Stability of Hoogsteen-type triplexes - Electrostatic attraction between duplex backbone and Triplex-Forming Oligonucleotide (TFO) using an intercalating conjugate

Article abstract of DOI:10.1002/hlca.200890082

Syntheses are described for two novel twisted intercalating nucleic acid (TINA) monomers where the intercalator comprises a benzene ring linked to a naphthalimide moiety via an ethynediyl bridge. The intercalators Y and Z have a 2-(dimethylamino)ethyl and a methyl residue on the naphthalimide moiety, respectively. When used as triplex-forming oligonucleotides (TFOs), the novel naphthalimide TINAs show extraordinary high thermal stability in Hoogsteen-type triplexes and duplexes with high discrimination of mismatch strands. DNA Strands containing the intercalator Y show higher thermal triplex stability than DNA strands containing the intercalator Z. This observation can be explained by the ionic interaction of the protonated dimethylamino group under physiological conditions, targeting the negatively charged phosphate backbone of the duplex. This interaction leads to an extra binding mode between the TFO and the duplex, in agreement with molecular-modeling studies. We believe that this is the first example of an intercalator linking the TFO to the phosphate backbone of the duplex by an ionic interaction, which is a promising tool to achieve a higher triplex stability.

Full text of DOI:10.1002/hlca.200890082

Helvetica Chimica Acta – Vol. 91 (2008)
805
Stability of Hoogsteen-Type Triplexes – Electrostatic Attraction between
Duplex Backbone and Triplex-Forming Oligonucleotide (TFO) Using an
Intercalating Conjugate
by Daniel Globischa), Niels Bomholta), Vyacheslav V. Filichevb), and Erik B. Pedersen*a)
a) Nucleic Acid Center, Department of Physics and Chemistry, University of Southern Denmark,
Campusvej 55, DK-5230 Odense M
(phone: þ 4565502555; fax: þ 4566158780; e-mail: ebp@ifk.sdu.dk)
b) Institute of Fundamental Sciences, Massey University, Private Bag 11-222, Palmerston North,
New Zealand
Syntheses are described for two novel twisted intercalating nucleic acid (TINA) monomers where
the intercalator comprises a benzene ring linked to a naphthalimide moiety via an ethynediyl bridge. The
intercalators Y and Z have a 2-(dimethylamino)ethyl and a methyl residue on the naphthalimide moiety,
respectively. When used as triplex-forming oligonucleotides (TFOs), the novel naphthalimide TINAs
show extraordinary high thermal stability in Hoogsteen-type triplexes and duplexes with high
discrimination of mismatch strands. DNA Strands containing the intercalator Y show higher thermal
triplex stability than DNA strands containing the intercalator Z. This observation can be explained by the
ionic interaction of the protonated dimethylamino group under physiological conditions, targeting the
negatively charged phosphate backbone of the duplex. This interaction leads to an extra binding mode
between the TFO and the duplex, in agreement with molecular-modeling studies. We believe that this is
the first example of an intercalator linking the TFO to the phosphate backbone of the duplex by an ionic
interaction, which is a promising tool to achieve a higher triplex stability.
Introduction. – Triplex-forming oligonucleotides (TFOs) are of great interest due
to their application in antigene strategy. The gene expression of malfunctioned genes is
blocked in this technology before the transcription step. Contrary to antisense agents,
there are no antigene agents in clinical trials. The advantage of the antigene strategy
should be the inhibition of gene expression at the first step, hence targeting of only two
copies instead of up to thousands of mRNA copies in the antisense approach. For the
success of antigene strategy, it is necessary to enhance stabilization of triplexes by novel
TFOs, due to low stability of natural TFOs, and the limitation of sufficiently long
homopurine/homopyrimidine stretches in the target DNA for triplex formation [1].
These facts and the quadruplex formation of G-rich strands are the major drawback for
the success of this strategy [2]. To increase the stability of triplexes, an enormous
number of chemically modified oligodeoxynucleotides (ONs) have been developed for
their use as TFOs. Improved stabilization can be achieved by either modifying the
nucleobase [3], the sugar part [4], or the phosphate backbone [5]. In some of these
approaches, enhanced triplex stabilities are possible by reducing the unfavorable
charge repulsion between the three negatively charged DNA strands under physio-
logical conditions. Exchange of negatively charged O-atoms in the phosphate backbone
by uncharged heteroatoms led to an increased stabilization of the triplex. In another
ꢀ 2008 Verlag Helvetica Chimica Acta AG, Zürich

806
Helvetica Chimica Acta – Vol. 91 (2008)
method, the repulsion was reduced by an amino side chain, which is protonated under
physiological conditions, forming ionic interactions with the phosphate backbone of the
TFO itself [3c,d][4a,e][5c].
Another method for stabilization of triplexes is the use of intercalating systems, in
which an aromatic system stabilizes the triplex by p – p interaction with the adjacent
nucleobases of the target duplex without disrupting the H-bonds. Pyrene is often used
as an aromatic system for intercalation due to its large p-surface and its fluorescence
properties [6][7]. An intercalating system, called twisted intercalating nucleic acid
(TINA), is named due to a benzene moiety and a large aromatic system, which are
supposed to twist around the CꢀC bond for better intercalating properties [6a,b].
Every part of this intercalator is able to stabilize the triplex by p – p interactions. The
benzene moiety is assumed to stabilize the nucleobases in the TFO, while the pyrene
moiety intercalates with the nucleobases of the duplex. Froehler et al. described that a
CꢀC bond can also enhance the triplex stability [8]. The intercalator is inserted as a
bulge without an opposite base.
Pyrene was found to give the best intercalating system X compared to other
aromatic systems like naphthalene W, 1,1’-biphenyl, and various 9-substituted acridines,
which showed similar properties as pyrene but with a lower stabilization of the triplexes
[6a,b][9] (Fig. 1). TFOs containing pyrene-TINA showed triplex stabilities with high
discrimination to mismatched Hoogsteen binding to duplexes. Another useful effect of
these TFOs was destabilization of antiparallel Watson – Crick duplexes which may
reduce the risk of unwanted side effects when TINA-TFOs eventually are used as
drugs. The most important advantage of TINAs has been described recently. It was
demonstrated that the intercalator is interfering with quadruplex formation of G-rich
oligonucleotides at physiological potassium concentrations releasing the oligonucleo-
tide for antiparallel triplex formation [2b]. This means a double effect of TINA: i.e.,
destabilization of the quadruplexes and increasing the triplex stability of G-rich TFOs.
In this study, an intercalating system Y with an amino side chain attached to the
aromatic system was prepared. Under physiological conditions, the positively charged
amine forms an ionic interaction to the negatively charged phosphate backbone of the
duplex, resulting in an additional binding mode to stabilize the triplex. This intercalator
showed an enhanced triplex stability compared to the similar intercalating system
without the ionic interaction Z. Molecular modeling is used to visualize that the
enhanced triplex stability is due to a combination of stacking with the intercalator, and
the ionic interaction between the duplex backbone and the amino group in the side
chain of the intercalator.
Results and Discussion. – Synthesis. The synthesis of both monomers for standard
phosphoramidite chemistry started with the condensation of 4-bromonaphthalene-1,8-
dicarboxylic anhydride (1) with N,N-dimethylethane-1,2-diamine in EtOH or MeNH2
in EtOH (33% v/v), to afford 4-bromo-N-[2-(dimethylamino)ethyl]-1,8-naphthalimide
(2a) and 4-bromo-N-methyl-1,8-naphthalimide (2b) in 74 and 89% yield, respectively
[10][11]. The precursor (2R)-3-[(4-ethynylbenzyl)oxy]propane-1,2-diol (3) [6a][9]
and the appropriate naphthalimides 2 were coupled under Sonogashira conditions [12]
to achieve the N-[2-(dimethylamino)ethyl]naphthalimide derivative 4a and the N-
methylnaphthalimide derivative 4b in 78 and 31% yield, respectively (Scheme 1).

Helvetica Chimica Acta – Vol. 91 (2008)
807
Fig. 1. The intercalators Y and Z, and the reference intercalators W and X
Scheme 1. Synthesis of Phosphoramidite Precursors
i) R1NH2, EtOH, reflux, 3.5 – 15 h; 2a: 74%, 2b: 89%. ii) Pd(PPh3)4, CuI, Et3N, DMF, 208, 20 – 45 h;
4a: 78%, 4b: 31%.
The synthesis of the phosphoramidite 7a was achieved in 38% yield by subsequent
DMT-protection of the primary alcohol of 4a to give 6a, which was phosphitylated with

808
Helvetica Chimica Acta – Vol. 91 (2008)
2-cyanoethyl N,N,N’,N’-(tetraisopropyl)phosphordiamidite in anhydrous CH2Cl2 using
diisopropylammonium tetrazolide as an activator (Scheme 2). When the same strategy
was applied for the synthesis of the amidite 7b, only low yields were obtained in the
synthesis of the DMT-protected compound 6b, although the protection reaction of 4b
was attempted with a variety of methods. Instead, DMT-protection of precursor 3,
followed by Sonogashira coupling, considerably improved the yield of the protected
compound 6b. Treatment of 3 with DMTCl in pyridine afforded (S)-1-[bis(4-
methoxyphenyl)(phenyl)methoxy]-3-[(4-ethynylbenzyl)oxy]propan-2-ol (5) in 73%
yield. Subsequent Sonogashira coupling with naphthalimide 2b afforded 6b in 65%
yield. The phosphitylated monomer 7b was then synthesized in 89% yield. Incorpo-
ration of both monomers 7a and 7b in several DNA sequences was achieved by
standard automated DNA syntheses. The synthesized oligonucleotides and their mass-
spectrometric analysis are listed in Table 1.
Thermal Stability Studies. The thermal stability (melting temperature/Tm,8) of
triplexes and different types of duplexes with the synthesized oligonucleotides
containing the novel intercalators Y and Z was assessed by thermal denaturation
experiments. The melting temperatures (Tm) are listed in Tables 2 – 4. All oligonucleo-
Scheme 2. Synthesis of Intercalator Y and Z
i) Dimethoxytrityl chloride (DMTCl), pyridine, 208; 6a: 60%, 5: 73%. ii) 2b, Pd(PPh3)4, CuI, Et3N, DMF,
208; 6b: 65%. iii) NC(CH2)2OP(NiPr2)2, diisopropylammonium tetrazolide, CH2Cl2, 0 – 208; 7a: 64%,
7b: 89%.

Helvetica Chimica Acta – Vol. 91 (2008)
809
Table 1. Calculated and Found Masses of Synthesized Oligonucleotides
No.
Sequence
m/z [Da]
calc.
founda)
ON4
ON7
ON13
ON16
5’-CCCCTTYTCTTTTTT-3’
5’-YCCCCTTTCTTTTTT-3’
5’-CTCAAGYCAAGCT-3’
5’-YCTCAAGCAAGCT-3’
4655.2
4655.2
4147.9
4147.9
4654.1
4658.1
4147.8
4147.3
ON5
ON8
ON14
ON17
5’-CCCCTTZTCTTTTTT-3’
5’-ZCCCCTTTCTTTTTT-3’
5’-CTCAAGZCAAGCT-3’
5’-ZCTCAAGCAAGCT-3’
4597.3
4597.3
4089.9
4089.9
4596.8
4597.9
4089.8
4091.4
a) Determined by MALDI-TOF-MS.
tides containing an intercalator in the middle of the sequence were constructed to
intercalate as a bulge without any opposite nucleobase.
The thermal stability of triplexes with the synthesized oligonucleotides towards the
complementary duplex, D1, was assessed at two different pH values. The main
difference between the Tm values at pH 6.0 and 7.2 is based on the protonation of
cytosine, because only the protonated cytosine is able to form Hoogsteen bonds. The
cytosine bases are protonated at lower pH value, which results in higher Tm values at
pH 6.0 compared to pH 7.2. The thermal stability of triplexes formed with the
naphthalimide TFOs ON4, ON5, ON7, and ON8 was compared with the unmodified
TFO ON1, with pyrene TFOs ON3 and ON6, and with the naphthalene TFO ON2.
All modified sequences showed high Tm values of the triplexes compared to the
unmodified ON1 at both pH values. At pH 6.0, the stability of the modified sequences
with the intercalators Y and Z were also measured at a wavelength of l 400 nm besides
l 260 nm due to overlapping duplex melting and triplex melting at the latter
wavelength. Since it was not attempted to perform curve resolution of duplex and
triplex melting curves at 260 nm, it is not a surprise that the triplex melting was slightly
lower (up to 3.58 for ON8) when the melting was measured at 400 nm. Therefore, the
values at l 400 nm were considered more accurate (Table 2).
The triplex stabilities of ON4/D1 (T2m60 nm ¼ 48.08; T4m00 nm ¼ 47.58) and ON5/D1
(T2m60 nm ¼ 45.08; T4m00 nm ¼ 43.08) are considerably increased compared to the unmodified
triplex ON1/D1 (T2m60 nm ¼ 27.08). Stabilization in the range of DTm ¼ 18.0 – 21.08 shows
that we were able to establish a novel intercalating system with excellent binding
properties. The difference of the Tm values between naphthalene in ON2, and the two
naphthalimide sequences ON4 and ON5 was in the range of DTm ¼ 8.0 – 12.58 in all
triplex measurements. The increased Tm values for ON4/ON5 can be explained by
stabilization via the imide function. The increased p-surface leads to additional p – p
stacking. The comparison of the Tm values of the sequences with the pyrene intercalator
X (ON3) and the simple naphthalimide intercalator Z (ON5) demonstrated a better
stabilization by the bigger aromatic pyrene system, at both pH values (DTm ¼ 3.0 –
6.08).

810
Helvetica Chimica Acta – Vol. 91 (2008)
Table 2. Melting-Temperature (Tm [8]) Data for Third-Strand and Mismatch Third-Strand Melting, Taken
from UV Melting Curves (l 260 nm)a)
TFO Sequences
Triplex
Triplex mismatchb)
3’-CTGCCCCTTTCTTTTTT
3’-CTGCCCCTTACTTTTTT
5’-GACGGGGAAAGAAAAAA 5’-GACGGGGAATGAAAAAA
(D1
(D2)
pH 6.0
pH 7.2
pH 6.0
ON1 5’-CCCCTTTCTTTTTT 27.0
ON2 5’-CCCCTTWTCTTTTTT 35.0
ON3 5’-CCCCTTXTCTTTTTT 46.0
ON4 5’-CCCCTTYTCTTTTTT 48.0 (47.5)d)
ON5 5’-CCCCTTZTCTTTTTT 45.0 (43.0)
ON6 5’-XCCCCTTTCTTTTTT 44.5
ON7 5’-YCCCCTTTCTTTTTT 46.5 (46.0)
ON8 5’-ZCCCCTTTCTTTTTT 47.5 (44.0)
< 5.0
13.5
28.0
27.0
22.0
20.5
21.0
21.0
< 5.0
n.d.c)
27.0
31.5
26.5
22.5
26.0
22.0
a) c ¼ 1.5 mm of ON1 – 8 and 1.0 mm of each strand of dsDNA (D1/D2) in 20 mm sodium cacodylate,
100 mm NaCl, 10 mm MgCl2, pH 6.0 and 7.2; duplex Tm ¼ 58.58 (pH 6.0) and 57.08 (pH 7.2). b) Bases in
c
italics are mismatched to TFOs. ) Not determined. d) Tm Values determined at 400 nm for sequences for
Y and Z are given in parentheses in cases of overlapped triplex and duplex melting.
For the naphthalimide intercalator Y (ON4) with the N-[2-(dimethylamino)ethyl]
group, the Tm value was higher in the triplex measurement compared to the intercalator
Z (ON5). A difference of DTm ¼ 4.58 was determined at pH 6.0 and DTm ¼ 5.08 at
pH 7.2, when the intercalator was incorporated in the middle of the pyrimidine strand.
When added to the 5’-end, the difference between these modifications (i.e., ON7 and
ON8) was much lower. At pH 6.0, a difference of DTm ¼ 2.08 was measured, while, at
pH 7.2, the Tm values were the same. The higher stability of ON4 compared with the
similar sequences ON5 can be explained by the protonated dimethylamino group which
forms a H-bond and partially quenches a negative charge on the phosphate backbone in
the pyrimidine strand of the duplex. Presumably, this led to a reduced electrostatic
repulsion between phosphates in the duplex backbone besides the electrostatic
attraction between the TFO and the duplex. At pH 6.0, sequence ON3 with pyrene had
a lower Tm value than ON4 (DTm ¼ 1.58).
The sensitivity to mismatches was studied for parallel triplexes using the duplex D2
in which a base pair A · T, neighboring the insertion of the intercalator in the middle of
the sequences, was reversed compared to the matching Hoogsteen-binding duplex D1
(Table 2). For the neighboring mismatch, DTm dropped by 16.5 – 19.08 for the
sequences ON3 – ON5. This is a high sensitivity to mismatch when compared to the
work of Zhou et al. who aimed at stabilizing triplex formation of mismatch sequences.
9-Amino-6-chloro-2-methoxyacridine was inserted in the middle of the TFOs as a
bulged insertion, and the DTm values were determined within 108 for mismatched
Hoogsteen bindings [13]. When the intercalator was attached at the 5’-end of ON6 –
ON8 for the same duplex target, the Tm values dropped for the triplex mismatch in
the range of 20.0 – 22.08 compared to the perfectly matched sequences.
For application of TFOs in the antigene strategy, not only a high sensitivity to
mismatched triplexes is crucial, but also a sensitivity that minimize the affinity to other

Helvetica Chimica Acta – Vol. 91 (2008)
811
possible targets where TFOs can form parallel or antiparallel duplexes with DNA or
RNA single strands. Therefore, thermal-stability studies of parallel Hoogsteen-type
duplexes with the modified sequences ON2 – ON8 towards ON9 were performed. The
sequence ON2 with intercalator W showed only a stabilization of DTm ¼ 3.08 at
pH 6.0 compared to the unmodified sequence ON1, whereas all other sequences
showed an increase of thermal stability in the range of DTm ¼ 10.0 – 17.08 when
compared with the wild-type parallel duplex (Table 3). However, this increase in
melting temperatures is lower than those observed for the corresponding triplexes,
which, in all cases, had considerably higher Tm values than the corresponding parallel
duplex.
Table 3. Melting-Temperature (Tm [8]) Data for Parallel Duplex and Parallel Mismatch Duplexa) Melt-
ing, Taken from UV Melting Curves (l 260 nm)
TFO Sequences
Parallel duplex
5’-GACGGGGAAAGAAAAAA 5’-GACGGGGAATGAAAAAA
Parallel mismatchb)
(ON9)
(ON10)
pH 6.0
pH 7.2
pH 6.0
ON1 5’-CCCCTTTCTTTTTT 19.0
ON2 5’-CCCCTTWTCTTTTTT 22.0
ON3 5’-CCCCTTXTCTTTTTT 33.5
ON4 5’-CCCCTTYTCTTTTTT 35.0
ON5 5’-CCCCTTZTCTTTTTT 29.0
ON6 5’-XCCCCTTTCTTTTTT 36.0
ON7 5’-YCCCCTTTCTTTTTT 33.5
ON8 5’-ZCCCCTTTCTTTTTT 32.5
< 5.0
n.d.c)
n.d.
27.5
21.5
n.d.
< 5.0
n.d.
21.5
17.0
17.0
n.d.
23.5
20.0
n.d.
n.d.
a) c ¼ 1.0 mm of each strand in 20 mm sodium cacodylate, 100 mm NaCl, 10 mm MgCl2, pH 6.0 or pH 7.2.
b) Bases in italic are mismatched to TFOs. c) Not determined.
The sensitivity to mismatches was routinely studied for parallel duplexes with
bulged insertion of Y and Z in the middle and at the 5’-end of the sequence at pH 6.0.
For these measurements, the unmodified ON10 was chosen, which is one of the strands
from the mismatching duplex D2. Also for parallel duplexes, high sensitivity to
mismatches was observed (Table 3).
The antiparallel-duplex measurements were performed with the mixed purine/
pyrimidine sequences containing intercalators Y and Z (i.e., ON13, ON14, ON16, and
ON17) towards the complementary ssDNA (i.e., ON18) and ssRNA (i.e., ON19) to
study hybridization affinity in Watson – Crick-type duplexes (Table 4). ON18 is a strand
of a mixed purine/pyrimidine duplex segment, which is highly conserved in 148 HIV-1
sequences [14], with the potential of improved antisense properties of oligonucleotides
comprising the intercalators Yand Z. The intercalators were incorporated in the middle
of a sequence as a bulge or at the 5’-end. The Tm values towards ON18 and ON19 were
compared with the unmodified sequence ON11, and with pyrene TFOs ON12 and
ON15. Compared to the unmodified duplexes, the 5’-end-modified sequences showed
an increase of the Tm values, towards DNA, of 4.5 – 7.08 and, towards RNA, of 5.0 –
6.58. This can be explained by the lid effect of modifications at the 5’-end, which is

812
Helvetica Chimica Acta – Vol. 91 (2008)
Table 4. Melting-Temperature (Tm [8]) Data for Antiparallel Duplexa) Melting, Taken from UV Melting
Curves (l 260 nm)
Sequences
DNA
RNA
5’-AGCTTGCTTGAG
(ON18)
5’-AGCUUGCUUGAG
(ON19)
ON11
ON12
ON13
ON14
ON15
ON16
ON17
3’-TCGAACGAACTC
3’-TCGAACXGAACTC
3’-TCGAACYGAACTC
3’-TCGAACZGAACTC
3’-TCGAACGAACTCX
3’-TCGAACGAACTCY
3’-TCGAACGAACTCZ
47.5
39.5
40.0
37.0
54.0
54.5
52.0
40.5
30.5
38.5
33.5
45.5
47.0
46.5
a) c ¼ 1.0 mm of each ON in 140 mm NaCl, 10 mm sodium phosphate buffer, 1 mm EDTA, pH 7.0.
ascribed to stacking of an aromatic system to the adjacent nucleobase [6c]. Sequences
with modifications inserted in the middle of the sequences showed a destabilization of
7.5 – 10.08 towards DNA for all sequences (i.e., ON12 – ON14). Towards RNA, a similar
destabilization of 7.0 – 10.08 was observed for the modified sequences ON12 and ON14,
whereas ON13 only destabilized the duplex with 2.08 when compared with the
unmodified duplex. These results eliminate the potential of using the new intercalators
in antisense oligonucleotides. Instead, the destabilization can be a desired discrim-
ination against antiparallel Watson – Crick-type duplexes when oligonucleotides are
used for triplex formation. The different melting-temperature discrimination of ON13
with 7.58 towards DNA and 2.08 towards RNA may be explained by the protonated
dimethylamino group which interacts differently with the backbone of the dsDNA (B-
type helix) and the DNA/RNA hybrid (between A- and B-type helix) [15].
Molecular-Modeling Studies. The molecular-modeling studies were performed on
truncated triplexes with four base triplets on both sides of the intercalator which is
incorporated in the middle of the sequence (i.e., ON4 and ON5) and the
complementary duplex D1. Representative low-energy structures were generated with
the AMBER* force field [16] in MacroModel 9.1 and selected after examination with
Xcluster from Schrçdinger.
The modeling study of intercalator Y and Z (Fig. 2) shows that the positions of the
intercalators are similar, and that the benzene and the naphthalimide moieties are
adding to the triplex stability via p – p interaction with the nucleobases of the TFO and
the purine strand of the duplex, respectively. In addition, the benzene moieties of the
intercalators are slightly twisted in comparison to the naphthalimide moiety which has
been indicated to be essential for stabilizing the adjacent nucleobases of the TFO [6a].
However, it seems that no or only low stabilizing interactions between the
naphthalimide moiety and the pyrimidine strand of the duplex is formed as well as
between the CꢀC bond and the surrounding nucleobases. The enhanced triplex
stability observed for intercalator Y might, therefore, be caused by protonation of the
dimethylamino group which forms an ionic interaction with the negatively charged
phosphate backbone of the pyrimidine strand of the duplex (Fig. 2). This interaction is
enhanced by a H-bond from the dimethylammonium ion oriented in the direction of

Helvetica Chimica Acta – Vol. 91 (2008)
813
Fig. 2. Modeling structure of protonated intercalator Y (left) and the neutral intercalator Z (right)
the O-atom of the phosphate. It should be noted that the H-bonding takes place with
the 3’-phosphate group of the neighboring thymine to the 3’-side in the pyrimidine
strand of the duplex, and not to the phosphate group directly opposite to the
intercalator. Even though this interaction seems to slightly distort the pyrimidine
phosphate backbone of the duplex compared to intercalator Z, the molecular-modeling
studies support the suggestion of an ionic interaction between the protonated
dimethylamino group of intercalator Y with the phosphate backbone of the pyrimidine
strand in the duplex, thereby adding to the stabilization of the triplex.
Molecular-modeling studies for the second conformation of the unsymmetrical
intercalator where the naphthalimide is twisted by 1808 around the CꢀC bond were
also performed. The position and interacting properties of the intercalator was almost
identical to those shown in Fig. 2, and no optimal conformation could be assigned.
Conclusions. – The synthesis of the novel intercalators Y and Z via Sonogashira
coupling, followed by phosphoramidite chemistry and incorporation into DNA
sequences, showed a highly effective possibility of triplex stabilization. Naphthalimide
intercalator Z in ON5 with a Me group on the imide function showed a good triplex
stability comparable to the corresponding pyrene intercalator. The naphthalimide
intercalator Y in ON4 with a side chain containing an amino function showed an even
better stabilization of the triplex at two different pH values (6.0 and 7.2) compared to Z.
The reason can be assigned to protonation of the amino group under physiological
conditions, which reduces the electrostatic repulsion of the phosphates in the negatively
charged backbone and forms a unique binding interaction to the duplex. TFOs
containing Y are the first example of a stabilizing effect of an intercalator combined
with an amino side chain that binds to the duplex backbone. This is supported by
molecular-modeling studies which showed the ionic interaction of the protonated
dimethylamino group of intercalator Y with the phosphate backbone of the duplex. In
spite of the stabilizing effect of the intercalator, considerable destabilization of
mismatch triplexes was detected by thermal-stability measurements. Therefore, a novel

814
Helvetica Chimica Acta – Vol. 91 (2008)
possibility for the design of strongly binding TFOs with good discriminating properties
between matched and mismatched triplexes was developed, and this may be useful in
the antigene strategy. To avoid the complication of quadruplex formation of G-rich
TFOs, we consider it possible to optimize the 7-deaza-2’-deoxyguanosine approach by
using our compatible intercalator approach in a proper combination [17]. In this
context, it could also be interesting to use 7-(aminoalkynyl)-7-deaza-2’-deoxyguano-
sine to take advantage of a possible ion-pair stabilization [18].
Experimental Part
General. Unmodified oligonucleotides were purchased from DNA Technology A/S (DK-Šrhus) and
from TAG Copenhagen A/S (DK-Copenhagen). THFand toluene were dried over Na, and pyridine over
KOH. DMFand MeCN were dried over 3-Š molecular sieves, and Et3N and CH2Cl2 were dried over 4-Š
molecular sieves. Solvents used for column chromatography (CC) were distilled prior to use. Petroleum
ether (PE): b.p. 60 – 808. Ar was used in the reactions involving inert atmosphere. The silica gel (0.040 –
0.063 mm) used for CC and anal. TLC analyses carried out on TLC plates 60 F254 (Merck). M.p.: Büchi
melting-point apparatus. NMR Spectra: Varian Gemini-2000 spectrometer at 300 MHz for 1H-, 75 MHz
for 13C-, and 121.5 MHz for 31P-NMR; with Me4Si as an internal standard for 1H-NMR, deuterated
solvents CDCl3 (d 77.16 ppm) and DMSO (d 39.44 ppm) for 13C-NMR, and 85% H3PO4 as an external
standard for 31P-NMR; d in ppm, J in Hz. MS: Ionspec 4.7 T HiResMALDI Ultima Fourier transform
(FT) mass spectrometer (Ion Spec, Irvine, CA). For accurate ion mass determination, the [M þ H]þ or
[M þ Na]þ ion peak was matched using ions derived from the 2,5-dihydroxybenzoic acid matrix.
Electrospray ionization (ESI)-MS: 4.7 T HiResESI Ultima (FT) mass spectrometer; both spectrometers
are controlled by the OMEGA data system. EI-MS: Finnigan SSQ 710 instrument; in m/z (rel. %).
MALDI-TOF-MS: of isolated oligonucleotides were determined on a Voyager Elite biospectrometry
research station (PerSeptive Biosystems); in m/z.
1. 6-{[4-({[(2R)-2,3-Dihydroxypropyl]oxy}methyl)phenyl]ethynyl}-2-[2-(dimethylamino)ethyl]-1H-
benzo[de]isoquinoline-1,3(2H)-dione; 4a). Diol 3 [6][9] (0.50 g, 2.42 mmol) was dissolved in anh. Et3N
(20 ml) and anh. DMF (20 ml), and flushed with Ar for 30 min. Then, 6-bromo-2-[2-(dimethylami-
no)ethyl]-1H-benzo[de]isoquinoline-1,3(2H)-dione (2a [10]; 1.04 g, 3.00 mmol), Pd(PPh3)4 (0.127 g,
0.110 mmol), and CuI (0.023 g, 0.121 mmol) were added under a flow of Ar to the brown soln., and the
soln. was stirred under Ar at 208. After 20 h, the precipitate was filtered off, and CH2Cl2 (130 ml) was
added to the orange soln., which was washed with 0.3m aq. soln. of EDTA ammonium salt (2 Â 100 ml)
and H2O (3 Â 70 ml). The org. layer was dried (Na2SO4), the solvent was evaporated under reduced
pressure, and the residue was co-evaporated with toluene/EtOH 3 :1 (2 Â 30 ml). The residue was
submitted to CC (CH2Cl2/MeOH 20 :1) to afford 4a (0.89 g, 78%). Yellow solid. M.p. 115 – 1178.
1H-NMR (CDCl3): 2.36 (s, Me2N); 2.56 (br. s, CHOH); 2.66 (t, J ¼ 6.9, CH2CH2NMe2); 3.58 – 3.77 (m,
CH2CH(OH)CH2OCH2); 3.92 – 3.96 (m, CHOH); 4.30 (t, J ¼ 7.1, CH2CH2NMe2); 4.61 (s, C6H4CH2);
7.38 (d, J ¼ 8.1, 2 arom. H); 7.60 (d, J ¼ 7.5, 2 arom. H); 7.79 (t, J ¼ 7.8, 1 arom. H); 7.87 (d, J ¼ 7.5, 1 arom.
H); 8.50 (d, J ¼ 7.5, 1 arom. H); 8.60 (d, J ¼ 8.1, 1 arom. H); 8.64 (d, J ¼ 8.4, 1 arom. H). 13C-NMR
(CDCl3): 38.27 (NMe2); 45.80 (CH2CH2NMe2); 56.97 (CH2CH2NMe2); 64.10 (CH2OH); 70.83 (CHOH);
72.22 (C6H4CH2OCH2); 73.14 (C6H4CH2OCH2); 86.48 (CꢀCÀC6H4); 98.99 (CꢀCÀC6H4); 121.66,
122.99, 127.53, 127.67, 127.85, 128.15, 128.19, 130.53, 130.84, 131.65, 131.78, 132.13, 132.49, 139.59 (arom.
C); 163.86, 164.13 (2 C¼O). HR-MALDI-MS: 473.2073 ([M þ H]þ, C28H29N2O5þ ; calc. 473.2076).
2. 6-{[4-({[(2R)-2,3-Dihydroxypropyl]oxy}methyl)phenyl]ethynyl}-2-methyl-1H-benzo[de]isoqui-
noline-1,3(2H)-dione; 4b). Diol 3 ([6][9], 0.50 g, 2.42 mmol) was dissolved in anh. Et3N (20 ml), and
anh. DMF (20 ml), and flushed with Ar for 30 min. Then, 6-bromo-2-methyl-1H-benzo[de]isoquinoline-
1,3(2H)-dione (2b [11]; 0.87 g, 3.00 mmol), Pd(PPh3)4 (0.127 g, 0.121 mmol), and CuI (0.023 g,
0.121 mmol) were added under a flow of Ar to the brown soln., and the soln. was stirred under Ar at
208. After 22 h, the precipitate was filtered off, CH2Cl2 (130 ml) was added to the orange mixture, which
was washed with 0.3m aq. soln. of EDTA ammonium salt (2 Â 100 ml) and with H2O (3 Â 70 ml). The org.

Helvetica Chimica Acta – Vol. 91 (2008)
815
layer was dried (Na2SO4), the solvent was evaporated under reduced pressure, and the residue was co-
evaporated with toluene/EtOH 3 :1 (2 Â 30 ml). The residue was submitted to CC (CH2Cl2/MeOH 20 :1)
to afford 4b (0.31 g, 31%). Yellow solid. M.p. 164 – 1668. 1H-NMR ((D6)DMSO): 3.35 (s, MeN); 3.40 –
3.43 (m, CH2CH(OH)CH2OCH2); 3.65 – 3.70 (m, CHOH); 4.53 – 4.57 (m, CH2CH(OH)CH2OCH2);
4.58 (s, C6H4CH2); 4.75 (br. s, CHOH); 7.47 (d, J ¼ 8.1, 2 arom. H); 7.74 (d, J ¼ 8.1, 2 arom. H); 7.91 (t, J ¼
8.1, 1 arom. H); 7.97 (d, J ¼ 7.8, 1 arom. H); 8.35 (d, J ¼ 7.5, 1 arom. H); 8.47 (d, J ¼ 7.2, 1 arom. H); 8.66
(d, J ¼ 7.5, 1 arom. H). 13C-NMR ((D6)DMSO): 26.58 (MeN); 63.00 (CH2OH); 70.51 (CHOH); 71.64
(C6H4CH2OCH2); 72.15 (C6H4CH2OCH2); 85.86 (CꢀCÀC6H4); 98.58 (CꢀCÀC6H4); 120.10, 121.67,
122.40, 126.05, 127.02, 127.51, 128.02, 129.40, 129.68, 130.68, 130.93, 131.69, 140.67 (arom. C); 162.93,
163.23 (2 C¼O). EI-MS: 415 (25, Mþ). HR-MALDI-MS: 438.1296 ([M þ Na]þ, C25H21NNaOþ5 ; calc.
438.1317).
3. 6-({4-[({(2S)-3-[Bis(4-methoxyphenyl)(phenyl)methoxy]-2-hydroxypropyl}oxy)methyl]phenyl}-
ethynyl)-2-[2-(dimethylamino)ethyl]-1H-benzo[de]isoquinoline-1,3(2H)-dione; 6a). Diol 4a (0.67 g,
1.42 mmol) was dissolved in anh. CH2Cl2 (40 ml) and anh. Et3N (0.22 ml, 1.56 mmol), and 4,4’-
dimethoxytrityl chloride (DMTCl; 0.53 g, 1.56 mmol) was added under a flow of Ar. The soln. was stirred
under Ar at 208. After 17 h, an extra portion of DMTCl (0.12 g, 0.36 mmol) and Et3N (0.05 ml,
0.36 mmol) was added, and, after another 7 h, another portion of DMTCl (0.05 g, 0.15 mmol) and Et3N
(0.02 ml, 0.14 mmol) was added. After a total reaction time of 45 h, H2O (30 ml) was added, the two
layers were separated, and the org. phase was washed with a sat. aq. soln. of NaHCO3 (2 Â 30 ml). The aq.
phase was extracted with CH2Cl2 (25 ml). The combined org. layers were dried (Na2SO4), filtered, and
evaporated under reduced pressure. The residue was co-evaporated with toluene/EtOH 1:1 (2 Â 25 ml).
The residue was submitted to CC (CH2Cl2/MeOH 1:1 with 0.5% Et3N) to afford 6a (0.657 g, 60%).
Yellow foam. M.p. 82 – 838. 1H-NMR (CDCl3): 2.37 (s, Me2N); 2.46 (br. s, CHOH); 2.68 (t, J ¼ 6.9,
CH2CH2NMe2); 3.23 – 3.26 (m, DMTOCH2); 3.59 – 3.64 (m, CH(OH)CH2OCH2); 3.78 (s, 2 MeO);
3.97 – 4.03 (m, CHOH); 4.33 (t, J ¼ 7.1, CH2CH2NMe2); 4.58 (s, C6H4CH2); 6.82 (d, J ¼ 9.0, 4 H, DMT);
7.20 – 7.45 (m, 11 H, C6H4, DMT); 7.63 (d, J ¼ 8.1, 2 arom. H); 7.83 (t, J ¼ 7.9, 1 arom. H); 7.94 (d, J ¼ 7.8, 1
arom. H); 8.55 (d, J ¼ 7.5, 1 arom. H); 8.64 (d, J ¼ 7.5, 1 arom. H); 8.72 (d, J ¼ 8.4, 1 arom. H). 13C-NMR
(CDCl3): 38.28 (Me2N); 45.81 (CH2CH2NMe2); 55.33 (MeO); 57.06 (CH2CH2NMe2); 64.44
(CH2ODMT); 70.09 (CHOH); 71.96 (C6H4CH2); 72.93 (C6H4CH2OCH2); 86.24 (Ph3C); 86.44
(CꢀCÀC6H4); 99.10 (CꢀCÀC6H4); 113.25, 121.53, 122.17, 123.06, 126.94, 127.19, 127.56, 127.78, 127.96,
128.25, 130.17, 130.56, 130.85, 131.79, 132.06, 132.54, 134.84, 136.08, 139.86, 144.92, 158.63 (arom. C);
163.92, 164.18 (2 C¼O). HR-MALDI-MS: 775.3362 ([M þ H]þ, C49H47N2O7þ ; calc. 775.3383).
4. (2S)-1-[Bis(4-methoxyphenyl)(phenyl)methoxy]-3-{[4-({2,3-dihydro-2-[2-(dimethylamino)ethyl]-
1,3-dioxo-1H-benzo[de]isoquinolin-6-yl}ethynyl)benzyl]oxy}propan-2-yl 2-Cyanoethyl Di(propan-2-yl)-
phosphoramidite; 7a). Alcohol 6a (0.43 g, 0.56 mmol) was dissolved in anh. CH2Cl2 (45 ml) and flushed
with Ar for 10 min, followed by addition of diisopropylammonium tetrazolide (0.12 g, 0.68 mmol). Then,
2-cyanoethyl tetraisopropylphosphordiamidite (0.08 g, 0.27 mmol) was added dropwise under external
cooling with an ice-water bath and stirred under Ar at 208. Extra addition of 2-cyanoethyl
tetraisopropylphosphordiamidite (0.08 g, 0.27 mmol) was performed after 21, 40, and 46 h, resp. After
46 h, also N,N-diisopropylammonium tetrazolide (0.05 g, 0.29 mmol) was added. After 61 h, anal. TLC
showed no more starting material, and the reaction was quenched with H2O (30 ml). The layers were
separated, and the org. phase was washed with H2O (30 ml). The combined aq. layers were extracted
with CH2Cl2 (25 ml). The combined org. layers were dried (Na2SO4), filtered, and silica gel (2.0 g) and
pyridine (1.5 ml) were added, and solvents were removed under reduced pressure. The residue was
purified by dry column vacuum chromatography (SiO2; AcOEt/cyclohexane 0 :10 ! 5 :5, with 0.5%
Et3N) to afford 7a (0.35 g, 64%). Yellow oil. 31P-NMR (CDCl3): 150.34, 150.57.
5. (2S)-1-[Bis(4-methoxyphenyl)(phenyl)methoxy]-3-[(4-ethynylbenzyl)oxy]propan-2-ol (5). Diol 3
([6][9], 0.23 g, 1.15 mmol) was dissolved in anh. pyridine (25 ml), and, under a flow of Ar, DMTCl
(0.47 g, 1.40 mmol) was added, followed by stirring under Ar at 208. After 36 h, the reaction was
quenched with MeOH (1 ml), and the mixture was diluted with AcOEt (50 ml) and washed with a sat. aq.
soln. of NaHCO3 (2 Â 30 ml). The H2O phase was extracted with AcOEt (30 ml). The combined org.
layers were dried (Na2SO4), filtered, and evaporated under reduced pressure. The residue was co-
evaporated with toluene/EtOH 1:1 (2 Â 25 ml), dissolved in anh. CH2Cl2 (20 ml) and anh. pyridine

816
Helvetica Chimica Acta – Vol. 91 (2008)
(2 ml), and adsorbed on silica gel (1.2 g) by evaporation under reduced pressure. The residue was
purified by dry column vacuum chromatography (SiO2; AcOEt/cyclohexane 0 :10 ! 5 :5, with 0.5%
Et3N) to afford 5 (0.43 g, 73%). Yellow foam. 1H-NMR (CDCl3): 2.40 (d, J ¼ 4.8, CHOH); 3.06 (s,
CꢀCH); 3.20 – 3.23 (m, DMTOCH2); 3.55 – 3.58 (m, CH(OH)CH2OCH2); 3.78 (s, 2 MeO); 3.97 (m,
CHOH); 4.52 (s, C6H4CH2); 6.81 (d, J ¼ 9.0, 4 H, DMT); 7.20 – 7.50 (m, 13 H, arom. H, DMT). 13C-NMR
(CDCl3): 55.34 (MeO); 64.43 (CH2ODMT); 70.10 (CHOH); 71.81 (C6H4CH2); 72.97 (C6H4CH2OCH2);
77.58 (CꢀCÀH); 83.63 (CꢀCÀH); 86.25 (Ph3C); 113.26, 121.49, 127.53, 127.97, 128.26, 130.17, 132.30,
136.09, 139.05, 144.93, 158.64 (arom. C). HR-MALDI-MS: 531.2164 ([M þ Na]þ, C33H32NaO5þ ; calc.
531.2147).
6. 6-({4-[({(2S)-3-[Bis(4-methoxyphenyl)(phenyl)methoxy]-2-hydroxypropyl}oxy)methyl]phenyl}-
ethynyl)-2-methyl-1H-benzo[de]isoquinoline-1,3(2H)-dione; 6b). Compound 5 (0.40 g, 0.79 mmol) was
dissolved in anh. Et3N (10 ml) and anh. DMF (10 ml), and flushed with Ar for 30 min. Then, 2b ([11],
0.29 g, 1.00 mmol), Pd(PPh3)4 (0.041 g, 0.035 mmol), and CuI (0.008 g, 0.042 mmol) were added under a
flow of Ar to the soln., and the soln. was stirred under Ar at 208. After 28 h, an extra portion of Pd(PPh3)4
(0.041 g, 0.035 mmol) and CuI (0.008 g, 0.042 mmol) were added, and, after a total reaction time of 40 h,
CH2Cl2 (75 ml) and Et3N (1 ml) were added to the mixture. The mixture was washed with 0.3m aq. soln.
of EDTA ammonium salt (2 Â 50 ml) and with H2O (3 Â 70 ml), followed by extraction of the combined
aq. layers with CH2Cl2 (40 ml). The org. layer was dried (Na2SO4), evaporated under reduced pressure,
and the residue was co-evaporated with toluene/EtOH 3 :1 containing 0.5% Et3N (2 Â 40 ml). The crude
material was dissolved in CH2Cl2 (20 ml) and pyridine (1.0 ml), and silica gel (1.6 g) was added, and the
solvents were removed under reduced pressure. The residue, adsorbed on silica gel, was purified by dry
column vacuum chromatography (SiO2; AcOEt/cyclohexane 0 :10 ! 5 :5 containing 0.5% Et3N) to
afford 6b (0.38 g, 65%). Yellow foam. M.p. 84 – 868. 1H-NMR (CDCl3): 2.47 (d, J ¼ 5.1, CHOH); 3.21 –
3.26 (m, DMTOCH2); 3.56 (s, MeN); 3.58 – 3.63 (m, CH2CH(OH)CH2OCH2); 3.77 (s, 2 MeO); 3.97 –
4.13 (m, CHOH); 4.58 (s, C6H4CH2); 6.82 (d, J ¼ 9.0, 4 H, DMT); 7.20 – 7.45 (m, 11 H, C6H4, DMT);
7.65 (d, J ¼ 8.1, 2 arom. H); 7.82 (t, J ¼ 7.8, 1 arom. H); 7.93 (d, J ¼ 7.5, 1 arom. H); 8.55 (d, J ¼ 7.5, 1 arom.
H); 8.64 (d, J ¼ 7.2, 1 arom. H); 8.72 (d, J ¼ 7.5, 1 arom. H). 13C-NMR (CDCl3): 27.19 (MeN); 55.33,
(MeO); 64.44 (CH2OH); 70.12 (CHOH); 71.96 (C6H4CH2OCH2); 72.94 (C6H4CH2OCH2); 86.27
(Ph3C); 86.43 (CꢀCÀC6H4); 99.12 (CꢀCÀC6H4); 113.25, 121.53, 122.09, 122.98, 126.95, 127.57, 127.78,
127.96, 128.25, 130.18, 130.51, 130.86, 131.73, 132.07, 132.55, 132.65, 136.07, 139.62, 139.86, 144.93 (arom.
C); 164.13, 164.39 (2 C¼O). HR-MALDI-MS: 740.2587 ([M þ Na]þ), C46H39NNaO7þ ; calc. 740.2624).
7. (2S)-1-[Bis(4-methoxyphenyl)(phenyl)methoxy]-3-({4-[(2,3-dihydro-2-methyl-1,3-dioxo-1H-ben-
zo[de]isoquinolin-6-yl)ethynyl]benzyl}oxy)propan-2-yl 2-Cyanoethyl Di(propan-2-yl)phosphoramidite;
7b). Alcohol 6b (0.30 g, 0.42 mmol) was dissolved in anh. CH2Cl2 (25 ml) and flushed with Ar for 10 min,
followed by addition of N,N-diisopropylammonium tetrazolide (0.09 g, 0.49 mmol). Then, 2-cyanoethyl
tetraisopropylphosphordiamidite (0.16 g, 0.54 mmol) was added dropwise under external cooling with an
ice-water bath, and the soln. was stirred under Ar at 208. An extra portion of 2-cyanoethyl
tetraisopropylphosphordiamidite (0.06 g, 0.21 mmol) and N,N-diisopropylammonium tetrazolide
(0.04 g, 0.21 mmol) were added after 21 h. After 44 h, anal. TLC showed no more starting material,
and the reaction was quenched with H2O (25 ml). The layers were separated, and the org. phase was
washed with H2O (25 ml). The combined aq. layers were extracted with CH2Cl2 (25 ml). The combined
org. layers were dried (Na2SO4), filtered, and silica gel (1.0 g) and anh. pyridine (1.0 ml) were added, and
solvents were removed under reduced pressure. The residue was purified by dry column vacuum
chromatography (SiO2; AcOEt/cyclohexane 0 :10 ! 3 :7 containing 0.5% Et3N) to afford 7b (0.34 g,
89%). Yellow foam. 31P-NMR (CDCl3): 150.26, 150.49.
8. Molecular Modeling. The molecular modeling was performed with the program Maestro v7.5 from
Schrçdinger. All calculations were conducted with AMBER* force field [16] and the GB/SAwater model
[19]. The dynamics simulations were performed with stochastic dynamics [20], a SHAKE algorithm to
constrain bonds to H-atoms [21], a time step of 1.5 fs, and simulation temp. of 300 K. Simulation for
0.5 ns with an equilibration time of 150 ps generated 250 structures, which were all minimized using the
PRCG method with convergence threshold of 0.05 kJ/mol. The minimized structures were examined
with Xcluster from Schrçdinger, and representative low-energy structures were selected. The starting

Helvetica Chimica Acta – Vol. 91 (2008)
817
structures were generated with Insight II v97.2 from MSI, followed by incorporation of the desired
intercalator between the two neighboring bases.
9. Synthesis of Oligonucleotides. DMT-On oligodeoxynucleotides were synthesized in a 0.2-mmol
scale on CPG supports using Expedite Nucleic Acid Synthesis System model 8909 (Applied Biosystems),
using 4,5-dicyanoimidazole as an activator and a 0.075 mm soln. of the corresponding phosphoramidites
7a and 7b in anh. CH2Cl2. The syringe was attached to the CPG support, and the soln. was pressed
through the support giving a coupling time of twice 20 min for 7a and of 20 min for 7b. Purification of 5’-
DMT-on oligonucleotides was accomplished by a reverse-phase semiprep. HPLC on a Waters Xterra MS
C18 column. DMT Groups were cleaved with 80% AcOH (100 ml) for 20 min. Afterwards, H2O (100 ml)
and 3m aq. AcONa (50 ml) were added, and oligonucleotides were precipitated from 99% EtOH (550 ml).
The modified oligonucleotides were confirmed by MALDI-TOF-MS analysis on a Voyager Elite
biospectrometry research station from PerSeptive Biosystems. The purity of the final TFOs was
controlled by ion-exchange chromatography using a LaChrom system from Merck Hitachi on a GenPak-
Fax column (Waters).
10. Thermal Stability Studies of Duplexes and Triplexes. Melting profiles were performed on a Perkin-
Elmer UV/VIS spectrometer Lambda 35 with a PTP-6 thermostat and fitted with a Perkin-Elmer
Templab 2.00 Software. The parallel triplexes were formed by mixing the two strands of the Watson –
Crick duplex each at a concentration of 1.0 mm and the TFO at a concentration of 1.5 mm in the
corresponding buffer soln. (pH 6.0 or 7.2). The soln. was heated to 808 for 5 min for removal of air
bubbles and afterwards cooled to 58 for 30 min. The duplexes were formed by mixing the two strands,
each at a concentration of 1.0 mm in the corresponding buffer soln., followed by heating to 808 for 5 min
and then cooling to 58 for 30 min. The absorbance of triplexes and duplexes was measured at l 260 nm
from 5 to 808 with a heating rate of 1.08/min. Experiments were also performed at l 400 nm. Each
measurement was performed twice. The melting temperatures (Tm, 8) were determined as the average
value of both maxima of the first derivative plots of the melting curves.
REFERENCES
[1] S. Obika, Chem. Pharm. Bull. 2004, 52, 1399.
[2] a) V. V. Soyfer, N. P. Potaman, ꢁTriple-Helical Nucleic Acidsꢂ, Springer Verlag, 1995, p. 166; b) E.
Schmitt, P. Muller, M. Paramavisam, V. V. Filichev, N. Bomholt, J. Schwarz-Finsterle, G.
Hildenbrand, E. B. Pedersen, L. E. Xodo, M. Hausmann, ꢁIntegrative Design of Selectively
Colocalizing Oligonucleotide Probesꢂ, Poster at the International Workshop, Integrative Bioinfor-
matics, 4th Annual Meeting, September 10 – 12, 2007, Ghent, Belgium, 2007.
[3] a) S. M. Gryaznov, J.-K. Chen, J. Am. Chem. Soc. 1994, 116, 3143; b) V. Tereshko, S. M. Gryaznov,
M. Egli, J. Am. Chem. Soc. 1998, 120, 269; c) F. Ehrenmann, J.-J. Vasseur, F. Debart, Nucleosides
Nucleotides Nucleic Acids 2001, 20, 797; d) T. Michel, F. Debart, F. Heitz, J.-J. Vasseur,
ChemBioChem 2005, 6, 1254.
[4] a) B. Cuenoud, F. Casset, D. Hüsken, F. Natt, R. M. Wolf, K.-H. Altmann, P. Martin, H. E. Moser,
Angew. Chem., Int. Ed. 1998, 37, 1288; b) J. Wengel, Acc. Chem. Res. 1999, 32, 301; c) B.-W. Sun,
B. R. Babu, M. D. Sørensen, K. Zakrzewska, J. Wengel, J.-S. Sun, Biochemistry 2004, 43, 4160; d) J.
Basye, J. O. Trent, D. Gao, S. W. Ebbinghaus, Nucleic Acids Res. 2001, 29, 4873; e) T. Carlomagno,
M. J. J. Blommers, J. Meiler, B. Cuenoud, C. Griesinger, J. Am. Chem. Soc. 2001, 123, 7364.
[5] a) L. E. Xodo, G. Manzini, F. Quadrifoglio, G. A. van der Marcel, J. H. van Boom, Nucleic Acids Res.
1991, 19, 5625; b) S. Hildbrand, A. Blaser, S. P. Parel, C. J. Leumann, J. Am. Chem. Soc. 1997, 119,
5499; c) V. Roig, U. Asseline, J. Am. Chem. Soc. 2003, 125, 4416.
[6] a) V. V. Filichev, E. B. Pedersen, J. Am. Chem. Soc. 2005, 127, 14849; b) V. V. Filichev, H. Gaber,
T. R. Olsen, P. T. Jørgensen, C. H. Jessen, E. B. Pedersen, Eur. J. Org. Chem. 2006, 17, 3960; c) R. X.-
F. Ren, N. C. Chaudhuri, P. L. Paris, I. S. Rumney, E. T. Kool, J. Am. Chem. Soc. 1996, 118, 7671.
[7] U. B. Christensen, E. B. Pedersen, Nucleic Acids Res. 2002, 30, 4918; A. A.-H. Abdel-Rahman, O. M.
Ali, E. B. Pedersen, Tetrahedron 2005, 52, 15311.
[8] B. C. Froehler, S. Wadwani, T. J. Terhorst, S. R. Gerrard, Tetrahedron Lett. 1992, 33, 5307.

818
Helvetica Chimica Acta – Vol. 91 (2008)
´
[9] I. Geci, V. V. Filichev, E. B. Pedersen, Bioconjugate Chem. 2006, 17, 950.
[10] P. Yang, Q. Yang, X. Qian, Tetrahedron 2005, 61, 11895.
[11] J.-X. Yang, X.-L. Wang, X.-M. Wang, L.-H. Xu, Dyes Pigm. 2005, 66, 83.
[12] K. Sonogashira, Y. Tohda, N. Hagihara, Tetrahedron Lett. 1975, 16, 4467.
´ `
[13] B.-W. Zhou, E. Puga, J.-S. Sun, T. Garestier, C. Helene, J. Am. Chem. Soc. 1995, 117, 10425.
[14] A. Mujeeb, S. M. Kerwin, G. L. Kenyon, T. L. James, Biochemistry 1993, 32, 13419.
[15] L. Stryer, J. L. Tymoczko, J. M. Berg, in ꢁBiochemistryꢂ, 5th. edn., W. H. Freeman and Company,
2001, p. 747 – 749.
[16] S. J. Weiner, P. A. Kollman, D. A. Case, U. C. Singh, C. Ghio, G. Alagona, S. Profeta Jr., P. Weiner, J.
Am. Chem. Soc. 1984, 106, 765; S. J. Weiner, P. A. Kollman, D. T. Nguyen, D. A. Case, J. Comput.
Chem. 1986, 7, 230.
[17] F. Seela, Y. Chen, C. Mittelbach, Helv. Chim. Acta 1998, 81, 570.
[18] H. Rosemeyer, N. Ramzaeva, E.-M. Becker, E. Feiling, F. Seela, Bioconjugate Chem. 2002, 13, 1274.
[19] W. C. Still, A. Tempczyk, R. C. Hawley, T. Hendrickson, J. Am. Chem. Soc. 1990, 112, 6127.
[20] W. F. Van Gunsteren, H. J. C. Berendsen, Mol. Simul. 1988, 1, 173.
[21] J.-P. Ryckaert, G. Ciccotti, H. J. C. Berendsen, J. Comput. Phys. 1977, 23, 327.
Received January 4, 2008