Highly efficient and selective cross-linking to cytidine based on a new strategy for auto-activation within a duplex [4]

Full text of DOI:10.1021/ja990356e

J. Am. Chem. Soc. 1999, 121, 6753-6754
6753
Scheme 1a
Highly Efficient and Selective Cross-Linking to
Cytidine Based on a New Strategy for
Auto-Activation within a Duplex
Fumi Nagatsugi, Takeshi Kawasaki, Daisaku Usui,
Minoru Maeda, and Shigeki Sasaki*
4
Graduate School of Pharmaceutical Sciences
Kyushu UniVersity, 3-1-1 Maidashi
Higashi-ku, Fukuoka 812-8582, Japan
ReceiVed February 5, 1999
Sequence-selective cross-linking between complementary du-
plexes or triplexes provides an important method for improving
the stability of the hybridized complexes.1-11 This strategy has
been expected to effect inhibition of gene expression in the
antisense and antigene methods. The basic concept of cross-linking
includes the incorporation of a reactive functional group into the
oligomers, and there are a number of reports on alkylating groups,
such as haloacetyl amide,4,5,9 aziridine,1,3,8 and psolaren deriva-
tives.7,10 Recently, interest in cross-linking has been expanding
as a tool for site-directed chemical modification which may induce
point mutation of a genetic code.12 However, the existing
aklylating agents do not seem to be satisfactory for general
application, and there is an urgent requirement for new cross-
linking agents that exhibit efficient and selective reactivity toward
a chosen target site. The use of functional groups with intrinsic
high reactivity may cause considerable instability, while those
requiring ultraviolet activation may limit the target site. In our
approach, we have developed a new strategy for cross-linking in
which a reactive species is auto-generated within a duplex. Here
we wish to report a highly efficient and selective cross-linking
reaction toward cytidine with the use of derivatives of a 2-amino-
6-vinylpurine nucleoside (1).
a (a) (1) PhSH, CH2Cl2 rt, 1 h, 74%, (2) PhOCH2COCl, 1-hydroxy-
benzotriazole, 78%, (3) nBu4NF, 76%, (4) DMTrCl, pyridine, 75%, (5)
iPr2EtN, CH2Cl2, iPr2NP(Cl)OCH2CH2CN, 64%. (b) (1) synthesis with
an automated DNA synthesizer, (2) 0.1 N NaOH, (3) neutralized with
CH3COOH, 70-80%, (3) 10% CH3COOH. (c) 2 equiv MMPP (magne-
sium monoperphthalate), pH 10, 30 min, quantitative. (d) 10 equiv MMPP,
pH 10, 1 d, quantitative, (e) 470 mM NaOH, 30 min, quantitative.
We have previously designed a 2-amino-6-vinylpurine nucleo-
side (1) as a selective alkylating agent to cytidine. The complex
resembling a natural G-C pair would favor the attack of the
4-amino group of cytidine, a weak nucleophile, leading to a cross-
linked product (2).13 Model experiments in organic solvents and
investigations with ODNs bearing 1 have demonstrated that 1 is
a cytidine-selective alkylating agent.14 To achieve high reactivity
without incurring chemical stability of 1, we next designed a new
strategy in which less reactive precursors (3, 4, or 5) would be
auto-activated within duplexes to generate 1.
The 2′-deoxy nucleoside derivative of 2-amino-6-vinylpurine
(1) was synthesized from 2′-deoxyguanosine as described previ-
ously,14 protected as phenylsulfide, and then transformed to the
amidite precursor (3). An example of the synthesis of function-
alized ODNs incorporating a 6-substituted 2-aminopurine nucleo-
side is illustrated in Scheme 1. The sulfide functional group of 6
was converted to the sulfoxide form (7) as a sole product by
oxidation with 2 equiv of magnesium monoperphthalate (MMPP).
The transformation of 6 to the sulfone form (8) was also a
quantitative reaction with an excess of MMPP. Both 7 and 8 were
smoothly transformed to 9, incorporating the 2-amino-6-vinylpu-
rine nucleoside under mild alkaline conditions. The structures of
the ODNs (6-9) were confirmed by UV and MALDI-TOF mass
measurements.15
The cross-linking was investigated with the functionalized
ODNs (6-9) and the target ODN (10) in the presence of 32P-
labeled 10 as a tracer and analyzed by gel electrophoresis with
(1) (a) Webb, T. R.; Matteucci, M. D. J. Am. Chem. Soc. 1986, 108, 2764-
2765. (b)Webb, T. R.; Matteucci, M. D. Nucleic Acids Res. 1986, 14, 7661-
7674.
(8) Shaw, J.-P.; Milligan, J. F.; Krawczyk, S. H.; Matteucci, M. D. J. Am.
Chem. Soc. 1991, 113, 7765-7666.
(9) (a) Grant, K. B.; Dervan, P. B. Biochemistry 1996, 65, 12313-12319.
(b) Povsic, T. J.; Strobel, S. A.; Dervan, P. B. J. Am. Chem. Soc. 1992, 114,
5934-5941. (c) Povsic, T. J.; Dervan, P. B. J. Am. Chem. Soc. 1990, 112,
9428-9430.
(2) (a) Zeng, Q.; Rokita, S. E. J. Org. Chem. 1996, 61, 9080-9081. (b)
Chatterjee, M.; Rokita, S. E. J. Am. Chem. Soc. 1994, 116, 1690-1697. (c)
Tianhu, Li and Rokita, S. E. J. Am. Chem. Soc. 1991, 113, 7771-7773. (d)
Chatterjee, M. and Rokita, S. E. J. Am. Chem. Soc. 1991, 113, 5116-5117.
(3) Cowart, M.; Benkovic, S. J. Biochemistry 1991, 30, 788-796.
(4) (a) Coleman, R. S.; Pires. R. M. Nucleic Acids Res. 1997, 25, 4771-
4777. (b) Coleman, R. S.; Kesicki, E. A. J. Org. Chem. 1995, 60, 6252-
6253.
(5) (a)Tabone, J. C.; Stamm, M. R.; Gamper H. B.; Meyer, R. B., Jr.
Biochemistry 1994, 33, 375-383. (b) Meyer, R. B. Jr.; Tabone, J. C.; Hurst,
G. D.; Smith, T. M.; Gamper, H. J. Am. Chem. Soc. 1989, 111, 8517-8519.
(6) Lukhtanov, E. A.; Podyminogin, M. A.; Kutyavin, I. V.; Meyer, R. B.,
Jr.; Gamper, H. B. Nucleic Acids Res. 1996, 24, 683-687.
(10) Takasugi, M.; Guenouz, A.; Chassignol, M.; Deout, J. L.; Lohmme,
J.; Thuong, N. T.; Helene, C. Proc. Natl. Acad. Sci. U.S.A. 1991, 88, 5602-
5606.
(11) Lukhtanov, E. A.; Mills, A. G.; Kutyavin, I. V.; Vladimir, V. V.; Reed,
M. W.; Meyer, R. B., Jr. Nucleic Acids Res. 1997, 25, 5077-5084.
(12) Woolf, T. M. Nature Biotech. 1998, 16, 341-344 and references
therein.
(13) The actual position of protonation is uncertain and could possibly be
either 1 or C.
(14) (a) Nagatsugi, F.; Uemura, K.; Nakashima, S.; Maeda, M.; Sasaki, S.
Tetrahedron 1997, 53, 3035-3044. (b) Nagatsugi, F.; Kawasaki, T.; Maeda,
M.; Sasaki, S. Nucleic Acids Symp. Ser. 1996, 35, 79-80. (c) Nagatsugi, F.;
Uemura, K.; Nakashima, S.; Maeda, M.; Sasaki, S. Nucleic Acids Symp. Ser.
1995, 34, 171-172. (d) Nagatsugi, F.; Uemura, K.; Nakashima, S.; Maeda,
M.; Sasaki, S. Tetrahedron Lett. 1995, 36, 421-424.
(7) (a) Chang, E. H.; Miller, P. S.; Cushman, C.; Devadas, K.; Pirollo, K.
F.; Ts’o, P. O. P.; Yu, Z. P. Biochemistry 1991, 30, 8283-8286. (b) Kean, J.
M.; Murakami, A.; Blake, K., R.; Cushman, C. D.; Miller, P. S. Biochemistry
1988, 27, 9113-9120. (c) Lee, B. L.; Murakami A.; Blake, K. R.; Lin, S.-B.;
Miller, P. S. Biochemistry 1988, 27, 3197-3203.
10.1021/ja990356e CCC: $18.00 © 1999 American Chemical Society
Published on Web 07/03/1999

6754 J. Am. Chem. Soc., Vol. 121, No. 28, 1999
Communications to the Editor
Figure 2. Comparison of the cross-linking reactivity. The reaction was
done under the same condition as described in Figure 1, and stopped by
adding formamide at the indicated time. Yields were determined by
quantitating the bands on the gels by BAS2500 using the imaging plate.
The yield of each reaction is indicated with the following marks: 6: 4,
7: b, 8: O, 9: 0.
Figure 1. Comparison of the base selectivity. Lanes 1-9, the reaction
with 9; lanes 10-18, the reaction with 7. The reaction was done using 7
µM ODNs (7 or 9), 3 µM target oligomer including 10 labeled with 32
P
at 5′ end as a tracer in 0.1 M NaCl, 50 mM MOPS, pH 5.0, 33 °C. (*)
Control in the absence 9 or 7.
most effectively achieved within the ODN (7) incorporating the
sulfoxide-substituted nucleoside. The vinyl functional group of
9 might suffer an undesired nucleophilic reaction, thereby causing
lower yield. Although actual activation mechanisms of the
functional ODNs (7-9) for cross-linking reaction are uncertain,
the present results have clearly supported the validity of the new
strategy.
In a preparative reaction using equimolar amounts of 7 and 10
(N ) C), the two peaks were transformed to a single peak in an
almost quantitative yield after 24 h at room temperature. The
isolated peak was proved to be the cross-linked product by the
MALDI-TOF mass measurement (calculated, 9768.76; found,
9770.97).17 Further proof of cross-linking at cytidine was
evidenced by observing peaks corresponding to the authentic
adduct (2) in addition to A, G, T, and C in an expected ratio in
the HPLC analysis after the hydrolysis with snake venom
phophodiestrease and alkaline phosphatase (Escherichia coli C75).
The cross-linked adduct between 9 and 10 was also confirmed
by the same method.
In conclusion, we have successfully demonstrated the validity
of the new strategy by achieving a highly efficient and selective
cross-linking reaction toward cytidine without incurring chemical
instability by using the derivatives of a 2-amino-6-vinylpurine
nucleoside (1). To our knowledge, the sulfoxide-substituted
nucleoside 4 is the best alkylating agent toward cytidine so far
reported from the viewpoints of reactivity and selectivity as well
chemical stability. The new cross-linking motifs based on 1 will
be generally useful in the antisense strategy as well as for site-
directed chemical modification of a cytidine within a selected
target. Further work is now ongoing in the search for sulfide
structures which are more susceptible to the activation within the
duplex.
20% denaturing gel.16 All of the functionalized ODNs produced
a higher molecular weight band, indicating the formation of a
cross-linked adduct. Figure 1 illustrates the reactivity of the ODNs
(7 and 9) toward the different target site (10, N ) C, G, A, or T).
The highest yields to cytidine (lanes 4 and 8) were obtained with
the ODN 9 incorporating the 2-amino-6-vinylpurine nucleoside,
clearly demonstrating cytidine selectivity. The reactivity is
decreased in the order of C > G > A . T, which is in good
agreement with the result obtained in an organic solvent.14d It
should be noted that 7 incorporating the sulfoxide-substituted
nucleoside underwent the reaction more efficiently than 9 while
retaining similar cytidine selectivity (lanes 4, 8 vs 13, 17).
The reactivities of the four functionalized ODNs (6-9) toward
the ODN bearing cytidine at the target site (10, N ) C) are
compared in Figure 2. The yields of the adduct with 9 were almost
steady between ca. 40 and 50% after 5 h. In contrast, 7 produced
the adduct in over 80% yield after 12 h, demonstrating the
remarkably efficient cross-linking reaction. The ODN (8) incor-
porating the sulfone-substituted nucleoside gave the adduct in a
somewhat slower reaction rate than 9. It is interesting that the
yield with the ODN (6) incorporating the sulfide-substituted
nucleoside gradually increased and reached ca. 15% after 36 h.
Neither the monomers of the sulfide-substituted (3), the
sulfoxide-substituted (4), nor the sulfone-substituted nucleoside
(5) formed adducts with cytidine in the model experiment in
organic solvents. Furthermore, the functionalized ODNs (6-8)
were stable and were almost unchanged in the buffer. These facts
have indicated that the vinyl group might be generated within
the duplex in the proximity of the target cytidine and followed
by the cross-linking. It turned out that such activation has been
(15) Retention times of oligomers (HPLC conditions: nacalai tesque
COSMOSIL 5C18-MS; buffer A: 0.1 M TEAA, B: CH3CN. B: 10% to 30%/
20 min, linear gradient; flow rate, 1 mL/min) and MALDI-TOF MS (negative
Acknowledgment. We are grateful for support by Grant-in-Aid for
Scientific Research (C) to S.S. and for the Young Scientist Award to
F.N. from the Ministry of Education, Science, Sports and Culture, Japan.
mode) of the ODNs were as follows: 6 at 17.0 min, MS m/z 4875.15 (M-1
,
calcd 4873.84), 7 at 14.0 min, MS m/z 4890.95 (M-1, calcd 4889.83), 8 at
14.3 min, MS m/z 4904.17 (M-1, calcd 4905.83), 9 at 13.0 min, MS m/z
4761.10 (M-1, calcd 4763.82).
JA990356E
(16) Melting temperature of the duplex between 6 and 10 (N ) C) at 0.2
µM is 35 °C, indicating the majority of the ODNs in the duplex form under
the reaction conditions. The reactions proceeded under physiological conditions
at pH 6.8 at a somewhat slower rate than at pH 5.0.
(17) Retention times in the HPLC15 of 7, 10, and the adduct were as
follows: 7 at 14.0 min, 10 at 9 min, and the adduct at 13 min.