Highly stable pyrimidine-motif triplex formation at physiological pH values by a bridged nucleic acid analogue

Article abstract of DOI:10.1002/anie.200604857

Good things come in threes: A novel bridged nucleic acid in which the furanose conformation was locked in the N form by a six-membered bridged moiety containing an N-O bond (see picture) has been developed. Triplex-forming oligonucleotides composed of thi

Full text of DOI:10.1002/anie.200604857

Communications
DOI: 10.1002/anie.200604857
Bridged Nucleic Acids
Highly Stable Pyrimidine-Motif Triplex Formation at Physiological pH
Values by a Bridged Nucleic Acid Analogue**
S. M. Abdur Rahman, Sayori Seki, Satoshi Obika, Sunao Haitani, Kazuyuki Miyashita, and
Takeshi Imanishi*
Formation of a stable triplex DNA molecule at physiological
pH values is a highly desirable phenomenon in molecular
biology and medicinal chemistry because of its great impor-
tance in regulation of gene expression, site-specific cleavage
of DNA, gene mapping and isolation, maintenance of folded
chromosome conformations, and gene-targeted mutagene-
sis.[1] In a pyrimidine-motif triplex DNA, the (homopyrimi-
dine) triplex-forming oligonucleotide (TFO) binds with the
homopurine tract of the target duplex DNA in a sequence-
specific manner through Hoogsteen hydrogen bonds to form
T·A:T and C+·G:C triads. However, formation of the C+·G:C
triad is dependent on the cytosine protonation, which is only
favorable at acidic pH values (pKa = 4.5) and, therefore,
homopyrimidine-motif triplexes are extremely unstable at
physiological pH values, which severely restricts their biolog-
ical application.
Although during the past few decades several efforts have
been directed to the formation of stable triplex DNA, most of
the investigations did not reach a practical level owing to
instability of the triplexes at physiological pH values.
Recently, our observation concluded that incorporation of a
bridged nucleic acid (BNA),[2] such as 2’,4’-BNA[3] (also
known as LNA;[4] Scheme 1) dramatically improved TFO
affinity for the target duplex and formed a stable triplex at
neutral pH values.[5] However, to our dismay, fully modified
TFO failed to bind with double-stranded DNA (dsDNA). The
optimum binding ability was found with TFOs composed of
alternating BNA and DNA monomers. Ethylene-bridged
nucleic acid (ENA), developed by us and Koizumi and co-
workers,[6] also exhibited comparable or better triplex-form-
Scheme 1. Structures of 2’,4’-BNA (LNA), ENA, and 2’,4’-BNANC
.
ing ability than that obtained with 2’,4’-BNA. Although
triplex formation with fully modified ENA was achieved at
neutral pH values, partially modified TFOs provided variable
results depending on the pH value (when compared with 2’,4’-
BNA). As a result of our continued investigations of the BNA
structure, we report herein a novel BNA molecule, 2’,4’-
[7]
BNANC
:
partially and fully modified TFOs[8] formed highly
stable triplexes at physiological pH values. Their overall
triplex-forming ability is superior to that of ENA and 2’,4’-
BNA, which is the most widely used BNA (as LNA) for
versatile genomic applications.[9]
As shown in Scheme 2, the 2’,4’-BNANC–thymine and –5-
methylcytosine phosphoroamidites 12 and 13, respectively,
were synthesized from the nucleoside derivative 1.[10] The
acetyl group was removed by aqueous methylamine and the
resultant alcohol 2 was converted to a mesylate 3, which was
treated with alkali to give the stereochemically inverted
alcohol 4 in very high yields. Debenzylation of 4 followed by
reprotection with a cyclic disiloxy group afforded the bicyclic
compound 5 in good yield.[11] The 2’-hydroxy group of 5 was
transformed to the triflate 6 by the treatment of trifluoro-
methanesulfonic anhydride in the presence of DMAP and
pyridine. The crude 6 was subjected to the SN2 reaction with
N-hydroxyphthalimide to yield the phthalimide derivative 7,
which was treated with hydrazine to deliver the aminoxy
compound 8. Exposure of 8 to phenoxyacetyl chloride in the
presence of pyridine provided the desired cyclized product 9
in one step. Deprotection of the silyl groups furnished our
target molecule, 2’,4’-BNANC–thymine monomer 10, in excel-
lent yields. Tritylation of the primary hydroxy group of 10
with 4,4’-dimethoxytrityl chloride gave 11. Then, phosphity-
lation of the secondary hydroxy group of 11 with 2-
cyanoethyl-N,N,N’,N’-tetraisopropylphosphordiamidite
yielded the desired thymine phosphoroamidite 12 in a very
good yield. On treatment with 1,2,4-triazole in the presence of
triethylamine and phosphoryl chloride,[12] compound 12
afforded the triazole derivative 13, which was successfully
used as a building block for the 5-methylcytidine unit of 2’,4’-
BNANC. Various TFOs were synthesized from these phos-
[*] Dr. S. M. A. Rahman,[+] S. Seki, S. Haitani, Prof. K. Miyashita,
Prof. T. Imanishi
Graduate School of Pharmaceutical Sciences
Osaka University
1-6 Yamadaoka, Suita, Osaka 565-0871 (Japan)
Fax: (+81)6-6879-8204
E-mail: imanishi@phs.osaka-u.ac.jp
Dr. S. Obika
Graduate School of Pharmaceutical Sciences
Osaka University
and
PRESTO, JST
4-1-8 Honcho Kawaguchi, Saitama 332-0012 (Japan)
[+] On leave from the Faculty of Pharmacy
Dhaka University, Dhaka-1000 (Bangladesh)
[**] S.M.A.R. thanks the Japan Society of Promotion of Science (JSPS)
for financial support.
Supporting information for this article is available on the WWW
under http://www.angewandte.org or from the author.
4306
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2007, 46, 4306 –4309

Angewandte
Chemie
Figure 1. UV melting curves (260 nm) for triplexes formed by 2’,4’-
BNANC-modified TFOs. ON-0 c, ON-1 c, ON-5 c, ON-7 c.
state for duplex dissociation and it might be concluded that
triplexes were simultaneously converted to three different
single strands.[13] Melting temperatures (Tm) of the triplexes
formed by 2’,4’-BNANC-modified TFOs were compared with
those formed by natural DNA, 2’,4’-BNA, and ENA-modified
TFOs, and the results are summarized in Table 1. Modifica-
tion of the natural DNA–TFO with a single 2’,4’-BNANC
Scheme 2. Synthesis of the 2’,4’-BNANC monomer 10 and the phos-
phoroamidites 12 and 13. Reagents and conditions: a) 40% aqueous
MeNH2, THF, RT, 3 h (99%); b) MsCl, pyridine, RT, 1 h; c) 1m NaOH,
EtOH, RT, 1 h (95% from 2); d) 20% Pd(OH)2–C, cyclohexene, EtOH,
reflux, 22 h; e) TIPDSCl2, imidazole, DMF, RT, 5.5 h (67% from 4);
f) Tf2O, DMAP, pyridine, RT, 7.5 h; g) N-hydroxyphthalimide, DBU,
MeCN, RT, 12 h (61% from 5); h) H2NNH2·H2O, EtOH, RT, 15 min
(73%); i) PhOCH2COCl, Et3N, CH2Cl2, RT, 2 h (57%); j) TBAF, THF,
RT, 5 min (96%); k) DMTrCl, DMAP, pyridine, RT, 7 h (87%);
Table 1: Tm values of triplexes containing 2,4,-BNANC (bold red), 2’,4’-
BNA (bold blue), and ENA (bold black).[a,b]
TFO
Sequence (5’…3’)
Tm [8C] DTm [8C] DTm/mod
ON-0
TTTTTmCTTTmCTmCTmCT
33
–
–
ON-1
BNA-1
ENA-1
TTTTTmCTTTmCTmCTmCT
TTTTTmCTTTmCTmCTmCT
TTTTTmCTTTmCTmCTmCT
44
44
42
+11
+11
+9
+11.0
+11.0
+9.0
l) (iPr2N)2PO(CH2)2CN, dicyanoimidazole, MeCN, RT, 4 h (85%);
m) 1,2,4-triazole, POCl3, Et3N, MeCN, 08C to RT, 5 h (95%). Bn=ben-
zyl, DBU=1,8-diazabicyclo[5.4.0]undec-7-ene, DMAP=4-dimethylami-
nopyridine, DMF=N,N-dimethylformamide, DMTr=dimethoxytri-
tyl=4,4’-dimethoxytriphenylmethyl, Ms=methanesulfonyl, Pac=phen-
oxyacetyl T=thymin-1-yl, TBAF=tetra-n-butylammonium fluoride,
Tf =trifluoromethanesulfonyl, TIPDS=tetraisopropyldisiloxane-l,3-diyl.
ON-2
BNA-2
ENA-2
TTTTTmCTTTmCTmCTmCT
TTTTTmCTTTmCTmCTmCT
TTTTTmCTTTmCTmCTmCT
60
59
56
+27
+26
+23
+9.0
+8.7
+7.7
ON-3
BNA-3
ENA-3
TTTTTmCTTTmCTmCTmCT
TTTTTmCTTTmCTmCTmCT
TTTTTmCTTTmCTmCTmCT
59
52
57
+26
+19
+24
+8.7
+6.3
+8.0
ON-4
BNA-4
ENA-4
TTTTTmCTTTmCTmCTmCT
TTTTTmCTTTmCTmCTmCT
TTTTTmCTTTmCTmCTmCT
58
57
57
+25
+24
+24
+6.3
+6.0
+6.0
phoroamidites and natural amidite building blocks on an
automated DNA synthesizer by using a conventional phos-
phoroamidite protocol. By using the usual workup procedure,
the phenoxyacetyl group was removed and the triazole group
ON-5
BNA-5
ENA-5
TTTTTmCTTTmCTmCTmCT
TTTTTmCTTTmCTmCTmCT
TTTTTmCTTTmCTmCTmCT
64
65
58
+31
+32
+25
+6.2
+6.4
+5.0
was converted to an amino group to give 2’,4’-BNANC
-
modified TFOs. The TFOs were characterized by MALDI-
TOF mass spectrometry (yields and the mass spectral data of
the TFOs are provided in the Supporting Information).
The triplex-forming ability of 2’,4’-BNANC-modified TFOs
against a 21-bp target DNA duplex at pH 7.0 was determined
through UV melting curves (Figure 1 and UV melting curves
in the Supporting Information). In the cases of the unmodi-
fied (ON-0) and slightly modified TFOs (such as ON-1), the
usual two-phase dissociation curves were obtained. In con-
trast, only a strong transition was obtained for the extensively
modified TFOs (ON-5, ON-7), which resulted from their very
high triplex stability (triplex stability is higher than the target
duplex stability). Unlike the usual curve, there is no transition
ON-6
BNA-6
ENA-6
TTTTTmCTTTmCTmCTmCT
TTTTTmCTTTmCTmCTmCT
TTTTTmCTTTmCTmCTmCT
78
67
72
+45
+34
+39
+6.4
+4.9
+5.6
ON-7
BNA-7
TTTTTmCTTTmCTmCTmCT
TTTTTmCTTTmCTmCTmCT
80
<5
+47
<À28
+3.1
<À2
[a] Target
duplex:
5’-d(GCTAAAAAGAAAGAGAGATCG)-3’/3’-
d(CGATTTTTCTTTCTCTCTAGC)-5’; underlined portion indicates the
target site for triplex formation. [b] Conditions: 7 mm Na2HPO4 buffer
solution containing 140 mm KCl; strand concentration=1.5 mm; scan
rate 0.58CminÀ1. Tm =melting temperatures, DTm =changes in melting
temperature, DTm/mod=changes in melting temperature per single
modification; mC=5-methylcytidine.
Angew. Chem. Int. Ed. 2007, 46, 4306 –4309ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 4307

Communications
monomer (ON-1) increased the Tm value by 118C, which is
equal to that of 2’,4’-BNA-modified TFO (BNA-1) and
slightly higher than that of the corresponding ENA-modified
TFO (ENA-1). Further modifications greatly enhanced the
triplex thermal stability. For example, by increasing the
number of modifications from one to three, Tm values
increased to 608C (DTm = + 278C) and 598C (DTm =
+ 268C) for ON-2 and ON-3, respectively. Therefore, it is
noteworthy that both the values of the 2’,4’-BNANC-modified
TFOs with either interrupted or continuous 2’,4’-BNANC
residues are very high and the same, whereas the Tm value
of BNA-3 (528C) containing three continuous modifications
was found to decrease by 78C compared with that of BNA-2
(598C). The corresponding ENA–TFOs (ENA-2 and ENA-3)
showed lower Tm values than the 2’,4’-BNANC–TFOs. Inter-
estingly, their triplex-forming behavior is in agreement with
that of 2’,4’-BNANC, clarifying that continuous six-membered
bridged structures are well tolerated by dsDNA. These
observations were also consistent with the results of other
TFOs (such as, ON-4 and ON-6 versus BNA-4, BNA-6 versus
ENA-4 and ENA-6, and ON-5 versus BNA-5) with the
exception of ENA-5. In the cases of ON-4/BNA-4/ENA-4 and
ON-5/BNA-5 where modifications are located far apart from
each other, all the 2’,4’-BNANC–, 2’,4’-BNA–, and ENA–TFOs
furnished similar Tm values for triplex dissociation. The
reason for the lower Tm value for ENA-5 is not clear. In
contrast, with relatively congested BNA residues, 2’,4’-
BNANC–TFO (ON-6) provided Tm values as high as 788C,
which is 118C and 68C higher than those provided by the
corresponding 2’,4’-BNA– and ENA–TFOs (BNA-6 and
ENA-6), respectively. Thus, with continuous modifications
Figure 2. Electrophoretic mobility shift assay of the triplexes on 20%
nondenaturing polyacrylamide gel at pH 6.8. Lanes 1, 2, and 3 in all
cases represent a target duplex (control), stoichiometric mixture
(6.5 pmol) of target duplex and TFO, and TFO with excess duplex
(TFO/duplex=1:2), respectively. T=triplex, D=duplex.
electrophoretic mobilities of the triplexes formed by 2’,4’-
BNANC-modified TFOs were slightly lower and that of the
fully modified TFO (ON-7) was remarkably lower.[17]
The extraordinarily high triplex-forming ability of 2’,4’-
BNANC might result from the combined effects of restricted
N conformation[5,18] and protonation of the N atom, which
might cause electrostatic interactions between the positively
charged TFO and the negatively charged phosphodiester
linkage of the target duplex.[8c,19–21] Moreover, in contrast with
the fully modified BNA–TFOs, which are too rigid in the
overall structure,[5b,22] the 2’,4’-BNANC-modified TFOs, bear-
ing a six-membered bridged structure like ENA, might pose
suitable conformational flexibility in their overall structure,
which facilitates stable triplex formation as well. The above
facts are the reason for which the 2’,4’-BNANC-modified TFOs
act as excellent TFOs for the recognition of the homopurine–
homopyrimidine tract of dsDNA. The predominance of 2’,4’-
BNANC over ENA and 2’,4’-BNA essentially lies with the role
of the protonated nitrogen to neutralize the negatively
charged phosphate backbone of the purine strand.
In conclusion, we have synthesized a novel bridged
nucleic acid analogue, 2’,4’-BNANC, and demonstrated that
the TFOs composed of 2’,4’-BNANC formed highly stable
pyrimidine-motif triplexes at physiological pH values. The
overall triplex-forming ability is higher than that of 2’,4’-
BNA/LNA- and ENA-modified TFOs. Unlike the 2’,4’-BNA-
modified TFOs, these TFOs eliminate the requirement of
placing alternating DNA monomers for optimum efficacy.[23]
More interestingly, fully modified TFOs still formed a highly
stable triplex. These promising properties of 2’,4’-BNANC will
be helpful for developing oligonucleotide-based technologies
for the postgenome era.
or with an increased number of modifications, 2’,4’-BNANC
-
modified TFOs showed higher Tm values than those of 2’,4’-
BNA- and ENA-modified TFOs. These interesting character-
istics of 2’,4’-BNANC prompted us to synthesize a fully
modified TFO, ON-7, which formed a very stable triplex
with a Tm value as high as 808C (DTm = + 478C; DTm per
modification = + 3.18C). The corresponding 2’,4’-BNA-modi-
fied TFO, BNA-7, failed to form a triplex.[14,15]
Next, triplex formation was evaluated by an electro-
phoretic mobility shift assay (EMSA) at pH 6.8 (Figure 2).
Each TFO was incubated with the target dsDNA at a ratio of
1:1 at 48C in 10 mm 2-[4-(2-hydroxyethyl)-1-piperazinyl]eth-
anesulfonic acid buffer solution (HEPES; pH 6.8) and sub-
jected to a 20% polyacrylamide gel electrophoresis at 48C
and room temperature.[16] To confirm the triplex formation,
the target duplex without TFO and TFO with excess duplex
(TFO/duplex = 1:2) were also run together (Figure 1, lanes 1
and 3, respectively). It was found that all the modified TFOs
formed stable triplexes at the stoichiometric ratio under the
experimental conditions. The natural DNA (ON-0) was
unable to form a triplex at room temperature even though
it showed triplex formation at 48C. In contrast, the TFO with
only a single 2’,4’-BNANC modification (ON-1) can form a
stable triplex at room temperature. These results correlate
with the Tm data and it might be expected that extensively
modified TFOs would promote stable triplex formation at
higher temperatures. The fully modified TFO, ON-7, also
formed a stable triplex with a clear and intense band. The
Experimental Section
UV melting experiments: UV melting experiments were carried out
on a Beckman DU-650 spectrometer equipped with a Tm analysis
4308 www.angewandte.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2007, 46, 4306 –4309

Angewandte
Chemie
accessory. Equimolecular amounts of the target duplex and TFO were
dissolved in 7 mm sodium phosphate buffer solution (pH 7.0)
containing 140 mm KCl to give a final strand concentration of
1.5 mm. The strands were annealed by heating the samples at 908C for
5 minutes followed by slow cooling to room temperature. Then the
samples were stored at 48C for 1 h. The melting profile was recorded
at 260 nm from 10 to 858C at a scan rate of 0.58CminÀ1. The Tm was
calculated as the temperature of the half dissociation of the formed
triplexes, which is determined by the first derivative of the melting
curve.
Sci. USA 1996, 93, 4365; c) B. Cuenoud, F. Casset, D. Husken, F.
Natt, R. M. Wolf, K.-H. Altmann, P. Martin, H. E. Moser,
Angew. Chem. 1998, 110, 1350; Angew. Chem. Int. Ed. 1998, 37,
1288; d) S. M. Gryaznov, H. Winter, Nucleic Acids Res. 1998, 26,
4160; e) N. Kumar, K. E. Nielsen, S. Maiti, M. Petersen, J. Am.
Chem. Soc. 2006, 128, 14.
[9]a) M. Petersen, J. Wengel, Trends Biotechnol. 2003, 21, 74;
b) J. S. Jespen, J. Wengel, Curr. Opin. Drug Discovery Dev. 2004,
7, 188; c) B. Vester, J. Wengel, Biochemistry 2004, 43, 13233;
d) L. Wang, C. J. Yang, C. D. Medley, S. A. Benner, W. Tan, J.
Am. Chem. Soc. 2005, 127, 15664.
Full experimental details are described in the Supporting
Information.
[10]A. A. Koshkin, V. K. Rajwanshi, J. Wengel, Tetrahedron Lett.
1998, 39, 4381.
[11]We had to replace the benzyl groups with a cyclic disiloxy group
Received: November 30, 2006
Revised: March 26, 2007
Published online: April 27, 2007
À
because we experienced cleavage of the N O bridged structure
during debenzylation of the corresponding dibenzyl derivative of
9.
[12]a) K. Shah, H. Wu, T. M. Rana, Bioconjugate Chem. 1994, 5, 508;
b) M. Scherr, C. Klebba, R. Häner, A. Ganser, J. W. Engels,
Bioorg. Med. Chem. Lett. 1997, 7, 1791.
[13]Similar triplex-to-single-strand dissociation was also reported
recently by Leumann, see: S. Buchini, C. J. Leumann, Angew.
Chem. 2004, 116, 4015; Angew. Chem. Int. Ed. 2004, 43, 3925.
[14]Similar results were obtained by using two different BNA–TFOs
in our previous investigation. The TFOs also failed to bind with a
30-bp target duplex.[5b]
[15]The Tm value of the corresponding ENA–TFO could not be
determined because of the difficulties in purification of the TFO.
[16]In the case of room temperature EMSA, the triplexes were
warmed to room temperature after incubating at 48C and kept at
room temperature for two hours before applying onto the gel.
The gel was run at room temperature at a constant voltage of
70 V.
Keywords: bridged nucleic acids · oligonucleotides ·
synthesis design · thermal stability · triplexes
.
[1]a) H. E. Moser, P. B. Dervan, Science 1987, 238, 645; b) M. D.
Frank-Kamenetskii, S. M. Mirkin, Annu. Rev. Biochem. 1995, 64,
65; c) V. N. Soyfer, V. N. Potaman, Triple-Helical Nucleic Acids,
Springer, New York, 1996, p. 220; d) M. M. Seidman, P. M.
Glazer, J. Clin. Invest. 2003, 112, 487; e) S. Buchini, C. J.
Leumann, Curr. Opin. Chem. Biol. 2003, 7, 717.
[2]T. Imanishi, S. Obika, Chem. Commun. 2002, 1653.
[3]S. Obika, D. Nandu, Y. Hari, K. Morio, Y. In, T. Ishida, T.
Imanishi, Tetrahedron Lett. 1997, 38, 8735.
[4]S. K. Singh, P. Nielsen, A. A. Koshkin, J. Wengel, Chem.
Commun. 1998, 455.
[5]a) S. Obika, Y. Hari, T. Sugimoto, M. Sekiguchi, T. Imanishi,
Tetrahedron Lett. 2000, 41, 8923; b) S. Obika, T. Uneda, T.
Sugimoto, D. Nanbu, T. Minami, T. Doi, T. Imanishi, Bioorg.
Med. Chem. 2001, 9, 1001; c) H. Torigoe, Y. Hari, M. Sekiguchi,
S. Obika, T. Imanishi, J. Biol. Chem. 2001, 276, 2354; d) S. Obika,
Y. Hari, M. Sekiguchi, T. Imanishi, Angew. Chem. 2001, 113,
2149; Angew. Chem. Int. Ed. 2001, 40, 2079; e) S. Obika, Y. Hari,
M. Sekiguchi, T. Imanishi, Chem. Eur. J. 2002, 8, 4796.
[17]This type of lower mobility was also found in the case of the
recently reported a-l-LNA-modified triplexes.[8e] Variation of
triplex mobility was also noted by Roberts and Crothers, see:
R. W. Roberts, D. M. Crothers, Science 1992, 258, 1463.
[18]J. L. Asensio, R. Carr, T. Brown, A. N. Lane, J. Am. Chem. Soc.
1999, 121, 11063.
[19]S. Rhee, Z.-J. Han, K. Liu, H. T. Miles, D. R. Davis, Biochemistry
1999, 38, 16810.
[6]a) K. Morita, C. Hasegawa, M. Kaneko, S. Tsutsumi, J. Sone, T.
Ishikawa, T. Imanishi, M. Koizumi, Bioorg. Med. Chem. Lett.
2002, 12, 73; b) M. Koizumi, K. Morita, M. Daigo, S. Tsutsumi,
K. Abe, S. Obika, T. Imanishi, Nucleic Acids Res. 2003, 31, 3267.
[7]This BNA is defined as 2 ’,4’-BNANC because the bridge between
2’-O and 4’-C is constituted by N and C atoms.
[20]T. P. Prakash, A. Puschl, E. Lesnik, V. Mohan, V. Tereshko, M.
Egli, M. Manoharan, Org. Lett. 2004, 6, 1971.
[21]A. Mayer, A. Häberli, C. Leumann, Org. Biomol. Chem. 2005, 3,
1653.
[22]C. H. Gotfredsen, P. Schultze, J. Feigon, J. Am. Chem. Soc. 1998,
120, 4281.
[8]The other fully modified TFOs formed triplexes at around the
physiological pH value, see: a) M. Bolli, C. Leumann, Angew.
Chem. 1995, 107, 694; Angew. Chem. Int. Ed. Engl. 1995, 34, 694;
b) C. EscudØ, C. Giovannangeli, J.-S. Sun, D. H. Lloyd, J-K.
Chen, S. M. Gryaznov, T. Garestier, C. HØl›ne, Proc. Natl. Acad.
[23]Rules for designing BNA–TFOs for optimum triplex-forming
ability were described, see: B. W. Sun, B. R. Babu, M. D.
Sørensen, K. Zakrzewska, J. Wengel, Biochemistry 2004, 43,
4160.
Angew. Chem. Int. Ed. 2007, 46, 4306 –4309ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 4309