Selective cross-linking to the adenine of the TA interrupting site within the triple helix

Article abstract of DOI:10.1016/S0960-894X(01)00783-1

The triplex-forming oligonucleotide incorporating the new nucleoside derivative (2) that connects the 2-amino-6-vinylpurine moiety to the 2-deoxyribose unit with an ethyl spacer has exhibited highly selective cross-linking reaction to the adenine of the TA interrupting site within the triple helix.

Full text of DOI:10.1016/S0960-894X(01)00783-1

Bioorganic & Medicinal Chemistry Letters 12 (2002) 487–489
Selective Cross-Linking to the Adenine of the
TA Interrupting Site within the Triple Helix
Fumi Nagatsugi, Yoshihisa Matsuyama, Minoru Maeda and Shigeki Sasaki*
Graduate School of Pharmaceutical Sciences, Kyushu University, 3-1-1 Maidashi, Higashi-ku, Fukuoka 812-8582, Japan
Received 25 September 2001;accepted 17 November 2001
Abstract—The triplex-forming oligonucleotide incorporating the new nucleoside derivative (2) that connects the 2-amino-6-vinyl-
purine moiety to the 2-deoxyribose unit with an ethyl spacer has exhibited highly selective cross-linking reaction to the adenine of
the TA interrupting site within the triple helix. # 2002 Elsevier Science Ltd. All rights reserved.
Triplex-forming oligonucleotides (TFOs) can bind to
homopurine-homopyrimidine regions in duplex DNA in
a sequence-specific manner.1,2 It has recently been
reported that a single insertion of an interrupting site
such as a TA or a CG pair can be recognized by TFO
having new nucleoside analogues,3 although a general
method for recognition of any duplex sequences has not
appeared to date.4 Triplex-based approaches are an
attractive means to achieve targeted gene regulation5ꢀ7
and gene manipulation8ꢀ10 both in vitro and in vivo.
Cross-linking reaction within the triplexes has been used
to ensure triplex formation and subsequent inhibition of
gene expression in the antigene method.11 Furthermore,
covalent modification of a single base has been expected
to alter the recognition mode of the base, and would be
useful as a new biotechnology.8ꢀ10 However, only a
limited number of cross-linking agents are currently
available, which include alkylating agents to N7 of gua-
nine as well as photo-reactive psolarens to a thymine of
an AT pair.12 For cross-linking agents within triplexes
to become a useful tool for gene manipulation, they
must exhibit high selectivity at the base level. Thus,
there is an urgent need for the development of a reliable
concept on which a new cross-linking molecule may be
designed to attain desired base selectivity.
the cytosine of the G-C target site (Fig. 1A).14 The high
selectivity to cytosine in the pyrimidine strand would be
due to a complex between the protonated form of 2-
amino-6-vinylpurine moiety (1) and a cytosine which is
flipped out from the GC pair.
Recently, we have demonstrated that the new nucleoside
derivative (1) with a butyl spacer between the sugar part
and the 2-amino-6-vinylpurine13 skeleton achieves tri-
plex-forming cross-linking with high selectively toward
Figure 1. The reaction of 1 with the flipping cytosine (A) leading to
the design of the new cross-linking agent 2 to the adenine within the
TA interrupting site (B).
*Corresponding author. Tel.: +81-92-642-6651;fax: +81-92-642-
6654;e-mail: sasaki@phar.kyushu-u.ac.jp
0960-894X/02/$ - see front matter # 2002 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(01)00783-1

488
F. Nagatsugi et al. / Bioorg. Med. Chem. Lett. 12 (2002) 487–489
Then, we proposed the new concept in which a flipping
base can be selectively reacted by the cross-linking
agent. As an application of this concept, we have
designed a new nucleoside derivative (2) having an ethyl
spacer between the sugar part and the 2-amino-6-vinyl-
purine skeleton as a cross-linking agent to an adenine at
the TA base pair. Here, we wish to demonstrate the
validity of the new concept by achieving the first selec-
tive cross-linking at the TA interrupting site.
The conformational change of the triplex during the
cross-linking of 1 with cytosine has been estimated by
MM3 calculation, and it has been suggested that only a
slight flipping of the cytosine may cause satisfactory
proximity between the vinyl and the amino groups, and
that almost no distortion is observed for the backbone
geometries of the triplex (Fig. 1A). The new cross-
linking agent 2 connects the 2-amino-6-vinylpurine
skeleton to the sugar part with an ethyl spacer (Fig. 1B). A
similar MM3 estimation has shown that a small flipping
of the adenine would be enough to bring the amino
group of the adenine into proximity of the vinyl group
of 2, without distortion of the triplex geometry (Fig.
1B). It has been also suggested that hydrogen bonds
between the three components, T, A and the 2-amino-
purine moiety would be maintained during the reaction.
In contrast, the reactions of 2 with GC, CG or AT base
pair were found to need a large conformational change.
Thus, it was anticipated that 2 would react selectively
with the adenine of the TA interrupting site.
Scheme 1. (a) TrCl, DMAP, pyridine 52%;(2) t-BuOK, (EtO)2-
P(O)CH2CO2Et, THF, 68%;(b) (1) i-Pr2NEt, MOMCl, 96%;
(2) LAH, THF, 96%;(3) TsCl, pyridine, 72%;(c) (1) 2-amino-6-
chloropurine, t-BuOK, DMSO, 73%;(2) n-Bu3SnCH=CH2,
Pd(PPh3)4, dioxane, 65%;(d) (1) MeSNa, CH 3CN, 73%;(2)
PhOCH2COCl, 1-HBT, CH3CN, pyridine, 89%;(3) BF 3–Et2O, Me2S,
63%, (4) DMTrCl, pyridine, 91%;(5)
i-Pr2NEt, CH2Cl2, i-
Pr2NP(Cl)OC2H4CN, 97%;(e) (1) synthesis with an automated DNA
synthesizer;(2) 28% NH 3, 20–30%, (3) 10% CH3COOH;(f) (1) 3
equiv MMPP pH 10;(2) 0.5 M NaOH.
The synthesis of the ODNs incorporating functionalized
nucleoside analogue (2) is summarized in Scheme 1. The
sugar part (3) was synthesized by b-selective C-glycoside
formation via Wittig–Horner–Emmons reaction from 2-
deoxy-d-ribose (Scheme 1).15 The alcohol of 3 was pro-
tected with a methoxymethyl (MOM) group, and the
following reduction and tosylation gave 4. The coupling
reaction between 2-amino-6-chloropurine16 and 4 yiel-
ded the desired product as a major isomer, which was
used for the Pd(II)-catalyzed cross-coupling reaction
with nBu3SnCH¼CH2 to afford 2-amino-6-vinylpurine
derivative (5). After protection of the vinyl group with
methylsulfide and the 2-amino group with phenoxy-
acetyl, the conventional procedure produced the phos-
phoramidite precursor (6) in good yield. The sulfide-
protected ODN (7) was obtained by applying 6 to an
automated DNA synthesizer in good yield after pur-
ification with RP-HPLC. The ODN (7) was then
smoothly converted to 8 by oxidation with magnesium
monoperphthalate (MMPP) following elimination
under an alkaline condition. The structure of ODN was
confirmed by MALDI-TOF mass measurements.
The cross-linking was investigated with the function-
alized ODNs (8) and the target duplexes (9–10). In
order to define the strand of the reaction, one of the two
strands (9 or 10) labeled with 32P at the 50-end was used
as a tracer. The results were analyzed by gel electro-
phoresis with 15% denaturing gel, in which the cross-
linked products were identified as less-mobility bands
relative to the unreacted radiolabeled single strand (Fig.
2). The less-mobility bands were observed in lane 7 (Fig.
2A and B), clearly indicating that the ODN (8) bearing
2 reacted only with the adenine within the pyrimidine
strand. The product yield was estimated by quantifica-
tion of each band to be ca. 50% after 24 h. Intermediary
formation of the triplex has been supported by the facts
that the reaction did not take place in the presence of
the competitive TFO 8 (Z=G instead of 2) or in the
absence of 9. Thus, high selectivity of 8 to the adenine in
Figure 2. Cross-linking selectivity with the TFO 8: (A) after 5 h, (B)
after 24 h. The reaction was done using 10 mM of ODNs (8: Z=2), 1
mM of target duplex 9–10 in a buffer including 10 mM cacodylate,
0.25 mM spermine, 100 mM NaCl, pH 4.5ꢁ0.3 at 30 ꢂC. The reaction
mixture contained either 50-32P-labeled 9 or 10 as a tracer. The reac-
tion was stopped by the addition of formamide after 20 h, and ana-
lyzed by gel electrophoresis with 15% denaturing polyacrylamide gel.

F. Nagatsugi et al. / Bioorg. Med. Chem. Lett. 12 (2002) 487–489
489
References and Notes
1. Soyfer, V. N.;Potaman, V. N. Triple-Helical Nucleic Acids;
Springer: Berlin, Heidelberg and New York, 1996.
2. Thuong, N. T.;He lene, C. Angew. Chem., Int. Ed. Engl.
1993, 32, 666.
3. (a) Prevot, I. H.;Leumann, C. J. Bioorg. Med. Chem. Lett.
1999, 9, 2657. (b) Obika S.;Hari, Y.;Sekiguchi, M.;Imanishi,
T. Angew. Chem., Int. Ed, Engl. 2001, 40, 2079. (c) Sasaki, S.;
Yamauchi, H.;Nagatsugi, F.;Takahasi, R.;Taniguchi, Y.;
Maeda, M. Tetrahedron Lett. In press.
4. (a) Selected reviews, Maher, L. J., III;Wold, B.;Dervan,
P. B. Antisense Res. Develop 1991, 1, 277. (b) Sasaki, S. J.
Synth Org. Chem. 1997, 55, 590. (c) Giovannangeli, C.;
Helene, C. Antisense & Nucleic Acid Drug Develop 1997, 7,
413. (d) Gowers, D. M.;Fox, K. R. Nucleic Acids Res. 1999,
27, 1569. (e) Gowers, D. M.;Bijapur, J.;Brown, T.;Fox, K. R.
Biochemistry 1999, 38, 13747.
Figure 3. Gel-shift analysis of triplexes with 15% non-denaturing
polyacrylamide gel at pH 5.5 and 7.7. The triplex formation was per-
formed at pH 4.5 for 24 h under the same condition as described in
Figure 2, and was analyzed by gel electrophoresis at indicated pH.
the pyrimidine strand at the TA base pair of the triplex
has been proven.
We next examined the effect of cross-link to the TA
interrupting site on the stability of the triplex by gel-
shift assay with non-denaturing gel (Fig. 3). The tri-
plexes containing C-GC base triplet were observed at
the concentrations higher than 1 mM of the TFO at pH
5.5 (lanes 2–6), whereas they dissociated at pH 7.7
(lanes 7–12). The less-mobility bands in lanes 15–18
showed the same gel-shift with the triplexes of lanes 2–6,
suggesting that the triplex is stabilized by the interstrand
cross-link between 8 and 10.17 In the investigation with
the monomer derivative of the vinyl compound (5) in
organic solvents, no adduct was formed with the ade-
nosine derivative.18 The cross-linking reaction with 2
and adenine may be effected by close proximity within
the triplex as expected.
5. Wang, Z.;Rana, T. M. Proc. Natl. Acad. Sci. U.S.A. 1997,
94, 6688.
6. Praseuth, D.;Guieysse, A. L.;He lene, C. Biochimica et
Biophys. Acta Gene Struct. Expr. 1999, 1489, 181.
7. Faria, M.;Wood, C. D.;Perrouault, L.;Nelson, J. S.;
Winter, A. H.;White, M. R.;He lene, C.;Giovannangeli, C.
Proc. Natl. Acad. Sci. U.S.A. 2000, 97, 3862.
8. Majumdar, A.;Khorlin, A. N.;Dyatkina, F. L. M.;Lin, J.;
Powell, J.;Liu, Z. Z.;Fei, Y.;Khripine, K. A.;Watanabe, J.;
George, P.;Glazer, M.;Seidman, M. M. Nat. Genet. 1998, 20,
212.
9. Chan, P. P.;Lin, M.;Faruqi, A. F.;Powell, J.;Seidman,
M. M.;Glazer, P. M. J. Biol. Chem. 1999, 274, 11541.
10. Vasquez, K. M.;Narayanan, L.;Glazer, P. M. Science
2000, 290, 530.
11. (a) Recent articles: Reed, M. W.;Wald, W.;Meyer, R. B.
J. Am. Chem. Soc. 1998, 120, 9729. (b) Belousov, E. S.;Afo-
nina, I. A.;Podyminogin, M. A.;Gamper, H. B.;Reed, M. W.;
Wydro, R. M.;Meyer, R. B. Nucleic Acids Res. 1997, 25, 3440.
(c) Lukhtanov, E. A.;Mills, A. G.;Kutyavin, I. V.;Gorn,
V. V.;Reed, M. W.;Meyer, R. B. Nucleic Acids Res. 1997, 25,
5077.
In conclusion, a highly selective cross-linking reaction
to the adenine at the TA interrupting site within the
triplex has been achieved with the use of the TFO
incorporating the new cross-linking agent 2. It should
be noted that only a minor change from a butyl to an
ethyl spacer caused a drastic change in base selectivity,
and the validity of the design concept has been con-
firmed. The new cross-liking reagent 2 is the first exam-
ple toward the TA interrupting site within the triplex.
We have recently reported that the ethylsulfoxide sub-
stitution on the vinyl group of the 2-amino-6-vinylpur-
ine skeleton improved its reactivity to a greater extent at
pH 7.13d Such modification of 2 would effect more effi-
cient reactivity within the triplex at pH 7, and produce a
useful cross-linking agent for application to site-directed
modification of an adenine within a selected target.
12. Some examples are reviewed;Sasaki, S. Eur. J. Pharm.
Sci. 2001, 13, 43.
13. (a) Nagatsugi, F.;Uemura, K.;Nakashima, S.;Maeda, S.
Sasaki, Tetrahedron Lett. 1995, 36, 421. (b) Nagatsugi, F.;
Uemura, K.;Nakashima, S.;Maeda, M.;Sasaki, S.
Tetra-
hedron 1997, 53, 3035. (c) Nagatsugi, F.;Kawasaki, T.;Usui,
D.;Maeda, M.;Sasaki, S. J. Am. Chem. Soc. 1999, 121, 6753.
(d) Nagatsugi, F.;Tokuda, N.;Maeda, M.;Sasaki, S. Bioorg.
Med. Chem. Lett. 2001, 11, 2577.
14. Nagatsugi, F.;Usui, D.;Kawasaki, T.;Maeda, M.;
Sasaki, S. Bioorg. Med. Chem. Lett. 2001, 11, 343.
15. Matsuura, N.;Yashiki, Y.;Nakashima, S.;Maeda, M.;
Sasaki, S. Heterocycles 1999, 51, 975.
16. Kjellberg, J.;Liljenberg, M.;Johansson, G. Tetrahedron
1986, 42, 6541.
Acknowledgements
17. Stabilization of a triple helix by cross-linking has been
demonstrated;Kool, E. T. Chem. Rev. 1997, 97, 1473.
18. As the authentic compound for the adduct between 2 and
adenosine could not be obtained, we are now investigating the
cross-linking reaction in large quantities to isolate and deter-
mine the cross-linked adduct.
This work was supported by a Grant-in-Aid for Scien-
tific Research (C), [Priority Area (A)(2)], from the Min-
istry of Education, Science, Sports, Science and
Technology, Japan.