room temp., 16 h, 69%; (e) 18-crown-6 (1.5 equiv.), KOH
(33 equiv.), MeOH, 1,4-dioxane, room temp., 16 h, 80%; (f) 2-cyan-
oethyl-N,N-diisopropylchlorophosphoramidite (1.5 equiv.), NEt3
(3 equiv.), CH2Cl2, room temp., 45 min, 95%.
mercially available 3-(2-hydroxyethyl)indole (10) was con-
verted into an activated ester by 1,1Ј-carbonyldiimidazole
and subsequent nucleophilic acyl substitution with 9 gave
conjugate 11 in 69% yield. The synthesis of phosphoramid-
ite 12 was accomplished by standard procedures.
All In- and InЈ-modified oligonucleotides presented in
this manuscript were prepared by automated solid-phase
synthesis on an Expedite 8909 DNA synthesizer. Near-
quantitative coupling of the indole building blocks was
achieved only when the coupling time was extended to
15 min for phosporamidite 8 and 30 min for phosphoram-
idite 12. The HPLC-purified oligonucleotides were iden-
tified by MS (ESI) and quantified by their UV/Vis absorp-
Scheme 4. Sequences of In-modified DNA1a–DNA2d, InЈ-modi-
fied DNA3a–DNA4d, InЈЈ-modified DNA5a–DNA6d[23] and un-
modified duplexes DNA7[23] and DNA8.[23]
tion by using ε260 = 3900 –1 cm–1 (In) and ε260
=
4000 –1 cm–1 (InЈ). We prepared two sets of duplexes for
To study the stacking situation and to evaluate the influ-
each modification that contained two sequential variations ence of the indole moieties we determined the thermal sta-
to study the In and InЈ modifications (Scheme 4): (i) The bility of all modified duplexes by measuring the melting
bases directly adjacent on each side of the modification site temperatures (Tm) at 260 nm and compared them with Tm
were either A (DNA1/DNA3) or T (DNA2/DNA4), as rep- values of the corresponding unmodified duplexes DNA7
resentatives for purines and pyrimidines. (ii) The and DNA8 (Table 1). Three major results can be drawn
counterbase opposite to the In or InЈ modification was from this data: (i) The melting temperature of all modified
either C, T, A or G (DNAXa–DNAXd). Additionally, we duplexes lie in a rather narrow range of 51–55 °C. (ii) All
366
www.eurjoc.org
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2009, 364–370

Comparison of a Nucleosidic with a Non-Nucleosidic DNA Base Substitution
modified duplexes show a strong destabilization by 11– indole moiety probably points into the major groove.
15 °C compared to the unmodified duplexes. (iii) None of Hence, the destabilization effect of the In-modified duplexes
the indole modifications exhibit any preferred pairing with has to be assigned to the lack of hydrogen-bonding interac-
one special counterbase in the complementary strand.
tions of the artificial DNA base In. This is an important
observation, because it rules out that the glycol linker,
Table 1. Melting temperatures (Tm)[a] of In-modified DNA1a– which contains one carbon atom less between the phospho-
DNA2d, InЈ-modified DNA3a–DNA4d, InЈЈ-modified DNA5a–
diester bridges (relative to 2Ј-deoxyribofuranoside), is the
single cause for destabilization of the duplex.
DNA6d[23] and unmodified duplexes DNA7[23] and DNA8.[23]
X
Y
With respect to the potential applicability of the In base
as a universal base analogue it is important to measure the
influence of the counterbase to the indole modification site
(In). In fact, all three indole modifications (In, InЈ and InЈЈ)
do not exhibit any preference in base-pairing selectivity ac-
cording to the melting temperatures. Hence, they all can be
considered as universal base analogues. For the duplexes
DNAXa and DNAXb this result was expected, because in-
dole represents an isosteric substitution for purines. How-
ever, for the duplexes DNAXc and DNAXb bearing purines
as counterbases a further destabilization was expected.
C Duplex
Tm [°C]
T Duplex
Tm [°C]
A Duplex
Tm [°C]
G Duplex
Tm [°C]
In
DNA1a
51.0
DNA2a
52.0
DNA1b
53.0
DNA2b
52.5
DNA1c
54.0
DNA2c
53.0
DNA1d
54.0
DNA2d
51.8
InЈ
InЈЈ
G
DNA3a
55.0
DNA4a
53.0
DNA3b
54.7
DNA4b
53.2
DNA3c
54.5
DNA4c
53.2
DNA3d
54.5
DNA4d
52.7
DNA5a[23]
52.0
DNA5b[23]
54.3
DNA5c[23]
54.2
DNA5d[23]
53.8
DNA6a[23]
52.9
DNA6b[23]
51.8
DNA6c[23]
53.5
DNA6d[23]
54.0
Conclusions
We showed the syntheses of two structurally different in-
corporations of indole as an artificial DNA base into oligo-
nucleotides. For both types, the indole moiety was attached
through the C-3 position to the oligonucleotides. As a
mimic of natural nucleosides we used the indole nucleoside
of β-2Ј-deoxyribofuranoside (In). The corresponding In-
modified duplexes were compared with duplexes that con-
nected the indole heterocycle with the use of (S)-3-amino-
1,2-propanediol as an acyclic linker between the phosphodi-
ester bridges of the oligonucleotides. This linker was teth-
ered to indole either directly (InЈЈ)[23] or through a carb-
amate function (InЈ). We measured and compared the melt-
ing temperatures of the corresponding modified DNA du-
plexes. The acyclic indole modifications InЈ and InЈЈ
strongly decrease the duplex stability. Surprisingly, the same
effect occurs by using the natural-like In nucleoside. Ac-
cording to these results it is obvious that the destabilizing
effect does not depend on the modality of the chromophore
attachment to the oligonucleotide. Consequently, in the fu-
ture the acyclic linker can preferably be chosen for the in-
corporation of artificial DNA bases, for example, fluoro-
phores. This is advantageous, because the corresponding
DNA building blocks are synthetically more easily and
readily accessible, as the preparation of the nucleosidic
bond and the separation and purification of the α- and β-
anomers can be avoided.
DNA7[23]
65.0
[23]
DNA8
65.8
[a] Conditions: λ = 260 nm, 10–90 °C, interval: 0.7 °Cmin–1, 2.5 µ
duplex in 10 m Na-Pi buffer, 250 m NaCl, pH 7.
For the InЈ (DNA3a–DNA4d) and InЈЈ modification
(DNA5a–DNA6d), the destabilizing effect is not surprising,
as the glycol linker substitutes the 2Ј-deoxyribofuranoside.
The major difference in this linker relative to the natural 2Ј-
deoxyribofuranoside is the number of carbon atoms be-
tween the phosphodiester bridges, which has been reduced
from three (in normal nucleosides) to two (in InЈ and InЈЈ).
The studies of glycol nucleic acids (GNA) by Meggers and
coworkers revealed also that a single glycol modification
typically destabilizes the duplex significantly.[29] According
to our experience, this depends on the size of the hydro-
phobic surface of the artificial DNA base surrogate if some
of the lost duplex stability that is caused by the glycol linker
can be regained by stacking interactions. For instance,
DNA duplexes that have been modified by perylene bis-
imide chromophores attached through the (S)-3-amino-1,2-
propanediol linker between the phosphodiester bridges
showed no significant destabilization.[25] This result was at-
tributed mainly to the hydrophobic interactions of those
chromophores.
More surprising is the observation that the duplexes with
the In modification (DNA1a–DNA2d) show a similarly
strong destabilization relative to the duplexes with the InЈ
or InЈЈ modifications. The nucleosidic In modification with
β-configuration at the anomeric centre resembles the struc-
ture of natural nucleosides best possible. It was designed as
an isosterical surrogate for purine DNA bases. In the anti
Experimental Section
Materials and Methods: 1H, 13C, 31P and 2D NMR spectra were
recorded with a Bruker Avance 300 or Avance 400 NMR spectrom-
eter. ESI and FAB mass spectra were measured in the analytical
facility of the institute with a Finnigan TSQ 7000 or Finnigan
MAT 95. HRMS (ESI) was measured by Coring System Diagnos-
conformation of the In nucleosides, the NH group of the tix GmbH, Gernsheim. Analytical chromatography was performed
Eur. J. Org. Chem. 2009, 364–370
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
367

J. Barbaric, C. Wanninger-Weiß, H.-A. Wagenknecht
FULL PAPER
on Merck silica gel 60 F254 plates. Flash chromatography was per-
formed on Merck silica gel (40–63 µm). Solvents were dried accord-
ing to standard procedures. All reactions were carried out under a
nitrogen atmosphere and protected from light. Reagents and chemi-
cals were obtained from Alfa Aesar, Fluka and Lancaster and were
used without further purification. The oligonucleotides were pre-
pared with an Expedite 8909 DNA synthesizer (ABI) by using CPG
(1 µmol) and chemicals from ABI and Proligo. The trityl-off oligo-
nucleotides were cleaved and deprotected by treatment with con-
centrated NH4OH at 60 °C for 10 h, dried and purified by HPLC
by using the following conditions: A = NH4OAc buffer (50 m),
pH 6.5; B = MeCN; gradient 0–20% over 70 min for In-modified
oligonucleotides and 5–15% over 55 min of InЈ-modified oligonu-
cleotides. C18-RP analytical and semipreparative HPLC columns
(300 Å) were purchased from Supelco. Duplexes were formed by
heating to 90 °C (10 min), followed by slow cooling. Melting tem-
perature measurements were performed with a Cary 100 (Varian)
with duplex samples (2.5 µ) in quartz glass cuvettes (1 cm) and
in Na-Pi buffer (10 m), NaCl (250 m), pH 7, 260 nm, interval
0.7 °Cmin–1.
CDCl3): δ = 7.99–7.97 (d, J = 8.2 Hz, 1 H, ArInd-H, H7), 7.97–
7.87 (m, 2 H, SO2-Ph, 2-H and H6), 7.6–7.57 (d, J = 7.8 Hz, ArInd
-
H, H4), 7.56–7.51 (m, 1 H, SO2-Ph, H3), 7.53 (s, ArInd-H, H2),
7.46–7.41 (m, 2 H, SO2-Ph, H2 and H4), 7.36–7.30 (dt, J = 1.7,
7.2 Hz, 1 H, ArInd-H, H6), 7.26–7.21 (dt, J = 0.8, 7.5 Hz, 1 H,
ArInd-H, H5), 5.36 (dd, J = 9, 6 Hz, 1 H, H1Ј), 4.5 (m, 1 H, H3Ј),
4.02 (dt, J = 3.4, 4.3 Hz, 1 H, H4Ј), 3.78 (m, 2 H, H5Ј), 2.3–2.23
(m, 2 H, H1Јα, H1Јβ) ppm. 13C NMR (100 MHz, CDCl3): δ =
135.5, 133.8, 130.1, 129.3, 128.9, 126.8, 125.0, 123.4, 122.9, 122.6,
120.3, 113.7, 87.1, 73.7, 73.6, 63.1, 41.8, 30.9, 29.7 ppm. MS (ESI):
m/z (%) = 374.1 (100) [M + H]+, 391.2 (90) [M + NH4]+.
3-(2-Deoxy-β-D-ribosyl)indole (5): A mixture of 4 (1.0 g,
1.65 mmol), 18-crown-6 (646 mg, 2.47 mmol), KOH (2.0 g,
35.8 mmol), dry MeOH (5 mL) and dry 1,4-dioxane (5 mL) was
stirred for 2 h at room temperature. The resulting mixture was ex-
tracted with CH2Cl2, dried with Na2SO4 and concentrated under
vacuum. Flash chromatography (Et2O/EtOAc, 2:1 to 1:3) yielded 5
1
(60 mg, 15%). H NMR (300 MHz, [D6]DMSO): δ = 10.91 (br. s,
1 H, NH), 7.56 (d, J = 7.9 Hz, Ar-H, H7), 7.33 (d, J = 8.1 Hz, 1
H, Ar-H, H4), 7.22 (d, J = 2.2 Hz, Ar-H, H2), 7.08 (dt, J = 7.9,
1.1 Hz, 1 H, H6), 6.95 (dt, J = 7.9, 0.9 Hz, 1 H, H5), 4.94 (dd, J
= 11.2, 1.6 Hz, 1 H, H1Ј), 4.0 (s, 1 H, H3Ј), 3.5 (m, 1 H, H4Ј),
3.49–3.39 (dq, J = 5, 17.7 Hz, 2 H, H5Ј), 2.15 (dd, J = 9.3, 19.5 Hz,
1 H, H2Јα), 1.95 (m, 1 H, H2Јβ) ppm. 13C NMR (151 MHz, [D6]-
DMSO): δ = 136.6, 125.9, 121.0, 119.5, 118.3, 116.4, 111.4, 72.2,
70.4, 68.6, 67.4, 66.3, 60.3 ppm. MS (ESI): m/z (%) = 234.1 (100)
[M + H]+, 251.1 (50) [M + NH4]+, 484.3 (30) [2M + NH4]+, 489.3
(25) [2M + Na]+.
1-Phenylsulfonyl-3-(2Ј-deoxy-3Ј,5Ј-di-O-para-toluoyl-β-ribofuran-
osyl)indole (3α/3β): To a solution of 1-O-methyl-3,5-di-O-toluoyl-
2-desoxyribose (1; 6.18 g, 16.09 mmol) and 1-phenylsulfonylindole
(2; 5.1 g, 19.32 mmol) in dry CH2Cl2 (15 mL) was added BF3·OEt2
(14 mL, 35.9 mmol) at –15 °C. After stirring for 1.5 h at –15 °C,
the reaction mixture was treated with aqueous NaHCO3 and ex-
tracted with CH2Cl2. The combined organic layer was dried with
Na2SO4 and concentrated under reduced pressure. The residue was
purified by flash chromatography (silica gel; hexanes/EtOAc, 8:1)
to yield 3α (0.83 g, 8.5%) and 3β (2.58 g, 26%) as white foams.
Data for 3α: Rf = 0.41 (hexanes/EtOAc, 3:1). 1H NMR (300 MHz,
[D6]DMSO): δ = 7.94–7.88 (m, 6 H, Ar-H), 7.60 (m, 2 H, Ar-H),
7.50 (m, 1 H, Ar-H), 7.39 (m, 2 H, Ar-H), 7.31–7.21 (m, 4 H, Ar-
H), 7.06 (m, 1 H, Ar-H), 5.61 (m, 1 H, H3Ј), 5.55 (t, J = 5.8 Hz, 1
H, H1Ј), 4.56 (m, 2 H, H5Ј), 4.53 (m, J = 4.5, J = 11.8 Hz, 1 H,
H5Ј), 4.51 (m, 1 H, H4Ј), 2.99 (m, H2Јβ), 2.50 (m, 1 H, H2Јα), 2.39
(s, 3 H, CH3), 2.36 (s, 3 H, CH3) ppm. 13C NMR (75.5 MHz, [D6]-
DMSO): δ = 166.7, 166.4, 144.2, 144.1, 137.8, 133.9, 129.45, 129.4,
129.1, 129.0, 128.8, 126.83, 126.6, 126.5, 124.7, 123.6, 123.2, 123.1,
120.2, 111.4, 82.1, 78.1, 77.7, 77.3, 76.5, 74.3, 64.2, 37.6, 20.6 ppm.
HRMS (ESI): calcd for C35H31NO7S 609.1821 [M + H]+ found
1-Phenylsulfonyl-3-[2Ј-deoxy-5Ј-O-(4,4Ј-dimethoxytrityl)-β-D-ribo-
furanosyl]indole (6): 4,4Ј-Dimethoxy-triphenylmethyl chloride
(411 mg, 1.21 mmol) and dry Et3N (392 µL, 2.79 mmol) was added
to a solution of 4 (350 mg, 0.93 mmol) in dry pyridine (10 mL). The
mixture was stirred overnight at room temperature. Subsequently,
MeOH (5 mL) was added. After 1 h at room temperature, the solu-
tion was concentrated to dryness. The crude product was purified
by flash chromatography (hexanes/EtOAc, 2:1 + 0.1% Et3N) to
yield 6 (440 mg, 69%) as a yellow oil. Rf = 0.5 (hexane/EtOAc,
1
1:1). H NMR (300 MHz, [D4]MeOD): δ = 7.95 (d, J = 8.5 Hz, 1
H, ArInd-H, H7), 7.73–7.66 (m, 4 H, Ar-H), 7.44–7.40 (dd, 2 H,
Ar-H), 7.32–7.25 (m, 10 H, Ar-H), 7.08–7.03 (t, 2 H, Ar-H), 6.70–
6.60 (2dd, J = 6.9, 2.2 Hz, 4 H, 2 ArDMT-H, H3 and H5), 5.29 (t,
J = 8.5 Hz, 1 H, H1Ј), 4.94 (d, J = 4 Hz, 1 H, H3Ј), 4.00 (br. s, 1
H, 3Ј-OH), 3.93 (m, 1 H, H4Ј), 3.69 (s, 3 H, OMe), 3.68 (s, 3 H,
OMe) 3.12–3.03 (m, 2 H, H5Ј), 2.2 (m, 2 H, H2Јα und H2Јβ) ppm.
13C NMR (100 MHz, [D4]MeOD): δ = 158.0, 157.9, 149.6, 144.9,
136.8, 136.0, 135.6, 135.5, 134.7, 134.5, 129.7, 129.6, 128.8, 127.7,
126.6, 126.5, 124.9, 124.0, 123.8, 123.2, 122.6, 121.1, 113.1, 113.1,
113.0, 85.9, 85.3, 73.1, 72.2, 64.3, 54.9, 41.3, 41.2, 29.5 ppm. MS
(ESI): m/z (%) = 734.3 (100) [M + CH3COO]–, 710.2 (25) [M +
Cl–]–.
1
610.1890. Data for 3β: Rf = 0.44 (hexanes/EtOAc, 3:1). H NMR
(600 MHz, [D6]DMSO): δ = 7.96–7.91 (m, 4 H, p-Tol, H2 and H6),
7.96–7.91 (m, 2 H, SO2-Ph, H2 and H6), 7.83 (s, 1 H, ArInd-H2),
7.69–61 (m, 1 H, ArInd-H6), 7.51 (t, J = 8.3 Hz, 1 H, ArInd-H5),
7.34–7.27 (m, 4 H, p-Tol, H3 and H5), 7.34–7.27 (m, 2 H, SO2-Ph,
H3 and H5), 7.06–7.03 (t, J = 7.4 Hz, 1 H, SO2-Ph, H4), 5.61 (d,
J = 5.8 Hz, 1 H, H3Ј), 5.43 (dd, J = 5.2, 10.4 Hz, 1 H, H1Ј), 4.61
(dd, J = 4, 11.8 Hz, 1 H, H5Ј), 4.53 (dd, J = 4.5, 11.8 Hz, 1 H,
H5Ј), 4.46 (m, 1 H, H4Ј), 2.55 (dd, J = 5.5, 13.3 Hz, 1 H, H2Јα),
2.46 (m, 1 H, H2Јβ), 2.39 (s, 3 H, CH3), 2.36 (s, 3 H, CH3) ppm.
13C NMR (151 MHz, [D6]DMSO): δ = 165.5, 165.4, 143.9, 143.8,
136.9, 134.7, 134.6, 129.8, 129.4, 129.3, 129.3, 129.2, 128.4, 126.7,
126.7, 126.6, 125.0, 123.7, 123.1, 122.2, 120.9, 113.2, 82.1, 76.6,
73.8, 64.3, 21.2, 21.1 ppm. HRMS (ESI): calcd. for C35H31NO7S
609.1821 [M + H]+; found 610.1890.
3-[2Ј-Deoxy-5Ј-O-(4,4Ј-dimethoxytrityl)-β-D-ribofuranosyl]indole
(7): A mixture of compound 6 (440 mg, 0.65 mmol), 18-crown-6
(262.5 mg, 0.975 mmol), KOH (1.21 g, 21.5 mmol), dry MeOH
(10 mL) and 1,4-dioxane (10 mL) was stirred overnight at room
temperature. The resulting mixture was evaporated to dryness.
1-Phenylsulfonyl-3-(2Ј-deoxy-β-D-ribofuranosyl)indole (4): NaOMe Flash chromatography (hexane/EtOAc, 1:1) yielded 7 (320 mg,
(183 mg, 3.39 mmol) was added to a solution of 3β (692 mg,
1.13 mmol) in dry MeOH (20 mL). After 6 h stirring at room tem-
perature, the solution was evaporated to dryness. The residue was
purified by flash chromatography (silica gel; toluene/CH2Cl2/
MeOH, 5:5:1 to 1:1:1) to yield 4 (350 mg, 83%) as a yellow powder.
Rf = 0.25 (toluene/CHCl2/MeOH, 5:5:1). 1H NMR (400 MHz,
80%) as a yellow oil. Rf = 0.41 (hexane/EtOAc, 1:1). 1H NMR
(300 MHz, CDCl3): δ = 7.74 (d, 1 H, ArInd-H, H7), 7.62–7.58 (m,
2 H, Ar-H), 7.47–7.19 (m, 14 H, Ar-H), 7.07 (m, 2 H, Ar-H), 6.9–
6.81 (m, 5 H, Ar-H), 5.29 (dd, J = 5.4, 10.2 Hz, 1 H, H1Ј), 4.54
(m, 1 H, H3Ј), 4.1 (m, 1 H, H4Ј), 3.75 (s, 3 H, OMe), 3.74 (s, 3 H,
OMe) 3.12–3.03 (m, 2 H, H5Ј), 2.2 (m, 1 H, H2Јα), 2.05 (m, 1 H,
368
www.eurjoc.org
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2009, 364–370

Comparison of a Nucleosidic with a Non-Nucleosidic DNA Base Substitution
H2Јβ) ppm. 13C NMR (75.5 MHz, CDCl3): δ = 157.9, 157.7, 144.9, [M – 3H]3–. Data for ss-DNA4: MS (ESI–): m/z = 1297.9 [M –
136.8, 137.0, 135.7, 135.5, 134.5, 129.8, 129.6, 128.8, 127.6, 127.5,
127.3, 126.6, 124.3, 123.2, 113.0, 112.7, 85.2, 85.1, 73.0, 68.6, 64.1,
56.5, 54.92, 29.5 ppm. MS (ESI): m/z (%) = 594.4 (100) [M +
CH3COO]–, 534.4 (28) [M – H]–.
4H]4–, 1730.1 [M – 3H]3–.
Acknowledgments
2Ј-Deoxy-5Ј-O-(4,4Ј-dimethoxytrityl)-1-(3-indolyl)-β-D-ribofuranose-
This work was supported by the Deutsche Forschungsgemeinschaft
(Wa 1386/7), the Fonds der Chemischen Industrie and the Univer-
sity of Regensburg.
3Ј-O-[(2-cyanoethyl)-N,N-diisopropyl]phosphoramidite (8): To a
solution of 7 (160 mg, 0.299 mmol) dissolved in dry CH2Cl2
(10 mL) was added Et3N (126.5 µL, 0.9 mmol) and 2-cyanoethyl-
N,NЈ-diisopropylchlorophosphoramidite (106 µL, 0.45 mmol), and
the solution was stirred for 1 h at room temperature. The reaction
was quenched by pouring it into aqueous saturated NaHCO3
(20 mL), and the solution was then extracted with CH2Cl2. The
organic layer was dried (Na2SO4) and evaporated to yield phos-
phoramidite 8 as a yellow foam, which was used directly for the
oligonucleotide synthesis. Rf = 0.85 (hexane/EtOAc, 2:1). 31P NMR
(121.5 MHz, CDCl3): δ = 148.8, 147.9 ppm.
[1] N. C. Chaudhuri, R. X.-F. Ren, E. T. Kool, Synlett 1997, 341–
347.
[2] a) E. T. Kool, Chem. Rev. 1997, 97, 1473–1487; b) E. T. Kool,
Acc. Chem. Res. 2002, 35, 936–943; c) A. T. Krueger, E. T.
Kool, Curr. Opin. Chem. Biol. 2007, 11, 588–594.
[3] A. A. Henry, F. E. Romesberg, Curr. Opin. Chem. Biol. 2003,
7, 727–733.
[4] I. Hirao, Curr. Opin. Chem. Biol. 2006, 10, 622–627.
[5] a) D. Loakes, D. M. Brown, Nucleic Acids Res. 1994, 22, 4039–
4043; b) D. Loakes, F. Hill, D. M. Brown, S. A. Salisbury, J.
Mol. Biol. 1997, 270, 426–435.
2-(1H-Indol-3-yl)ethyl {(S)-3-[Bis(4-methoxyphenyl)phenylmeth-
oxy]-2-hydroxypropyl}carbamate (11): 3-(2-Hydroxyethyl)indole
(10; 80.5 mg, 0.50 mmol) and 1,1Ј-carbonyldiimidazole (89.2 mg,
0.55 mmol) were dissolved in dry DMF (8 mL) and stirred for 2 h
at room temperature. Compound 9[24] (196.5 mg, 0.50 mmol) was
then added. The reaction mixture was stirred at room temperature
for 3 d. The solvent was removed under vacuum. The crude prod-
uct was purified by flash chromatography (EtOAc/CH2Cl2, 1:1 +
0.4% NEt3) to yield 11 (200 mg, 69%) as a colourless solid. Rf =
0.65 (EtOAc/CH2Cl2, 10:3). 1H NMR (400 MHz, [D6]DMSO): δ =
10.82 (br. s, 1 H, NH), 7.53 (d, 1 H, ArInd-H2), 7.39 (psd, 2 H,
ArInd-H), 7.34–7.19 (m, 9 H, Ar-H), 7.05 (pst, 1 H, ArInd-H), 6.79
[6] T. H. Dinh, M.-J. Bayard, J. Igolen, C R. Acad. Sci. Paris 1976,
283, 227–228.
[7] N. S. Girgis, H. B. Cottam, R. L. Robins, J. Heterocycl. Chem.
1988, 25, 361–366.
[8] M. Cornia, G. Casiraghi, L. Zetta, J. Org. Chem. 1991, 56,
5466–5468.
[9] J. S. Lai, E. T. Kool, J. Am. Chem. Soc. 2004, 126, 3040–3041.
[10] S. Moran, R. X.-F. Ren, C. J. Sheils, S. Rumney IV, E. T. Kool,
Nucleic Acids Res. 1996, 24, 2044–2052.
[11] X. Zhang, I. Lee, A. J. Berdis, Org. Biomol. Chem. 2004, 2,
1703–1711.
(pst, 1 H, ArInd-H), 6.86 (m, 4 H, Ar-H), 4.88 (d, 1 H, OH), 4.14 [12] S. Matsuda, A. A. Henry, P. G. Schultz, F. E. Romesberg, J.
Am. Chem. Soc. 2003, 125, 6134–6139.
(t, 2 H, CH2-indole), 3.71 (m, 7 H, 2 OCH3 and H-COH), 3.15 (m,
1 H, COO-CH), 2.92 [m, 5 H, COO-CH, 2-CH2 (linker)] ppm. 13
C
[13] a) D. GuianvarcЈh, R. Benhida, J.-L. Fourrey, Tetrahedron Lett.
2001, 42, 647–650; b) D. GuianvarcЈh, J.-L. Fourrey, M.-
E. T. H. Dau, V. Guérineau, J. Org. Chem. 2002, 67, 3724–
3732.
NMR (100 MHz, [D6]DMSO): δ = 158.0, 156.5, 145.1, 136.12,
135.8, 135.8, 129.7 (Ar-H), 127.7 (Ar-H), 127.7 (Ar-H), 127.1 (Ar-
Ind-H), 126.5, 123.0 (Ar-H), 120.9 (ArInd-H), 118.3 (ArInd-H), 118.2
(ArInd-H2), 113.0 (Ar-H), 111.3 (Ar-H), 110.2, 85.14, 68.8, 65.8,
63.9, 55.0 (OCH3), 44.3, 24.9 ppm. MS (ESI): m/z (%) = 579.4 (100)
[M – H]–, 615.4 (13) [M + Cl]–, 639.5 (35) [M + CH3COO]–. HRMS
(FAB): calcd. for C35H35N2O6 579.2495 [M – H]– found 579.2499.
[14] M. Yokoyama, H. Toyoshima, M. Shimizu, H. Togo, J. Chem.
Soc. Perkin Trans. 1 1997, 29–33.
[15] T. N. Sokolova, I. V. Yartseva, M. N. Preobrazhenskaya, Car-
bohydr. Res. 1981, 93, 19–34.
[16] a) M. Yokoyama, M. Nomura, T. Tanabe, H. Togo, Heteroat.
Chem. 1995, 6, 189–193; b) M. Yokoyama, M. Nomura, H.
Togo, H. Seki, J. Chem. Soc. Perkin Trans. 1 1996, 2145–2149;
c) W. Hei, H. Togo, M. Yokoyama, Tetrahedron Lett. 1997, 38,
5541–5544; d) W. Hei, H. Togo, Y. Waki, M. Yokoyama, J.
Chem. Soc. Perkin Trans. 1 1998, 2425–2433.
2-(1H-Indol-3-yl)ethyl {(S)-3-[Bis(4-methoxyphenyl)phenylmeth-
oxy]-2-[(2-cyanoethoxy)diisopropylaminophosphanyloxy]propyl}-
carbamate (12): To a solution of 11 (104.4 mg, 0.18 mmol) dis-
solved in dry CH2Cl2 (6 mL) was added dry EtN(iPr)2 (90 µL,
[17] a) H.-A. Wagenknecht, E. D. A. Stemp, J. K. Barton, J. Am.
Chem. Soc. 2000, 122, 1–7; b) H.-A. Wagenknecht, E. D. A.
Stemp, J. K. Barton, Biochemistry 2000, 39, 5483–5491.
[18] a) E. Mayer-Enthart, P. Kaden, H.-A. Wagenknecht, Biochem-
istry 2005, 44, 11749–11757; b) L. Valis, E. Mayer-Enthart, H.-
A. Wagenknecht, Bioorg. Med. Chem. Lett. 2006, 16, 3184–
3187.
0.648 mmol)
and
2-cyanoethyl-N,NЈ-diisopropylchlorophos-
phoramidite (40.3 µL, 0.18 mmol), and the mixture was stirred at
room temperature for 30 min. The reaction was quenched with dry
EtOH (100 µL) and quickly washed with freshly prepared aqueous
NaHCO3 solution. The organic phase was dried with Na2SO4, and
the solvent was removed under vacuum to yield 12 (135 mg, 96%)
as a pale brown, viscous liquid. Rf = 0.60 (CH2Cl2/MeOH, 100:3).
31P NMR (121.5 MHz, [D6]DMSO): δ = 149.7, 149.1 ppm.
[19] a) S. R. Rajski, S. Kumar, R. J. Roberts, J. K. Barton, J. Am.
Chem. Soc. 1999, 121, 5615–5616; b) H.-A. Wagenknecht, S. R.
Rajski, M. Pascaly, E. D. A. Stemp, J. K. Barton, J. Am. Chem.
Soc. 2001, 123, 4400–4407; c) S. R. Rajski, J. K. Barton, Bio-
chemistry 2001, 40, 5556–5564.
General Procedure for the Solid-phase Synthesis of the In- and InЈ-
Modified Oligonucleotides: The synthesis were performed on a
1 µmol scale (CPG 500 Å, Proligo) by using standard phosphoram-
idite protocols. Quantitative coupling of 8 was achieved by using a
coupling time of 15 min. The coupling time of 14 was 30 min. After
preparation, the oligonucleotides were quantified by their ab-
sorbance at 260 nm and by using ε (In) = 3900 –1 cm–1, ε (InЈ) =
4000 –1 cm–1. Data for ss-DNA1: MS (ESI–): m/z = 1289.3 [M –
4H]4–. Data for ss-DNA2: MS (ESI–): m/z = 1284.6 [M – 4H]4–.
Data for ss-DNA3: MS (ESI–): m/z = 1300.4 [M – 4H]4–, 1734.1
[20] M. Pascaly, J. Yoo, J. K. Barton, J. Am. Chem. Soc. 2002, 124,
9083–9092.
[21] C. Wanninger-Weiβ, L. Valis, H.-A. Wagenknecht, Bioorg.
Med. Chem. 2008, 16, 100–106.
[22] W. A. Prütz, E. J. Land, Int. J. Radiat. Biol. 1979, 36, 513–520.
[23] C. Wanninger, H.-A. Wagenknecht, Synlett 2006, 2051–2054.
[24] a) R. Huber, N. Amann, H.-A. Wagenknecht, J. Org. Chem.
2004, 69, 744–751; b) N. Amann, R. Huber, H.-A. Wagen-
knecht, Angew. Chem. Int. Ed. 2004, 43, 1845–1847.
Eur. J. Org. Chem. 2009, 364–370
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
369

J. Barbaric, C. Wanninger-Weiß, H.-A. Wagenknecht
FULL PAPER
[25] a) C. Wagner, H.-A. Wagenknecht, Org. Lett. 2006, 8, 4191–
4194; b) D. Baumstark, H.-A. Wagenknecht, Angew. Chem. Int.
Ed. 2008, 47, 2652–2654.
[26] C. Wagner, H.-A. Wagenknecht, Org. Biomol. Chem. 2008, 6,
48–50.
[30] a) V. V. Filichev, U. B. Christensen, E. B. Pedersen, B. R. Babu,
J. Wengel, ChemBioChem 2004, 5, 1673–1679; b) V. V. Filichev,
E. B. Pedersen, J. Am. Chem. Soc. 2005, 127, 14849–14858; c)
I. Geci, V. V. Filichev, E. B. Pedersen, Chem. Eur. J. 2007, 13,
6379–6386.
[27] F. Menacher, M. Rubner, S. Berndl, H.-A. Wagenknecht, J.
Org. Chem. 2008, 73, 4263–4266.
[31] P. M. E. Gierlich, S. Warncke, J. Gierlich, T. Carell, Angew.
Chem. Int. Ed. 2008, 47, 3442–3444.
[28] N. Amann, H.-A. Wagenknecht, Tetrahedron Lett. 2003, 44, [32] S. Hainke, S. Arndt, O. Seitz, Org. Biomol. Chem. 2005, 3,
1685–1690.
4233–4238.
[29] a) L. Zhang, A. Peritz, E. Meggers, J. Am. Chem. Soc. 2005,
127, 4174–4175; b) L. Zhang, A. E. Peritz, P. J. Carroll, E.
Meggers, Synthesis 2006, 645–653; c) M. K. Schlegel, L.-O.
Essen, E. Meggers, J. Am. Chem. Soc. 2008, 130, 8158–8159.
[33] S. A. Nadeem Hashmi, X. Hu, C. E. Immoos, S. J. Lee, M. W.
Grinstaff, Org. Lett. 2002, 4, 4571–4574.
Received: September 7, 2008
Published Online: December 3, 2008
370
www.eurjoc.org
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2009, 364–370