The first postsynthetic 5′-5′ intercalators in triplex DNA - Solid-phase postsynthetic Sonogashira reaction and homocouplings on arylacetylenes

Article abstract of DOI:10.1002/hlca.200800397

New Zealand We report herein a new technique for postsynthetic modifications of on-column oligonucleotides. CPG-Supported nonamer polypyrimidine oligodeoxynucleotides comprising 2-(3-iodobenzyloxy)ethyl phosphate at the terminal 5′-ends were successfully

Full text of DOI:10.1002/hlca.200800397

716
Helvetica Chimica Acta – Vol. 92 (2009)
The First Postsynthetic 5’-5’ Intercalators in Triplex DNA – Solid-Phase
Postsynthetic Sonogashira Reaction and Homocouplings on Arylacetylenes
by 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 45-65-502555; fax: þ 45-66-158780;
e-mail: ebp@ifk.sdu.dk)
b) Institute of Fundamental Sciences, Massey University, Private Bag 11-222, Palmerston North,
New Zealand
We report herein a new technique for postsynthetic modifications of on-column oligonucleotides.
CPG-Supported nonamer polypyrimidine oligodeoxynucleotides comprising 2-(3-iodobenzyloxy)ethyl
phosphate at the terminal 5’-ends were successfully capped with 1,3-diethynylbenzene under Sonogashira
reaction conditions, resulting in novel intercalating 5’-5’ linker in alternate strand Hoogsteen triplex-
forming oligonucleotidej (TFO) X. In addition, novel 5’-5’-linked alternate strand TFOs, Y and Z, were
isolated as a result of on-column Sonogashira coupling with 1,3-diethynylbenzene, followed by Cu-
catalyzed oxidative arylacetylenic homocoupling(s) between neighboring terminal arylacetylenes.
Linkers X, Y, and Z represent the first postsynthesis 5’-5’ intercalating linkers for alternate strand
TFOs. Reproducibility and verification by independent syntheses using conventional phosphoramidite
chemistry and DNA synthesis, as well as thermal stability and molecular modelling studies are included.
Introduction. – With the decoding of the human genome and with an increased
knowledge of the gene functions, the necessity for sequence-specific recognition and
targeting of double-stranded DNA (dsDNA) for diagnostic applications became of
high interest. The formation of triple DNA helices via the binding of DNA probes to
DNA duplexes through H-bonding is the key to sequence-selective dsDNA targeting
[1][2]. Triplex-forming oligonucleotides (TFOs) can bind to dsDNA in either a parallel
or antiparallel motif with respect to the homopurine strand of the duplex [3]. The main
disadvantage, employing TFOs as a diagnostic tool, is the instability of the
corresponding triplex under physiological conditions, since cytosine requires proto-
nation at N(3) to form the Cþ · GꢀC Hoogsteen base triplet. In addition, G-rich TFOs
are unlikely to form antiparallel triplexes under physiological ion concentrations,
because, under these conditions, they form more stable aggregates, e.g., quadruplexes,
promoted by Kþ ions [4][5]; however, insertion of TINA (twisted intercalating nucleic
acid) favors triplex formation [6][7]. Another major limitation for the triplex
technology is the requirement of a homopurine tract of more than 15 purines in order
to achieve high target-binding affinity [8].
Alternate-strand TFOs that target homopurine tracts located on each strand of the
dsDNA have been designed to increase the number of targets in dsDNA (Fig. 1).
Alternate-strand TFO consists of minimum two independent oligonucleotides cova-
lently linked at their 3’- or the 5’-ends [9][10]. Since each oligonucleotide of the probe
ꢀ 2009 Verlag Helvetica Chimica Acta AG, Zꢁrich

Helvetica Chimica Acta – Vol. 92 (2009)
717
should bind simultaneously to the target, in combination with specific point of strand
alternation, a better specificity to dsDNA can be assumed when compared to targeting
a homopurine sequence.
Fig. 1. 5’-5’ Intercalating linkers for alternate-strand TFOs (X, Y, Z, and W) and a sketch of the 5’-5’
alternate-strand triplex concept
Herein, we report the synthesis of novel 5’-5’ linkers obtained via on-column
postsynthetic derivatization of DNA. The postsynthetically Pd0-catalyzed reaction
between 2-(3-iodobenzyloxy)ethyl phosphate linked to the 5’-end of DNA and 1,3-
diethynylbenzene resulted in different types of on-column oligonucleotide cross-
linking (Fig. 1). The linker X was formed by Sonogashira reactions between two
separate terminal 2-(3-iodobenzyloxy)ethyl phosphates and 1,3-diethynylbenzene. In
addition, the linkers Y and Z were presumably formed by Sonogashira reactions
between two separate terminal 2-(3-iodobenzyloxy)ethyl phosphates and two or six
1,3-diethynylbenzenes via oxidative arylacetylenic homocoupling(s). To our knowl-
edge, this is the first observation of one-pot postsynthetic Sonogashira coupling and

718
Helvetica Chimica Acta – Vol. 92 (2009)
acetylenic homocoupling between neighboring on-column oligonucleotides, represent-
ing a new technique in postsynthetic modifications of on-column oligonucleotides. In
continuation of this work, the linker W was anticipated as an interesting target, and it
was synthesized by usual amidite chemistry.
Results and Discussion. – Postsynthetic 5’-5’ Intercalators. Recently, we have shown
the possibility of achieving high thermal stability of a matched 5’-5’ linked alternate-
strand triplex under physiological conditions [11]. Therefore, we decided to investigate
the possibility of synthesizing a novel 5’-5’ linker in an alternate-strand TFO, by capping
the modified 5’-ends of a nonamer oligonucleotide together through two separate solid-
phase postsynthetic Sonogashira reactions. Postsynthetic modification of CPG-
supported oligonucleotides has shown to be an effective and time-saving alternative
to standard phosphoramidite synthesis, and several reports on postsynthetically Pd0-
catalyzed reactions have been reported [6][12 – 15].
A nonamer oligodeoxynucleotide was synthesized by standard DNA automated
phosphoramidite approach in a 1.0-mmol scale on 500-ꢂ CPG-support and subsequent
coupled at the 5’-end with the phosphoramidite of 2-(3-iodobenzyloxy)ethanol
(Scheme 1). In the first postsynthesis, the modified oligonucleotide was treated
afterwards with a reaction mixture containing 1,3-diethynylbenzene (38 mmol),
Pd(PPh3)4 (22 mmol), and CuI (26 mmol) in dry DMF/Et3N (350 ml/150 ml) under Ar
and at room temperature over 3 h. After washing the CPG-support with DMF (2 ꢁ
2 ml) and MeCN (2 ꢁ 2 ml), the oligonucleotide derivative was cleaved from the
support and deprotected by 32% aqueous NH3 (558, overnight), followed by removal
of excess reagents and impurities from the postsynthesis by NAPTM 10 column. Major
fractions from the RP-HPLC purification were collected and analyzed by MALDI-
TOF-MS (Table 1).
Scheme 1. Synthesis of Phosphoramidite and Oligonucleotides (ONs) ON1 and ON2
i) NC(CH2)2OP(NiPr2)2, diisopropylammonium tetrazolide, CH2Cl2, 19 h; 25%. ii) Incorporation of 2 at
the 5’-end of nonamer oligodeoxynucleotide synthesized by standard DNA synthesis on a 500-ꢂ 1.0-
mmol CPG-support. iii) Postsynthesis: 1,3-diethynylbenzene (38 mmol), Pd(PPh3)4 (22 mmol), and CuI
(26 mmol) in dry DMF/Et3N (350 ml/150 ml) under Ar. iv) 32% aq. NH3, 558, overnight.
Surprisingly, from MALDI-TOFanalysis, two different 5’-5’ linked oligonucleotides
(ONs) were isolated, ON1 and ON2. ON1 was isolated as the major product and
identified from the molecular mass as the possible product of Sonogashira couplings
between 1,3-diethynylbenzene and two neighboring oligonucleotides possessing a

Helvetica Chimica Acta – Vol. 92 (2009)
719
Table 1. Reproducibility and Verification Study: Calculated and Found Masses of Synthesized ONs
Entry
Synthesisa)
Sequence
Mass [m/z]
Foundb)
Calc.
ON1
ON3
ON7
1st Postsynth.
2nd Postsynth.
Amidite chem.
3’-TTTTCTTTT-5’-X-5’-TTTTCTTTT-3’
5868.7
5872.5
5867.5
5871.1
–
–
ON2
ON4
ON5
ON8
1st Postsynth.
2nd Postsynth.
3rd Postsynth.
Amidite chem.
3’-TTTTCTTTT-5’-Y-5’-TTTTCTTTT-3’
5995.0
5991.1
5995.8
5996.5
5996.3
–
–
–
ON6
4th Postsynth.
3’-TTTTCTTTT-5’-Z-5’-TTTTCTTTT-3’
6487.1
6491.8
a) Reaction conditions are discussed in the text. b) Determined by MALDI-TOF-MS.
terminal 2-(3-iodobenzyloxy)ethyl phosphate resulting in the formation of the 5’-5’
linker X. ON2 was isolated as the minor product, and from the molecular mass, we
assumed that it contained a larger 5’-5’ linker corresponding to the structure Y. Traces
of atmospheric O2 during postsynthesis presumably catalyzed an arylacetylenic
homocoupling either before or after Sonogashira coupling with the CPG-supported
oligonucleotide [16 – 18]. Minakawa et al. [19] showed in 2003 the possibility of a
postsynthetically Cu-catalyzed oxidative acetylenic homocoupling between the 5’-ends
of on-column hexamer ONs performed on a 500-ꢂ CPG-support. But to our
knowledge, this is the first observation of one-pot postsynthetic Sonogashira and
homocoupling between neighboring on-column ONs. The possibility of synthesizing
novel 5’-5’ linkers through two different carbon – carbon reactions performed
postsynthetically encouraged us to investigate reproducibility of X and Y, and to
verify their structures by conventional chemistry.
Reproducibility and Verification Study. Reproducibility of ON1 and ON2 were
performed via a second postsynthesis using equal amounts of reactant and reagents, but
under atmospheric O2 in order to ensure sufficient amounts of O2 for the oxidative
acetylenic homocoupling of ON2. As seen from Table 1, oligonucleotides with masses
equal to ON1 and ON2 were isolated, i.e., ON3 and ON4, respectively, still with the
same distribution in yield. To investigate the possibility of increasing the yields of the
oligonucleotide comprising the linkage Y, we performed two additional postsyntheses
(third postsynthesis: the CPG-supported oligonucleotide was treated twice with equal
Sonogashira mixtures both performed under Ar; fourth postsynsthesis: the CPG-
supported oligonucleotide was treated twice with two different Sonogashira mixtures,
first with both reactant and reagents under Ar, followed with only a reagent mixture
under atmospheric O2, as to ensure sufficient O2 for the oxidative acetylenic
homocoupling). In the case of the third postsynthesis, only an ON with a mass equal
to ON2 was isolated, i.e., ON5. It is plausible that by saturating the CPG-supported
oligonucleotide with 1,3-diethynylbenzene, a full conversion of 2-(3-iodobenzyloxy)-
ethyl into 2-{3-[(3-ethynylphenyl)ethynyl]benzyloxy}ethyl is achieved before the on-
column cross-linking takes place. This results in the exclusive formation of the linkage
Y. In the case of the fourth postsynthesis, an oligonucleotide, ON6, with surprisingly

720
Helvetica Chimica Acta – Vol. 92 (2009)
high molecular weight of 6487.1 g/mol was isolated. From the MALDI-TOF analyses, a
5’-5’ linker Z, which contains five homocoupled aryldiacetylenes with a calculated mass
of 6491.8 g/mol, was proposed. All novel 5’-5’ linked alternate-strand TFOs, i.e., ON1 –
ON6, were used in thermal-stability studies to evaluate their ability to stabilize
matched and mismatched alternate-strand triplexes (Table 2). In addition, we wanted
to use these Tm values along with MALDI-TOF data for comparison with the values
obtained using the same 5’-5’-linked alternate-strand TFOs synthesized via conven-
tional phosphoramidite chemistry as a way to verify the structures of linkers X and Y.
Table 2. Reproducibility and Verification Study: Tm [8] Data for Third-Strand Melting, Taken from UV
Melting Curves (l ¼ 260 nm)a)
Entry Synthesisb)
Sequence
D1
D1
D2
D3
pH 6.0 pH 7.2 pH 6.0 pH 6.0
ON1 1st Postsynth.
ON3 2nd Postsynth.
ON7 Amidite chem.
3’-TTTTCTTTT-5’-X-5’-TTTTCTTTT-3’ 55.5
46.5
46.0
45.5
28.5
28.5
27.5
25.5
25.5
25.0
55.5
54.5
ON2 1st Postsynth.
ON4 2nd Postsynth.
ON5 3rd Postsynth.
ON8 Amidite chem.
3’-TTTTCTTTT-5’-Y-5’-TTTTCTTTT-3’ 51.5
43.5
43.5
44.0
43.0
23.0
22.5
24.5
24.0
21.0
20.0
21.5
21.5
52.0
53.0
51.5
ON6 4th Postsynth. 3’-TTTTCTTTT-5’-Z-5’-TTTTCTTTT-3’ 54.5
46.0
30.0
24.5
D1
D2
D3
3’-GGTCAGGGAAAAGAAAATTTTCTTTTCACC-5’
5’-CCAGTCCCTTTTCTTTTAAAAGAAAAGTGG-3’
3’-GGTCAGGGAACACAAAATTTTCTTTTCACC-5’
5’-CCAGTCCCTTGTGTTTTAAAAGAAAAGTGG-3’
3’-GGTCAGGGAAAAGAAAATTGGCTTTTCACC-5’
5’-CCAGTCCCTTTTCTTTTAACCGAAAAGTGG-3’
a) c ¼ 1.5 mm of ON1 – 8 and 1.0 mm of each strand of dsDNA (D1 – D3) in 20 mm sodium cacodylate,
100 mm NaCl, 10 mm MgCl2, pH 6.0 and 7.2; duplex Tm 67.58 (D1; pH 6.0), 70.58 (D2; pH 6.0), 71.08
(D3; pH 6.0), and 68.08 (D1; pH 7.2); target regions are underlined, and bases in italics are mismatch for
TFO hybridization. b) Reaction conditions are discussed in the text.
Therefore, the phosphoramidites 5 and 10 of X and Y, respectively, were
synthesized (Scheme 2). Phosphoramidite 5 was synthesized starting from the central
1,3-diethynylbenzene, which was coupled with 2-(3-iodobenzyloxy)ethanol (1) under
Sonogashira conditions to produce the diol 3 of linker X. Phosphoramidite 10 was
synthesized starting from the central 1,4-bis(3-bromophenyl)buta-1,3-diyne (6) [20],
which was synthesized from 1-bromo-3-iodobenzene via one-pot synthesis involving a
Sonogashira reaction with (trimethylsilyl)acetylene, followed by a cleavage of the silyl
protection group to achieve an oxidative homocoupling between the arylacetylenes.
The phosphoramidites were used in standard DNA syntheses to obtain ON7 and
ON8, corresponding to ON1 and ON2, respectively. As can be seen from Table 1, equal
masses for ON1, ON3, and ON7, and ON2, ON4, ON5, and ON8, respectively, were
determined. Final verification of the linker structures X and Y was achieved via
identical melting curves of thermal denaturation experiments of triplexes, shown as
first derivative plots in Fig. 2. Obtained Tm values (within the uncertainty of the first

Helvetica Chimica Acta – Vol. 92 (2009)
721
Scheme 2. Synthesis of the phosphoramidite of X and Y
i) 2-(3-Iodobenzyloxy)ethanol (1), Pd(PPh3)2Cl2, CuI, Et3N, overnight. ii) Dimethoxytrityl chloride
(¼ bis(4-methoxyphenyl)(phenyl)methyl; DMT-Cl), Et3N, CH2Cl2, overnight. iii) NC(CH2)2OP-
(NiPr2)2, diisopropylammonium tetrazolide, CH2Cl2, overnight. iv) Pd(PPh3)2Cl2, CuI, PPh3, (trime-
thylsilyl)acetylene (TMSA), Et3N, 17 h; followed by Et3N · 3 HF, 2 h.
derivative method for determination of Tm [21]) for Hoogsteen-matched (i.e., D1) and
Hoogsteen-mismatched triplexes (i.e., D2 and D3) are shown in Table 2.
As it can be seen from Table 2, all TFOs formed highly stabile triplexes with the
complementary duplex D1 at pH 6.0 and 7.2, respectively. In addition, it was established
that all TFOs formed 5’-5’-linked alternate-strand triplexes at pH 6.0, as a large drop in
Tm was observed upon insertion of pyrimidine mismatches in the homopurine tracts,
e.g., in D2 and D3, corresponding to loss of stabilizing Hoogsteen hybridization.
TFOs with the linker X were found to form more stabile triplexes than TFOs with
the linker Y (Tm(ON7/D1) ¼ 54.58 and Tm(ON8/D1) ¼ 51.58, resp., at pH 6.0. Jessen
and Pedersen [10] showed in 2004 that the 1,3-bis(phenylethynyl)benzene moiety of
linker X is able to form p – p interaction with all four strands at the specific point of
alternation. Therefore, the lower triplex stability observed for the use of Y could be the
result of poorer p – p interaction caused by the larger structure of the 1,4-bis(3-
(phenylethynyl)phenyl)buta-1,3-diyne moiety in addition to its lower rigidity when

722
Helvetica Chimica Acta – Vol. 92 (2009)
Fig. 2. First-derivative plots of thermal denaturation experiments of triplexes: a) ON1, ON3, and ON7/D1,
b) ON2, ON4, ON5, and ON8/D1 (recorded in 20 mm sodium cacodylate, 100 mm NaCl, 10 mm MgCl2,
pH 6.0 and 7.2, at 260 nm vs. temperature, with a heating of 1.08/min)
compared with X. Surprisingly, the much larger linkage Z (in ON6) was able to stabilize
the triplex better than Y, the Tm values being nearly identical to those found for X.
Molecular Modelling. To assess the novel linkersꢃ ability to intercalate the triplex, as
well as to evaluate the variations in Tm values for Y and the unexpected high Tm values
observed for Z, we performed molecular-modelling studies using a modified AMBER*
force field to generate representative low-energy structures with linkers X, Y, and Z
(Fig. 3,a – h). As can be seen from Fig. 3,a and e, linker X intercalates the triplex,
positioning the central 1,3-diethynylbenzene moiety and the peripheral aromatic rings
so that they form p – p interaction with the Watson – Crick duplex and the adjacent
nucleobases of the TFO, respectively. From molecular dynamics of linker Y, three low-
energy structures were found (Fig. 3,b). Even though the intercalating part of Y was
found to be similar to that of linker X (Fig. 3,f), the diacetylene moiety resulted in two
benzene rings to be positioned outside the triplex where no interaction with the
surrounding nucleobases could be formed. We believe that the unfixed part of Y will
wobble increasingly with elevated temperatures during thermal denaturation measure-
ments, resulting in a lowering of thermal stability (Table 2). To our surprise, the much
larger structure of linker Z was able to form a helical macrocycle via intramolecular p –
p interactions in addition to triplex intercalation (Figs. 3,c and g). Hence, the
formation of intramolecular p – p interactions in the linker would prevent wobbling and
thereby lowering of Tm during thermal denaturation measurements as asserted for the
linker Y. As can be seen from Fig. 3,g, the intercalating part of Z is equal to that of X,
which could explain why similar Tm values for Z and X were observed.
Optimized 5’-5’ Intercalator. These findings encouraged us to synthesize a novel
linker, W, where the central diacetylene of Y was replaced with a single CꢂC bond

Helvetica Chimica Acta – Vol. 92 (2009)
723

724
Helvetica Chimica Acta – Vol. 92 (2009)
(Fig. 3,d and e). By shortening the diacetylene moiety of linker Y, a linker, which
would fit better within the base triplets and thereby eliminate the possibility of
wobbling, could be obtained. The phosphoramidites 14 and 10 (Schemes 3 and 2, resp.)
of linkers Wand Y, respectively, were used for the synthesis of TFOs complementary to
the HIV-1 sequence D4 (Table 3) [22][23].
Scheme 3. Synthesis of the Phosphoramidite of W
i) 2-(3-Iodobenzyloxy)ethanol (1), Pd(PPh3)2Cl2, CuI, Et3N, 18 h. ii) DMT-Cl, Et3N, CH2Cl2, 15 h. iii)
NC(CH2)2OP(NiPr2)2, diisopropylammonium tetrazolide, CH2Cl2, 18 h.
Table 3. Tm [8] Data for Third-Strand Melting, Taken from UV Melting Curves (l ¼ 260 nm)a)
Entry
Sequence
Mass [m/z]
D4
D4
D5
D6
Found Calc.
pH 6.0 pH 7.2 pH 6.0 pH 6.0
ON9
3’-TTTTCTTTT-5’-Y-5’-TTTTT-3’
3’-TTTTCTTTT-5’-Y-5’
3’-TTTTCTTTT-5’-W-5’-TTTTT-3’ 4783.2 4784.4 46.0
3’-TTTTCTTTT-5’-W-5’ 3263.1 3263.4 31.0
16.0
4806.8 4808.4 39.5
3287.2 3287.5 23.5
35.0
15.5
42.0
23.5
< 5.0
16.0
< 5.0
22.0
19.5
–
26.0
–
ON10
ON11
ON12
23.0
ON13b) 3’-TTTTCTTTT-5’
–
–
n.d.c)
–
D4
D5
D6
3’-GGTCAGGGGGGAAAAGAAAATTTTTCACC-5’
5’-CCAGTCCCCCCTTTTCTTTTAAAAAGTGG-3’
3’-GGTCAGGGGGGAAAAGAAAGTTTTTCACC-5’
5’-CCAGTCCCCCCTTTTCTTTCAAAAAGTGG-3’
3’-GGTCAGGGGGGAAAAGAAAATTGGTCACC-5’
5’-CCAGTCCCCCCTTTTCTTTTAACCAGTGG-3’
a) c ¼ 1.5 mm of ON9 – 13 and 1.0 mm of each strand of dsDNA (D4 – D6) in 20 mm sodium cacodylate,
100 mm NaCl, 10 mm MgCl2, pH 6.0 and 7.2; duplex Tm 68.08 (D4; pH 6.0), 70.08 (D5; pH 6.0), 70.08
(D6; pH 6.0), and 68.08 (D4, pH 7.2); target regions are underlined, and bases in italics are mismatch for
TFO hybridization; C is 5-methylcytosine. b) Oligonucleotides and Tm data are from [11]. c) n.d. ¼ Not
determined.
This sequence is normally targeted at the 16-mer G-rich polypurine tract (PPT)
which, for the parallel motif, is strongly pH-dependent. The dsDNA HIV-1 target
allows us to use an asymmetric alternate-strand TFO consisting of a 14-mer pyrimidine
sequence containing only one 5-methylcytosine to be protonated for triplex formation.

Helvetica Chimica Acta – Vol. 92 (2009)
725
Indeed, only a moderate decrease in the triplex melting temperature was observed for
ON9 and ON11 going from pH 6.0 to 7.2 (DTm ¼ ꢀ4.5 and ꢀ 4.08, resp.). The 5-Me
derivative was selected because, at physiological pH, it is easier to protonate than the
native cytosine [4]. Mismatch studies using the dsDNAs D5 and D6 were used to
confirm that both alternate strands hybridized to purine regions of dsDNA (Table 3).
As it was expected from molecular-modelling studies, the use of linker W instead of Y
led to increased thermal stability at pH 6.0 and 7.2: DTm(ON11/D4 ꢀ ON9/D4) ¼ þ6.58
and DTm(ON11/D4 ꢀ ON9/D4) ¼ þ7.08, respectively (Table 3). Furthermore, im-
proved intercalating properties of W in comparison with Y was observed as an
increase in Tm value when W and Y were used as terminal 5’-intercalators: DTm(ON12/
D4 ꢀ ON13/D4) ¼ þ15.08 compared to DTm(ON10/D4 ꢀ ON13/D4) ¼ þ7.58. Upon
insertion of mismatch in either side of the purine tract (i.e., D5 and D6), thermal
stability dropped significantly for ON9 and ON11 at pH 6.0. In addition, thermal
stability dropped considerably for the two non-alternating triplexes ON10/D5 and
ON12/D5. These observations are a clear indication of alternate-strand triplex
formation for ON9 and ON11. Besides, it is worth mentioning that the mismatch
studies also show that ON9 and ON11 are sequence-specific with a good mismatch
discrimination (DTm(ON9/D5 ꢀ ON9/D4) ¼ ꢀ23.58 and DTm(ON11/D5 ꢀ ON11/
D4) ¼ ꢀ24.08).
Conclusions. – We have successfully demonstrated a new technique in postsyn-
thetical modifications of on-column oligonucleotides. Via Pd0-catalyzed postsynthetic
reaction with CPG-supported oligodeoxynucleotides possessing 2-(3-iodobenzyloxy)-
ethyl phosphate at the terminal 5’-ends and 1,3-diethynylbenzene, three novel
intercalating 5’-5’ linkers, X, Y, and Z, in alternate-strand TFOs were produced.
Reproducibility, and verification of X and Y were achieved through independent
postsyntheses and conventional phosphoramidite chemistry in DNA synthesis,
respectively. All novel 5’-5’ linkers were shown to form stabile alternate-strand
Hoogsteen-type triplexes at physiological pH. Molecular modelling was used to study
the p – p interaction between linker and surrounding nucleobases leading to the
structurally optimized linker W. An interesting feature of using the alternate-strand
triplexes is the shift from a G-rich target sequence used in the HIV-1 polypurine tract to
an alternate sequence with only one G which makes the TFO less pH-dependent in the
parallel motif. A 14-mer oligonucleotide with the linker W showed in the latter case low
pH dependence (DTm(pH 7.2 – pH 6.0) ¼ ꢀ4.08) and high thermal stability at physio-
logical pH (Tm(pH 7.2) ¼ 42.08).
Experimental Part
General. M.p.: Bꢀchi melting-point apparatus; uncorrected. TLC: Plates 60 F254 (Merck);
visualization with UV light (254 nm). Column chromatography (CC): silica gel (SiO2 ; 0.040 –
0.063 mm, Merck). Solvents for CC and reagents were used as purchased without further purification.
NMR Spectra: Varian Gemini 2000 spectrometer in CDCl3 at 300 (1H), 75 (13C), and 121.5 MHz (31P),
resp.; d in ppm rel. to Me4Si (d 0.00; 1H) and CDCl3 (d 77.00; 13C) as internal standards, and H3PO4 (d
0.00, 31P) as external standard; J in Hz. Accurate ion mass determinations: 4.7 Tesla Ultima Fourier
Transform (FT) mass spectrometer (Ion Spec, Irvine, CA). The [M þ Na]þ ions were peaks matched

726
Helvetica Chimica Acta – Vol. 92 (2009)
using ions derived from the 2,5-dihydroxybenzoic acid matrix. EI-MS: Finnigan SSQ 710; in m/z (rel. %).
MALDI-TOF-MS: MicroFlex LT from Bruker Daltronics.
1. 2-(3-Iodobenzyloxy)ethanol (1). Compound 1 was synthesized according to the procedure in [10]
using 3-iodobenzyl bromide as reactant instead of 4-iodobenzyl bromide. Yield 86%. Yellow oil.
1H-NMR: 2.44 (br. s, OH); 3.58 – 3.61 (m, CH2OH); 3.75 (t, J ¼ 4.5, CH2CH2OH); 4.48 (s, C6H4CH2O);
7.07 (t, J ¼ 7.8, 1 arom. H); 7.29 (d, J ¼ 7.5, 1 arom. H); 7.61 (d, J ¼ 7.8, 1 arom. H); 7.69 (s, 1 arom. H).
13C-NMR: 61.7 (CH2OH); 71.6 (OCH2CH2OH); 72.2 (C6H4CH2O); 94.3, 126.7, 130.1, 136.5, 136.7, 140.3
(C6H4). HR-MALDI-MS: 300.9700 ([M þ Na]þ, C9H11INaO2þ ; calc. 300.9701).
2. 2-({[3-(3-{3-[(2-Hydroxyethoxy)methyl]phenyl}ethynyl)phenyl]ethynyl}benzyloxy)ethanol (3).
1,3-Diethynylbenzene (360 mg, 2.86 mmol) was dissolved in dry Et3N (50 ml) and flushed with Ar
before CuI (56 mg, 0.29 mmol) and Pd(PPh3)2Cl2 (130 mg, 0.19 mmol) were added. Compound 1 (1.77 g,
6.37 mmol) was added via a syringe, and the mixture was stirred at r.t. under Ar for 22 h. The solvent was
removed under reduced pressure, and the residue was dissolved in CH2Cl2 (100 ml) which was washed
with 0.3m aq. EDTA (2 ꢁ 100 ml). After back-extraction with CH2Cl2 (100 ml), the combined org. phases
were washed with brine (100 ml), dried (MgSO4), filtered, and concentrated under reduced pressure.
The residue was purified by CC (SiO2; 0 – 85% AcOEt in cyclohexane (v/v)) to afford 3 (1.16 g, 43%).
Yellow oil. 1H-NMR: 3.62 (t, J ¼ 4.6, 4 H, 2 OCH2CH2OH); 3.78 (t, J ¼ 4.6, 4 H, 2 OCH2CH2OH); 4.57 (s,
4 H, 2 C6H4CH2O); 7.28 – 7.71 (m, 12 arom. H). 13C-NMR: 61.8 (CH2OH); 71.6 (OCH2CH2OH); 72.7
(C6H4CH2O); 88.6, 89.8 (C ꢂ C); 123.1, 123.5, 127.7, 128.5, 130.8, 130.9, 131.3, 134.6, 138.4 (3 C6H4). HR-
MALDI-MS: 449.1707 ([M þ Na]þ, C28H26NaO4þ ; calc. 449.1729).
3. 1,4-Bis(3-bromophenyl)buta-1,3-diyne (6). 1-Bromo-3-iodobenzene (2.40 g, 8.5 mmol) was
dissolved in dry THF (30 ml) and dry Et3N (2 ml), and flushed with Ar, before Pd(PPh3)2Cl2 (122 mg,
0.17 mmol) and CuI (89 mg, 0.47 mmol) were added. (Trimethylsilyl)acetylene (1.00 g, 10.5 mmol) was
added via a syringe, and the mixture was stirred at r.t. under Ar for 21 h, before Et3N · 3 HF (1.20 g,
7.7 mmol) was added. After 30 min, additional CuI (80 mg, 0.42 mmol) was added, and the mixture was
flushed with O2, closed with a septum, and stirred at r.t. under O2 for 22 h. The solvent was removed
under reduced pressure, and the residue was dissolved in CH2Cl2 (50 ml), which was washed with 0.3m aq.
EDTA (100 ml), H2O (100 ml), brine (100 ml), and 10% HCl (100 ml). The org. phase was dried
(MgSO4), filtered, and concentrated under reduced pressure. The residue was purified by dry column
vacuum chromatography (CVC; SiO2; 0 – 50% AcOEt in cyclohexane (v/v)) to afford 6 (1.09 g, 71%).
Pale yellow solid. M.p. 97 – 988. 1H- and 13C-NMR: identical to those reported in [20]. EI-MS: 362 (50,
[M þ 2]þ), 360 (100, [C16H8Br2]þ), 358 (50, [M ꢀ 2]þ), 200 (70, [M ꢀ 2 Br]þ).
4. 1,4-Bis(3-ethynylphenyl)buta-1,3-diyne (7). Compound 6 (1.99 g, 5.5 mmol), Pd(PPh3)2Cl2
(287 mg, 0.41 mmol), CuI (130 mg, 0.68 mmol), and powdered PPh3 (331 mg, 1.13 mmol) were dissolved
in dry Et3N (75 ml). The mixture was flushed with Ar before closing with a septum. After addition of
(trimethylsilyl)acetylene (3.30 g, 33.6 mmol), the mixture was stirred and refluxed under Ar for 17 h. The
solvent was removed under reduced pressure, and the residue was dissolved in CH2Cl2 (100 ml), which
was washed with H2O (100 ml) and 0.3m aq. EDTA (2 ꢁ 100 ml). After back-extraction of aq. layers with
CH2Cl2 (100 ml), the combined org. phases were washed with brine (100 ml), dried (MgSO4), filtered,
and concentrated under reduced pressure. The residue was dissolved in dry CH2Cl2 (50 ml), and Et3N ·
3 HF (948 mg, 5.88 mmol) was added. After 2 h, the mixture was concentrated under reduced pressure
and purified by dry CVC (SiO2; cyclohexane) to afford 7 (902 mg, 65%). Pale yellow solid. M.p. 122 –
1
1268. H-NMR: 3.10 (s, 2 C ꢂ CH); 7.27 – 7.32 (m, 2 arom. H); 7.46 – 7.51 (m, 4 arom. H); 7.64 (t, J ¼
1.5, 2 arom. H). 13C-NMR: 74.4 (C ꢂ CꢀC ꢂ C); 78.2 (C ꢂ CH); 80.7 (C ꢂ CH); 82.4 (C ꢂ CꢀC ꢂ C);
122.0, 122.7, 128.6, 132.7, 132.8, 135.9 (2 C6H4). EI-MS: 250 (100, Mþ), 248 (30, [M ꢀ 2 H]þ).
5. 2-{3-[(3-{4-[3-({3-(2-Hydroxyethoxy)methyl]phenyl}ethynyl)phenyl]buta-1,3-diynyl}phenyl)ethy-
nyl]benzyloxy}ethanol (8). Compound 7 (550 mg, 2.20 mmol) was dissolved in dry Et3N (50 ml) and
flushed with Ar, before CuI (32 mg, 0.17 mmol) and Pd(PPh3)2Cl2 (84 mg, 0.12 mmol) were added.
Compound 1 (1.44 g, 5.2 mmol) was added via a syringe, and the mixture was stirred at r.t. under Ar for
17 h. The solvent was removed under reduced pressure, and the residue was dissolved in CH2Cl2 (100 ml)
which was washed with 0.3m aq. EDTA (2 ꢁ 100 ml). After back-extraction with CH2Cl2 (100 ml), the
combined org. phases were washed with brine (100 ml), dried (MgSO4), filtered, and concentrated under
reduced pressure. Crystallization from MeCN gave 8 (1.04 g, 86%). Yellow solid. M.p. 130 – 1358.

Helvetica Chimica Acta – Vol. 92 (2009)
727
1H-NMR: 2.15 (br. s, 2 CH2OH); 3.60 – 3.63 (m, 2 CH2OH); 3.78 (br. s, 2 CH2CH2OH); 4.56 (s,
C6H4CH2O); 7.30 – 7.37 (m, 6 arom. H); 7.45 – 7.69 (m, 8 arom. H); 7.69 (s, 2 arom. H). 13C-NMR: 61.9
(CH2OH); 71.5 (CH2CH2OH); 72.7 (C6H4CH2O); 74.3 (C ꢂ CꢀC ꢂ C); 80.9, 88.3 (C6H4ꢀC ꢂ CꢀC6H4);
90.1 (C ꢂ CꢀC ꢂ C); 122.0, 122.9, 123.7, 127.8, 128.6, 128.6, 130.8, 131.0, 132.2, 132.3, 135.4, 138.4
(4 C6H4). EI-MS: 550 (100, Mþ), 440 (60, [M ꢀ C4H10O3]þ).
6. 2-{3-[(3-{[3-({3-[(2-Hydroxyethoxy)methyl]phenyl}ethynyl)phenyl]ethynyl}phenyl)ethynyl]ben-
zyloxy}ethanol (12). 1,2-Bis(3-ethynyl-phenyl)ethyne (11 [24]; 455 mg, 2.01 mmol) was dissolved in dry
Et3N (50 ml) and flushed with Ar, before CuI (34 mg, 0.18 mmol) and Pd(PPh3)2Cl2 (72 mg, 0.10 mmol)
were added. Compound 1 (1.34 g, 4.8 mmol) was added via a syringe, and the mixture was stirred at r.t.
under Ar for 18 h. The solvent was removed under reduced pressure, and the residue was dissolved in
CH2Cl2 (100 ml) which was washed with 0.3m aq. EDTA (2 ꢁ 100 ml). After back-extraction with CH2Cl2
(100 ml), the combined org. phase was washed with brine (100 ml), dried (MgSO4), filtered, and
concentrated under reduced pressure. The residue was purified by dry CVC (SiO2; 0 – 50% AcOEt in
1
cyclohexane (v/v)) to afford 12 (674 mg, 64%). Yellow solid. M.p. 124 – 1308. H-NMR: 3.60 – 3.63 (m,
2 CH2OH); 3.76 – 3.81 (m, 2 CH2CH2OH); 4.56 (s, 2 C6H4CH2O); 7.31 – 7.37 (m, 6 arom. H); 7.45 – 7.71
(m, 8 arom. H); 7.71 (s, 2 arom. H). 13C-NMR: 61.9 (CH2OH); 71.5 (CH2CH2OH); 72.8 (C6H4CH2O);
88.7, 89.1, 89.9 (C ꢂ C); 123.2, 123.4, 123.6, 127.8, 128.5, 128.6, 130.9, 131.0, 131.4, 131.5, 134.7, 138.4
(4 C6H4). EI-MS: 526 (30, Mþ), 420 (35, [M ꢀ C4H10O3]þ), 44 (100, [CH2OCH2]þ).
7. General Procedure for DMT Protection of the Diol. The diol was dissolved in dry CH2Cl2 and dry
Et3N (2.5 equiv.), and the mixture was flushed with Ar. DMT-Cl (1.2 equiv.), dissolved in dry CH2Cl2 and
dry Et3N (0.5 equiv. in relation to DMT-Cl), was added dropwise under vigorous stirring. The mixture
was stirred at r.t. under Ar overnight, before the reaction was quenched with H2O. The org. phase was
dried (MgSO4), filtered, and concentrated under reduced pressure, before the residue was purified by dry
CVC (SiO2; 1% Et3N v/v, 0 – 50% AcOEt in cyclohexane) to afford the desired DMT-protected alcohol.
8. 2-{3-[(3-{[3-({2-[Bis(4-methoxyphenyl)phenylmethoxy]ethoxy}methyl)phenyl]ethynyl}phenyl)-
ethynyl]benzyloxy}ethanol (4). Yield: 38%. Yellow oil. 1H-NMR: 3.28 (t, J ¼ 5.1, CH2O-DMT); 3.61
(t, J ¼ 4.5, OCH2CH2OH); 3.69 (t, J ¼ 5.1, OCH2CH2O-DMT); 3.76 (s, 2 MeO); 3.76 – 3.80 (m, CH2OH);
4.57 (s, C6H4CH2O); 4.61 (s, C6H4CH2O); 6.80, 6.83, 7.19 – 7.70 (m, 21 arom. H). 13C-NMR: 55.30, 55.31
(2 CH2O); 62.1 (CH2OH); 63.5 (CH2O-DMT); 70.1 (CH2CH2O-DMT); 71.6 (CH2CH2OH); 72.7
(C6H4CH2O); 72.9 (C6H4CH2O); 86.1 (Ar3C); 88.7, 88.9, 89.9, 90.2 (2 C ꢂ C); 113.2, 123.2, 123.3, 123.7,
123.8, 126.8, 127.7, 127.9, 128.4, 128.6, 128.6, 128.7, 130.2, 130.7, 130.8, 131.0, 131.1, 131.4, 131.5, 134.8, 136.5,
138.5, 139.2, 145.2, 158.5 (36 arom. C). HR-MALDI-MS: 751.3012 ([M þ Na]þ, C49H44NaO6þ ; calc.
751.3036).
9. 2-[3-({3-[4-(3-{[3-({2-[Bis(4-methoxyphenyl)phenylmethoxy]ethoxy}methyl)phenyl]ethynyl}phe-
nyl)buta-1,3-diynyl]phenyl}ethynyl)benzyloxy]ethanol (9). Yield: 43%. Yellow oil. 1H-NMR: 3.29 – 3.30
(m, CH2O-DMT); 3.60 – 3.63 (m, CH2OH); 3.67 – 3.71 (m, CH2CH2OH); 3.76 (s, 2 MeO); 3.76 – 3.80 (m,
CH2CH2O-DMT); 4.57, 4.61 (s, 2 C6H4CH2O); 6.82 (d, J ¼ 7.1, 4 H, DMT); 7.25 – 7.69 (m, 25 arom. H).
13C-NMR: 55.1 (2 MeO); 61.9 (CH2OH); 63.3 (CH2O-DMT); 70.0 (CH2CH2O-DMT); 71.5
(CH2CH2OH); 72.5, 72.7 (2 C6H4CH2O); 74.3 (C ꢂ CꢀC ꢂ C); 80.8, 80.9, 88.1, 88.3 (2 C6H4ꢀC ꢂ
CꢀC6H4); 86.0 (Ar3C); 90.1, 90.4 (C ꢂ CꢀC ꢂ C); 113.0, 126.7, 127.6, 127.7, 127.8, 130.1, 136.3, 144.1,
158.4 (DMT); 122.0, 122.1, 123.8, 123.8, 128.2, 128.4, 128.6, 128.6, 130.6, 130.7, 130.8, 131.0, 132.1, 132.2,
132.3, 135.4, 138.4, 139.1 (4 C6H4). HR-MALDI-MS: 875.3323 ([M þ Na]þ, C59H48NaO6þ ; calc.
875.3349).
10. 2-[3-({3-[(3-{[3-({2-[Bis-(4-methoxyphenyl)phenylmethoxy]ethoxy}methyl)phenyl]ethynyl}phe-
nyl)ethynyl]phenyl}ethynyl)benzyloxy]ethanol (13). Yield: 46%. Yellow oil. 1H-NMR: 3.28 – 3.31 (m,
CH2O-DMT); 3.61 – 3.78 (m, CH2OH, CH2CH2OH, CH2CH2O-DMT); 3.76 (s, 2 MeO); 4.57, 4.61 (s,
2 C6H4CH2O); 6.82 (d, J ¼ 6.8, 4 H, DMT); 7.29 – 7.71 (m, 25 arom. H). 13C-NMR: 55.2 (2 MeO); 61.9
(CH2OH); 63.3 (CH2O-DMT); 70.0 (CH2CH2O-DMT); 71.5 (CH2CH2OH); 72.6, 72.8 (2 C6H4CH2O);
86.0 (Ar3C); 88.6, 89.2, 90.1 (C ꢂ C); 113.1, 126.7, 127.5, 130.1, 136.3, 145.1, 158.4 (DMT); 123.0, 123.2,
123.3, 123.4, 123.6, 123.7, 127.8, 128.3, 128.4, 128.5, 128.5, 130.5, 130.7, 130.9, 131.0, 131.3, 131.4, 131.4,
131.5, 132.0, 132.2, 134.6, 138.4, 139.0 (C6H4). HR-MALDI-MS: 851.3369 ([M þ Na]þ, C57H48NaO6þ ; calc.
851.3349).

728
Helvetica Chimica Acta – Vol. 92 (2009)
11. General Procedure for the Formation of Phosphoramidites. The alcohol and diisopropylammo-
nium tetrazolide (1.5 equiv.) were dissolved under Ar in dry CH2Cl2 or dry MeCN, followed by dropwise
addition of 2-cyanoethyl N,N,N’,N’-tetraisopropylphosphordiamidite (2.1 equiv.) via a syringe at 08. The
mixture was stirred under Ar at r.t. overnight, before the reaction was quenched with H2O, and the org.
phase was washed with H2O. After back-extraction with CH2Cl2, the combined org. phases were dried
(MgSO4), filtered, and concentrated under reduced pressure. The residue was purified by dry CVC
(SiO2; Et3N 0.05% (v/v), 0 – 100% AcOEt in cyclohexane) to afford the desired phosphoramidite.
12. 1-{2-[(2-cyanoethoxy)(diisopropylamino)phosphinoxy]ethoxy}-3-iodobenzene (2). Yield: 25%.
Clear oil. 13C-NMR: 20.3, 20.4 (CH2CN); 24.5, 24.6, 24.6, 24.7 (2 Me2CH); 43.0, 43.1 (2 Me2CH); 58.3,
58.6 (OCH2CH2CN); 62.5, 62.8 (OCH2CH2OP); 70.4, 70.5 (CH2CH2OP); 72.2, (C6H4CH2O); 117.6
(CN); 94.3, 126.6, 130.1, 136.4, 136.6, 104.7 (C6H4). 31P-NMR: 149.0. HR-MALDI-MS: 501.0761 ([M þ
Na]þ, C18H28IN2NaO3Pþ; calc. 501.0780).
13. 1-[(3-{2-[(2-Cyanoethyl)(diisopropyl)phosphinoxy]ethoxy}benzyloxy)ethynyl]-3-[(3-{2-[bis(4-
methoxyphenyl)phenylmethoxy]ethoxy}benzyloxy)ethynyl]benzene (5). Yield: 45%. Yellow oil.
13C-NMR: 20.3, 20.4 (CH2CN); 24.5, 24.6, 24.6, 24.7 (2 Me2CH); 43.0, 43.2 (2 Me2CH); 55.1 (2 MeO);
58.3, 58.6 (OCH2CH2CN); 62.6, 62.8 (OCH2CH2OP); 63.3 (CH2O-DMT); 70.0 (CH2CH2O-DMT); 70.3,
70.4 (CH2CH2OP); 72.57, 72.63 (2 C6H4CH2O); 86.0 (Ar3C); 88.5, 88.6, 89.9, 90.0 (2 C6H4ꢀC ꢂ
CꢀC6H4); 113.0, 126.6, 127.5, 127.7, 127.7, 130.1, 136.3, 145.1, 158.4 (DMT); 123.0, 123.5, 123.6, 128.2,
128.4, 128.4, 130.5, 130.7, 130.7, 130.8, 131.2, 131.3, 134.6, 138.7, 139.0 (3 C6H4). 31P-NMR: 149.6.
14. 1-{3-[(3-{2-[Bis(4-methoxyphenyl)phenylmethoxy]ethoxy}benzyloxy)ethynyl]phenyl}-4-{3-[(3-
{2-[(2-cyanoethoxy)(diisopropyl)phosphinoxy]ethoxy}benzyloxy)ethynyl]phenyl}buta-1,3-diyne (10).
Yield: 51%. Yellow oil. 13C-NMR: 20.2, 20.4 (CH2CN); 24.5, 24.6, 24.7 (2 Me2CH); 43.0, 43.2
(2 Me2CH); 55.2 (2 MeO); 58.3, 58.7 (OCH2CH2CN); 62.6, 62.9 (OCH2CH2OP); 63.3 (CH2O-DMT);
70.0 (CH2CH2O-DMT); 70.3, 70.4 (CH2CH2OP); 72.6, 72.6 (2 C6H4CH2O); 74.4 (C ꢂ CꢀC ꢂ C); 81.0,
88.2 (C6H4ꢀC ꢂ CꢀC6H4); 86.0 (Ar3C); 90.3 (C ꢂ CꢀC ꢂ C); 117.6 (CN); 113.1, 126.7, 127.6, 127.7, 130.1,
136.3, 145.1, 158.4 (DMT); 122.0, 122.1, 122.9, 123.8, 123.9, 128.2, 128.4, 128.6, 128.6, 130.6, 130.7, 130.8,
130.8, 132.1, 132.3, 132.3, 135.4, 138.7, 139.1 (4 C6H4). 31P-NMR: 149.7.
15. 1-{3-[(3-{2-[Bis(4-methoxyphenyl)phenylmethoxy]ethoxy}benzyloxy)ethynyl]phenyl}-2-{3-[(3-
{2-[(2-cyanoethoxy)(diisopropyl)phosphinoxy]ethoxy}benzyloxy)ethynyl]phenyl}ethyne (14). Yield:
50%. Yellow oil. 13C-NMR: 20.3, 20.4 (CH2CN); 24.5, 24.6, 24.6, 24.7 (2 Me2CH); 43.0, 43.2 (2 Me2CH);
55.1 (2 MeO); 58.3, 58.6 (OCH2CH2CN); 62.6, 62.9 (OCH2CH2OP); 63.3 (CH2O-DMT); 70.0
(CH2CH2O-DMT); 70.3, 70.4 (CH2CH2OP); 72.6, 72.6 (2 C6H4CH2O); 86.0 (Ar3C); 88.5, 88.5, 89.1,
89.2, 90.0, 90.1 (C ꢂ C); 113.0, 126.7, 127.5, 127.7, 127.8, 130.1, 136.3, 145.1, 158.4 (DMT); 123.0, 123.3,
123.4, 123.6, 123.7, 128.2, 128.4, 128.5, 128.5, 128.5, 130.6, 130.7, 130.8, 130.8, 131.3, 131.3, 131.4, 131.5,
134.6, 138.7, 139.0 (4 C6H4). 31P-NMR: 149.6.
16. Molecular Modelling. Molecular modelling was performed with Macro Model v9.1 from
Schrçdinger. All calculations were conducted with AMBER* force field and the GB/SA water model. A
parameter for the single bond in the diacetylene was added to the AMBER* force field using the same
binding constants as for the C ꢂ C bond but with a bond length of 1.3837 ꢂ [25][26]. The dynamic
simulations were performed with stochastic dynamics, a SHAKE algorithm to constrain bonds to H-
atoms, 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 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 structures were generated with Insight II
v97.2 from MSI, followed by incorporation of the modified nucleotide.
17. Synthesis and Purification of Modified Oligonucleotides. Modified oligonucleotides were
synthesized in a 0.2-mmol or 1-mmol scale for postsynthesis on 500-ꢂ CPG supports using Expedite
Nucleic Acid Synthesis System model 8909 (Applied Biosystems). Standard procedures were used for the
coupling of commercial phosphoramidites, whereas modified phosphoramidites, 2, 5, 10, and 14, were
coupled with 1H-tetrazole as an activator and an extended coupling time (10 min). ONs were cleaved
from the CPG support with 32% aq. NH3 (1.2 ml) and deprotected at 558 overnight. ON1 – ON5 were
purified with NAPTM 10 columns from GE Healthcare Bio-Sciences AB before RP-HPLC. Purification of
ONs was carried out on semiprep. RP-HPLC on a Waters Xterra MS C18 column (10 mm, 7.8 ꢁ 150 mm).

Helvetica Chimica Acta – Vol. 92 (2009)
729
ON6 and ON7 were desalted with NAPTM 10 columns from GE Healthcare Bio-Sciences AB. ON8 –
ON13 were submitted to DMT-deprotection with 80% aq. AcOH (100 ml) for 20 min, followed by
addition of H2O (100 ml) and 3m aq. AcONa (50 ml), before precipitation from 99.9% EtOH (550 ml) at
ꢀ 188. The purity of the obtained ONs was checked by ion-exchange chromatography on a LaChrom
system (Merck Hitachi) using a GenPak-Fax column (Waters). Verification was achieved by MALDI-
TOF analysis on a Voyager Elite Bio spectrometry research station (Perspective Biosystems).
18. Postsynthesis. In the case of ON1 – ON3, a fresh mixture of Pd(PPh3)4 (22 mmol), CuI (26 mmol),
and 1,3-diethynylbenzene (38 mmol) in dry DMF/Et3N (350 ml/150 ml) was prepared in a 1-ml syringe,
which was flushed with Ar prior use. The CPG support was flushed with Ar before the syringe containing
the Sonogashira reagent mixture was attached to one end, and an empty 1-ml syringe was attached to the
other end. The CPG support was treated several times with the reagent mixture every 45 min. After a
coupling time of 3 h, the CPG support was washed with dry DMF (2 ꢁ 2 ml) and dry MeCN (2 ꢁ 2 ml),
before it was dried with a flow of Ar. In the case of ON4 and ON5, the CPG support was treated with an
equal mixture as used for ON1 – ON3 but without the use of Ar. In the case of ON6 and ON7, the
postsynthesis were carried out in a similar way as for ON1 – ON3, using a fresh mixture of Pd(PPh3)4
(22 mmol), CuI (26 mmol), and 1,3-diethynylbenzene (38 mmol) in dry DMF/Et3N (350 ml/150 ml). The
CPG supports were treated again under the same conditions and with equiv. reagent mixtures with the
exception of 1,3-diethynylbenzene and the use of Ar in the case of ON7.
19. Melting-Temperature Measurements. Tm Measurements were performed on a Perkin-Elmer
Lambda 35 UV/VIS spectrometer with a PTP 6 thermostat and Perkin-Elmer Templab 2.00 Software.
Triplexes were formed by mixing 1.0 mm of each ssDNA and 1.5 mm of the TFO in the corresponding
buffer soln. The solns. were heated to 808 for 5 min, and afterward cooled to 58 and kept at this temp. for
30 min. The absorbance of triplexes was measured at 260 nm from 5 to 808 with a heating rate of 1.08/min
or 0.58/min. The Tm [8] values were determined as the maximum of the first-derivative plots of the melting
curves. All Tm values are within the uncertainty ꢃ 0.58 as determined by repetitive experiments.
REFERENCES
[1] R. V. Guntaka, B. R. Varma, K. T. Weber, Int. J. Biochem. Cell. Biol. 2003, 35, 22.
´ `
[2] D. Praseuth, A. L. Guieysse, C. Helene, Biochim. Biophys. Acta 1999, 1489, 181.
[3] M. Faria, C. Giovannangeli, J. Gene Med. 2001, 3, 299.
[4] L. E. Xodo, G. Manzini, F. Quadrifoglio, G. A. van der Marcel, J. H. van Boom, Nucleic Acids Res.
1991, 19, 5625.
[5] P. L. James, T. Brown, K. R. Fox, Nucleic Acids Res. 2003, 31, 5598.
[6] V. V. Filichev, E. B. Pedersen, J. Am. Chem. Soc. 2005, 127, 14849.
[7] M. Paramavisam, C. Susanna, V. V. Filichev, N. Bomholt, E. B. Pedersen, L. E. Xodo, Nucleic Acid
Res. 2008, 36, 3494.
[8] S. Buchini, C. J. Leumann, Curr. Opin. Chem. Biol. 2003, 7, 717.
[9] S. Hoshika, Y. Ueno, A. Matsuda, Bioconjugate Chem. 2003, 14, 607.
[10] H. C. Jessen, E. B. Pedersen, Helv. Chim. Acta 2004, 87, 2465.
[11] V. V. Filichev, M. C. Nielsen, N. Bomholt, C. H. Jessen, E. B. Pedersen, Angew. Chem., Int. Ed. 2006,
45, 5311.
´
[12] I. Geci, V. V. Filichev, E. B. Pedersen, Bioconjugate Chem. 2006, 17, 950.
[13] S. I. Khan, M. W. Grinstaff, J. Am. Chem. Soc. 1999, 121, 4704.
[14] A. E. Beilstein, M. W. Grinstaff, Chem. Commun. 2000, 509.
[15] M. Rist, N. Amann, H.-A. Wagenknecht, Eur. J. Org. Chem. 2003, 2498.
[16] P. Siemsen, R. C. Livingston, F. Diederich, Angew. Chem., Int. Ed. 2000, 39, 2632.
[17] A. Elangovan, Y.-H. Wang, T.-I. Ho, Org. Lett. 2003, 5, 1841.
[18] H. G. Viehe, ꢄChemistry of Acetylenesꢃ, Marcel Dekker Inc., New York, 1969, p. 599.
[19] N. Minakawa, Y. Ono, A. Matsuda, J. Am. Chem. Soc. 2003, 125, 11545.
[20] T. Matsuda, S. Kadowaki, M. Murakami, Chem. Commun. 2007, 2627.
[21] J.-L. Mergny, L. Lacroix, Oligonucleotides 2003, 13, 515.

730
Helvetica Chimica Acta – Vol. 92 (2009)
´ `
´
[22] C. Giovannangeli, L. Perrouault, C. Escude, N. Thuong, C. Helene, Biochemistry 1996, 35, 10539.
´
´ `
[23] C. Giovannangeli, L. Perrouault, C. Escude, S. Gryaznov, C. Helene, J. Mol. Biol. 1996, 261,
386.
[24] A. Błaszczyk, M. Chadim, C. von Hꢅnisch, M. Mayor, Eur. J. Org. Chem. 2006, 3809.
[25] M. Tanimoto, K. Kuchitsu, Y. Morino, Bull. Chem. Soc. Jpn. 1971, 44, 386.
[26] G. A. Williams, J. N. Macdonald, J. Mol. Struct. 1994, 320, 217.
Received November 4, 2008